1
|
Karimi D, Calixto C, Snoussi H, Cortes-Albornoz MC, Velasco-Annis C, Rollins C, Jaimes C, Gholipour A, Warfield SK. Detailed delineation of the fetal brain in diffusion MRI via multi-task learning. BIORXIV : THE PREPRINT SERVER FOR BIOLOGY 2024:2024.08.29.609697. [PMID: 39257731 PMCID: PMC11383702 DOI: 10.1101/2024.08.29.609697] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 09/12/2024]
Abstract
Diffusion-weighted MRI is increasingly used to study the normal and abnormal development of fetal brain inutero. Recent studies have shown that dMRI can offer invaluable insights into the neurodevelopmental processes in the fetal stage. However, because of the low data quality and rapid brain development, reliable analysis of fetal dMRI data requires dedicated computational methods that are currently unavailable. The lack of automated methods for fast, accurate, and reproducible data analysis has seriously limited our ability to tap the potential of fetal brain dMRI for medical and scientific applications. In this work, we developed and validated a unified computational framework to (1) segment the brain tissue into white matter, cortical/subcortical gray matter, and cerebrospinal fluid, (2) segment 31 distinct white matter tracts, and (3) parcellate the brain's cortex and delineate the deep gray nuclei and white matter structures into 96 anatomically meaningful regions. We utilized a set of manual, semi-automatic, and automatic approaches to annotate 97 fetal brains. Using these labels, we developed and validated a multi-task deep learning method to perform the three computations. Our evaluations show that the new method can accurately carry out all three tasks, achieving a mean Dice similarity coefficient of 0.865 on tissue segmentation, 0.825 on white matter tract segmentation, and 0.819 on parcellation. The proposed method can greatly advance the field of fetal neuroimaging as it can lead to substantial improvements in fetal brain tractography, tract-specific analysis, and structural connectivity assessment.
Collapse
Affiliation(s)
- Davood Karimi
- Boston Children's Hospital and Harvard Medical School, Boston, MA
| | - Camilo Calixto
- Boston Children's Hospital and Harvard Medical School, Boston, MA
- Elmhurst Hospital Center and Icahn School of Medicine at Mount Sinai, New York, NY
| | - Haykel Snoussi
- Boston Children's Hospital and Harvard Medical School, Boston, MA
| | | | | | - Caitlin Rollins
- Boston Children's Hospital and Harvard Medical School, Boston, MA
| | - Camilo Jaimes
- Massachusetts General Hospital and Harvard Medical School, Boston, MA
| | - Ali Gholipour
- Boston Children's Hospital and Harvard Medical School, Boston, MA
- Department of Radiological Sciences, University of California Irvine, Irvine, CA
| | - Simon K Warfield
- Boston Children's Hospital and Harvard Medical School, Boston, MA
| |
Collapse
|
2
|
Noorizadeh N, Kazemi K, Taji SM, Danyali H, Aarabi A. Subject-specific atlas for automatic brain tissue segmentation of neonatal magnetic resonance images. Sci Rep 2024; 14:19114. [PMID: 39155321 PMCID: PMC11330982 DOI: 10.1038/s41598-024-69995-z] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/29/2023] [Accepted: 08/12/2024] [Indexed: 08/20/2024] Open
Abstract
Developing advanced systems for 3D brain tissue segmentation from neonatal magnetic resonance (MR) images is vital for newborn structural analysis. However, automatic segmentation of neonatal brain tissues is challenging due to smaller head size and inverted T1/T2 tissue contrast compared to adults. In this work, a subject-specific atlas based technique is presented for segmentation of gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF) from neonatal MR images. It involves atlas selection, subject-specific atlas creation using random forest (RF) classifier, and brain tissue segmentation using the expectation maximization-Markov random field (EM-MRF) method. To increase the segmentation accuracy, different tissue intensity- and gradient-based features were used. Evaluation on 40 neonatal MR images (gestational age of 37-44 weeks) demonstrated an overall accuracy of 94.3% and an average Dice similarity coefficient (DSC) of 0.945 (GM), 0.947 (WM), and 0.912 (CSF). Compared to multi-atlas segmentation methods like SEGMA and EM-MRF with multiple atlases, our method improved accuracy by up to 4%, particularly in complex tissue regions. Our proposed method allows accurate brain tissue segmentation, a crucial step in brain magnetic resonance imaging (MRI) applications including brain surface reconstruction and realistic head model creation in neonates.
Collapse
Affiliation(s)
- Negar Noorizadeh
- Department of Pediatrics, University of Tennessee Health Science Center, Memphis, TN, USA
| | - Kamran Kazemi
- Department of Electrical Engineering, Shiraz University of Technology, Shiraz, Iran.
| | | | - Habibollah Danyali
- Department of Electrical Engineering, Shiraz University of Technology, Shiraz, Iran
| | - Ardalan Aarabi
- Laboratory of Functional Neuroscience and Pathologies (UR UPJV 4559), University Research Center (CURS), University of Picardie Jules Verne, Amiens, France
| |
Collapse
|
3
|
Calixto C, Jaimes C, Soldatelli MD, Warfield SK, Gholipour A, Karimi D. Anatomically constrained tractography of the fetal brain. Neuroimage 2024; 297:120723. [PMID: 39029605 PMCID: PMC11382095 DOI: 10.1016/j.neuroimage.2024.120723] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/22/2024] [Accepted: 07/03/2024] [Indexed: 07/21/2024] Open
Abstract
Diffusion-weighted Magnetic Resonance Imaging (dMRI) is increasingly used to study the fetal brain in utero. An important computation enabled by dMRI is streamline tractography, which has unique applications such as tract-specific analysis of the brain white matter and structural connectivity assessment. However, due to the low fetal dMRI data quality and the challenging nature of tractography, existing methods tend to produce highly inaccurate results. They generate many false streamlines while failing to reconstruct the streamlines that constitute the major white matter tracts. In this paper, we advocate for anatomically constrained tractography based on an accurate segmentation of the fetal brain tissue directly in the dMRI space. We develop a deep learning method to compute the segmentation automatically. Experiments on independent test data show that this method can accurately segment the fetal brain tissue and drastically improve the tractography results. It enables the reconstruction of highly curved tracts such as optic radiations. Importantly, our method infers the tissue segmentation and streamline propagation direction from a diffusion tensor fit to the dMRI data, making it applicable to routine fetal dMRI scans. The proposed method can facilitate the study of fetal brain white matter tracts with dMRI.
Collapse
Affiliation(s)
- Camilo Calixto
- Boston Children's Hospital, 300 Longwood Ave, Boston, MA 02115, USA
| | - Camilo Jaimes
- Massachusetts General Hospital, 55 Fruit St, Boston, MA 02114, USA
| | | | - Simon K Warfield
- Boston Children's Hospital, 300 Longwood Ave, Boston, MA 02115, USA
| | - Ali Gholipour
- Boston Children's Hospital, 300 Longwood Ave, Boston, MA 02115, USA
| | - Davood Karimi
- Boston Children's Hospital, 300 Longwood Ave, Boston, MA 02115, USA.
| |
Collapse
|
4
|
Calixto C, Taymourtash A, Karimi D, Snoussi H, Velasco-Annis C, Jaimes C, Gholipour A. Advances in Fetal Brain Imaging. Magn Reson Imaging Clin N Am 2024; 32:459-478. [PMID: 38944434 PMCID: PMC11216711 DOI: 10.1016/j.mric.2024.03.004] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 07/01/2024]
Abstract
Over the last 20 years, there have been remarkable developments in fetal brain MR imaging analysis methods. This article delves into the specifics of structural imaging, diffusion imaging, functional MR imaging, and spectroscopy, highlighting the latest advancements in motion correction, fetal brain development atlases, and the challenges and innovations. Furthermore, this article explores the clinical applications of these advanced imaging techniques in comprehending and diagnosing fetal brain development and abnormalities.
Collapse
Affiliation(s)
- Camilo Calixto
- Computational Radiology Laboratory, Department of Radiology, Boston Children's Hospital, 401 Park Dr, 7th Floor West, Boston, MA 02215, USA; Harvard Medical School, 25 Shattuck Street, Boston, MA 02115, USA.
| | - Athena Taymourtash
- Department of Biomedical Imaging and Image-guided Therapy, Medical University of Vienna, Spitalgasse 23, Wien 1090, Austria
| | - Davood Karimi
- Computational Radiology Laboratory, Department of Radiology, Boston Children's Hospital, 401 Park Dr, 7th Floor West, Boston, MA 02215, USA; Harvard Medical School, 25 Shattuck Street, Boston, MA 02115, USA
| | - Haykel Snoussi
- Computational Radiology Laboratory, Department of Radiology, Boston Children's Hospital, 401 Park Dr, 7th Floor West, Boston, MA 02215, USA; Harvard Medical School, 25 Shattuck Street, Boston, MA 02115, USA
| | - Clemente Velasco-Annis
- Computational Radiology Laboratory, Department of Radiology, Boston Children's Hospital, 401 Park Dr, 7th Floor West, Boston, MA 02215, USA; Harvard Medical School, 25 Shattuck Street, Boston, MA 02115, USA
| | - Camilo Jaimes
- Harvard Medical School, 25 Shattuck Street, Boston, MA 02115, USA; Department of Radiology, Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02215, USA
| | - Ali Gholipour
- Computational Radiology Laboratory, Department of Radiology, Boston Children's Hospital, 401 Park Dr, 7th Floor West, Boston, MA 02215, USA; Harvard Medical School, 25 Shattuck Street, Boston, MA 02115, USA
| |
Collapse
|
5
|
Turesky TK, Escalante E, Loh M, Gaab N. Longitudinal trajectories of brain development from infancy to school age and their relationship to literacy development. BIORXIV : THE PREPRINT SERVER FOR BIOLOGY 2024:2024.06.29.601366. [PMID: 39005343 PMCID: PMC11244924 DOI: 10.1101/2024.06.29.601366] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 07/16/2024]
Abstract
Reading is one of the most complex skills that we utilize daily, and it involves the early development and interaction of various lower-level subskills, including phonological processing and oral language. These subskills recruit brain structures, which begin to develop long before the skill manifests and exhibit rapid development during infancy. However, how longitudinal trajectories of early brain development in these structures supports long-term acquisition of literacy subskills and subsequent reading is unclear. Children underwent structural and diffusion MRI scanning at multiple timepoints between infancy and second grade and were tested for literacy subskills in preschool and decoding and word reading in early elementary school. We developed and implemented a reproducible pipeline to generate longitudinal trajectories of early brain development to examine associations between these trajectories and literacy (sub)skills. Furthermore, we examined whether familial risk of reading difficulty and a child's home literacy environment, two common literacy-related covariates, influenced those trajectories. Results showed that individual differences in curve features (e.g., intercepts and slopes) for longitudinal trajectories of volumetric, surface-based, and white matter organization measures in left-hemispheric reading-related regions and tracts were linked directly to phonological processing and indirectly to second-grade decoding and word reading skills via phonological processing. Altogether, these findings suggest that the brain bases of phonological processing, previously identified as the strongest behavioral predictor of reading and decoding skills, may already begin to develop early in infancy but undergo further refinement between birth and preschool. The present study underscores the importance of considering academic skill acquisition from the very beginning of life.
Collapse
|
6
|
Lautarescu A, Bonthrone AF, Bos B, Barratt B, Counsell SJ. Advances in fetal and neonatal neuroimaging and everyday exposures. Pediatr Res 2024:10.1038/s41390-024-03294-1. [PMID: 38877283 DOI: 10.1038/s41390-024-03294-1] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 02/13/2024] [Revised: 04/25/2024] [Accepted: 04/29/2024] [Indexed: 06/16/2024]
Abstract
The complex, tightly regulated process of prenatal brain development may be adversely affected by "everyday exposures" such as stress and environmental pollutants. Researchers are only just beginning to understand the neural sequelae of such exposures, with advances in fetal and neonatal neuroimaging elucidating structural, microstructural, and functional correlates in the developing brain. This narrative review discusses the wide-ranging literature investigating the influence of parental stress on fetal and neonatal brain development as well as emerging literature assessing the impact of exposure to environmental toxicants such as lead and air pollution. These 'everyday exposures' can co-occur with other stressors such as social and financial deprivation, and therefore we include a brief discussion of neuroimaging studies assessing the effect of social disadvantage. Increased exposure to prenatal stressors is associated with alterations in the brain structure, microstructure and function, with some evidence these associations are moderated by factors such as infant sex. However, most studies examine only single exposures and the literature on the relationship between in utero exposure to pollutants and fetal or neonatal brain development is sparse. Large cohort studies are required that include evaluation of multiple co-occurring exposures in order to fully characterize their impact on early brain development. IMPACT: Increased prenatal exposure to parental stress and is associated with altered functional, macro and microstructural fetal and neonatal brain development. Exposure to air pollution and lead may also alter brain development in the fetal and neonatal period. Further research is needed to investigate the effect of multiple co-occurring exposures, including stress, environmental toxicants, and socioeconomic deprivation on early brain development.
Collapse
Affiliation(s)
- Alexandra Lautarescu
- Department of Perinatal Imaging and Health, Centre for the Developing Brain, School of Biomedical Engineering and Imaging Sciences, King's College London, London, UK
- Department of Psychology, Institute of Psychiatry, Psychology and Neuroscience, King's College London, London, UK
| | - Alexandra F Bonthrone
- Department of Perinatal Imaging and Health, Centre for the Developing Brain, School of Biomedical Engineering and Imaging Sciences, King's College London, London, UK
| | - Brendan Bos
- MRC Centre for Environment and Health, Imperial College London, London, UK
| | - Ben Barratt
- MRC Centre for Environment and Health, Imperial College London, London, UK
| | - Serena J Counsell
- Department of Perinatal Imaging and Health, Centre for the Developing Brain, School of Biomedical Engineering and Imaging Sciences, King's College London, London, UK.
| |
Collapse
|
7
|
Fidon L, Aertsen M, Kofler F, Bink A, David AL, Deprest T, Emam D, Guffens F, Jakab A, Kasprian G, Kienast P, Melbourne A, Menze B, Mufti N, Pogledic I, Prayer D, Stuempflen M, Van Elslander E, Ourselin S, Deprest J, Vercauteren T. A Dempster-Shafer Approach to Trustworthy AI With Application to Fetal Brain MRI Segmentation. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 2024; 46:3784-3795. [PMID: 38198270 DOI: 10.1109/tpami.2023.3346330] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 01/12/2024]
Abstract
Deep learning models for medical image segmentation can fail unexpectedly and spectacularly for pathological cases and images acquired at different centers than training images, with labeling errors that violate expert knowledge. Such errors undermine the trustworthiness of deep learning models for medical image segmentation. Mechanisms for detecting and correcting such failures are essential for safely translating this technology into clinics and are likely to be a requirement of future regulations on artificial intelligence (AI). In this work, we propose a trustworthy AI theoretical framework and a practical system that can augment any backbone AI system using a fallback method and a fail-safe mechanism based on Dempster-Shafer theory. Our approach relies on an actionable definition of trustworthy AI. Our method automatically discards the voxel-level labeling predicted by the backbone AI that violate expert knowledge and relies on a fallback for those voxels. We demonstrate the effectiveness of the proposed trustworthy AI approach on the largest reported annotated dataset of fetal MRI consisting of 540 manually annotated fetal brain 3D T2w MRIs from 13 centers. Our trustworthy AI method improves the robustness of four backbone AI models for fetal brain MRIs acquired across various centers and for fetuses with various brain abnormalities.
Collapse
|
8
|
Ciceri T, Casartelli L, Montano F, Conte S, Squarcina L, Bertoldo A, Agarwal N, Brambilla P, Peruzzo D. Fetal brain MRI atlases and datasets: A review. Neuroimage 2024; 292:120603. [PMID: 38588833 DOI: 10.1016/j.neuroimage.2024.120603] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/03/2023] [Revised: 03/28/2024] [Accepted: 04/05/2024] [Indexed: 04/10/2024] Open
Abstract
Fetal brain development is a complex process involving different stages of growth and organization which are crucial for the development of brain circuits and neural connections. Fetal atlases and labeled datasets are promising tools to investigate prenatal brain development. They support the identification of atypical brain patterns, providing insights into potential early signs of clinical conditions. In a nutshell, prenatal brain imaging and post-processing via modern tools are a cutting-edge field that will significantly contribute to the advancement of our understanding of fetal development. In this work, we first provide terminological clarification for specific terms (i.e., "brain template" and "brain atlas"), highlighting potentially misleading interpretations related to inconsistent use of terms in the literature. We discuss the major structures and neurodevelopmental milestones characterizing fetal brain ontogenesis. Our main contribution is the systematic review of 18 prenatal brain atlases and 3 datasets. We also tangentially focus on clinical, research, and ethical implications of prenatal neuroimaging.
Collapse
Affiliation(s)
- Tommaso Ciceri
- NeuroImaging Lab, Scientific Institute IRCCS Eugenio Medea, Bosisio Parini, Italy; Department of Information Engineering, University of Padua, Padua, Italy
| | - Luca Casartelli
- Theoretical and Cognitive Neuroscience Unit, Scientific Institute IRCCS Eugenio Medea, Bosisio Parini, Italy
| | - Florian Montano
- Diagnostic Imaging and Neuroradiology Unit, Scientific Institute IRCCS Eugenio Medea, Bosisio Parini, Italy
| | - Stefania Conte
- Psychology Department, State University of New York at Binghamton, New York, USA
| | - Letizia Squarcina
- Department of Pathophysiology and Transplantation, University of Milan, Milan, Italy
| | - Alessandra Bertoldo
- Department of Information Engineering, University of Padua, Padua, Italy; Padova Neuroscience Center, University of Padua, Padua, Italy
| | - Nivedita Agarwal
- Diagnostic Imaging and Neuroradiology Unit, Scientific Institute IRCCS Eugenio Medea, Bosisio Parini, Italy
| | - Paolo Brambilla
- Department of Pathophysiology and Transplantation, University of Milan, Milan, Italy; Department of Neurosciences and Mental Health, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, Milan, Italy.
| | - Denis Peruzzo
- NeuroImaging Lab, Scientific Institute IRCCS Eugenio Medea, Bosisio Parini, Italy
| |
Collapse
|
9
|
Valabregue R, Girka F, Pron A, Rousseau F, Auzias G. Comprehensive analysis of synthetic learning applied to neonatal brain MRI segmentation. Hum Brain Mapp 2024; 45:e26674. [PMID: 38651625 PMCID: PMC11036377 DOI: 10.1002/hbm.26674] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/02/2023] [Revised: 03/09/2024] [Accepted: 03/18/2024] [Indexed: 04/25/2024] Open
Abstract
Brain segmentation from neonatal MRI images is a very challenging task due to large changes in the shape of cerebral structures and variations in signal intensities reflecting the gestational process. In this context, there is a clear need for segmentation techniques that are robust to variations in image contrast and to the spatial configuration of anatomical structures. In this work, we evaluate the potential of synthetic learning, a contrast-independent model trained using synthetic images generated from the ground truth labels of very few subjects. We base our experiments on the dataset released by the developmental Human Connectome Project, for which high-quality images are available for more than 700 babies aged between 26 and 45 weeks postconception. First, we confirm the impressive performance of a standard UNet trained on a few volumes, but also confirm that such models learn intensity-related features specific to the training domain. We then confirm the robustness of the synthetic learning approach to variations in image contrast. However, we observe a clear influence of the age of the baby on the predictions. We improve the performance of this model by enriching the synthetic training set with realistic motion artifacts and over-segmentation of the white matter. Based on extensive visual assessment, we argue that the better performance of the model trained on real T2w data may be due to systematic errors in the ground truth. We propose an original experiment allowing us to show that learning from real data will reproduce any systematic bias affecting the training set, while synthetic models can avoid this limitation. Overall, our experiments confirm that synthetic learning is an effective solution for segmenting neonatal brain MRI. Our adapted synthetic learning approach combines key features that will be instrumental for large multisite studies and clinical applications.
Collapse
Affiliation(s)
- R. Valabregue
- CENIR, Institut du Cerveau (ICM)—Paris Brain Institute, Inserm U 1127, CNRS UMR 7225, Sorbonne UniversitéParisFrance
| | - F. Girka
- CENIR, Institut du Cerveau (ICM)—Paris Brain Institute, Inserm U 1127, CNRS UMR 7225, Sorbonne UniversitéParisFrance
| | - A. Pron
- Aix‐Marseille Université, CNRS, Institut de Neurosciences de la Timone, UMR 7289MarseilleFrance
| | - F. Rousseau
- IMT Atlantique, LaTIM INSERM U1101BrestFrance
| | - G. Auzias
- Aix‐Marseille Université, CNRS, Institut de Neurosciences de la Timone, UMR 7289MarseilleFrance
| |
Collapse
|
10
|
Chen J, Lu R, Jing B, Zhang H, Chen G, Shen D. One model, two brains: Automatic fetal brain extraction from MR images of twins. Comput Med Imaging Graph 2024; 112:102330. [PMID: 38262133 DOI: 10.1016/j.compmedimag.2024.102330] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/19/2023] [Revised: 11/27/2023] [Accepted: 12/13/2023] [Indexed: 01/25/2024]
Abstract
Fetal brain extraction from magnetic resonance (MR) images is of great importance for both clinical applications and neuroscience studies. However, it is a challenging task, especially when dealing with twins, which are commonly existing in pregnancy. Currently, there is no brain extraction method dedicated to twins, raising significant demand to develop an effective twin fetal brain extraction method. To this end, we propose the first twin fetal brain extraction framework, which possesses three novel features. First, to narrow down the region of interest and preserve structural information between the two brains in twin fetal MR images, we take advantage of an advanced object detector to locate all the brains in twin fetal MR images at once. Second, we propose a Twin Fetal Brain Extraction Network (TFBE-Net) to further suppress insignificant features for segmenting brain regions. Finally, we propose a Two-step Training Strategy (TTS) to learn correlation features of the single fetal brain for further improving the performance of TFBE-Net. We validate the proposed framework on a twin fetal brain dataset. The experiments show that our framework achieves promising performance on both quantitative and qualitative evaluations, and outperforms state-of-the-art methods for fetal brain extraction.
Collapse
Affiliation(s)
- Jian Chen
- School of Electronic, Electrical Engineering and Physics, Fujian University of Technology, Fuzhou, 350118, Fujian, China; Beijing Key Laboratory of Fundamental Research on Biomechanics in Clinical Application, Beijing, 100069, China
| | - Ranlin Lu
- School of Electronic, Electrical Engineering and Physics, Fujian University of Technology, Fuzhou, 350118, Fujian, China
| | - Bin Jing
- Beijing Key Laboratory of Fundamental Research on Biomechanics in Clinical Application, Beijing, 100069, China; School of Biomedical Engineering, Capital Medical University, Beijing, 100069, China
| | - He Zhang
- Department of Radiology, Obstetrics and Gynecology Hospital, Fudan University, Shanghai, 200011, China
| | - Geng Chen
- National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology, School of Computer Science and Engineering, Northwestern Polytechnical University, Xi'an 710072, China.
| | - Dinggang Shen
- School of Biomedical Engineering & State Key Laboratory of Advanced Medical Materials and Devices, ShanghaiTech University, Shanghai, 201210, China; Shanghai Clinical Research and Trial Center, Shanghai, 201210, China; Shanghai United Imaging Intelligence Co., Ltd., Shanghai, 200230, China.
| |
Collapse
|
11
|
Chen JV, Li Y, Tang F, Chaudhari G, Lew C, Lee A, Rauschecker AM, Haskell-Mendoza AP, Wu YW, Calabrese E. Automated neonatal nnU-Net brain MRI extractor trained on a large multi-institutional dataset. Sci Rep 2024; 14:4583. [PMID: 38403673 PMCID: PMC10894871 DOI: 10.1038/s41598-024-54436-8] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/29/2023] [Accepted: 02/13/2024] [Indexed: 02/27/2024] Open
Abstract
Brain extraction, or skull-stripping, is an essential data preprocessing step for machine learning approaches to brain MRI analysis. Currently, there are limited extraction algorithms for the neonatal brain. We aim to adapt an established deep learning algorithm for the automatic segmentation of neonatal brains from MRI, trained on a large multi-institutional dataset for improved generalizability across image acquisition parameters. Our model, ANUBEX (automated neonatal nnU-Net brain MRI extractor), was designed using nnU-Net and was trained on a subset of participants (N = 433) enrolled in the High-dose Erythropoietin for Asphyxia and Encephalopathy (HEAL) study. We compared the performance of our model to five publicly available models (BET, BSE, CABINET, iBEATv2, ROBEX) across conventional and machine learning methods, tested on two public datasets (NIH and dHCP). We found that our model had a significantly higher Dice score on the aggregate of both data sets and comparable or significantly higher Dice scores on the NIH (low-resolution) and dHCP (high-resolution) datasets independently. ANUBEX performs similarly when trained on sequence-agnostic or motion-degraded MRI, but slightly worse on preterm brains. In conclusion, we created an automatic deep learning-based neonatal brain extraction algorithm that demonstrates accurate performance with both high- and low-resolution MRIs with fast computation time.
Collapse
Affiliation(s)
- Joshua V Chen
- Department of Radiology, University of California San Francisco, San Francisco, CA, USA
| | - Yi Li
- Department of Radiology, University of California San Francisco, San Francisco, CA, USA
| | - Felicia Tang
- Department of Radiology, University of California San Francisco, San Francisco, CA, USA
| | - Gunvant Chaudhari
- Department of Radiology, University of California San Francisco, San Francisco, CA, USA
| | - Christopher Lew
- Division of Neuroradiology, Department of Radiology, Duke University Medical Center, Durham, NC, 27710, USA
| | - Amanda Lee
- Division of Neuroradiology, Department of Radiology, Duke University Medical Center, Durham, NC, 27710, USA
| | - Andreas M Rauschecker
- Department of Radiology, University of California San Francisco, San Francisco, CA, USA
| | | | - Yvonne W Wu
- University of California San Francisco Weill Institute for Neurosciences, San Francisco, CA, USA
| | - Evan Calabrese
- Division of Neuroradiology, Department of Radiology, Duke University Medical Center, Durham, NC, 27710, USA.
- Duke Center for Artificial Intelligence in Radiology (DAIR), Durham, NC, USA.
| |
Collapse
|
12
|
Chen J, Lu R, Ye S, Guang M, Tassew TM, Jing B, Zhang G, Chen G, Shen D. Image Recovery Matters: A Recovery-Extraction Framework for Robust Fetal Brain Extraction From MR Images. IEEE J Biomed Health Inform 2024; 28:823-834. [PMID: 37995170 DOI: 10.1109/jbhi.2023.3333953] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/25/2023]
Abstract
The extraction of the fetal brain from magnetic resonance (MR) images is a challenging task. In particular, fetal MR images suffer from different kinds of artifacts introduced during the image acquisition. Among those artifacts, intensity inhomogeneity is a common one affecting brain extraction. In this work, we propose a deep learning-based recovery-extraction framework for fetal brain extraction, which is particularly effective in handling fetal MR images with intensity inhomogeneity. Our framework involves two stages. First, the artifact-corrupted images are recovered with the proposed generative adversarial learning-based image recovery network with a novel region-of-darkness discriminator that enforces the network focusing on artifacts of the images. Second, we propose a brain extraction network for more effective fetal brain segmentation by strengthening the association between lower- and higher-level features as well as suppressing task-irrelevant features. Thanks to the proposed recovery-extraction strategy, our framework is able to accurately segment fetal brains from artifact-corrupted MR images. The experiments show that our framework achieves promising performance in both quantitative and qualitative evaluations, and outperforms state-of-the-art methods in both image recovery and fetal brain extraction.
Collapse
|
13
|
Peng Y, Xu Y, Wang M, Zhang H, Xie J. The nnU-Net based method for automatic segmenting fetal brain tissues. Health Inf Sci Syst 2023; 11:17. [PMID: 36998806 PMCID: PMC10043149 DOI: 10.1007/s13755-023-00220-3] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/24/2022] [Accepted: 03/09/2023] [Indexed: 04/01/2023] Open
Abstract
The magnetic resonance (MR) images of fetuses make it possible for doctors to detect out pathological fetal brains in early stages. Brain tissue segmentation is prerequisite for making brain morphology and volume analyses. nnU-Net is an automatic segmentation method based on deep learning. It can adaptively configure itself, so as to adapt to a specific task via preprocessing, network architecture, training, and post-processing. Therefore, we adapt nnU-Net to segment seven types of fetal brain tissues, including external cerebrospinal fluid, gray matter, white matter, ventricle, cerebellum, deep gray matter, and brainstem. With regard to the characteristics of the FeTA 2021 data, some adjustments are made to the original nnU-Net, so that it can segment seven types of fetal brain tissues precisely as far as possible. The average segmentation results on FeTA 2021 training data demonstrate that our advanced nnU-Net is superior to the peers including SegNet, CoTr, AC U-Net and ResUnet. Its average segmentation results are 0.842, 11.759 and 0.957 in terms of Dice, HD95 and VS criteria. Moreover, the experimental results on FeTA 2021 test data further demonstrate that our advanced nnU-Net has obtained good segmentation performance of 0.774, 14.699 and 0.875 in terms of Dice, HD95 and VS, ranked the third in FeTA 2021 challenge. Our advanced nnU-Net realized the task for segmenting the fetal brain tissues using MR images of different gestational ages, which can help doctors to make correct and seasonable diagnoses.
Collapse
Affiliation(s)
- Ying Peng
- School of Computer Science, Shaanxi Normal University, West Chang’an Avenue, Chang’an District, Xi’an, 710119 Shaanxi People’s Republic of China
| | - Yandi Xu
- School of Medicine, Tulane University, 1430 Tulane Ave, New Orleans, LA 70112 USA
| | - Mingzhao Wang
- School of Computer Science, Shaanxi Normal University, West Chang’an Avenue, Chang’an District, Xi’an, 710119 Shaanxi People’s Republic of China
| | - Huiquan Zhang
- School of Computer Science, Shaanxi Normal University, West Chang’an Avenue, Chang’an District, Xi’an, 710119 Shaanxi People’s Republic of China
| | - Juanying Xie
- School of Computer Science, Shaanxi Normal University, West Chang’an Avenue, Chang’an District, Xi’an, 710119 Shaanxi People’s Republic of China
| |
Collapse
|
14
|
Russo C, Pirozzi MA, Mazio F, Cascone D, Cicala D, De Liso M, Nastro A, Covelli EM, Cinalli G, Quarantelli M. Fully automated measurement of intracranial CSF and brain parenchyma volumes in pediatric hydrocephalus by segmentation of clinical MRI studies. Med Phys 2023; 50:7921-7933. [PMID: 37166045 DOI: 10.1002/mp.16445] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/08/2022] [Revised: 03/29/2023] [Accepted: 04/18/2023] [Indexed: 05/12/2023] Open
Abstract
BACKGROUND Brain parenchyma (BP) and intracranial cerebrospinal fluid (iCSF) volumes measured by fully automated segmentation of clinical brain MRI studies may be useful for the diagnosis and follow-up of pediatric hydrocephalus. However, previously published segmentation techniques either rely on dedicated sequences, not routinely used in clinical practice, or on spatial normalization, which has limited accuracy when severe brain distortions, such as in hydrocephalic patients, are present. PURPOSE We developed a fully automated method to measure BP and iCSF volumes from clinical brain MRI studies of pediatric hydrocephalus patients, exploiting the complementary information contained in T2- and T1-weighted images commonly used in clinical practice. METHODS The proposed procedure, following skull-stripping of the combined volumes, performed using a multiparametric method to obtain a reliable definition of the inner skull profile, maximizes the CSF-to-parenchyma contrast by dividing the T2w- by the T1w- volume after full-scale dynamic rescaling, thus allowing separation of iCSF and BP through a simple thresholding routine. RESULTS Validation against manual tracing on 23 studies (four controls and 19 hydrocephalic patients) showed excellent concordance (ICC > 0.98) and spatial overlap (Dice coefficients ranging from 77.2% for iCSF to 96.8% for intracranial volume). Accuracy was comparable to the intra-operator reproducibility of manual segmentation, as measured in 14 studies processed twice by the same experienced neuroradiologist. Results of the application of the algorithm to a dataset of 63 controls and 57 hydrocephalic patients (19 with parenchymal damage), measuring volumes' changes with normal development and in hydrocephalic patients, are also reported for demonstration purposes. CONCLUSIONS The proposed approach allows fully automated segmentation of BP and iCSF in clinical studies, also in severely distorted brains, enabling to assess age- and disease-related changes in intracranial tissue volume with an accuracy comparable to expert manual segmentation.
Collapse
Affiliation(s)
- Carmela Russo
- Neuroradiology Unit, Department of Neuroscience, Santobono-Pausilipon Children's Hospital, Naples, Italy
| | - Maria Agnese Pirozzi
- Institute of Biostructures and Bioimaging, National Research Council, Naples, Italy
- Department of Advanced Medical and Surgical Sciences, University of Campania "Luigi Vanvitelli", Naples, Italy
| | - Federica Mazio
- Neuroradiology Unit, Department of Neuroscience, Santobono-Pausilipon Children's Hospital, Naples, Italy
| | - Daniele Cascone
- Neuroradiology Unit, Department of Neuroscience, Santobono-Pausilipon Children's Hospital, Naples, Italy
| | - Domenico Cicala
- Neuroradiology Unit, Department of Neuroscience, Santobono-Pausilipon Children's Hospital, Naples, Italy
| | - Maria De Liso
- Neuroradiology Unit, Department of Neuroscience, Santobono-Pausilipon Children's Hospital, Naples, Italy
| | - Anna Nastro
- Neuroradiology Unit, Department of Neuroscience, Santobono-Pausilipon Children's Hospital, Naples, Italy
| | - Eugenio Maria Covelli
- Neuroradiology Unit, Department of Neuroscience, Santobono-Pausilipon Children's Hospital, Naples, Italy
| | - Giuseppe Cinalli
- Pediatric Neurosurgery Unit, Department of Neuroscience, Santobono-Pausilipon Children's Hospital, Naples, Italy
| | - Mario Quarantelli
- Institute of Biostructures and Bioimaging, National Research Council, Naples, Italy
| |
Collapse
|
15
|
Shrot S, Hadi E, Barash Y, Hoffmann C. Effect of magnet strength on fetal brain biometry - a single-center retrospective MRI-based cohort study. Neuroradiology 2023; 65:1517-1525. [PMID: 37436475 DOI: 10.1007/s00234-023-03193-y] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/03/2023] [Accepted: 07/05/2023] [Indexed: 07/13/2023]
Abstract
PURPOSE Abnormal fetal brain measurements might affect clinical management and parental counseling. The effect of between-field-strength differences was not evaluated in quantitative fetal brain imaging until now. Our study aimed to compare fetal brain biometry measurements in 3.0 T with 1.5 T scanners. METHODS A retrospective cohort of 1150 low-risk fetuses scanned between 2012 and 2021, with apparently normal brain anatomy, were retrospectively evaluated for biometric measurements. The cohort included 1.5 T (442 fetuses) and 3.0 T scans (708 fetuses) of populations with comparable characteristics in the same tertiary medical center. Manually measured biometry included bi-parietal, fronto-occipital and trans-cerebellar diameters, length of the corpus-callosum, vermis height, and width. Measurements were then converted to centiles based on previously reported biometric reference charts. The 1.5 T centiles were compared with the 3.0 T centiles. RESULTS No significant differences between centiles of bi-parietal diameter, trans-cerebellar diameter, or length of the corpus callosum between 1.5 T and 3.0 T scanners were found. Small absolute differences were found in the vermis height, with higher centiles in the 3.0 T, compared to the 1.5 T scanner (54.6th-centile, vs. 39.0th-centile, p < 0.001); less significant differences were found in vermis width centiles (46.9th-centile vs. 37.5th-centile, p = 0.03). Fronto-occipital diameter was higher in 1.5 T than in the 3.0 T scanner (66.0th-centile vs. 61.8th-centile, p = 0.02). CONCLUSIONS The increasing use of 3.0 T MRI for fetal imaging poses a potential bias when using 1.5 T-based charts. We elucidate those biometric measurements are comparable, with relatively small between-field-strength differences, when using manual biometric measurements. Small inter-magnet differences can be related to higher spatial resolution with 3 T scanners and may be substantial when evaluating small brain structures, such as the vermis.
Collapse
Affiliation(s)
- Shai Shrot
- Section of Neuroradiology, Division of Diagnostic Imaging, Sheba Medical Center, Tel Hashomer, 2 Sheba Rd, 52621, Ramat Gan, Israel.
- Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel.
| | - Efrat Hadi
- Diagnostic Ultrasound Unit of the Institute of Obstetrical and Gynecological Imaging, Department of Obstetrics and Gynecology, Sheba Medical Center, 52621, Ramat Gan, Israel
| | - Yiftach Barash
- Section of Neuroradiology, Division of Diagnostic Imaging, Sheba Medical Center, Tel Hashomer, 2 Sheba Rd, 52621, Ramat Gan, Israel
- Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel
| | - Chen Hoffmann
- Section of Neuroradiology, Division of Diagnostic Imaging, Sheba Medical Center, Tel Hashomer, 2 Sheba Rd, 52621, Ramat Gan, Israel
- Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel
| |
Collapse
|
16
|
Vahedifard F, Liu X, Adepoju JO, Zhao S, Ai HA, Marathu KK, Supanich M, Byrd SE, Deng J. Automatic Localization of the Pons and Vermis on Fetal Brain MR Imaging Using a U-Net Deep Learning Model. AJNR Am J Neuroradiol 2023; 44:1191-1200. [PMID: 37652583 PMCID: PMC10549940 DOI: 10.3174/ajnr.a7978] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/06/2023] [Accepted: 08/02/2023] [Indexed: 09/02/2023]
Abstract
BACKGROUND AND PURPOSE An MRI of the fetus can enhance the identification of perinatal developmental disorders, which improves the accuracy of ultrasound. Manual MRI measurements require training, time, and intra-variability concerns. Pediatric neuroradiologists are also in short supply. Our purpose was developing a deep learning model and pipeline for automatically identifying anatomic landmarks on the pons and vermis in fetal brain MR imaging and suggesting suitable images for measuring the pons and vermis. MATERIALS AND METHODS We retrospectively used 55 pregnant patients who underwent fetal brain MR imaging with a HASTE protocol. Pediatric neuroradiologists selected them for landmark annotation on sagittal single-shot T2-weighted images, and the clinically reliable method was used as the criterion standard for the measurement of the pons and vermis. A U-Net-based deep learning model was developed to automatically identify fetal brain anatomic landmarks, including the 2 anterior-posterior landmarks of the pons and 2 anterior-posterior and 2 superior-inferior landmarks of the vermis. Four-fold cross-validation was performed to test the accuracy of the model using randomly divided and sorted gestational age-divided data sets. A confidence score of model prediction was generated for each testing case. RESULTS Overall, 85% of the testing results showed a ≥90% confidence, with a mean error of <2.22 mm, providing overall better estimation results with fewer errors and higher confidence scores. The anterior and posterior pons and anterior vermis showed better estimation (which means fewer errors in landmark localization) and accuracy and a higher confidence level than other landmarks. We also developed a graphic user interface for clinical use. CONCLUSIONS This deep learning-facilitated pipeline practically shortens the time spent on selecting good-quality fetal brain images and performing anatomic measurements for radiologists.
Collapse
Affiliation(s)
- Farzan Vahedifard
- From the Department of Diagnostic Radiology and Nuclear Medicine (F.V., X.L., J.O.A., K.K.M., S.E.B.), Rush Medical College, Chicago, Illinois
| | - Xuchu Liu
- From the Department of Diagnostic Radiology and Nuclear Medicine (F.V., X.L., J.O.A., K.K.M., S.E.B.), Rush Medical College, Chicago, Illinois
| | - Jubril O Adepoju
- From the Department of Diagnostic Radiology and Nuclear Medicine (F.V., X.L., J.O.A., K.K.M., S.E.B.), Rush Medical College, Chicago, Illinois
| | - Shiqiao Zhao
- Department of Biostatistics (S.Z.), Yale School of Public Health, New Haven, Connecticut
| | - H Asher Ai
- Division for Diagnostic Medical Physics (H.A.A., M.S.), Department of Radiology and Nuclear Medicine, Rush University Medical Center, Chicago, Illinois
| | - Kranthi K Marathu
- From the Department of Diagnostic Radiology and Nuclear Medicine (F.V., X.L., J.O.A., K.K.M., S.E.B.), Rush Medical College, Chicago, Illinois
| | - Mark Supanich
- Division for Diagnostic Medical Physics (H.A.A., M.S.), Department of Radiology and Nuclear Medicine, Rush University Medical Center, Chicago, Illinois
| | - Sharon E Byrd
- From the Department of Diagnostic Radiology and Nuclear Medicine (F.V., X.L., J.O.A., K.K.M., S.E.B.), Rush Medical College, Chicago, Illinois
| | - Jie Deng
- Department of Radiation Oncology (J.D.), Division of Medical Physics & Engineering, University of Texas Southwestern Medical Center, Dallas, Texas
| |
Collapse
|
17
|
Qin Y, Liu Y, Cao C, Ouyang L, Ding Y, Wang D, Zheng M, Liao Z, Yue S, Liao W. A Novel Nomogram Based on Quantitative MRI and Clinical Features for the Prediction of Neonatal Intracranial Hypertension. CHILDREN (BASEL, SWITZERLAND) 2023; 10:1582. [PMID: 37892245 PMCID: PMC10605298 DOI: 10.3390/children10101582] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 08/18/2023] [Revised: 09/05/2023] [Accepted: 09/15/2023] [Indexed: 10/29/2023]
Abstract
Intracranial hypertension (ICH) is a serious threat to the health of neonates. However, early and accurate diagnosis of neonatal intracranial hypertension remains a major challenge in clinical practice. In this study, a predictive model based on quantitative magnetic resonance imaging (MRI) data and clinical parameters was developed to identify neonates with a high risk of ICH. Newborns who were suspected of having intracranial lesions were included in our study. We utilized quantitative MRI to obtain the volumetric data of gray matter, white matter, and cerebrospinal fluid. After the MRI examination, a lumbar puncture was performed. The nomogram was constructed by incorporating the volumetric data and clinical features by multivariable logistic regression. The performance of the nomogram was evaluated by discrimination, calibration curve, and decision curve. Clinical parameters and volumetric quantitative MRI data, including postmenstrual age (p = 0.06), weight (p = 0.02), mode of delivery (p = 0.01), and gray matter volume (p = 0.003), were included in and significantly associated with neonatal intracranial hypertension risk. The nomogram showed satisfactory discrimination, with an area under the curve of 0.761. Our results demonstrated that decision curve analysis had promising clinical utility of the nomogram. The nomogram, incorporating clinical and quantitative MRI features, provided an individualized prediction of neonatal intracranial hypertension risk and facilitated decision making guidance for the early diagnosis and treatment for neonatal ICH. External validation from studies using a larger sample size before implementation in the clinical decision-making process is needed.
Collapse
Affiliation(s)
- Yan Qin
- Department of Radiology, Xiangya Hospital, Central South University, 87 Xiangya Road, Changsha 410008, China
- National Clinical Research Center for Geriatric Disorders, Xiangya Hospital, Central South University, 87 Xiangya Road, Changsha 410008, China
| | - Yang Liu
- Department of Pediatrics, Xiangya Hospital, Central South University, 87 Xiangya Road, Changsha 410008, China (S.Y.)
| | - Chuanding Cao
- Department of Pediatrics, Xiangya Hospital, Central South University, 87 Xiangya Road, Changsha 410008, China (S.Y.)
| | - Lirong Ouyang
- Department of Radiology, Xiangya Hospital, Central South University, 87 Xiangya Road, Changsha 410008, China
| | - Ying Ding
- Department of Pediatrics, Xiangya Hospital, Central South University, 87 Xiangya Road, Changsha 410008, China (S.Y.)
| | - Dongcui Wang
- Department of Radiology, Xiangya Hospital, Central South University, 87 Xiangya Road, Changsha 410008, China
| | - Mengqiu Zheng
- Department of Pediatrics, Xiangya Hospital, Central South University, 87 Xiangya Road, Changsha 410008, China (S.Y.)
| | - Zhengchang Liao
- Department of Pediatrics, Xiangya Hospital, Central South University, 87 Xiangya Road, Changsha 410008, China (S.Y.)
| | - Shaojie Yue
- Department of Pediatrics, Xiangya Hospital, Central South University, 87 Xiangya Road, Changsha 410008, China (S.Y.)
| | - Weihua Liao
- Department of Radiology, Xiangya Hospital, Central South University, 87 Xiangya Road, Changsha 410008, China
- National Clinical Research Center for Geriatric Disorders, Xiangya Hospital, Central South University, 87 Xiangya Road, Changsha 410008, China
| |
Collapse
|
18
|
Ciceri T, Squarcina L, Giubergia A, Bertoldo A, Brambilla P, Peruzzo D. Review on deep learning fetal brain segmentation from Magnetic Resonance images. Artif Intell Med 2023; 143:102608. [PMID: 37673558 DOI: 10.1016/j.artmed.2023.102608] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/21/2022] [Revised: 05/31/2023] [Accepted: 06/06/2023] [Indexed: 09/08/2023]
Abstract
Brain segmentation is often the first and most critical step in quantitative analysis of the brain for many clinical applications, including fetal imaging. Different aspects challenge the segmentation of the fetal brain in magnetic resonance imaging (MRI), such as the non-standard position of the fetus owing to his/her movements during the examination, rapid brain development, and the limited availability of imaging data. In recent years, several segmentation methods have been proposed for automatically partitioning the fetal brain from MR images. These algorithms aim to define regions of interest with different shapes and intensities, encompassing the entire brain, or isolating specific structures. Deep learning techniques, particularly convolutional neural networks (CNNs), have become a state-of-the-art approach in the field because they can provide reliable segmentation results over heterogeneous datasets. Here, we review the deep learning algorithms developed in the field of fetal brain segmentation and categorize them according to their target structures. Finally, we discuss the perceived research gaps in the literature of the fetal domain, suggesting possible future research directions that could impact the management of fetal MR images.
Collapse
Affiliation(s)
- Tommaso Ciceri
- NeuroImaging Laboratory, Scientific Institute IRCCS Eugenio Medea, Bosisio Parini, Italy; Department of Information Engineering, University of Padua, Padua, Italy
| | - Letizia Squarcina
- Department of Pathophysiology and Transplantation, University of Milan, Milan, Italy
| | - Alice Giubergia
- NeuroImaging Laboratory, Scientific Institute IRCCS Eugenio Medea, Bosisio Parini, Italy; Department of Information Engineering, University of Padua, Padua, Italy
| | - Alessandra Bertoldo
- Department of Information Engineering, University of Padua, Padua, Italy; University of Padua, Padova Neuroscience Center, Padua, Italy
| | - Paolo Brambilla
- Department of Pathophysiology and Transplantation, University of Milan, Milan, Italy; Department of Neurosciences and Mental Health, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, Milan, Italy.
| | - Denis Peruzzo
- NeuroImaging Laboratory, Scientific Institute IRCCS Eugenio Medea, Bosisio Parini, Italy
| |
Collapse
|
19
|
Vahedifard F, Ai HA, Supanich MP, Marathu KK, Liu X, Kocak M, Ansari SM, Akyuz M, Adepoju JO, Adler S, Byrd S. Automatic Ventriculomegaly Detection in Fetal Brain MRI: A Step-by-Step Deep Learning Model for Novel 2D-3D Linear Measurements. Diagnostics (Basel) 2023; 13:2355. [PMID: 37510099 PMCID: PMC10378043 DOI: 10.3390/diagnostics13142355] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/15/2023] [Revised: 07/07/2023] [Accepted: 07/09/2023] [Indexed: 07/30/2023] Open
Abstract
In this study, we developed an automated workflow using a deep learning model (DL) to measure the lateral ventricle linearly in fetal brain MRI, which are subsequently classified into normal or ventriculomegaly, defined as a diameter wider than 10 mm at the level of the thalamus and choroid plexus. To accomplish this, we first trained a UNet-based deep learning model to segment the brain of a fetus into seven different tissue categories using a public dataset (FeTA 2022) consisting of fetal T2-weighted images. Then, an automatic workflow was developed to perform lateral ventricle measurement at the level of the thalamus and choroid plexus. The test dataset included 22 cases of normal and abnormal T2-weighted fetal brain MRIs. Measurements performed by our AI model were compared with manual measurements performed by a general radiologist and a neuroradiologist. The AI model correctly classified 95% of fetal brain MRI cases into normal or ventriculomegaly. It could measure the lateral ventricle diameter in 95% of cases with less than a 1.7 mm error. The average difference between measurements was 0.90 mm in AI vs. general radiologists and 0.82 mm in AI vs. neuroradiologists, which are comparable to the difference between the two radiologists, 0.51 mm. In addition, the AI model also enabled the researchers to create 3D-reconstructed images, which better represent real anatomy than 2D images. When a manual measurement is performed, it could also provide both the right and left ventricles in just one cut, instead of two. The measurement difference between the general radiologist and the algorithm (p = 0.9827), and between the neuroradiologist and the algorithm (p = 0.2378), was not statistically significant. In contrast, the difference between general radiologists vs. neuroradiologists was statistically significant (p = 0.0043). To the best of our knowledge, this is the first study that performs 2D linear measurement of ventriculomegaly with a 3D model based on an artificial intelligence approach. The paper presents a step-by-step approach for designing an AI model based on several radiological criteria. Overall, this study showed that AI can automatically calculate the lateral ventricle in fetal brain MRIs and accurately classify them as abnormal or normal.
Collapse
Affiliation(s)
- Farzan Vahedifard
- Department of Diagnostic Radiology and Nuclear Medicine, Rush University Medical Center, Rush Medical College, Chicago, IL 60612, USA
| | - H Asher Ai
- Division for Diagnostic Medical Physics, Department of Radiology and Nuclear Medicine, Rush University Medical Center, Rush Medical College, Chicago, IL 60612, USA
| | - Mark P Supanich
- Division for Diagnostic Medical Physics, Department of Radiology and Nuclear Medicine, Rush University Medical Center, Rush Medical College, Chicago, IL 60612, USA
| | - Kranthi K Marathu
- Department of Diagnostic Radiology and Nuclear Medicine, Rush University Medical Center, Rush Medical College, Chicago, IL 60612, USA
| | - Xuchu Liu
- Department of Diagnostic Radiology and Nuclear Medicine, Rush University Medical Center, Rush Medical College, Chicago, IL 60612, USA
| | - Mehmet Kocak
- Department of Diagnostic Radiology and Nuclear Medicine, Rush University Medical Center, Rush Medical College, Chicago, IL 60612, USA
| | - Shehbaz M Ansari
- Department of Diagnostic Radiology and Nuclear Medicine, Rush University Medical Center, Rush Medical College, Chicago, IL 60612, USA
| | - Melih Akyuz
- Department of Diagnostic Radiology and Nuclear Medicine, Rush University Medical Center, Rush Medical College, Chicago, IL 60612, USA
| | - Jubril O Adepoju
- Department of Diagnostic Radiology and Nuclear Medicine, Rush University Medical Center, Rush Medical College, Chicago, IL 60612, USA
| | - Seth Adler
- Department of Diagnostic Radiology and Nuclear Medicine, Rush University Medical Center, Rush Medical College, Chicago, IL 60612, USA
| | - Sharon Byrd
- Department of Diagnostic Radiology and Nuclear Medicine, Rush University Medical Center, Rush Medical College, Chicago, IL 60612, USA
| |
Collapse
|
20
|
Shen DD, Bao SL, Wang Y, Chen YC, Zhang YC, Li XC, Ding YC, Jia ZZ. An automatic and accurate deep learning-based neuroimaging pipeline for the neonatal brain. Pediatr Radiol 2023; 53:1685-1697. [PMID: 36884052 DOI: 10.1007/s00247-023-05620-x] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 10/20/2022] [Revised: 01/26/2023] [Accepted: 01/27/2023] [Indexed: 03/09/2023]
Abstract
BACKGROUND Accurate segmentation of neonatal brain tissues and structures is crucial for studying normal development and diagnosing early neurodevelopmental disorders. However, there is a lack of an end-to-end pipeline for automated segmentation and imaging analysis of the normal and abnormal neonatal brain. OBJECTIVE To develop and validate a deep learning-based pipeline for neonatal brain segmentation and analysis of structural magnetic resonance images (MRI). MATERIALS AND METHODS Two cohorts were enrolled in the study, including cohort 1 (582 neonates from the developing Human Connectome Project) and cohort 2 (37 neonates imaged using a 3.0-tesla MRI scanner in our hospital).We developed a deep leaning-based architecture capable of brain segmentation into 9 tissues and 87 structures. Then, extensive validations were performed for accuracy, effectiveness, robustness and generality of the pipeline. Furthermore, regional volume and cortical surface estimation were measured through in-house bash script implemented in FSL (Oxford Centre for Functional MRI of the Brain Software Library) to ensure reliability of the pipeline. Dice similarity score (DSC), the 95th percentile Hausdorff distance (H95) and intraclass correlation coefficient (ICC) were calculated to assess the quality of our pipeline. Finally, we finetuned and validated our pipeline on 2-dimensional thick-slice MRI in cohorts 1 and 2. RESULTS The deep learning-based model showed excellent performance for neonatal brain tissue and structural segmentation, with the best DSC and the 95th percentile Hausdorff distance (H95) of 0.96 and 0.99 mm, respectively. In terms of regional volume and cortical surface analysis, our model showed good agreement with ground truth. The ICC values for the regional volume were all above 0.80. Considering the thick-slice image pipeline, the same trend was observed for brain segmentation and analysis. The best DSC and H95 were 0.92 and 3.00 mm, respectively. The regional volumes and surface curvature had ICC values just below 0.80. CONCLUSIONS We propose an automatic, accurate, stable and reliable pipeline for neonatal brain segmentation and analysis from thin and thick structural MRI. The external validation showed very good reproducibility of the pipeline.
Collapse
Affiliation(s)
- Dan Dan Shen
- Department of Medical Imaging, Affiliated Hospital and Medical School of Nantong University, NO.20 Xisi Road, Nantong, Jiangsu, 226001, People's Republic of China
| | - Shan Lei Bao
- Department of Nuclear Medicine, Affiliated Hospital and Medical School of Nantong University, Jiangsu, People's Republic of China
| | - Yan Wang
- Department of Medical Imaging, Affiliated Hospital and Medical School of Nantong University, NO.20 Xisi Road, Nantong, Jiangsu, 226001, People's Republic of China
| | - Ying Chi Chen
- Department of Medical Imaging, Affiliated Hospital and Medical School of Nantong University, NO.20 Xisi Road, Nantong, Jiangsu, 226001, People's Republic of China
| | - Yu Cheng Zhang
- Department of Medical Imaging, Affiliated Hospital and Medical School of Nantong University, NO.20 Xisi Road, Nantong, Jiangsu, 226001, People's Republic of China
| | - Xing Can Li
- Department of Medical Imaging, Affiliated Hospital and Medical School of Nantong University, NO.20 Xisi Road, Nantong, Jiangsu, 226001, People's Republic of China
| | - Yu Chen Ding
- Department of Medical Imaging, Affiliated Hospital and Medical School of Nantong University, NO.20 Xisi Road, Nantong, Jiangsu, 226001, People's Republic of China
| | - Zhong Zheng Jia
- Department of Medical Imaging, Affiliated Hospital and Medical School of Nantong University, NO.20 Xisi Road, Nantong, Jiangsu, 226001, People's Republic of China.
| |
Collapse
|
21
|
Vahedifard F, Adepoju JO, Supanich M, Ai HA, Liu X, Kocak M, Marathu KK, Byrd SE. Review of deep learning and artificial intelligence models in fetal brain magnetic resonance imaging. World J Clin Cases 2023; 11:3725-3735. [PMID: 37383127 PMCID: PMC10294149 DOI: 10.12998/wjcc.v11.i16.3725] [Citation(s) in RCA: 3] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 01/04/2023] [Revised: 01/30/2023] [Accepted: 05/06/2023] [Indexed: 06/02/2023] Open
Abstract
Central nervous system abnormalities in fetuses are fairly common, happening in 0.1% to 0.2% of live births and in 3% to 6% of stillbirths. So initial detection and categorization of fetal Brain abnormalities are critical. Manually detecting and segmenting fetal brain magnetic resonance imaging (MRI) could be time-consuming, and susceptible to interpreter experience. Artificial intelligence (AI) algorithms and machine learning approaches have a high potential for assisting in the early detection of these problems, improving the diagnosis process and follow-up procedures. The use of AI and machine learning techniques in fetal brain MRI was the subject of this narrative review paper. Using AI, anatomic fetal brain MRI processing has investigated models to predict specific landmarks and segmentation automatically. All gestation age weeks (17-38 wk) and different AI models (mainly Convolutional Neural Network and U-Net) have been used. Some models' accuracy achieved 95% and more. AI could help preprocess and post-process fetal images and reconstruct images. Also, AI can be used for gestational age prediction (with one-week accuracy), fetal brain extraction, fetal brain segmentation, and placenta detection. Some fetal brain linear measurements, such as Cerebral and Bone Biparietal Diameter, have been suggested. Classification of brain pathology was studied using diagonal quadratic discriminates analysis, K-nearest neighbor, random forest, naive Bayes, and radial basis function neural network classifiers. Deep learning methods will become more powerful as more large-scale, labeled datasets become available. Having shared fetal brain MRI datasets is crucial because there aren not many fetal brain pictures available. Also, physicians should be aware of AI's function in fetal brain MRI, particularly neuroradiologists, general radiologists, and perinatologists.
Collapse
Affiliation(s)
- Farzan Vahedifard
- Department of Diagnostic Radiology and Nuclear Medicine, Rush Medical College, Chicago, IL 606012, United States
| | - Jubril O Adepoju
- Department of Diagnostic Radiology and Nuclear Medicine, Rush Medical College, Chicago, IL 606012, United States
| | - Mark Supanich
- Division for Diagnostic Medical Physics, Department of Radiology and Nuclear Medicine, Rush University Medical Center, Chicago, IL 606012, United States
| | - Hua Asher Ai
- Division for Diagnostic Medical Physics, Department of Radiology and Nuclear Medicine, Rush University Medical Center, Chicago, IL 606012, United States
| | - Xuchu Liu
- Department of Diagnostic Radiology and Nuclear Medicine, Rush Medical College, Chicago, IL 606012, United States
| | - Mehmet Kocak
- Department of Diagnostic Radiology and Nuclear Medicine, Rush Medical College, Chicago, IL 606012, United States
| | - Kranthi K Marathu
- Department of Diagnostic Radiology and Nuclear Medicine, Rush Medical College, Chicago, IL 606012, United States
| | - Sharon E Byrd
- Department of Diagnostic Radiology and Nuclear Medicine, Rush Medical College, Chicago, IL 606012, United States
| |
Collapse
|
22
|
de Vareilles H, Rivière D, Mangin JF, Dubois J. Development of cortical folds in the human brain: An attempt to review biological hypotheses, early neuroimaging investigations and functional correlates. Dev Cogn Neurosci 2023; 61:101249. [PMID: 37141790 PMCID: PMC10311195 DOI: 10.1016/j.dcn.2023.101249] [Citation(s) in RCA: 6] [Impact Index Per Article: 6.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/14/2022] [Revised: 03/28/2023] [Accepted: 04/21/2023] [Indexed: 05/06/2023] Open
Abstract
The folding of the human brain mostly takes place in utero, making it challenging to study. After a few pioneer studies looking into it in post-mortem foetal specimen, modern approaches based on neuroimaging have allowed the community to investigate the folding process in vivo, its normal progression, its early disturbances, and its relationship to later functional outcomes. In this review article, we aimed to first give an overview of the current hypotheses on the mechanisms governing cortical folding. After describing the methodological difficulties raised by its study in fetuses, neonates and infants with magnetic resonance imaging (MRI), we reported our current understanding of sulcal pattern emergence in the developing brain. We then highlighted the functional relevance of early sulcal development, through recent insights about hemispheric asymmetries and early factors influencing this dynamic such as prematurity. Finally, we outlined how longitudinal studies have started to relate early folding markers and the child's sensorimotor and cognitive outcome. Through this review, we hope to raise awareness on the potential of studying early sulcal patterns both from a fundamental and clinical perspective, as a window into early neurodevelopment and plasticity in relation to growth in utero and postnatal environment of the child.
Collapse
Affiliation(s)
- H de Vareilles
- Université Paris-Saclay, NeuroSpin-BAOBAB, CEA, CNRS, Gif-sur-Yvette, France.
| | - D Rivière
- Université Paris-Saclay, NeuroSpin-BAOBAB, CEA, CNRS, Gif-sur-Yvette, France
| | - J F Mangin
- Université Paris-Saclay, NeuroSpin-BAOBAB, CEA, CNRS, Gif-sur-Yvette, France
| | - J Dubois
- Université Paris Cité, NeuroDiderot, Inserm, Paris, France; Université Paris-Saclay, NeuroSpin-UNIACT, CEA, Gif-sur-Yvette, France
| |
Collapse
|
23
|
Conole ELS, Vaher K, Cabez MB, Sullivan G, Stevenson AJ, Hall J, Murphy L, Thrippleton MJ, Quigley AJ, Bastin ME, Miron VE, Whalley HC, Marioni RE, Boardman JP, Cox SR. Immuno-epigenetic signature derived in saliva associates with the encephalopathy of prematurity and perinatal inflammatory disorders. Brain Behav Immun 2023; 110:322-338. [PMID: 36948324 DOI: 10.1016/j.bbi.2023.03.011] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 10/19/2022] [Revised: 02/12/2023] [Accepted: 03/16/2023] [Indexed: 03/24/2023] Open
Abstract
BACKGROUND Preterm birth is closely associated with a phenotype that includes brain dysmaturation and neurocognitive impairment, commonly termed Encephalopathy of Prematurity (EoP), of which systemic inflammation is considered a key driver. DNA methylation (DNAm) signatures of inflammation from peripheral blood associate with poor brain imaging outcomes in adult cohorts. However, the robustness of DNAm inflammatory scores in infancy, their relation to comorbidities of preterm birth characterised by inflammation, neonatal neuroimaging metrics of EoP, and saliva cross-tissue applicability are unknown. METHODS Using salivary DNAm from 258 neonates (n = 155 preterm, gestational age at birth 23.28 - 34.84 weeks, n = 103 term, gestational age at birth 37.00 - 42.14 weeks), we investigated the impact of a DNAm surrogate for C-reactive protein (DNAm CRP) on brain structure and other clinically defined inflammatory exposures. We assessed i) if DNAm CRP estimates varied between preterm infants at term equivalent age and term infants, ii) how DNAm CRP related to different types of inflammatory exposure (maternal, fetal and postnatal) and iii) whether elevated DNAm CRP associated with poorer measures of neonatal brain volume and white matter connectivity. RESULTS Higher DNAm CRP was linked to preterm status (-0.0107 ± 0.0008, compared with -0.0118 ± 0.0006 among term infants; p < 0.001), as well as perinatal inflammatory diseases, including histologic chorioamnionitis, sepsis, bronchopulmonary dysplasia, and necrotising enterocolitis (OR range |2.00 | to |4.71|, p < 0.01). Preterm infants with higher DNAm CRP scores had lower brain volume in deep grey matter, white matter, and hippocampi and amygdalae (β range |0.185| to |0.218|). No such associations were observed for term infants. Association magnitudes were largest for measures of white matter microstructure among preterms, where elevated epigenetic inflammation associated with poorer global measures of white matter integrity (β range |0.206| to |0.371|), independent of other confounding exposures. CONCLUSIONS Inflammatory-related DNAm captures the allostatic load of inflammatory burden in preterm infants. Such DNAm measures complement biological and clinical metrics when investigating the determinants of neurodevelopmental differences.
Collapse
Affiliation(s)
- Eleanor L S Conole
- Lothian Birth Cohorts group, Department of Psychology, University of Edinburgh, Edinburgh EH8 9JZ, UK; Centre for Genomic and Experimental Medicine, Institute of Genetics and Cancer, University of Edinburgh, Edinburgh EH4 2XU, UK; Centre for Clinical Brain Sciences, University of Edinburgh, Edinburgh EH16 4SB, UK.
| | - Kadi Vaher
- MRC Centre for Reproductive Health, Queen's Medical Research Institute, Edinburgh BioQuarter, University of Edinburgh, Edinburgh EH16 4TJ, UK
| | - Manuel Blesa Cabez
- MRC Centre for Reproductive Health, Queen's Medical Research Institute, Edinburgh BioQuarter, University of Edinburgh, Edinburgh EH16 4TJ, UK
| | - Gemma Sullivan
- MRC Centre for Reproductive Health, Queen's Medical Research Institute, Edinburgh BioQuarter, University of Edinburgh, Edinburgh EH16 4TJ, UK
| | - Anna J Stevenson
- Centre for Genomic and Experimental Medicine, Institute of Genetics and Cancer, University of Edinburgh, Edinburgh EH4 2XU, UK
| | - Jill Hall
- MRC Centre for Reproductive Health, Queen's Medical Research Institute, Edinburgh BioQuarter, University of Edinburgh, Edinburgh EH16 4TJ, UK
| | - Lee Murphy
- Edinburgh Clinical Research Facility, University of Edinburgh, Edinburgh EH4 2XU, UK
| | - Michael J Thrippleton
- Centre for Clinical Brain Sciences, University of Edinburgh, Edinburgh EH16 4SB, UK; Edinburgh Clinical Research Facility, University of Edinburgh, Edinburgh EH4 2XU, UK
| | - Alan J Quigley
- Imaging Department, Royal Hospital for Children and Young People, Edinburgh, EH16 4TJ, UK
| | - Mark E Bastin
- Lothian Birth Cohorts group, Department of Psychology, University of Edinburgh, Edinburgh EH8 9JZ, UK; Centre for Clinical Brain Sciences, University of Edinburgh, Edinburgh EH16 4SB, UK
| | - Veronique E Miron
- MRC Centre for Reproductive Health, Queen's Medical Research Institute, Edinburgh BioQuarter, University of Edinburgh, Edinburgh EH16 4TJ, UK
| | - Heather C Whalley
- Centre for Clinical Brain Sciences, University of Edinburgh, Edinburgh EH16 4SB, UK
| | - Riccardo E Marioni
- Centre for Genomic and Experimental Medicine, Institute of Genetics and Cancer, University of Edinburgh, Edinburgh EH4 2XU, UK
| | - James P Boardman
- Centre for Clinical Brain Sciences, University of Edinburgh, Edinburgh EH16 4SB, UK; MRC Centre for Reproductive Health, Queen's Medical Research Institute, Edinburgh BioQuarter, University of Edinburgh, Edinburgh EH16 4TJ, UK
| | - Simon R Cox
- Lothian Birth Cohorts group, Department of Psychology, University of Edinburgh, Edinburgh EH8 9JZ, UK.
| |
Collapse
|
24
|
Deprest T, Fidon L, De Keyzer F, Ebner M, Deprest J, Demaerel P, De Catte L, Vercauteren T, Ourselin S, Dymarkowski S, Aertsen M. Application of Automatic Segmentation on Super-Resolution Reconstruction MR Images of the Abnormal Fetal Brain. AJNR Am J Neuroradiol 2023; 44:486-491. [PMID: 36863845 PMCID: PMC10084897 DOI: 10.3174/ajnr.a7808] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/06/2022] [Accepted: 02/06/2023] [Indexed: 03/04/2023]
Abstract
BACKGROUND AND PURPOSE Fetal brain MR imaging is clinically used to characterize fetal brain abnormalities. Recently, algorithms have been proposed to reconstruct high-resolution 3D fetal brain volumes from 2D slices. By means of these reconstructions, convolutional neural networks have been developed for automatic image segmentation to avoid labor-intensive manual annotations, usually trained on data of normal fetal brains. Herein, we tested the performance of an algorithm specifically developed for segmentation of abnormal fetal brains. MATERIALS AND METHODS This was a single-center retrospective study on MR images of 16 fetuses with severe CNS anomalies (gestation, 21-39 weeks). T2-weighted 2D slices were converted to 3D volumes using a super-resolution reconstruction algorithm. The acquired volumetric data were then processed by a novel convolutional neural network to perform segmentations of white matter and the ventricular system and cerebellum. These were compared with manual segmentation using the Dice coefficient, Hausdorff distance (95th percentile), and volume difference. Using interquartile ranges, we identified outliers of these metrics and further analyzed them in detail. RESULTS The mean Dice coefficient was 96.2%, 93.7%, and 94.7% for white matter and the ventricular system and cerebellum, respectively. The Hausdorff distance was 1.1, 2.3, and 1.6 mm, respectively. The volume difference was 1.6, 1.4, and 0.3 mL, respectively. Of the 126 measurements, there were 16 outliers among 5 fetuses, discussed on a case-by-case basis. CONCLUSIONS Our novel segmentation algorithm obtained excellent results on MR images of fetuses with severe brain abnormalities. Analysis of the outliers shows the need to include pathologies underrepresented in the current data set. Quality control to prevent occasional errors is still needed.
Collapse
Affiliation(s)
- T Deprest
- From the Department of Radiology (T.D., F.D.K., P.D., S.D., M.A.)
| | - L Fidon
- School of Biomedical Engineering and Imaging Sciences (L.F., M.E., T.V., S.O.), King's College London, London, UK
| | - F De Keyzer
- From the Department of Radiology (T.D., F.D.K., P.D., S.D., M.A.)
| | - M Ebner
- School of Biomedical Engineering and Imaging Sciences (L.F., M.E., T.V., S.O.), King's College London, London, UK
- Department of Medical Physics and Biomedical Engineering (M.E., T.V.), University College London, London, UK
| | - J Deprest
- Gynaecology and Obstetrics (J.D., L.D.C., T.V.), University Hospitals Leuven, Belgium
- Institute for Women's Health (J.D.)
| | - P Demaerel
- From the Department of Radiology (T.D., F.D.K., P.D., S.D., M.A.)
| | - L De Catte
- Gynaecology and Obstetrics (J.D., L.D.C., T.V.), University Hospitals Leuven, Belgium
| | - T Vercauteren
- Gynaecology and Obstetrics (J.D., L.D.C., T.V.), University Hospitals Leuven, Belgium
- School of Biomedical Engineering and Imaging Sciences (L.F., M.E., T.V., S.O.), King's College London, London, UK
- Department of Medical Physics and Biomedical Engineering (M.E., T.V.), University College London, London, UK
| | - S Ourselin
- School of Biomedical Engineering and Imaging Sciences (L.F., M.E., T.V., S.O.), King's College London, London, UK
| | - S Dymarkowski
- From the Department of Radiology (T.D., F.D.K., P.D., S.D., M.A.)
| | - M Aertsen
- From the Department of Radiology (T.D., F.D.K., P.D., S.D., M.A.)
| |
Collapse
|
25
|
Karimi D, Rollins CK, Velasco-Annis C, Ouaalam A, Gholipour A. Learning to segment fetal brain tissue from noisy annotations. Med Image Anal 2023; 85:102731. [PMID: 36608414 PMCID: PMC9974964 DOI: 10.1016/j.media.2022.102731] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/12/2022] [Revised: 11/17/2022] [Accepted: 12/23/2022] [Indexed: 01/03/2023]
Abstract
Automatic fetal brain tissue segmentation can enhance the quantitative assessment of brain development at this critical stage. Deep learning methods represent the state of the art in medical image segmentation and have also achieved impressive results in brain segmentation. However, effective training of a deep learning model to perform this task requires a large number of training images to represent the rapid development of the transient fetal brain structures. On the other hand, manual multi-label segmentation of a large number of 3D images is prohibitive. To address this challenge, we segmented 272 training images, covering 19-39 gestational weeks, using an automatic multi-atlas segmentation strategy based on deformable registration and probabilistic atlas fusion, and manually corrected large errors in those segmentations. Since this process generated a large training dataset with noisy segmentations, we developed a novel label smoothing procedure and a loss function to train a deep learning model with smoothed noisy segmentations. Our proposed methods properly account for the uncertainty in tissue boundaries. We evaluated our method on 23 manually-segmented test images of a separate set of fetuses. Results show that our method achieves an average Dice similarity coefficient of 0.893 and 0.916 for the transient structures of younger and older fetuses, respectively. Our method generated results that were significantly more accurate than several state-of-the-art methods including nnU-Net that achieved the closest results to our method. Our trained model can serve as a valuable tool to enhance the accuracy and reproducibility of fetal brain analysis in MRI.
Collapse
Affiliation(s)
- Davood Karimi
- Department of Radiology, Boston Children's Hospital, Harvard Medical School, Boston, MA, USA.
| | - Caitlin K Rollins
- Department of Neurology, Boston Children's Hospital, Harvard Medical School, Boston, MA, USA
| | - Clemente Velasco-Annis
- Department of Radiology, Boston Children's Hospital, Harvard Medical School, Boston, MA, USA
| | - Abdelhakim Ouaalam
- Department of Radiology, Boston Children's Hospital, Harvard Medical School, Boston, MA, USA
| | - Ali Gholipour
- Department of Radiology, Boston Children's Hospital, Harvard Medical School, Boston, MA, USA
| |
Collapse
|
26
|
Zoetmulder R, Baak L, Khalili N, Marquering HA, Wagenaar N, Benders M, van der Aa NE, Išgum I. Brain segmentation in patients with perinatal arterial ischemic stroke. Neuroimage Clin 2023; 38:103381. [PMID: 36965456 PMCID: PMC10074207 DOI: 10.1016/j.nicl.2023.103381] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/02/2022] [Revised: 02/20/2023] [Accepted: 03/14/2023] [Indexed: 03/19/2023]
Abstract
BACKGROUND Perinatal arterial ischemic stroke (PAIS) is associated with adverse neurological outcomes. Quantification of ischemic lesions and consequent brain development in newborn infants relies on labor-intensive manual assessment of brain tissues and ischemic lesions. Hence, we propose an automatic method utilizing convolutional neural networks (CNNs) to segment brain tissues and ischemic lesions in MRI scans of infants suffering from PAIS. MATERIALS AND METHODS This single-center retrospective study included 115 patients with PAIS that underwent MRI after the stroke onset (baseline) and after three months (follow-up). Nine baseline and 12 follow-up MRI scans were manually annotated to provide reference segmentations (white matter, gray matter, basal ganglia and thalami, brainstem, ventricles, extra-ventricular cerebrospinal fluid, and cerebellum, and additionally on the baseline scans the ischemic lesions). Two CNNs were trained to perform automatic segmentation on the baseline and follow-up MRIs, respectively. Automatic segmentations were quantitatively evaluated using the Dice coefficient (DC) and the mean surface distance (MSD). Volumetric agreement between segmentations that were manually and automatically obtained was computed. Moreover, the scan quality and automatic segmentations were qualitatively evaluated in a larger set of MRIs without manual annotation by two experts. In addition, the scan quality was qualitatively evaluated in these scans to establish its impact on the automatic segmentation performance. RESULTS Automatic brain tissue segmentation led to a DC and MSD between 0.78-0.92 and 0.18-1.08 mm for baseline, and between 0.88-0.95 and 0.10-0.58 mm for follow-up scans, respectively. For the ischemic lesions at baseline the DC and MSD were between 0.72-0.86 and 1.23-2.18 mm, respectively. Volumetric measurements indicated limited oversegmentation of the extra-ventricular cerebrospinal fluid in both the follow-up and baseline scans, oversegmentation of the ischemic lesions in the left hemisphere, and undersegmentation of the ischemic lesions in the right hemisphere. In scans without imaging artifacts, brain tissue segmentation was graded as excellent in more than 85% and 91% of cases, respectively for the baseline and follow-up scans. For the ischemic lesions at baseline, this was in 61% of cases. CONCLUSIONS Automatic segmentation of brain tissue and ischemic lesions in MRI scans of patients with PAIS is feasible. The method may allow evaluation of the brain development and efficacy of treatment in large datasets.
Collapse
Affiliation(s)
- Riaan Zoetmulder
- Department of Biomedical Engineering and Physics, Amsterdam University Medical Center, Location University of Amsterdam, Amsterdam, the Netherlands; Department of Radiology and Nuclear Medicine, Amsterdam University Medical Center, Location University of Amsterdam, Amsterdam, the Netherlands; Informatics Institute, University of Amsterdam, Amsterdam, the Netherlands
| | - Lisanne Baak
- Department of Neonatology and Utrecht Brain Center, University Medical Center Utrecht, Utrecht, the Netherlands
| | - Nadieh Khalili
- Image Sciences Institute, University Medical Center Utrecht, Utrecht, the Netherlands
| | - Henk A Marquering
- Department of Biomedical Engineering and Physics, Amsterdam University Medical Center, Location University of Amsterdam, Amsterdam, the Netherlands; Department of Radiology and Nuclear Medicine, Amsterdam University Medical Center, Location University of Amsterdam, Amsterdam, the Netherlands
| | - Nienke Wagenaar
- Department of Neonatology and Utrecht Brain Center, University Medical Center Utrecht, Utrecht, the Netherlands
| | - Manon Benders
- Department of Neonatology and Utrecht Brain Center, University Medical Center Utrecht, Utrecht, the Netherlands
| | - Niek E van der Aa
- Department of Neonatology and Utrecht Brain Center, University Medical Center Utrecht, Utrecht, the Netherlands
| | - Ivana Išgum
- Department of Biomedical Engineering and Physics, Amsterdam University Medical Center, Location University of Amsterdam, Amsterdam, the Netherlands; Department of Radiology and Nuclear Medicine, Amsterdam University Medical Center, Location University of Amsterdam, Amsterdam, the Netherlands; Informatics Institute, University of Amsterdam, Amsterdam, the Netherlands; Amsterdam Cardiovascular Sciences, Amsterdam, the Netherlands.
| |
Collapse
|
27
|
Zeng Z, Zhao T, Sun L, Zhang Y, Xia M, Liao X, Zhang J, Shen D, Wang L, He Y. 3D-MASNet: 3D mixed-scale asymmetric convolutional segmentation network for 6-month-old infant brain MR images. Hum Brain Mapp 2023; 44:1779-1792. [PMID: 36515219 PMCID: PMC9921327 DOI: 10.1002/hbm.26174] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/24/2022] [Revised: 11/04/2022] [Accepted: 11/25/2022] [Indexed: 12/15/2022] Open
Abstract
Precise segmentation of infant brain magnetic resonance (MR) images into gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF) are essential for studying neuroanatomical hallmarks of early brain development. However, for 6-month-old infants, the extremely low-intensity contrast caused by inherent myelination hinders accurate tissue segmentation. Existing convolutional neural networks (CNNs) based segmentation models for this task generally employ single-scale symmetric convolutions, which are inefficient for encoding the isointense tissue boundaries in baby brain images. Here, we propose a 3D mixed-scale asymmetric convolutional segmentation network (3D-MASNet) framework for brain MR images of 6-month-old infants. We replaced the traditional convolutional layer of an existing to-be-trained network with a 3D mixed-scale convolution block consisting of asymmetric kernels (MixACB) during the training phase and then equivalently converted it into the original network. Five canonical CNN segmentation models were evaluated using both T1- and T2-weighted images of 23 6-month-old infants from iSeg-2019 datasets, which contained manual labels as ground truth. MixACB significantly enhanced the average accuracy of all five models and obtained the most considerable improvement in the fully convolutional network model (CC-3D-FCN) and the highest performance in the Dense U-Net model. This approach further obtained Dice coefficient accuracies of 0.931, 0.912, and 0.961 in GM, WM, and CSF, respectively, ranking first among 30 teams on the validation dataset of the iSeg-2019 Grand Challenge. Thus, the proposed 3D-MASNet can improve the accuracy of existing CNNs-based segmentation models as a plug-and-play solution that offers a promising technique for future infant brain MRI studies.
Collapse
Affiliation(s)
- Zilong Zeng
- State Key Laboratory of Cognitive Neuroscience and LearningBeijing Normal UniversityBeijingChina
- Beijing Key Laboratory of Brain Imaging and ConnectomicsBeijing Normal UniversityBeijingChina
- IDG/McGovern Institute for Brain Research, Beijing Normal UniversityBeijingChina
| | - Tengda Zhao
- State Key Laboratory of Cognitive Neuroscience and LearningBeijing Normal UniversityBeijingChina
- Beijing Key Laboratory of Brain Imaging and ConnectomicsBeijing Normal UniversityBeijingChina
- IDG/McGovern Institute for Brain Research, Beijing Normal UniversityBeijingChina
| | - Lianglong Sun
- State Key Laboratory of Cognitive Neuroscience and LearningBeijing Normal UniversityBeijingChina
- Beijing Key Laboratory of Brain Imaging and ConnectomicsBeijing Normal UniversityBeijingChina
- IDG/McGovern Institute for Brain Research, Beijing Normal UniversityBeijingChina
| | - Yihe Zhang
- State Key Laboratory of Cognitive Neuroscience and LearningBeijing Normal UniversityBeijingChina
- Beijing Key Laboratory of Brain Imaging and ConnectomicsBeijing Normal UniversityBeijingChina
- IDG/McGovern Institute for Brain Research, Beijing Normal UniversityBeijingChina
| | - Mingrui Xia
- State Key Laboratory of Cognitive Neuroscience and LearningBeijing Normal UniversityBeijingChina
- Beijing Key Laboratory of Brain Imaging and ConnectomicsBeijing Normal UniversityBeijingChina
- IDG/McGovern Institute for Brain Research, Beijing Normal UniversityBeijingChina
| | - Xuhong Liao
- School of Systems ScienceBeijing Normal UniversityBeijingChina
| | - Jiaying Zhang
- State Key Laboratory of Cognitive Neuroscience and LearningBeijing Normal UniversityBeijingChina
- Beijing Key Laboratory of Brain Imaging and ConnectomicsBeijing Normal UniversityBeijingChina
- IDG/McGovern Institute for Brain Research, Beijing Normal UniversityBeijingChina
| | - Dinggang Shen
- School of Biomedical EngineeringShanghaiTech UniversityShanghaiChina
- Shanghai Clinical Research and Trial CenterShanghaiChina
- Department of Research and DevelopmentShanghai United Imaging Intelligence Co., Ltd.ShanghaiChina
| | - Li Wang
- Department of Radiology and BRICUniversity of North Carolina at Chapel HillChapel HillNorth CarolinaUSA
| | - Yong He
- State Key Laboratory of Cognitive Neuroscience and LearningBeijing Normal UniversityBeijingChina
- Beijing Key Laboratory of Brain Imaging and ConnectomicsBeijing Normal UniversityBeijingChina
- IDG/McGovern Institute for Brain Research, Beijing Normal UniversityBeijingChina
- Chinese Institute for Brain ResearchBeijingChina
| |
Collapse
|
28
|
Sullivan G, Vaher K, Blesa M, Galdi P, Stoye DQ, Quigley AJ, Thrippleton MJ, Norrie J, Bastin ME, Boardman JP. Breast Milk Exposure is Associated With Cortical Maturation in Preterm Infants. Ann Neurol 2023; 93:591-603. [PMID: 36412221 DOI: 10.1002/ana.26559] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/10/2022] [Revised: 11/14/2022] [Accepted: 11/18/2022] [Indexed: 11/23/2022]
Abstract
OBJECTIVE Breast milk exposure is associated with improved neurocognitive outcomes following preterm birth but the neural substrates linking breast milk with outcome are uncertain. We tested the hypothesis that high versus low breast milk exposure in preterm infants results in cortical morphology that more closely resembles that of term-born infants. METHODS We studied 135 preterm (<32 weeks' gestation) and 77 term infants. Feeding data were collected from birth until hospital discharge and brain magnetic resonance imaging (MRI) was performed at term-equivalent age. Cortical indices (volume, thickness, surface area, gyrification index, sulcal depth, and curvature) and diffusion parameters (fractional anisotropy [FA], mean diffusivity [MD], radial diffusivity [RD], axial diffusivity [AD], neurite density index [NDI], and orientation dispersion index [ODI]) were compared between preterm infants who received exclusive breast milk for <75% of inpatient days, preterm infants who received exclusive breast milk for ≥75% of inpatient days and term-born controls. To investigate a dose response effect, we performed linear regression using breast milk exposure quartile weighted by propensity scores. RESULTS In preterm infants, high breast milk exposure was associated with reduced cortical gray matter volume (d = 0.47, 95% confidence interval [CI] = 0.14 to 0.94, p = 0.014), thickness (d = 0.42, 95% CI = 0.08 to 0.84, p = 0.039), and RD (d = 0.38, 95% CI = 0.002 to 0.77, p = 0.039), and increased FA (d = -0.38, 95% CI = -0.74 to -0.01, p = 0.037) after adjustment for age at MRI, which was similar to the cortical phenotype observed in term-born controls. Breast milk exposure quartile was associated with cortical volume (ß = -0.192, 95% CI = -0.342 to -0.042, p = 0.017), FA (ß = 0.223, 95% CI = 0.075 to 0.372, p = 0.007), and RD (ß = -0.225, 95% CI = -0.373 to -0.076, p = 0.007) following adjustment for age at birth, age at MRI, and weighted by propensity scores, suggesting a dose effect. INTERPRETATION High breast milk exposure following preterm birth is associated with a cortical imaging phenotype that more closely resembles the brain morphology of term-born infants and effects appear to be dose-dependent. ANN NEUROL 2023;93:591-603.
Collapse
Affiliation(s)
- Gemma Sullivan
- MRC Centre for Reproductive Health, Queen's Medical Research Institute, University of Edinburgh, Edinburgh, UK
| | - Kadi Vaher
- MRC Centre for Reproductive Health, Queen's Medical Research Institute, University of Edinburgh, Edinburgh, UK
| | - Manuel Blesa
- MRC Centre for Reproductive Health, Queen's Medical Research Institute, University of Edinburgh, Edinburgh, UK
| | - Paola Galdi
- MRC Centre for Reproductive Health, Queen's Medical Research Institute, University of Edinburgh, Edinburgh, UK
| | - David Q Stoye
- MRC Centre for Reproductive Health, Queen's Medical Research Institute, University of Edinburgh, Edinburgh, UK
| | - Alan J Quigley
- Department of Radiology, Royal Hospital for Children and Young People, Edinburgh, UK
| | - Michael J Thrippleton
- Edinburgh Imaging, University of Edinburgh, Edinburgh, UK.,Centre for Clinical Brain Sciences, Chancellor's Building, University of Edinburgh, Edinburgh, UK
| | - John Norrie
- Usher Institute, Edinburgh Clinical Trials Unit, University of Edinburgh, Edinburgh, UK
| | - Mark E Bastin
- Centre for Clinical Brain Sciences, Chancellor's Building, University of Edinburgh, Edinburgh, UK
| | - James P Boardman
- MRC Centre for Reproductive Health, Queen's Medical Research Institute, University of Edinburgh, Edinburgh, UK.,Centre for Clinical Brain Sciences, Chancellor's Building, University of Edinburgh, Edinburgh, UK
| |
Collapse
|
29
|
Bastiaansen WAP, Klein S, Koning AHJ, Niessen WJ, Steegers-Theunissen RPM, Rousian M. Computational methods for the analysis of early-pregnancy brain ultrasonography: a systematic review. EBioMedicine 2023; 89:104466. [PMID: 36796233 PMCID: PMC9958260 DOI: 10.1016/j.ebiom.2023.104466] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/10/2022] [Revised: 01/09/2023] [Accepted: 01/23/2023] [Indexed: 02/16/2023] Open
Abstract
BACKGROUND Early screening of the brain is becoming routine clinical practice. Currently, this screening is performed by manual measurements and visual analysis, which is time-consuming and prone to errors. Computational methods may support this screening. Hence, the aim of this systematic review is to gain insight into future research directions needed to bring automated early-pregnancy ultrasound analysis of the human brain to clinical practice. METHODS We searched PubMed (Medline ALL Ovid), EMBASE, Web of Science Core Collection, Cochrane Central Register of Controlled Trials, and Google Scholar, from inception until June 2022. This study is registered in PROSPERO at CRD42020189888. Studies about computational methods for the analysis of human brain ultrasonography acquired before the 20th week of pregnancy were included. The key reported attributes were: level of automation, learning-based or not, the usage of clinical routine data depicting normal and abnormal brain development, public sharing of program source code and data, and analysis of the confounding factors. FINDINGS Our search identified 2575 studies, of which 55 were included. 76% used an automatic method, 62% a learning-based method, 45% used clinical routine data and in addition, for 13% the data depicted abnormal development. None of the studies shared publicly the program source code and only two studies shared the data. Finally, 35% did not analyse the influence of confounding factors. INTERPRETATION Our review showed an interest in automatic, learning-based methods. To bring these methods to clinical practice we recommend that studies: use routine clinical data depicting both normal and abnormal development, make their dataset and program source code publicly available, and be attentive to the influence of confounding factors. Introduction of automated computational methods for early-pregnancy brain ultrasonography will save valuable time during screening, and ultimately lead to better detection, treatment and prevention of neuro-developmental disorders. FUNDING The Erasmus MC Medical Research Advisor Committee (grant number: FB 379283).
Collapse
Affiliation(s)
- Wietske A P Bastiaansen
- Department of Obstetrics and Gynecology, Erasmus MC, University Medical Center, Rotterdam, the Netherlands; Biomedical Imaging Group Rotterdam, Department of Radiology and Nuclear Medicine, Erasmus MC, University Medical Center, Rotterdam, the Netherlands
| | - Stefan Klein
- Biomedical Imaging Group Rotterdam, Department of Radiology and Nuclear Medicine, Erasmus MC, University Medical Center, Rotterdam, the Netherlands
| | - Anton H J Koning
- Department of Pathology, Erasmus MC, University Medical Center, Rotterdam, the Netherlands
| | - Wiro J Niessen
- Biomedical Imaging Group Rotterdam, Department of Radiology and Nuclear Medicine, Erasmus MC, University Medical Center, Rotterdam, the Netherlands
| | | | - Melek Rousian
- Department of Obstetrics and Gynecology, Erasmus MC, University Medical Center, Rotterdam, the Netherlands.
| |
Collapse
|
30
|
Wang X, Yang Y, Wu T, Zhu H, Yu J, Tian J, Li H. Energy minimization segmentation model based on MRI images. Front Neurosci 2023; 17:1175451. [PMID: 37123357 PMCID: PMC10140403 DOI: 10.3389/fnins.2023.1175451] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/27/2023] [Accepted: 03/24/2023] [Indexed: 05/02/2023] Open
Abstract
Introduction Medical image segmentation is an important tool for doctors to accurately analyze the volume of brain tissue and lesions, which is important for the correct diagnosis of brain diseases. However, manual image segmentation methods are time-consuming, subjective and lack of repeatability, it needs to develop automatic and reliable methods for image segmentation. Methods Magnetic Resonance Imaging (MRI), a non-invasive imaging technique, is commonly used to detect, characterize and quantify tissues and lesions in the brain. Partial volume effect, gray scale in homogeneity, and lesions presents a great challenge for automatic medical image segmentation methods. So, the paper is dedicated to address the impact of partial volume effect and multiple sclerosis lesions on the segmentation accuracy in MRI. The objective function of the improved model and the post-processing method of lesion filling are researched based on the fuzzy clustering space and energy model. Results In particular, an energy-minimized segmentation algorithm is proposed. Through experimental verification, the AR-FCM algorithm can better overcome the problem of low segmentation accuracy of the RFCM algorithm for tissue boundary voxels and improve the segmentation accuracy of this algorithm. Meanwhile, a multi-channel input energy-minimization segmentation method with lesion filling and anatomical mapping is further proposed. Discussion The feasibility of the lesion filling strategy using post-processing can be confirmed and the segmentation accuracy is increased by comparison experiments.
Collapse
Affiliation(s)
- Xiuxin Wang
- Chongqing University of Posts and Telecommunications, Chongqing, China
- Institute for Advanced Sciences, Chongqing University of Posts and Telecommunications, Chongqing, China
- *Correspondence: Xiuxin Wang,
| | - Yuling Yang
- Chongqing University of Posts and Telecommunications, Chongqing, China
| | - Ting Wu
- Chongqing University of Posts and Telecommunications, Chongqing, China
| | - Hao Zhu
- Chongqing University of Posts and Telecommunications, Chongqing, China
| | - Jisheng Yu
- Chongqing University of Posts and Telecommunications, Chongqing, China
| | - Jian Tian
- Chongqing University of Posts and Telecommunications, Chongqing, China
| | - Hongzhong Li
- School of Civil Engineering, Guangdong Communication Polytechnic, Guangzhou, China
- Hongzhong Li,
| |
Collapse
|
31
|
RS-FetMRI: a MATLAB-SPM Based Tool for Pre-processing Fetal Resting-State fMRI Data. Neuroinformatics 2022; 20:1137-1154. [PMID: 35834105 DOI: 10.1007/s12021-022-09592-5] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Accepted: 05/25/2022] [Indexed: 12/31/2022]
Abstract
Resting-state functional magnetic resonance imaging (rs-fMRI) most recently has proved to open a measureless window on functional neurodevelopment in utero. Fetal brain activation and connectivity maps can be heavily influenced by 1) fetal-specific motion effects on the time-series and 2) the accuracy of time-series spatial normalization to a standardized gestational-week (GW) specific fetal template space.Due to the absence of a standardized and generalizable image processing protocol, the objective of the present work was to implement a validated fetal rs-fMRI preprocessing pipeline (RS-FetMRI) divided into 6 inter-dependent preprocessing modules (i.e., M1 to M6) and designed to work entirely as an extension for Statistical Parametric Mapping (SPM).RS-FetMRI pipeline output analyses on rs-fMRI time-series sampled from a cohort of fetuses acquired on both 1.5 T and 3 T MRI scanning systems showed increased efficacy of estimation of the degree of movement coupled with an efficient motion censoring procedure, resulting in increased number of motion-uncorrupted volumes and temporal continuity in fetal rs-fMRI time-series data. Moreover, a "structural-free" SPM-based spatial normalization procedure granted a high degree of spatial overlap with high reproducibility and a significant improvement in whole-brain and parcellation-specific Temporal Signal-to-Noise Ratio (TSNR) mirrored by functional connectivity analysis.To our knowledge, the RS-FetMRI pipeline is the first semi-automatic and easy-to-use standardized fetal rs-fMRI preprocessing pipeline completely integrated in MATLAB-SPM able to remove entry barriers for new research groups into the field of fetal rs-fMRI, for both research or clinical purposes, and ultimately to make future fetal brain connectivity investigations more suitable for comparison and cross-validation.
Collapse
|
32
|
Richter L, Fetit AE. Accurate segmentation of neonatal brain MRI with deep learning. Front Neuroinform 2022; 16:1006532. [PMID: 36246394 PMCID: PMC9554654 DOI: 10.3389/fninf.2022.1006532] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/29/2022] [Accepted: 09/06/2022] [Indexed: 11/13/2022] Open
Abstract
An important step toward delivering an accurate connectome of the human brain is robust segmentation of 3D Magnetic Resonance Imaging (MRI) scans, which is particularly challenging when carried out on perinatal data. In this paper, we present an automated, deep learning-based pipeline for accurate segmentation of tissues from neonatal brain MRI and extend it by introducing an age prediction pathway. A major constraint to using deep learning techniques on developing brain data is the need to collect large numbers of ground truth labels. We therefore also investigate two practical approaches that can help alleviate the problem of label scarcity without loss of segmentation performance. First, we examine the efficiency of different strategies of distributing a limited budget of annotated 2D slices over 3D training images. In the second approach, we compare the segmentation performance of pre-trained models with different strategies of fine-tuning on a small subset of preterm infants. Our results indicate that distributing labels over a larger number of brain scans can improve segmentation performance. We also show that even partial fine-tuning can be superior in performance to a model trained from scratch, highlighting the relevance of transfer learning strategies under conditions of label scarcity. We illustrate our findings on large, publicly available T1- and T2-weighted MRI scans (n = 709, range of ages at scan: 26–45 weeks) obtained retrospectively from the Developing Human Connectome Project (dHCP) cohort.
Collapse
Affiliation(s)
- Leonie Richter
- Department of Computing, Imperial College London, London, United Kingdom
- *Correspondence: Leonie Richter
| | - Ahmed E. Fetit
- Department of Computing, Imperial College London, London, United Kingdom
- UKRI CDT in Artificial Intelligence for Healthcare, Imperial College London, London, United Kingdom
| |
Collapse
|
33
|
Romberg J, Wilke M, Allgaier C, Nägele T, Engel C, Poets CF, Franz A. MRI-based brain volumes of preterm infants at term: a systematic review and meta-analysis. Arch Dis Child Fetal Neonatal Ed 2022; 107:520-526. [PMID: 35078779 PMCID: PMC9411894 DOI: 10.1136/archdischild-2021-322846] [Citation(s) in RCA: 11] [Impact Index Per Article: 5.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 07/22/2021] [Accepted: 11/30/2021] [Indexed: 12/17/2022]
Abstract
BACKGROUND MRI allows a detailed assessment of brain structures in preterm infants, outperforming cranial ultrasound. Neonatal MR-based brain volumes of preterm infants could serve as objective, quantitative and reproducible surrogate parameters of early brain development. To date, there are no reference values for preterm infants' brain volumes at term-equivalent age. OBJECTIVE Systematic review of the literature to determine reference ranges for MRI-based brain volumes of very preterm infants at term-equivalent age. METHODS PubMed Database was searched on 6 April 2020 for studies reporting MR-based brain volumes on representative unselected populations of very preterm and/or very low birthweight infants examined at term equivalent age (defined as 37-42 weeks mean postmenstrual age at MRI). Analyses were limited to volumetric parameters reported in >3 studies. Weighted mean volumes and SD were both calculated and simulated for each parameter. RESULTS An initial 367 publications were identified. Following application of exclusion criteria, 13 studies from eight countries were included for analysis, yielding four parameters. Weighted mean total brain volume was 379 mL (SD 72 mL; based on n=756). Cerebellar volume was 21 mL (6 mL; n=791), cortical grey matter volume 140 mL (47 mL; n=572) and weighted mean volume of unmyelinated white matter was 195 mL (38 mL; n=499). CONCLUSION This meta-analysis reports pooled data on several brain and cerebellar volumes which can serve as reference for future studies assessing MR-based volumetric parameters as a surrogate outcome for neurodevelopment and for the interpretation of individual or cohort MRI-based volumetric findings.
Collapse
Affiliation(s)
- Julia Romberg
- Department of Pediatrics, University Hospital Tuebingen, Tuebingen, Germany
| | - Marko Wilke
- Pediatric Neurology & Developmental Medicine, University Hospital Tuebingen, Tuebingen, Germany
| | - Christoph Allgaier
- Department of Pediatrics, Center for Pediatric Clinical Studies, University Hospital Tuebingen, Tuebingen, Germany
| | - Thomas Nägele
- Department of Neuroradiology, University Hospital Tuebingen, Tuebingen, Germany
| | - Corinna Engel
- Department of Pediatrics, Center for Pediatric Clinical Studies, University Hospital Tuebingen, Tuebingen, Germany
| | - Christian F Poets
- Department of Neonatology, University Hospital Tuebingen, Tuebingen, Germany
| | - Axel Franz
- Department of Neonatology, University Hospital Tuebingen, Tuebingen, Germany
| |
Collapse
|
34
|
Shoeibi A, Moridian P, Khodatars M, Ghassemi N, Jafari M, Alizadehsani R, Kong Y, Gorriz JM, Ramírez J, Khosravi A, Nahavandi S, Acharya UR. An overview of deep learning techniques for epileptic seizures detection and prediction based on neuroimaging modalities: Methods, challenges, and future works. Comput Biol Med 2022; 149:106053. [DOI: 10.1016/j.compbiomed.2022.106053] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/12/2022] [Revised: 08/17/2022] [Accepted: 08/17/2022] [Indexed: 02/01/2023]
|
35
|
Fidon L, Viola E, Mufti N, David AL, Melbourne A, Demaerel P, Ourselin S, Vercauteren T, Deprest J, Aertsen M. A spatio-temporal atlas of the developing fetal brain with spina bifida aperta. OPEN RESEARCH EUROPE 2022; 1:123. [PMID: 37645096 PMCID: PMC10445840 DOI: 10.12688/openreseurope.13914.2] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Accepted: 08/25/2022] [Indexed: 08/31/2023]
Abstract
Background: Spina bifida aperta (SBA) is a birth defect associated with severe anatomical changes in the developing fetal brain. Brain magnetic resonance imaging (MRI) atlases are popular tools for studying neuropathology in the brain anatomy, but previous fetal brain MRI atlases have focused on the normal fetal brain. We aimed to develop a spatio-temporal fetal brain MRI atlas for SBA. Methods: We developed a semi-automatic computational method to compute the first spatio-temporal fetal brain MRI atlas for SBA. We used 90 MRIs of fetuses with SBA with gestational ages ranging from 21 to 35 weeks. Isotropic and motion-free 3D reconstructed MRIs were obtained for all the examinations. We propose a protocol for the annotation of anatomical landmarks in brain 3D MRI of fetuses with SBA with the aim of making spatial alignment of abnormal fetal brain MRIs more robust. In addition, we propose a weighted generalized Procrustes method based on the anatomical landmarks for the initialization of the atlas. The proposed weighted generalized Procrustes can handle temporal regularization and missing annotations. After initialization, the atlas is refined iteratively using non-linear image registration based on the image intensity and the anatomical land-marks. A semi-automatic method is used to obtain a parcellation of our fetal brain atlas into eight tissue types: white matter, ventricular system, cerebellum, extra-axial cerebrospinal fluid, cortical gray matter, deep gray matter, brainstem, and corpus callosum. Results: An intra-rater variability analysis suggests that the seven anatomical land-marks are sufficiently reliable. We find that the proposed atlas outperforms a normal fetal brain atlas for the automatic segmentation of brain 3D MRI of fetuses with SBA. Conclusions: We make publicly available a spatio-temporal fetal brain MRI atlas for SBA, available here: https://doi.org/10.7303/syn25887675. This atlas can support future research on automatic segmentation methods for brain 3D MRI of fetuses with SBA.
Collapse
Affiliation(s)
- Lucas Fidon
- School of Biomedical Engineering & Imaging Sciences, King’s College London, London, SE1 7EU, UK
| | - Elizabeth Viola
- School of Biomedical Engineering & Imaging Sciences, King’s College London, London, SE1 7EU, UK
| | - Nada Mufti
- School of Biomedical Engineering & Imaging Sciences, King’s College London, London, SE1 7EU, UK
- Elizabeth Garrett Anderson Institute for Women’s Health, University College London, London, WC1E 6DB, UK
| | - Anna L. David
- Elizabeth Garrett Anderson Institute for Women’s Health, University College London, London, WC1E 6DB, UK
- Department of Obstetrics and Gynaecology, University Hospitals Leuven, 3000 Leuven, Belgium
| | - Andrew Melbourne
- School of Biomedical Engineering & Imaging Sciences, King’s College London, London, SE1 7EU, UK
| | - Philippe Demaerel
- Department of Radiology, University Hospitals Leuven, 3000 Leuven, Belgium
| | - Sébastien Ourselin
- School of Biomedical Engineering & Imaging Sciences, King’s College London, London, SE1 7EU, UK
| | - Tom Vercauteren
- School of Biomedical Engineering & Imaging Sciences, King’s College London, London, SE1 7EU, UK
| | - Jan Deprest
- Elizabeth Garrett Anderson Institute for Women’s Health, University College London, London, WC1E 6DB, UK
- Department of Obstetrics and Gynaecology, University Hospitals Leuven, 3000 Leuven, Belgium
- Department of Radiology, University Hospitals Leuven, 3000 Leuven, Belgium
| | - Michael Aertsen
- Department of Radiology, University Hospitals Leuven, 3000 Leuven, Belgium
| |
Collapse
|
36
|
Moser F, Huang R, Papież BW, Namburete AIL. BEAN: Brain Extraction and Alignment Network for 3D Fetal Neurosonography. Neuroimage 2022; 258:119341. [PMID: 35654376 DOI: 10.1016/j.neuroimage.2022.119341] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/27/2021] [Revised: 04/08/2022] [Accepted: 05/28/2022] [Indexed: 01/18/2023] Open
Abstract
Brain extraction (masking of extra-cerebral tissues) and alignment are fundamental first steps of most neuroimage analysis pipelines. The lack of automated solutions for 3D ultrasound (US) has therefore limited its potential as a neuroimaging modality for studying fetal brain development using routinely acquired scans. In this work, we propose a convolutional neural network (CNN) that accurately and consistently aligns and extracts the fetal brain from minimally pre-processed 3D US scans. Our multi-task CNN, Brain Extraction and Alignment Network (BEAN), consists of two independent branches: 1) a fully-convolutional encoder-decoder branch for brain extraction of unaligned scans, and 2) a two-step regression-based branch for similarity alignment of the brain to a common coordinate space. BEAN was tested on 356 fetal head 3D scans spanning the gestational range of 14 to 30 weeks, significantly outperforming all current alternatives for fetal brain extraction and alignment. BEAN achieved state-of-the-art performance for both tasks, with a mean Dice Similarity Coefficient (DSC) of 0.94 for the brain extraction masks, and a mean DSC of 0.93 for the alignment of the target brain masks. The presented experimental results show that brain structures such as the thalamus, choroid plexus, cavum septum pellucidum, and Sylvian fissure, are consistently aligned throughout the dataset and remain clearly visible when the scans are averaged together. The BEAN implementation and related code can be found under www.github.com/felipemoser/kelluwen.
Collapse
Affiliation(s)
- Felipe Moser
- Oxford Machine Learning in Neuroimaging laboratory, OMNI, Department of Computer Science, University of Oxford, Oxford, UK.
| | - Ruobing Huang
- Institute of Biomedical Engineering, Department of Engineering Science, University of Oxford, Oxford, UK
| | -
- Nuffield Department of Women's and Reproductive Health, John Radcliffe Hospital, University of Oxford, Oxford, UK
| | - Bartłomiej W Papież
- Institute of Biomedical Engineering, Department of Engineering Science, University of Oxford, Oxford, UK; Big Data Institute, Li Ka Shing Centre for Health Information and Discovery, University of Oxford, Oxford, UK
| | - Ana I L Namburete
- Oxford Machine Learning in Neuroimaging laboratory, OMNI, Department of Computer Science, University of Oxford, Oxford, UK; Wellcome Centre for Integrative Neuroimaging, FMRIB, Nuffield Department of Clinical Neurosciences, University of Oxford, United Kingdom
| |
Collapse
|
37
|
A Fetal Brain magnetic resonance Acquisition Numerical phantom (FaBiAN). Sci Rep 2022; 12:8682. [PMID: 35606398 PMCID: PMC9127105 DOI: 10.1038/s41598-022-10335-4] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/13/2021] [Accepted: 04/05/2022] [Indexed: 11/28/2022] Open
Abstract
Accurate characterization of in utero human brain maturation is critical as it involves complex and interconnected structural and functional processes that may influence health later in life. Magnetic resonance imaging is a powerful tool to investigate equivocal neurological patterns during fetal development. However, the number of acquisitions of satisfactory quality available in this cohort of sensitive subjects remains scarce, thus hindering the validation of advanced image processing techniques. Numerical phantoms can mitigate these limitations by providing a controlled environment with a known ground truth. In this work, we present FaBiAN, an open-source Fetal Brain magnetic resonance Acquisition Numerical phantom that simulates clinical T2-weighted fast spin echo sequences of the fetal brain. This unique tool is based on a general, flexible and realistic setup that includes stochastic fetal movements, thus providing images of the fetal brain throughout maturation comparable to clinical acquisitions. We demonstrate its value to evaluate the robustness and optimize the accuracy of an algorithm for super-resolution fetal brain magnetic resonance imaging from simulated motion-corrupted 2D low-resolution series compared to a synthetic high-resolution reference volume. We also show that the images generated can complement clinical datasets to support data-intensive deep learning methods for fetal brain tissue segmentation.
Collapse
|
38
|
Hesse LS, Aliasi M, Moser F, theINTERGROWTH-Twenty First Consortium, Haak MC, Xie W, Jenkinson M, Namburete AIL. Subcortical Segmentation of the Fetal Brain in 3D Ultrasound using Deep Learning. Neuroimage 2022; 254:119117. [PMID: 35331871 DOI: 10.1016/j.neuroimage.2022.119117] [Citation(s) in RCA: 7] [Impact Index Per Article: 3.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/04/2022] [Revised: 02/25/2022] [Accepted: 03/17/2022] [Indexed: 12/24/2022] Open
Abstract
The quantification of subcortical volume development from 3D fetal ultrasound can provide important diagnostic information during pregnancy monitoring. However, manual segmentation of subcortical structures in ultrasound volumes is time-consuming and challenging due to low soft tissue contrast, speckle and shadowing artifacts. For this reason, we developed a convolutional neural network (CNN) for the automated segmentation of the choroid plexus (CP), lateral posterior ventricle horns (LPVH), cavum septum pellucidum et vergae (CSPV), and cerebellum (CB) from 3D ultrasound. As ground-truth labels are scarce and expensive to obtain, we applied few-shot learning, in which only a small number of manual annotations (n = 9) are used to train a CNN. We compared training a CNN with only a few individually annotated volumes versus many weakly labelled volumes obtained from atlas-based segmentations. This showed that segmentation performance close to intra-observer variability can be obtained with only a handful of manual annotations. Finally, the trained models were applied to a large number (n = 278) of ultrasound image volumes of a diverse, healthy population, obtaining novel US-specific growth curves of the respective structures during the second trimester of gestation.
Collapse
Affiliation(s)
- Linde S Hesse
- Institute of Biomedical Engineering, Department of Engineering Science, University of Oxford, United Kingdom.
| | - Moska Aliasi
- Department of Obstetrics and Fetal Medicine, Leiden University Medical Center, The Netherlands
| | - Felipe Moser
- Institute of Biomedical Engineering, Department of Engineering Science, University of Oxford, United Kingdom
| | - theINTERGROWTH-Twenty First Consortium
- Institute of Biomedical Engineering, Department of Engineering Science, University of Oxford, United Kingdom; Department of Obstetrics and Fetal Medicine, Leiden University Medical Center, The Netherlands; Visual Geometry Group, Department of Engineering Science, University of Oxford, United Kingdom; Wellcome center for Integrative NeuroImaging, FMRIB, University of Oxford, United Kingdom; Australian Institute for Machine Learning (AIML), Australia; South Australian Health and Medical Research Institute (SAHMRI), Australia
| | - Monique C Haak
- Department of Obstetrics and Fetal Medicine, Leiden University Medical Center, The Netherlands
| | - Weidi Xie
- Visual Geometry Group, Department of Engineering Science, University of Oxford, United Kingdom
| | - Mark Jenkinson
- Wellcome center for Integrative NeuroImaging, FMRIB, University of Oxford, United Kingdom; Australian Institute for Machine Learning (AIML), Australia; South Australian Health and Medical Research Institute (SAHMRI), Australia
| | - Ana I L Namburete
- Institute of Biomedical Engineering, Department of Engineering Science, University of Oxford, United Kingdom
| |
Collapse
|
39
|
Torres HR, Morais P, Oliveira B, Birdir C, Rüdiger M, Fonseca JC, Vilaça JL. A review of image processing methods for fetal head and brain analysis in ultrasound images. COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE 2022; 215:106629. [PMID: 35065326 DOI: 10.1016/j.cmpb.2022.106629] [Citation(s) in RCA: 15] [Impact Index Per Article: 7.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 03/19/2021] [Revised: 12/20/2021] [Accepted: 01/08/2022] [Indexed: 06/14/2023]
Abstract
BACKGROUND AND OBJECTIVE Examination of head shape and brain during the fetal period is paramount to evaluate head growth, predict neurodevelopment, and to diagnose fetal abnormalities. Prenatal ultrasound is the most used imaging modality to perform this evaluation. However, manual interpretation of these images is challenging and thus, image processing methods to aid this task have been proposed in the literature. This article aims to present a review of these state-of-the-art methods. METHODS In this work, it is intended to analyze and categorize the different image processing methods to evaluate fetal head and brain in ultrasound imaging. For that, a total of 109 articles published since 2010 were analyzed. Different applications are covered in this review, namely analysis of head shape and inner structures of the brain, standard clinical planes identification, fetal development analysis, and methods for image processing enhancement. RESULTS For each application, the reviewed techniques are categorized according to their theoretical approach, and the more suitable image processing methods to accurately analyze the head and brain are identified. Furthermore, future research needs are discussed. Finally, topics whose research is lacking in the literature are outlined, along with new fields of applications. CONCLUSIONS A multitude of image processing methods has been proposed for fetal head and brain analysis. Summarily, techniques from different categories showed their potential to improve clinical practice. Nevertheless, further research must be conducted to potentiate the current methods, especially for 3D imaging analysis and acquisition and for abnormality detection.
Collapse
Affiliation(s)
- Helena R Torres
- Algoritmi Center, School of Engineering, University of Minho, Guimarães, Portugal; Life and Health Sciences Research Institute (ICVS), School of Medicine, University of Minho, Braga, Portugal; ICVS/3B's - PT Government Associate Laboratory, Braga/Guimarães, Portugal; 2Ai - School of Technology, IPCA, Barcelos, Portugal.
| | - Pedro Morais
- 2Ai - School of Technology, IPCA, Barcelos, Portugal
| | - Bruno Oliveira
- Algoritmi Center, School of Engineering, University of Minho, Guimarães, Portugal; Life and Health Sciences Research Institute (ICVS), School of Medicine, University of Minho, Braga, Portugal; ICVS/3B's - PT Government Associate Laboratory, Braga/Guimarães, Portugal; 2Ai - School of Technology, IPCA, Barcelos, Portugal
| | - Cahit Birdir
- Department of Gynecology and Obstetrics, University Hospital Carl Gustav Carus, TU Dresden, Germany; Saxony Center for Feto-Neonatal Health, TU Dresden, Germany
| | - Mario Rüdiger
- Department for Neonatology and Pediatric Intensive Care, University Hospital Carl Gustav Carus, TU Dresden, Germany
| | - Jaime C Fonseca
- Algoritmi Center, School of Engineering, University of Minho, Guimarães, Portugal
| | - João L Vilaça
- 2Ai - School of Technology, IPCA, Barcelos, Portugal
| |
Collapse
|
40
|
Zhao L, Asis-Cruz JD, Feng X, Wu Y, Kapse K, Largent A, Quistorff J, Lopez C, Wu D, Qing K, Meyer C, Limperopoulos C. Automated 3D Fetal Brain Segmentation Using an Optimized Deep Learning Approach. AJNR Am J Neuroradiol 2022; 43:448-454. [PMID: 35177547 PMCID: PMC8910820 DOI: 10.3174/ajnr.a7419] [Citation(s) in RCA: 10] [Impact Index Per Article: 5.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/25/2021] [Accepted: 12/06/2021] [Indexed: 01/01/2023]
Abstract
BACKGROUND AND PURPOSE MR imaging provides critical information about fetal brain growth and development. Currently, morphologic analysis primarily relies on manual segmentation, which is time-intensive and has limited repeatability. This work aimed to develop a deep learning-based automatic fetal brain segmentation method that provides improved accuracy and robustness compared with atlas-based methods. MATERIALS AND METHODS A total of 106 fetal MR imaging studies were acquired prospectively from fetuses between 23 and 39 weeks of gestation. We trained a deep learning model on the MR imaging scans of 65 healthy fetuses and compared its performance with a 4D atlas-based segmentation method using the Wilcoxon signed-rank test. The trained model was also evaluated on data from 41 fetuses diagnosed with congenital heart disease. RESULTS The proposed method showed high consistency with the manual segmentation, with an average Dice score of 0.897. It also demonstrated significantly improved performance (P < .001) based on the Dice score and 95% Hausdorff distance in all brain regions compared with the atlas-based method. The performance of the proposed method was consistent across gestational ages. The segmentations of the brains of fetuses with high-risk congenital heart disease were also highly consistent with the manual segmentation, though the Dice score was 7% lower than that of healthy fetuses. CONCLUSIONS The proposed deep learning method provides an efficient and reliable approach for fetal brain segmentation, which outperformed segmentation based on a 4D atlas and has been used in clinical and research settings.
Collapse
Affiliation(s)
- L Zhao
- From the Department of Diagnostic Imaging and Radiology (L.Z., J.D.A.-C., Y.W., K.K., A.L., J.Q., C. Lopez, C. Limperopoulos), Developing Brain Institute, Children's National, Washington, DC
- Department of Biomedical Engineering (L.Z., D.W.), Key Laboratory for Biomedical Engineering of Ministry of Education, College of Biomedical Engineering & Instrument Science, Zhejiang University, China
| | - J D Asis-Cruz
- From the Department of Diagnostic Imaging and Radiology (L.Z., J.D.A.-C., Y.W., K.K., A.L., J.Q., C. Lopez, C. Limperopoulos), Developing Brain Institute, Children's National, Washington, DC
| | - X Feng
- Department of Biomedical Engineering (X.F., C.M.), University of Virginia, Charlottesville, Virginia
| | - Y Wu
- From the Department of Diagnostic Imaging and Radiology (L.Z., J.D.A.-C., Y.W., K.K., A.L., J.Q., C. Lopez, C. Limperopoulos), Developing Brain Institute, Children's National, Washington, DC
| | - K Kapse
- From the Department of Diagnostic Imaging and Radiology (L.Z., J.D.A.-C., Y.W., K.K., A.L., J.Q., C. Lopez, C. Limperopoulos), Developing Brain Institute, Children's National, Washington, DC
| | - A Largent
- From the Department of Diagnostic Imaging and Radiology (L.Z., J.D.A.-C., Y.W., K.K., A.L., J.Q., C. Lopez, C. Limperopoulos), Developing Brain Institute, Children's National, Washington, DC
| | - J Quistorff
- From the Department of Diagnostic Imaging and Radiology (L.Z., J.D.A.-C., Y.W., K.K., A.L., J.Q., C. Lopez, C. Limperopoulos), Developing Brain Institute, Children's National, Washington, DC
| | - C Lopez
- From the Department of Diagnostic Imaging and Radiology (L.Z., J.D.A.-C., Y.W., K.K., A.L., J.Q., C. Lopez, C. Limperopoulos), Developing Brain Institute, Children's National, Washington, DC
| | - D Wu
- Department of Biomedical Engineering (L.Z., D.W.), Key Laboratory for Biomedical Engineering of Ministry of Education, College of Biomedical Engineering & Instrument Science, Zhejiang University, China
| | - K Qing
- Department of Radiation Oncology (K.Q.), City of Hope National Center, Duarte, California
| | - C Meyer
- Department of Biomedical Engineering (X.F., C.M.), University of Virginia, Charlottesville, Virginia
| | - C Limperopoulos
- From the Department of Diagnostic Imaging and Radiology (L.Z., J.D.A.-C., Y.W., K.K., A.L., J.Q., C. Lopez, C. Limperopoulos), Developing Brain Institute, Children's National, Washington, DC
| |
Collapse
|
41
|
Coupet M, Urruty T, Leelanupab T, Naudin M, Bourdon P, Maloigne CF, Guillevin R. A multi-sequences MRI deep framework study applied to glioma classfication. MULTIMEDIA TOOLS AND APPLICATIONS 2022; 81:13563-13591. [PMID: 35250358 PMCID: PMC8882719 DOI: 10.1007/s11042-022-12316-1] [Citation(s) in RCA: 6] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 07/31/2020] [Revised: 09/02/2021] [Accepted: 01/17/2022] [Indexed: 06/14/2023]
Abstract
Glioma is one of the most important central nervous system tumors, ranked 15th in the most common cancer for men and women. Magnetic Resonance Imaging (MRI) represents a common tool for medical experts to the diagnosis of glioma. A set of multi-sequences from an MRI is selected according to the severity of the pathology. Our proposed approach aims moreto create a computer-aided system that is capable of helping morethe expert diagnose the brain gliomas. moreWe propose a supervised learning regime based on a convolutional neural network based framework and transfer learning techniques. Our research morefocuses on the performance of different pre-trained deep learning models with respect to different MRI sequences. We highlight the best combinations of such model-MRI sequence couple for our specific task of classifying healthy brain against brain with glioma. moreWe also propose to visually analyze the extracted deep features for studying the existing relation of the MRI sequences and models. This interpretability analysis gives some hints for medical expert to understand the diagnosis made by the models. Our study is based on the well-known BraTS datasets including multi-sequence images and expert diagnosis.
Collapse
Affiliation(s)
- Matthieu Coupet
- XLIM Laboratory, University of Poitiers, UMR CNRS 7252, Poitiers, France
- I3M, Common Laboratory CNRS-Siemens, University and Hospital of Poitiers, Poitiers, France
| | - Thierry Urruty
- XLIM Laboratory, University of Poitiers, UMR CNRS 7252, Poitiers, France
- I3M, Common Laboratory CNRS-Siemens, University and Hospital of Poitiers, Poitiers, France
| | - Teerapong Leelanupab
- Faculty of Information Technology, King Mongkut’s Institute of Technology Ladkrabang (KMITL), Bangkok, 10520 Thailand
| | - Mathieu Naudin
- I3M, Common Laboratory CNRS-Siemens, University and Hospital of Poitiers, Poitiers, France
- Poitiers University Hospital, CHU, Poitiers, France
| | - Pascal Bourdon
- I3M, Common Laboratory CNRS-Siemens, University and Hospital of Poitiers, Poitiers, France
- Poitiers University Hospital, CHU, Poitiers, France
| | - Christine Fernandez Maloigne
- I3M, Common Laboratory CNRS-Siemens, University and Hospital of Poitiers, Poitiers, France
- Poitiers University Hospital, CHU, Poitiers, France
| | - Rémy Guillevin
- I3M, Common Laboratory CNRS-Siemens, University and Hospital of Poitiers, Poitiers, France
- Poitiers University Hospital, CHU, Poitiers, France
- DACTIM-MIS/LMA Laboratory University of Poitiers, UMR CNRS 7348, Poitiers, France
| |
Collapse
|
42
|
Akiyama S, Madan N, Graham G, Samura O, Kitano R, Yun HJ, Craig A, Nakamura T, Hozawa A, Grant E, Im K, Tarui T. Regional brain development in fetuses with Dandy-Walker malformation: A volumetric fetal brain magnetic resonance imaging study. PLoS One 2022; 17:e0263535. [PMID: 35202430 PMCID: PMC8870580 DOI: 10.1371/journal.pone.0263535] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/19/2021] [Accepted: 01/20/2022] [Indexed: 11/18/2022] Open
Abstract
Dandy-Walker malformation (DWM) is a common prenatally diagnosed cerebellar malformation, characterized by cystic dilatation of the fourth ventricle, upward rotation of the hypoplastic vermis, and posterior fossa enlargement with torcular elevation. DWM is associated with a broad spectrum of neurodevelopmental abnormalities such as cognitive, motor, and behavioral impairments, which cannot be explained solely by cerebellar malformations. Notably, the pathogenesis of these symptoms remains poorly understood. This study investigated whether fetal structural developmental abnormalities in DWM extended beyond the posterior fossa to the cerebrum even in fetuses without apparent cerebral anomalies. Post-acquisition volumetric fetal magnetic resonance imaging (MRI) analysis was performed in 12 fetuses with DWM and 14 control fetuses. Growth trajectories of the volumes of the cortical plate, subcortical parenchyma, cerebellar hemispheres, and vermis between 18 and 33 weeks of gestation were compared. The median (interquartile range) gestational ages at the time of MRI were 22.4 (19.4-24.0) and 23.9 (20.6-29.2) weeks in the DWM and control groups, respectively (p = 0.269). Eight of the 12 fetuses with DWM presented with associated cerebral anomalies, including hydrocephalus (n = 3), cerebral ventriculomegaly (n = 3), and complete (n = 2) and partial (n = 2) agenesis of the corpus callosum (ACC); 7 presented with extracerebral abnormalities. Chromosomal abnormalities were detected by microarray analysis in 4 of 11 fetuses with DWM, using amniocentesis. Volumetric analysis revealed that the cortical plate was significantly larger in fetuses with DWM than in controls (p = 0.040). Even without ACC, the subcortical parenchyma, whole cerebrum, cerebellar hemispheres, and whole brain were significantly larger in fetuses with DWM (n = 8) than in controls (p = 0.004, 0.025, 0.033, and 0.026, respectively). In conclusion, volumetric fetal MRI analysis demonstrated that the development of DWM extends throughout the brain during the fetal period, even without apparent cerebral anomalies.
Collapse
Affiliation(s)
- Shizuko Akiyama
- Mother Infant Research Institute, Tufts Medical Center, Boston, Massachusetts, United States of America
| | - Neel Madan
- Radiology, Tufts Medical Center, Boston, Massachusetts, United States of America
| | - George Graham
- Obstetrics and Gynecology, Tufts Medical Center, Boston, Massachusetts, United States of America
| | - Osamu Samura
- Obstetrics and Gynecology, Jikei University School of Medicine, Tokyo, Japan
| | - Rie Kitano
- Mother Infant Research Institute, Tufts Medical Center, Boston, Massachusetts, United States of America
| | - Hyuk Jin Yun
- Fetal-Neonatal Neuroimaging and Developmental Science Center, Boston Children’s Hospital, Boston, Massachusetts, United States of America
| | - Alexa Craig
- Pediatric Neurology, Maine Medical Center, Portland, Oregan, United States of America
| | - Tomohiro Nakamura
- Department of Preventive Medicine and Epidemiology, Tohoku Medical Megabank Organization, Tohoku University, Sendai, Japan
| | - Atsushi Hozawa
- Department of Preventive Medicine and Epidemiology, Tohoku Medical Megabank Organization, Tohoku University, Sendai, Japan
| | - Ellen Grant
- Fetal-Neonatal Neuroimaging and Developmental Science Center, Boston Children’s Hospital, Boston, Massachusetts, United States of America
| | - Kiho Im
- Fetal-Neonatal Neuroimaging and Developmental Science Center, Boston Children’s Hospital, Boston, Massachusetts, United States of America
| | - Tomo Tarui
- Mother Infant Research Institute, Tufts Medical Center, Boston, Massachusetts, United States of America
- Pediatric Neurology, Tufts Children’s Hospital, Boston, Massachusetts, United States of America
| |
Collapse
|
43
|
Coupeau P, Fasquel JB, Mazerand E, Menei P, Montero-Menei CN, Dinomais M. Patch-based 3D U-Net and transfer learning for longitudinal piglet brain segmentation on MRI. COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE 2022; 214:106563. [PMID: 34890993 DOI: 10.1016/j.cmpb.2021.106563] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 07/15/2021] [Revised: 11/26/2021] [Accepted: 11/26/2021] [Indexed: 06/13/2023]
Abstract
BACKGROUND AND OBJECTIVES In order to study neural plasticity in immature brain following early brain lesion, large animal model are needed. Because of its morphological similarities with the human developmental brain, piglet is a suitable but little used one. Its study from Magnetic Resonance Imaging (MRI) requires the development of automatic algorithms for the segmentation of the different structures and tissues. A crucial preliminary step consists in automatically segmenting the brain. METHODS We propose a fully automatic brain segmentation method applied to piglets by combining a 3D patch-based U-Net and a post-processing pipeline for spatial regularization and elimination of false positives. Our approach also integrates a transfer-learning strategy for managing an automated longitudinal monitoring evaluated for four developmental stages (2, 6, 10 and 18 weeks), facing the issue of MRI changes resulting from the rapid brain development. It is compared to a 2D approach and the Brain Extraction Tool (BET) as well as techniques adapted to other animals (rodents, macaques). The influence of training patches size and distribution is studied as well as the benefits of spatial regularization. RESULTS Results show that our approach is efficient in terms of average Dice score (0.952) and Hausdorff distance (8.51), outperforming the use of a 2D U-Net (Dice: 0.919, Hausdorff distance: 11.06) and BET (Dice: 0.764, Hausdorff distance: 25.91). The transfer-learning strategy achieves a good performance on older piglets (Dice of 0.934 at 6 weeks, 0.956 at 10 weeks and 0.958 at 18 weeks) compared to a standard training strategy with few data (Dice of 0.636 at 6 weeks, 0.907 at 10 weeks, not calculable at 18 weeks because of too few training piglets). CONCLUSIONS In conclusion, we provide a method for longitudinal MRI piglet brain segmentation based on 3D U-Net and transfer learning which can be used for future morphometric studies and applied to other animals.
Collapse
Affiliation(s)
- P Coupeau
- Université d'Angers, LARIS, SFR MATHSTIC, F-49000 Angers, France.
| | - J-B Fasquel
- Université d'Angers, LARIS, SFR MATHSTIC, F-49000 Angers, France
| | - E Mazerand
- CRCINA, UMR 1232, INSERM, Université de Nantes, Université d'Angers, F-49933 Angers, France; Département de neurochirurgie, Centre Hospitalier Universitaire d'Angers, France
| | - P Menei
- CRCINA, UMR 1232, INSERM, Université de Nantes, Université d'Angers, F-49933 Angers, France; Département de neurochirurgie, Centre Hospitalier Universitaire d'Angers, France
| | - C N Montero-Menei
- CRCINA, UMR 1232, INSERM, Université de Nantes, Université d'Angers, F-49933 Angers, France
| | - M Dinomais
- Université d'Angers, LARIS, SFR MATHSTIC, F-49000 Angers, France; Département de médecine physique et de réadaptation, Centre Hospitalier Universitaire d'Angers, France
| |
Collapse
|
44
|
Pseudo-Label-Assisted Self-Organizing Maps for Brain Tissue Segmentation in Magnetic Resonance Imaging. J Digit Imaging 2022; 35:180-192. [PMID: 35018537 PMCID: PMC8921351 DOI: 10.1007/s10278-021-00557-9] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/23/2021] [Revised: 11/26/2021] [Accepted: 11/27/2021] [Indexed: 10/19/2022] Open
Abstract
Brain tissue segmentation in magnetic resonance imaging volumes is an important image processing step for analyzing the human brain. This paper presents a novel approach named Pseudo-Label Assisted Self-Organizing Map (PLA-SOM) that enhances the result produced by a base segmentation method. Using the output of a base method, PLA-SOM calculates pseudo-labels in order to keep inter-class separation and intra-class compactness in the training phase. For the mapping phase, PLA-SOM uses a novel fuzzy function that combines feature space learned by the SOM's prototypes, topological ordering from the map, and spatial information from a brain atlas. We assessed PLA-SOM performance on synthetic and real MRIs of the brain, obtained from the BrainWeb and the Internet Brain Image Repository datasets. The experimental results showed evidence of segmentation improvement achieved by the proposed method over six different base methods. The best segmentation improvements reported by PLA-SOM on synthetic brain scans are 11%, 6%, and 4% for the tissue classes cerebrospinal fluid, gray matter, and white matter, respectively. On real brain scans, PLA-SOM achieved segmentation enhancements of 15%, 5%, and 12% for cerebrospinal fluid, gray matter, and white matter, respectively.
Collapse
|
45
|
Song J, Yuan L. Brain tissue segmentation via non-local fuzzy c-means clustering combined with Markov random field. MATHEMATICAL BIOSCIENCES AND ENGINEERING : MBE 2022; 19:1891-1908. [PMID: 35135234 DOI: 10.3934/mbe.2022089] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/14/2023]
Abstract
The segmentation and extraction of brain tissue in magnetic resonance imaging (MRI) is a meaningful task because it provides a diagnosis and treatment basis for observing brain tissue development, delineating lesions, and planning surgery. However, MRI images are often damaged by factors such as noise, low contrast and intensity brightness, which seriously affect the accuracy of segmentation. A non-local fuzzy c-means clustering framework incorporating the Markov random field for brain tissue segmentation is proposed in this paper. Firstly, according to the statistical characteristics that MRF can effectively describe the local spatial correlation of an image, a new distance metric with neighborhood constraints is constructed by combining probabilistic statistical information. Secondly, a non-local regularization term is integrated into the objective function to utilize the global structure feature of the image, so that both the local and global information of the image can be taken into account. In addition, a linear model of inhomogeneous intensity is also built to estimate the bias field in brain MRI, which has achieved the goal of overcoming the intensity inhomogeneity. The proposed model fully considers the randomness and fuzziness in the image segmentation problem, and obtains the prior knowledge of the image reasonably, which reduces the influence of low contrast in the MRI images. Then the experimental results demonstrate that the proposed method can eliminate the noise and intensity inhomogeneity of the MRI image and effectively improve the image segmentation accuracy.
Collapse
Affiliation(s)
- Jianhua Song
- The Key Laboratory of Intelligent Optimization and Information Processing, Minnan Normal University, Zhangzhou, 363000, China
- College of Physics and Information Engineering, Minnan Normal University, Zhangzhou, 363000, China
| | - Lei Yuan
- College of Physics and Information Engineering, Minnan Normal University, Zhangzhou, 363000, China
| |
Collapse
|
46
|
Abstract
While medical advancements have led to improved survival of extremely premature infants, children remain at risk for brain injury and neurodevelopmental impairment. Brain imaging can offer insight into an infant's acute and long-term outcome; however, counseling parents about the results and implications of brain imaging remains challenging. The purpose of this article is to review the current literature and describe the challenges associated with counseling families of premature infants on neuroimaging findings. We propose a framework to guide clinicians in counseling parents about brain imaging results, informed by best practices in other disciplines: (FIGURE): 1) Formulate a plan 2) Identify parental needs and values 3) Give information 4) Acknowledge Uncertainty 5) Recognize and Respond to emotions 6) Discuss Expectations and Establish follow-up.
Collapse
Affiliation(s)
- Sarah M Bernstein
- Department of Pediatrics, Duke University Medical Center, Durham, NC, United States
| | | | - Monica E Lemmon
- Departments of Pediatrics and Population Health Sciences, Duke University Medical Center, Duke-Margolis Center for Health Policy, DUMC 3936, Durham, NC 27710, United States.
| |
Collapse
|
47
|
Single-Input Multi-Output U-Net for Automated 2D Foetal Brain Segmentation of MR Images. J Imaging 2021; 7:jimaging7100200. [PMID: 34677286 PMCID: PMC8536962 DOI: 10.3390/jimaging7100200] [Citation(s) in RCA: 4] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/30/2021] [Revised: 09/14/2021] [Accepted: 09/26/2021] [Indexed: 11/16/2022] Open
Abstract
In this work, we develop the Single-Input Multi-Output U-Net (SIMOU-Net), a hybrid network for foetal brain segmentation inspired by the original U-Net fused with the holistically nested edge detection (HED) network. The SIMOU-Net is similar to the original U-Net but it has a deeper architecture and takes account of the features extracted from each side output. It acts similar to an ensemble neural network, however, instead of averaging the outputs from several independently trained models, which is computationally expensive, our approach combines outputs from a single network to reduce the variance of predications and generalization errors. Experimental results using 200 normal foetal brains consisting of over 11,500 2D images produced Dice and Jaccard coefficients of 94.2 ± 5.9% and 88.7 ± 6.9%, respectively. We further tested the proposed network on 54 abnormal cases (over 3500 images) and achieved Dice and Jaccard coefficients of 91.2 ± 6.8% and 85.7 ± 6.6%, respectively.
Collapse
|
48
|
Peterson MR, Cherukuri V, Paulson JN, Ssentongo P, Kulkarni AV, Warf BC, Monga V, Schiff SJ. Normal childhood brain growth and a universal sex and anthropomorphic relationship to cerebrospinal fluid. J Neurosurg Pediatr 2021; 28:458-468. [PMID: 34243147 PMCID: PMC8594737 DOI: 10.3171/2021.2.peds201006] [Citation(s) in RCA: 14] [Impact Index Per Article: 4.7] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 12/29/2020] [Accepted: 02/19/2021] [Indexed: 11/23/2022]
Abstract
OBJECTIVE The study of brain size and growth has a long and contentious history, yet normal brain volume development has yet to be fully described. In particular, the normal brain growth and cerebrospinal fluid (CSF) accumulation relationship is critical to characterize because it is impacted in numerous conditions of early childhood in which brain growth and fluid accumulation are affected, such as infection, hemorrhage, hydrocephalus, and a broad range of congenital disorders. The authors of this study aim to describe normal brain volume growth, particularly in the setting of CSF accumulation. METHODS The authors analyzed 1067 magnetic resonance imaging scans from 505 healthy pediatric subjects from birth to age 18 years to quantify component and regional brain volumes. The volume trajectories were compared between the sexes and hemispheres using smoothing spline ANOVA. Population growth curves were developed using generalized additive models for location, scale, and shape. RESULTS Brain volume peaked at 10-12 years of age. Males exhibited larger age-adjusted total brain volumes than females, and body size normalization procedures did not eliminate this difference. The ratio of brain to CSF volume, however, revealed a universal age-dependent relationship independent of sex or body size. CONCLUSIONS These findings enable the application of normative growth curves in managing a broad range of childhood diseases in which cognitive development, brain growth, and fluid accumulation are interrelated.
Collapse
Affiliation(s)
- Mallory R. Peterson
- Center for Neural Engineering, The Pennsylvania State University, University Park
- Department of Engineering Science and Mechanics, The Pennsylvania State University, University Park
- The Pennsylvania State University College of Medicine, Hershey, Pennsylvania
| | - Venkateswararao Cherukuri
- Center for Neural Engineering, The Pennsylvania State University, University Park
- School of Electrical Engineering and Computer Science, The Pennsylvania State University, University Park
| | - Joseph N. Paulson
- Department of Biostatistics, Product Development, Genentech Inc., South San Francisco, California
| | - Paddy Ssentongo
- Department of Engineering Science and Mechanics, The Pennsylvania State University, University Park
| | - Abhaya V. Kulkarni
- Department of Neurosurgery, University of Toronto
- Department of Neurosurgery, The Hospital for Sick Children, Toronto, Ontario, Canada
| | - Benjamin C. Warf
- Department of Neurosurgery, Harvard Medical School
- Department of Neurosurgery, Boston Children’s Hospital, Boston, Massachusetts
| | - Vishal Monga
- School of Electrical Engineering and Computer Science, The Pennsylvania State University, University Park
| | - Steven J. Schiff
- Center for Neural Engineering, The Pennsylvania State University, University Park
- Department of Engineering Science and Mechanics, The Pennsylvania State University, University Park
- Department of Neurosurgery, The Pennsylvania State University, University Park
- Department of Physics, The Pennsylvania State University, University Park
| |
Collapse
|
49
|
Li B, You X, Wang J, Peng Q, Yin S, Qi R, Ren Q, Hong Z. IAS-NET: Joint intraclassly adaptive GAN and segmentation network for unsupervised cross-domain in neonatal brain MRI segmentation. Med Phys 2021; 48:6962-6975. [PMID: 34494276 DOI: 10.1002/mp.15212] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/04/2021] [Revised: 07/15/2021] [Accepted: 08/15/2021] [Indexed: 11/11/2022] Open
Abstract
PURPOSE In neonatal brain magnetic resonance image (MRI) segmentation, the model we trained on the training set (source domain) often performs poorly in clinical practice (target domain). As the label of target-domain images is unavailable, this cross-domain segmentation needs unsupervised domain adaptation (UDA) to make the model adapt to the target domain. However, the shape and intensity distribution of neonatal brain MRI images across the domains are largely different from adults'. Current UDA methods aim to make synthesized images similar to the target domain as a whole. But it is impossible to synthesize images with intraclass similarity because of the regional misalignment caused by the cross-domain difference. This will result in generating intraclassly incorrect intensity information from target-domain images. To address this issue, we propose an IAS-NET (joint intraclassly adaptive generative adversarial network (GAN) (IA-NET) and segmentation) framework to bridge the gap between the two domains for intraclass alignment. METHODS Our proposed IAS-NET is an elegant learning framework that transfers the appearance of images across the domains from both image and feature perspectives. It consists of the proposed IA-NET and a segmentation network (S-NET). The proposed IA-NET is a GAN-based adaptive network that contains one generator (including two encoders and one shared decoder) and four discriminators for cross-domain transfer. The two encoders are implemented to extract original image, mean, and variance features from source and target domains. The proposed local adaptive instance normalization algorithm is used to perform intraclass feature alignment to the target domain in the feature-map level. S-NET is a U-net structure network that is used to provide semantic constraint by a segmentation loss for the training of IA-NET. Meanwhile, it offers pseudo-label images for calculating intraclass features of the target domain. Source code (in Tensorflow) is available at https://github.com/lb-whu/RAS-NET/. RESULTS Extensive experiments are carried out on two different data sets (NeoBrainS12 and dHCP), respectively. There exist great differences in the shape, size, and intensity distribution of magnetic resonance (MR) images in the two databases. Compared to baseline, we improve the average dice score of all tissues on NeoBrains12 by 6% through adaptive training with unlabeled dHCP images. Besides, we also conduct experiments on dHCP and improved the average dice score by 4%. The quantitative analysis of the mean and variance of the synthesized images shows that the synthesized image by the proposed is closer to the target domain both in the full brain or within each class than that of the compared methods. CONCLUSIONS In this paper, the proposed IAS-NET can improve the performance of the S-NET effectively by its intraclass feature alignment in the target domain. Compared to the current UDA methods, the synthesized images by IAS-NET are more intraclassly similar to the target domain for neonatal brain MR images. Therefore, it achieves state-of-the-art results in the compared UDA models for the segmentation task.
Collapse
Affiliation(s)
- Bo Li
- School of Electronic Information and Communication, Huazhong University of Science and Technology, Wuhan, China
| | - Xinge You
- School of Electronic Information and Communication, Huazhong University of Science and Technology, Wuhan, China.,Shenzhen Research Institute, Huazhong University of Science and Technology, Shenzhen, China
| | - Jing Wang
- Department of Radiology, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China
| | - Qinmu Peng
- School of Electronic Information and Communication, Huazhong University of Science and Technology, Wuhan, China.,Shenzhen Research Institute, Huazhong University of Science and Technology, Shenzhen, China
| | - Shi Yin
- School of Electronic Information and Communication, Huazhong University of Science and Technology, Wuhan, China
| | - Ruinan Qi
- Department of Radiology, Huazhong University of Science and Technology Hospital, Wuhan, China
| | - Qianqian Ren
- Department of Radiology, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China.,Hubei Province Key Laboratory of Molecular Imaging, Wuhan, China
| | - Ziming Hong
- School of Electronic Information and Communication, Huazhong University of Science and Technology, Wuhan, China
| |
Collapse
|
50
|
Basnet R, Ahmad MO, Swamy M. A deep dense residual network with reduced parameters for volumetric brain tissue segmentation from MR images. Biomed Signal Process Control 2021. [DOI: 10.1016/j.bspc.2021.103063] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/29/2022]
|