Multi-Atlas Segmentation with Joint Label Fusion

Neuroanatomical changes observed over the course of a human pregnancy

Pregnancy is a period of profound hormonal and physiological changes experienced by millions of women annually, yet the neural changes unfolding in the maternal brain throughout gestation are not well studied in humans. Leveraging precision imaging, we mapped neuroanatomical changes in an individual from preconception through 2 years postpartum. Pronounced decreases in gray matter volume and cortical thickness were evident across the brain, standing in contrast to increases in white matter microstructural integrity, ventricle volume and cerebrospinal fluid, with few regions untouched by the transition to motherhood. This dataset serves as a comprehensive map of the human brain across gestation, providing an open-access resource for the brain imaging community to further explore and understand the maternal brain.

A roadmap towards standardized neuroimaging approaches for human thalamic nuclei

The thalamus has a key role in mediating cortical-subcortical interactions but is often neglected in neuroimaging studies, which mostly focus on changes in cortical structure and activity. One of the main reasons for the thalamus being overlooked is that the delineation of individual thalamic nuclei via neuroimaging remains controversial. Indeed, neuroimaging atlases vary substantially regarding which thalamic nuclei are included and how their delineations were established. Here, we review current and emerging methods for thalamic nuclei segmentation in neuroimaging data and consider the limitations of existing techniques in terms of their research and clinical applicability. We address these challenges by proposing a roadmap to improve thalamic nuclei segmentation in human neuroimaging and, in turn, harmonize research approaches and advance clinical applications. We believe that a collective effort is required to achieve this. We hope that this will ultimately lead to the thalamic nuclei being regarded as key brain regions in their own right and not (as often currently assumed) as simply a gateway between cortical and subcortical regions.

BIBSNet: A Deep Learning Baby Image Brain Segmentation Network for MRI Scans

Objectives

Brain segmentation of infant magnetic resonance (MR) images is vitally important in studying developmental mental health and disease. The infant brain undergoes many changes throughout the first years of postnatal life, making tissue segmentation difficult for most existing algorithms. Here, we introduce a deep neural network BIBSNet (Baby and Infant Brain Segmentation Neural Network), an open-source, community-driven model that relies on data augmentation and a large sample size of manually annotated images to facilitate the production of robust and generalizable brain segmentations.

Experimental Design

Included in model training and testing were MR brain images on 84 participants with an age range of 0-8 months (median postmenstrual ages of 13.57 months). Using manually annotated real and synthetic segmentation images, the model was trained using a 10-fold cross-validation procedure. Testing occurred on MRI data processed with the DCAN labs infant-ABCD-BIDS processing pipeline using segmentations produced from gold standard manual annotation, joint-label fusion (JLF), and BIBSNet to assess model performance.

Principal Observations

Using group analyses, results suggest that cortical metrics produced using BIBSNet segmentations outperforms JLF segmentations. Additionally, when analyzing individual differences, BIBSNet segmentations perform even better.

Conclusions

BIBSNet segmentation shows marked improvement over JLF segmentations across all age groups analyzed. The BIBSNet model is 600x faster compared to JLF and can be easily included in other processing pipelines.

Lifespan differences in hippocampal subregion connectivity patterns during movie watching

Age-related episodic memory decline is attributed to functional alternations in the hippocampus. Less clear is how aging affects the functional connections of the hippocampus to the rest of the brain during episodic memory processing. We examined fMRI data from the CamCAN dataset, in which a large cohort of participants watched a movie (N = 643; 18-88 years), a proxy for naturalistic episodic memory encoding. We examined connectivity profiles across the lifespan both within the hippocampus (anterior, posterior), and between the hippocampal subregions and cortical networks. Aging was associated with reductions in contralateral (left, right) but not ipsilateral (anterior, posterior) hippocampal subregion connectivity. Aging was primarily associated with increased coupling between the anterior hippocampus and regions affiliated with Control, Dorsal Attention and Default Mode networks, yet decreased coupling between the posterior hippocampus and a selection of these regions. Differences in age-related hippocampal-cortical, but not within-hippocampus circuitry selectively predicted worse memory performance. Our findings comprehensively characterize hippocampal functional topography in relation to cognition in older age, suggesting that shifts in cortico-hippocampal connectivity may be sensitive markers of age-related episodic memory decline.

Deep network and multi-atlas segmentation fusion for delineation of thigh muscle groups in three-dimensional water-fat separated MRI

Purpose

Segmentation is essential for tissue quantification and characterization in studies of aging and age-related and metabolic diseases and the development of imaging biomarkers. We propose a multi-method and multi-atlas methodology for automated segmentation of functional muscle groups in three-dimensional (3D) thigh magnetic resonance images. These groups lie anatomically adjacent to each other, rendering their manual delineation a challenging and time-consuming task.

Approach

We introduce a framework for automated segmentation of the four main functional muscle groups of the thigh, gracilis, hamstring, quadriceps femoris, and sartorius, using chemical shift encoded water-fat magnetic resonance imaging (CSE-MRI). We propose fusing anatomical mappings from multiple deformable models with 3D deep learning model-based segmentation. This approach leverages the generalizability of multi-atlas segmentation (MAS) and accuracy of deep networks, hence enabling accurate assessment of volume and fat content of muscle groups.

Results

For segmentation performance evaluation, we calculated the Dice similarity coefficient (DSC) and Hausdorff distance 95th percentile (HD-95). We evaluated the proposed framework, its variants, and baseline methods on 15 healthy subjects by threefold cross-validation and tested on four patients. Fusion of multiple atlases, deformable registration models, and deep learning segmentation produced the top performance with an average DSC of 0.859 and HD-95 of 8.34 over all muscles.

Conclusions

Fusion of multiple anatomical mappings from multiple MAS techniques enriches the template set and improves the segmentation accuracy. Additional fusion with deep network decisions applied to the subject space offers complementary information. The proposed approach can produce accurate segmentation of individual muscle groups in 3D thigh MRI scans.

Menstrual cycle-driven hormone concentrations co-fluctuate with white and gray matter architecture changes across the whole brain

Cyclic fluctuations in hypothalamic-pituitary-gonadal axis (HPG-axis) hormones exert powerful behavioral, structural, and functional effects through actions on the mammalian central nervous system. Yet, very little is known about how these fluctuations alter the structural nodes and information highways of the human brain. In a study of 30 naturally cycling women, we employed multidimensional diffusion and T1-weighted imaging during three estimated menstrual cycle phases (menses, ovulation, and mid-luteal) to investigate whether HPG-axis hormone concentrations co-fluctuate with alterations in white matter (WM) microstructure, cortical thickness (CT), and brain volume. Across the whole brain, 17β-estradiol and luteinizing hormone (LH) concentrations were directly proportional to diffusion anisotropy (μFA; 17β-estradiol: β1 = 0.145, highest density interval (HDI) = [0.211, 0.4]; LH: β1 = 0.111, HDI = [0.157, 0.364]), while follicle-stimulating hormone (FSH) was directly proportional to CT (β1 = 0 .162, HDI = [0.115, 0.678]). Within several individual regions, FSH and progesterone demonstrated opposing relationships with mean diffusivity (Diso) and CT. These regions mainly reside within the temporal and occipital lobes, with functional implications for the limbic and visual systems. Finally, progesterone was associated with increased tissue (β1 = 0.66, HDI = [0.607, 15.845]) and decreased cerebrospinal fluid (CSF; β1 = -0.749, HDI = [-11.604, -0.903]) volumes, with total brain volume remaining unchanged. These results are the first to report simultaneous brain-wide changes in human WM microstructure and CT coinciding with menstrual cycle-driven hormone rhythms. Effects were observed in both classically known HPG-axis receptor-dense regions (medial temporal lobe, prefrontal cortex) and in other regions located across frontal, occipital, temporal, and parietal lobes. Our results suggest that HPG-axis hormone fluctuations may have significant structural impacts across the entire brain.

Connectivity of the Piriform Cortex and its Implications in Temporal Lobe Epilepsy

Background

The piriform cortex has been implicated in the initiation, spread and termination of epileptic seizures. This understanding has extended to surgical management of epilepsy, where it has been shown that resection or ablation of the piriform cortex can result in better outcomes. How and why the piriform cortex may play such a crucial role in seizure networks is not well understood. To answer these questions, we investigated the functional and structural connectivity of the piriform cortex in both healthy controls and temporal lobe epilepsy (TLE) patients.

Methods

We studied a retrospective cohort of 55 drug-resistant unilateral TLE patients and 26 healthy controls who received structural and functional neuroimaging. Using seed-to-voxel connectivity we compared the normative whole-brain connectivity of the piriform to that of the hippocampus, a region commonly involved in epilepsy, to understand the differential contribution of the piriform to the epileptogenic network. We subsequently measured the inter-piriform coupling (IPC) to quantify similarities in the inter-hemispheric cortical functional connectivity profile between the two piriform cortices. We related differences in IPC in TLE back to aberrations in normative piriform connectivity, whole brain functional properties, and structural connectivity.

Results

We find that relative to the hippocampus, the piriform is functionally connected to the anterior insula and the rest of the salience ventral attention network (SAN). We also find that low IPC is a sensitive metric of poor surgical outcome (sensitivity: 85.71%, 95% CI: [19.12%, 99.64%]); and differences in IPC within TLE were related to disconnectivity and hyperconnectivity to the anterior insula and the SAN. More globally, we find that low IPC is associated with whole-brain functional and structural segregation, marked by decreased functional small-worldness and fractional anisotropy.

Conclusions

Our study presents novel insights into the functional and structural neural network alterations associated with this structure, laying the foundation for future work to carefully consider its connectivity during the presurgical management of epilepsy.

Medial temporal lobe gray matter microstructure in preclinical Alzheimer's disease

INTRODUCTION

Typical MRI measures of neurodegeneration have limited sensitivity in early disease stages. Diffusion MRI (dMRI) microstructural measures may allow for detection in preclinical stages.

METHODS

Participants had dMRI and either beta-amyloid PET or plasma biomarkers of Alzheimer's pathology within 18 months of MRI. Microstructure was measured in portions of the medial temporal lobe (MTL) with high neurofibrillary tangle (NFT) burden based on a previously developed post mortem 3D-map. Regressions examined relationships between microstructure and markers of Alzheimer's pathology in preclinical disease and then across disease stages.

RESULTS

There was higher isometric volume fraction in amyloid-positive compared to amyloid-negative cognitively unimpaired individuals in high tangle MTL regions. Similarly, plasma biomarkers and 18F-flortaucipir were associated with microstructural changes in preclinical disease. Additional microstructural effects were seen across disease stages.

DISCUSSION

Combining a post mortem atlas of NFT pathology with microstructural measures allows for detection of neurodegeneration in preclinical Alzheimer's disease. Highlights Typical markers of neurodegeneration are not sensitive in preclinical Alzheimer's. dMRI measured microstructure in regions with high NFT. Microstructural changes occur in medial temporal regions in preclinical disease. Microstructural changes occur in other typical Alzheimer's regions in later stages. Combining post mortem pathology atlases with in vivo MRI is a powerful framework.

Automated protocols for delineating human hippocampal subfields from 3 Tesla and 7 Tesla magnetic resonance imaging data

Researchers who study the human hippocampus are naturally interested in how its subfields function. However, many researchers are precluded from examining subfields because their manual delineation from magnetic resonance imaging (MRI) scans (still the gold standard approach) is time consuming and requires significant expertise. To help ameliorate this issue, we present here two protocols, one for 3T MRI and the other for 7T MRI, that permit automated hippocampus segmentation into six subregions, namely dentate gyrus/cornu ammonis (CA)4, CA2/3, CA1, subiculum, pre/parasubiculum, and uncus along the entire length of the hippocampus. These protocols are particularly notable relative to existing resources in that they were trained and tested using large numbers of healthy young adults (n = 140 at 3T, n = 40 at 7T) whose hippocampi were manually segmented by experts from MRI scans. Using inter-rater reliability analyses, we showed that the quality of automated segmentations produced by these protocols was high and comparable to expert manual segmenters. We provide full open access to the automated protocols, and anticipate they will save hippocampus researchers a significant amount of time. They could also help to catalyze subfield research, which is essential for gaining a full understanding of how the hippocampus functions.

Artificial Intelligence in Urologic Robotic Oncologic Surgery: A Narrative Review

With the rapid increase in computer processing capacity over the past two decades, machine learning techniques have been applied in many sectors of daily life. Machine learning in therapeutic settings is also gaining popularity. We analysed current studies on machine learning in robotic urologic surgery. We searched PubMed/Medline and Google Scholar up to December 2023. Search terms included "urologic surgery", "artificial intelligence", "machine learning", "neural network", "automation", and "robotic surgery". Automatic preoperative imaging, intraoperative anatomy matching, and bleeding prediction has been a major focus. Early artificial intelligence (AI) therapeutic outcomes are promising. Robot-assisted surgery provides precise telemetry data and a cutting-edge viewing console to analyse and improve AI integration in surgery. Machine learning enhances surgical skill feedback, procedure effectiveness, surgical guidance, and postoperative prediction. Tension-sensors on robotic arms and augmented reality can improve surgery. This provides real-time organ motion monitoring, improving precision and accuracy. As datasets develop and electronic health records are used more and more, these technologies will become more effective and useful. AI in robotic surgery is intended to improve surgical training and experience. Both seek precision to improve surgical care. AI in ''master-slave'' robotic surgery offers the detailed, step-by-step examination of autonomous robotic treatments.

Volumetric segmentation in the context of posterior fossa-related pathologies: a systematic review

BACKGROUND

Segmentation tools continue to advance, evolving from manual contouring to deep learning. Researchers have utilized segmentation to study a myriad of posterior fossa-related conditions, such as Chiari malformation, trigeminal neuralgia, post-operative pediatric cerebellar mutism syndrome, and Crouzon syndrome. Herein, we present a summary of the current literature on segmentation of the posterior fossa. The review highlights the various segmentation techniques, and their respective strengths and weaknesses, employed along with objectives and outcomes of the various studies reported in the literature.

METHODS

A literature search was conducted in PubMed, Embase, Cochrane, and Web of Science up to November 2023 for articles on segmentation techniques of posterior fossa. The two senior authors searched through databases based on the keywords of the article separately and then enrolled joint articles that met the inclusion and exclusion criteria.

RESULTS

The initial search identified 2205 articles. After applying inclusion and exclusion criteria, 77 articles were selected for full-text review after screening of titles/abstracts. 52 articles were ultimately included in the review. Segmentation techniques included manual, semi-automated, and fully automated (atlas-based, convolutional neural networks). The most common pathology investigated was Chiari malformation.

CONCLUSIONS

Various forms of segmentation techniques have been used to assess posterior fossa volumes/pathologies and each has its advantages and disadvantages. We discuss these nuances and summarize the current state of literature in the context of posterior fossa-associated pathologies.

ANTsX neuroimaging-derived structural phenotypes of UK Biobank

UK Biobank is a large-scale epidemiological resource for investigating prospective correlations between various lifestyle, environmental, and genetic factors with health and disease progression. In addition to individual subject information obtained through surveys and physical examinations, a comprehensive neuroimaging battery consisting of multiple modalities provides imaging-derived phenotypes (IDPs) that can serve as biomarkers in neuroscience research. In this study, we augment the existing set of UK Biobank neuroimaging structural IDPs, obtained from well-established software libraries such as FSL and FreeSurfer, with related measurements acquired through the Advanced Normalization Tools Ecosystem. This includes previously established cortical and subcortical measurements defined, in part, based on the Desikan-Killiany-Tourville atlas. Also included are morphological measurements from two recent developments: medial temporal lobe parcellation of hippocampal and extra-hippocampal regions in addition to cerebellum parcellation and thickness based on the Schmahmann anatomical labeling. Through predictive modeling, we assess the clinical utility of these IDP measurements, individually and in combination, using commonly studied phenotypic correlates including age, fluid intelligence, numeric memory, and several other sociodemographic variables. The predictive accuracy of these IDP-based models, in terms of root-mean-squared-error or area-under-the-curve for continuous and categorical variables, respectively, provides comparative insights between software libraries as well as potential clinical interpretability. Results demonstrate varied performance between package-based IDP sets and their combination, emphasizing the need for careful consideration in their selection and utilization.

A New Multi-Atlas Based Deep Learning Segmentation Framework With Differentiable Atlas Feature Warping

Deep learning based multi-atlas segmentation (DL-MA) has achieved the state-of-the-art performance in many medical image segmentation tasks, e.g., brain parcellation. In DL-MA methods, atlas-target correspondence is the key for accurate segmentation. In most existing DL-MA methods, such correspondence is usually established using traditional or deep learning based registration methods at image level with no further feature level adaption. This could cause possible atlas-target feature inconsistency. As a result, the information from atlases often has limited positive and even counteractive impact on the final segmentation results. To tackle this issue, in this paper, we propose a new DL-MA framework, where a novel differentiable atlas feature warping module with a new smooth regularization term is presented to establish feature level atlas-target correspondence. Comparing with the existing DL-MA methods, in our framework, atlas features containing anatomical prior knowledge are more relevant to the target image feature, leading the final segmentation results to a high accuracy level. We evaluate our framework in the context of brain parcellation using two public MR brain image datasets: LPBA40 and NIREP-NA0. The experimental results demonstrate that our framework outperforms both traditional multi-atlas segmentation (MAS) and state-of-the-art DL-MA methods with statistical significance. Further ablation studies confirm the effectiveness of the proposed differentiable atlas feature warping module.

Learning Better Registration to Learn Better Few-Shot Medical Image Segmentation: Authenticity, Diversity, and Robustness

In this work, we address the task of few-shot medical image segmentation (MIS) with a novel proposed framework based on the learning registration to learn segmentation (LRLS) paradigm. To cope with the limitations of lack of authenticity, diversity, and robustness in the existing LRLS frameworks, we propose the better registration better segmentation (BRBS) framework with three main contributions that are experimentally shown to have substantial practical merit. First, we improve the authenticity in the registration-based generation program and propose the knowledge consistency constraint strategy that constrains the registration network to learn according to the domain knowledge. It brings the semantic-aligned and topology-preserved registration, thus allowing the generation program to output new data with great space and style authenticity. Second, we deeply studied the diversity of the generation process and propose the space-style sampling program, which introduces the modeling of the transformation path of style and space change between few atlases and numerous unlabeled images into the generation program. Therefore, the sampling on the transformation paths provides much more diverse space and style features to the generated data effectively improving the diversity. Third, we first highlight the robustness in the learning of segmentation in the LRLS paradigm and propose the mix misalignment regularization, which simulates the misalignment distortion and constrains the network to reduce the fitting degree of misaligned regions. Therefore, it builds regularization for these regions improving the robustness of segmentation learning. Without any bells and whistles, our approach achieves a new state-of-the-art performance in few-shot MIS on two challenging tasks that outperform the existing LRLS-based few-shot methods. We believe that this novel and effective framework will provide a powerful few-shot benchmark for the field of medical image and efficiently reduce the costs of medical image research. All of our code will be made publicly available online.

Brain multi-contrast, multi-atlas segmentation of diffusion tensor imaging and ensemble learning automatically diagnose late-life depression

We investigated the potential of machine learning for diagnostic classification in late-life major depression based on an advanced whole brain white matter segmentation framework. Twenty-six late-life depression and 12 never depressed individuals aged > 55 years, matched for age, MMSE, and education underwent brain diffusion tensor imaging and a multi-contrast, multi-atlas segmentation in MRIcloud. Fractional anisotropy volume, mean fractional anisotropy, trace, axial and radial diffusivity (RD) extracted from 146 white matter parcels for each subject were used to train and test the AdaBoost classifier using stratified 12-fold cross validation. Performance was evaluated using various measures. The statistical power of the classifier was assessed using label permutation test. Statistical analysis did not yield significant differences in DTI measures between the groups. The classifier achieved a balanced accuracy of 71% and an Area Under the Receiver Operator Characteristic Curve (ROC-AUC) of 0.81 by trace, and a balanced accuracy of 70% and a ROC-AUC of 0.80 by RD, in limbic, cortico-basal ganglia-thalamo-cortical loop, brainstem, external and internal capsules, callosal and cerebellar structures. Both indices shared important structures for classification, while fornix was the most important structure for classification by both indices. The classifier proved statistically significant, as trace and RD ROC-AUC scores after permutation were lower than those obtained with the actual data (P = 0.022 and P = 0.024, respectively). Similar results were obtained with the Gradient Boosting classifier, whereas the RBF-kernel Support Vector Machine with k-best feature selection did not exceed the chance level. Finally, AdaBoost significantly predicted the class using all features together. Limitations are discussed. The results encourage further investigation of the implemented methods for computer aided diagnostics and anatomically informed therapeutics.

Tau PET burden in Brodmann areas 35 and 36 is associated with individual differences in cognition in non-demented older adults

Introduction

The accumulation of neurofibrillary tau tangles, a neuropathological hallmark of Alzheimer's disease (AD), occurs in medial temporal lobe (MTL) regions early in the disease process, with some of the earliest deposits localized to subregions of the entorhinal cortex. Although functional specialization of entorhinal cortex subregions has been reported, few studies have considered functional associations with localized tau accumulation.

Methods

In this study, stepwise linear regressions were used to examine the contributions of regional tau burden in specific MTL subregions, as measured by 18F-MK6240 PET, to individual variability in cognition. Dependent measures of interest included the Clinical Dementia Rating Sum of Boxes (CDR-SB), Mini Mental State Examination (MMSE), and composite scores of delayed episodic memory and language. Other model variables included age, sex, education, APOE4 status, and global amyloid burden, indexed by 11C-PiB.

Results

Tau burden in right Brodmann area 35 (BA35), left and right Brodmann area 36 (BA36), and age each uniquely contributed to the proportion of explained variance in CDR-SB scores, while right BA36 and age were also significant predictors of MMSE scores, and right BA36 was significantly associated with delayed episodic memory performance. Tau burden in both left and right BA36, along with education, uniquely contributed to the proportion of explained variance in language composite scores. Importantly, the addition of more inclusive ROIs, encompassing less granular segmentation of the entorhinal cortex, did not significantly contribute to explained variance in cognition across any of the models.

Discussion

These findings suggest that the ability to quantify tau burden in more refined MTL subregions may better account for individual differences in cognition, which may improve the identification of non-demented older adults who are on a trajectory of decline due to AD.

Hippocampus substructure segmentation using morphological vision transformer learning

The hippocampus plays a crucial role in memory and cognition. Because of the associated toxicity from whole brain radiotherapy, more advanced treatment planning techniques prioritize hippocampal avoidance, which depends on an accurate segmentation of the small and complexly shaped hippocampus. To achieve accurate segmentation of the anterior and posterior regions of the hippocampus from T1 weighted (T1w) MR images, we developed a novel model, Hippo-Net, which uses a cascaded model strategy. The proposed model consists of two major parts: (1) a localization model is used to detect the volume-of-interest (VOI) of hippocampus. (2) An end-to-end morphological vision transformer network (Franchietal2020Pattern Recognit.102107246, Ranemetal2022 IEEE/CVF Conf. on Computer Vision and Pattern Recognition Workshops (CVPRW) pp 3710-3719) is used to perform substructures segmentation within the hippocampus VOI. The substructures include the anterior and posterior regions of the hippocampus, which are defined as the hippocampus proper and parts of the subiculum. The vision transformer incorporates the dominant features extracted from MR images, which are further improved by learning-based morphological operators. The integration of these morphological operators into the vision transformer increases the accuracy and ability to separate hippocampus structure into its two distinct substructures. A total of 260 T1w MRI datasets from medical segmentation decathlon dataset were used in this study. We conducted a five-fold cross-validation on the first 200 T1w MR images and then performed a hold-out test on the remaining 60 T1w MR images with the model trained on the first 200 images. In five-fold cross-validation, the Dice similarity coefficients were 0.900 ± 0.029 and 0.886 ± 0.031 for the hippocampus proper and parts of the subiculum, respectively. The mean surface distances (MSDs) were 0.426 ± 0.115 mm and 0.401 ± 0.100 mm for the hippocampus proper and parts of the subiculum, respectively. The proposed method showed great promise in automatically delineating hippocampus substructures on T1w MR images. It may facilitate the current clinical workflow and reduce the physicians' effort.

Deep Learning Approaches for Quantifying Ventilation Defects in Hyperpolarized Gas Magnetic Resonance Imaging of the Lung: A Review

This paper provides an in-depth overview of Deep Neural Networks and their application in the segmentation and analysis of lung Magnetic Resonance Imaging (MRI) scans, specifically focusing on hyperpolarized gas MRI and the quantification of lung ventilation defects. An in-depth understanding of Deep Neural Networks is presented, laying the groundwork for the exploration of their use in hyperpolarized gas MRI and the quantification of lung ventilation defects. Five distinct studies are examined, each leveraging unique deep learning architectures and data augmentation techniques to optimize model performance. These studies encompass a range of approaches, including the use of 3D Convolutional Neural Networks, cascaded U-Net models, Generative Adversarial Networks, and nnU-net for hyperpolarized gas MRI segmentation. The findings highlight the potential of deep learning methods in the segmentation and analysis of lung MRI scans, emphasizing the need for consensus on lung ventilation segmentation methods.

A Segmentation Framework With Unsupervised Learning-Based Label Mapper for the Ventricular Target of Intracranial Germ Cell Tumor

Intracranial germ cell tumors are rare tumors that mainly affect children and adolescents. Radiotherapy is the cornerstone of interdisciplinary treatment methods. Radiation of the whole ventricle system and the local tumor can reduce the complications in the late stage of radiotherapy while ensuring the curative effect. However, manually delineating the ventricular system is labor-intensive and time-consuming for physicians. The diverse ventricle shape and the hydrocephalus-induced ventricle dilation increase the difficulty of automatic segmentation algorithms. Therefore, this study proposed a fully automatic segmentation framework. Firstly, we designed a novel unsupervised learning-based label mapper, which is used to handle the ventricle shape variations and obtain the preliminary segmentation result. Then, to boost the segmentation performance of the framework, we improved the region growth algorithm and combined the fully connected conditional random field to optimize the preliminary results from both regional and voxel scales. In the case of only one set of annotated data is required, the average time cost is 153.01 s, and the average target segmentation accuracy can reach 84.69%. Furthermore, we verified the algorithm in practical clinical applications. The results demonstrate that our proposed method is beneficial for physicians to delineate radiotherapy targets, which is feasible and clinically practical, and may fill the gap of automatic delineation methods for the ventricular target of intracranial germ celltumors.

HumanBrainAtlas: an in vivo MRI dataset for detailed segmentations

We introduce HumanBrainAtlas, an initiative to construct a highly detailed, open-access atlas of the living human brain that combines high-resolution in vivo MR imaging and detailed segmentations previously possible only in histological preparations. Here, we present and evaluate the first step of this initiative: a comprehensive dataset of two healthy male volunteers reconstructed to a 0.25 mm isotropic resolution for T1w, T2w, and DWI contrasts. Multiple high-resolution acquisitions were collected for each contrast and each participant, followed by averaging using symmetric group-wise normalisation (Advanced Normalisation Tools). The resulting image quality permits structural parcellations rivalling histology-based atlases, while maintaining the advantages of in vivo MRI. For example, components of the thalamus, hypothalamus, and hippocampus are often impossible to identify using standard MRI protocols-can be identified within the present data. Our data are virtually distortion free, fully 3D, and compatible with the existing in vivo Neuroimaging analysis tools. The dataset is suitable for teaching and is publicly available via our website (hba.neura.edu.au), which also provides data processing scripts. Instead of focusing on coordinates in an averaged brain space, our approach focuses on providing an example segmentation at great detail in the high-quality individual brain. This serves as an illustration on what features contrasts and relations can be used to interpret MRI datasets, in research, clinical, and education settings.

ANTsX neuroimaging-derived structural phenotypes of UK Biobank

UK Biobank is a large-scale epidemiological resource for investigating prospective correlations between various lifestyle, environmental, and genetic factors with health and disease progression. In addition to individual subject information obtained through surveys and physical examinations, a comprehensive neuroimaging battery consisting of multiple modalities provides imaging-derived phenotypes (IDPs) that can serve as biomarkers in neuroscience research. In this study, we augment the existing set of UK Biobank neuroimaging structural IDPs, obtained from well-established software libraries such as FSL and FreeSurfer, with related measurements acquired through the Advanced Normalization Tools Ecosystem. This includes previously established cortical and subcortical measurements defined, in part, based on the Desikan-Killiany-Tourville atlas. Also included are morphological measurements from two recent developments: medial temporal lobe parcellation of hippocampal and extra-hippocampal regions in addition to cerebellum parcellation and thickness based on the Schmahmann anatomical labeling. Through predictive modeling, we assess the clinical utility of these IDP measurements, individually and in combination, using commonly studied phenotypic correlates including age, fluid intelligence, numeric memory, and several other sociodemographic variables. The predictive accuracy of these IDP-based models, in terms of root-mean-squared-error or area-under-the-curve for continuous and categorical variables, respectively, provides comparative insights between software libraries as well as potential clinical interpretability. Results demonstrate varied performance between package-based IDP sets and their combination, emphasizing the need for careful consideration in their selection and utilization.

Recurrent Self Fusion: Iterative Denoising for Consistent Retinal OCT Segmentation

Optical coherence tomography (OCT) is a valuable imaging technique in ophthalmology, providing high-resolution, cross-sectional images of the retina for early detection and monitoring of various retinal and neurological diseases. However, discrepancies in retinal layer thickness measurements among different OCT devices pose challenges for data comparison and interpretation, particularly in longitudinal analyses. This work introduces the idea of a recurrent self fusion (RSF) algorithm to address this issue. Our RSF algorithm, built upon the self fusion methodology, iteratively denoises retinal OCT images. A deep learning-based retinal OCT segmentation algorithm is employed for downstream analyses. A large dataset of paired OCT scans acquired on both a Spectralis and Cirrus OCT device are used for validation. The results demonstrate that the RSF algorithm effectively reduces speckle contrast and enhances the consistency of retinal OCT segmentation.

Mask Region-Based Convolutional Neural Network segmentation of the humerus and scapula from proton density-weighted axial shoulder magnetic resonance images

OBJECTIVES

This study aimed to evaluate the effectiveness of Mask Region-Based Convolutional Neural Network (R-CNN) in humerus and scapula segmentation.

PATIENTS AND METHODS

The study included 665 axial proton density (PD)-weighted magnetic resonance images of 665 consecutive shoulder instability patients (412 males, 253 females; mean age: 27±5.2 years; range, 18 to 42 years) between January 2011 and December 2014. Mask R-CNN was used to automatically segment humerus and scapula regions simultaneously. Segmentation success of Mask R-CNN was compared to the manual segmentation results of an orthopedic surgeon. Statistical evaluation was done with the Dice coefficient and the mean average precision) score. According to the humeral head structure three groups were generated: the healthy humeral head group, the edematous humeral head group, and the Hill-Sachs group (humeral heads with Hill-Sachs lesions).

RESULTS

In the test images, 81 humeral heads were healthy, 100 were edematous, and 38 had a Hill-Sachs lesion. According to the Dice metric, the overall success rate of Mask R-CNN configuration was 96.47 and 93.87% for the segmentation of the humeral head and scapula, respectively, and 95.86 and 92.35% for an intersection over union of 0.5 according to the mean average precision. According to the Dice metric, the segmentation success of the humerus and scapula of the healthy group was 94.58 and 97.42%, the segmentation success of the edematous humerus group was 93.56 and 96.53%, and the segmentation success of the Hill-Sachs group was 93.47 to 95.48%. The segmentation success of scapula in the case of discontinuity was 92.86% according to Dice metric.

CONCLUSION

Mask R-CNN-based humerus and scapula segmentation provided promising results compared to manual segmentation of an expert. Mask R-CNN overcomes the problem of discontinuous edges and Rician noise in axial PD-weighted shoulder magnetic resonance imaging.

Deep convolutional neural network for hippocampus segmentation with boundary region refinement

Accurately segmenting the hippocampus from magnetic resonance (MR) brain images is a crucial step in studying brain disorders. However, this task is challenging due to the low signal contrast of hippocampal images, the irregular shape, and small structural size of the hippocampi. In recent years, several deep convolutional networks have been proposed for hippocampus segmentation, which have achieved state-of-the-art performance. These methods typically use large image patches for training the network, as larger patches are beneficial for capturing long-range contextual information. However, this approach increases the computational burden and overlooks the significance of the boundary region. In this study, we propose a deep learning-based method for hippocampus segmentation with boundary region refinement. Our method involves two main steps. First, we propose a convolutional network that takes large image patches as input for initial segmentation. Then, we extract small image patches around the hippocampal boundary for training the second convolutional neural network, which refines the segmentation in the boundary regions. We validate our proposed method on a publicly available dataset and demonstrate that it significantly improves the performance of convolutional neural networks that use single-size image patches as input. In conclusion, our study proposes a novel method for hippocampus segmentation, which improves upon the current state-of-the-art methods. By incorporating a boundary refinement step, our approach achieves higher accuracy in hippocampus segmentation and may facilitate research on brain disorders.

Deep fusion of multi-modal features for brain tumor image segmentation

Accurate segmentation of pathological regions in brain magnetic resonance images (MRI) is essential for the diagnosis and treatment of brain tumors. Multi-modality MRIs, which offer diverse feature information, are commonly utilized in brain tumor image segmentation. Deep neural networks have become prevalent in this field; however, many approaches simply concatenate different modalities and input them directly into the neural network for segmentation, disregarding the unique characteristics and complementarity of each modality. In this study, we propose a brain tumor image segmentation method that leverages deep residual learning with multi-modality image feature fusion. Our approach involves extracting and fusing distinct and complementary features from various modalities, fully exploiting the multi-modality information within a deep convolutional neural network to enhance the performance of brain tumor image segmentation. We evaluate the effectiveness of our proposed method using the BraTS2021 dataset and demonstrate that deep residual learning with multi-modality image feature fusion significantly improves segmentation accuracy. Our method achieves competitive segmentation results, with Dice values of 83.3, 89.07, and 91.44 for enhanced tumor, tumor core, and whole tumor, respectively. These findings highlight the potential of our method in improving brain tumor diagnosis and treatment through accurate segmentation of pathological regions in brain MRIs.

Perspective of artificial intelligence in healthcare data management: A journey towards precision medicine

Mounting evidence has highlighted the implementation of big data handling and management in the healthcare industry to improve the clinical services. Various private and public companies have generated, stored, and analyzed different types of big healthcare data, such as omics data, clinical data, electronic health records, personal health records, and sensing data with the aim to move in the direction of precision medicine. Additionally, with the advancement in technologies, researchers are curious to extract the potential involvement of artificial intelligence and machine learning on big healthcare data to enhance the quality of patient's lives. However, seeking solutions from big healthcare data requires proper management, storage, and analysis, which imposes hinderances associated with big data handling. Herein, we briefly discuss the implication of big data handling and the role of artificial intelligence in precision medicine. Further, we also highlighted the potential of artificial intelligence in integrating and analyzing the big data that offer personalized treatment. In addition, we briefly discuss the applications of artificial intelligence in personalized treatment, especially in neurological diseases. Lastly, we discuss the challenges and limitations imposed by artificial intelligence in big data management and analysis to hinder precision medicine.

Dopaminergic Dysfunction Is More Symmetric in Dementia with Lewy Bodies Compared to Parkinson's Disease

BACKGROUND

The α-syn Origin site and Connectome model (SOC) proposes that α-synucleinopathies can be divided into two categories: the asymmetrical brain-first, and more symmetrical body-first Lewy body disease. We have hypothesized that most patients with dementia with Lewy bodies (DLB) belong to the body-first subtype, whereas patients with Parkinson's disease (PD) more often belong to the brain-first subtype.

OBJECTIVE

To compare asymmetry of striatal dopaminergic dysfunction in DLB and PD patients using [18F]-FE-PE2I positron emission tomography (PET).

METHODS

We analyzed [18F]-FE-PE2I PET data from 29 DLB patients and 76 PD patients who were identified retrospectively during a 5-year period at Dept. of Neurology, Aarhus University Hospital. Additionally, imaging data from 34 healthy controls was used for age-correction and visual comparison.

RESULTS

PD patients showed significantly more asymmetry in specific binding ratios between the most and least affected putamen (p < 0.0001) and caudate (p = 0.003) compared to DLB patients. PD patients also had more severe degeneration in the putamen compared to the caudate in comparison to DLB patients (p < 0.0001) who had a more universal pattern of striatal degeneration.

CONCLUSION

Patients with DLB show significantly more symmetric striatal degeneration on average compared to PD patients. These results support the hypothesis that DLB patients may be more likely to conform to the body-first subtype characterized by a symmetrical spread of pathology, whereas PD patients may be more likely to conform to the brain-first subtype with more lateralized initial propagation of pathology.

Antemortem network analysis of spreading pathology in autopsy-confirmed frontotemporal degeneration

Despite well-articulated hypotheses of spreading pathology in animal models of neurodegenerative disease, the basis for spreading neurodegenerative pathology in humans has been difficult to ascertain. In this study, we used graph theoretic analyses of structural networks in antemortem, multimodal MRI from autopsy-confirmed cases to examine spreading pathology in sporadic frontotemporal lobar degeneration. We defined phases of progressive cortical atrophy on T1-weighted MRI using a published algorithm in autopsied frontotemporal lobar degeneration with tau inclusions or with transactional DNA binding protein of ∼43 kDa inclusions. We studied global and local indices of structural networks in each of these phases, focusing on the integrity of grey matter hubs and white matter edges projecting between hubs. We found that global network measures are compromised to an equal degree in patients with frontotemporal lobar degeneration with tau inclusions and frontotemporal lobar degeneration-transactional DNA binding protein of ∼43 kDa inclusions compared to healthy controls. While measures of local network integrity were compromised in both frontotemporal lobar degeneration with tau inclusions and frontotemporal lobar degeneration-transactional DNA binding protein of ∼43 kDa inclusions, we discovered several important characteristics that distinguished between these groups. Hubs identified in controls were degraded in both patient groups, but degraded hubs were associated with the earliest phase of cortical atrophy (i.e. epicentres) only in frontotemporal lobar degeneration with tau inclusions. Degraded edges were significantly more plentiful in frontotemporal lobar degeneration with tau inclusions than in frontotemporal lobar degeneration-transactional DNA binding protein of ∼43 kDa inclusions, suggesting that the spread of tau pathology involves more significant white matter degeneration. Weakened edges were associated with degraded hubs in frontotemporal lobar degeneration with tau inclusions more than in frontotemporal lobar degeneration-transactional DNA binding protein of ∼43 kDa inclusions, particularly in the earlier phases of the disease, and phase-to-phase transitions in frontotemporal lobar degeneration with tau inclusions were characterized by weakened edges in earlier phases projecting to diseased hubs in subsequent phases of the disease. When we examined the spread of pathology from a region diseased in an earlier phase to physically adjacent regions in subsequent phases, we found greater evidence of disease spreading to adjacent regions in frontotemporal lobar degeneration-transactional DNA binding protein of ∼43 kDa inclusions than in frontotemporal lobar degeneration with tau inclusions. We associated evidence of degraded grey matter hubs and weakened white matter edges with quantitative measures of digitized pathology from direct observations of patients' brain samples. We conclude from these observations that the spread of pathology from diseased regions to distant regions via weakened long-range edges may contribute to spreading disease in frontotemporal dementia-tau, while spread of pathology to physically adjacent regions via local neuronal connectivity may play a more prominent role in spreading disease in frontotemporal lobar degeneration-transactional DNA binding protein of ∼43 kDa inclusions.

A calibrated SVM based on weighted smooth GL1/2 for Alzheimer’s disease prediction

Alzheimer's disease (AD) is currently one of the mainstream senile diseases in the world. It is a key problem predicting the early stage of AD. Low accuracy recognition of AD and high redundancy brain lesions are the main obstacles. Traditionally, Group Lasso method can achieve good sparseness. But, redundancy inside group is ignored. This paper proposes an improved smooth classification framework which combines the weighted smooth GL_1/2 (wSGL_1/2) as feature selection method and a calibrated support vector machine (cSVM) as the classifier. wSGL_1/2 can make intra-group and inner-group features sparse, in which the group weights can further improve the efficiency of the model. cSVM can enhance the speed and stability of model by adding calibrated hinge function. Before feature selecting, an anatomical boundary-based clustering, called as ac-SLIC-AAL, is designed to make adjacent similar voxels into one group for accommodating the overall differences of all data. The cSVM model is fast convergence speed, high accuracy and good interpretability on AD classification, AD early diagnosis and MCI transition prediction. In experiments, all steps are tested respectively, including classifiers' comparison, feature selection verification, generalization verification and comparing with state-of-the-art methods. The results are supportive and satisfactory. The superior of the proposed model are verified globally. At the same time, the algorithm can point out the important brain areas in the MRI, which has important reference value for the doctor's predictive work. The source code and data is available at http://github.com/Hu-s-h/c-SVMForMRI.

White Matter Hyperintensity Volume and Amyloid-PET Synergistically Impact Memory Independent of Tau-PET in Older Adults Without Dementia

BACKGROUND

Alzheimer's disease (AD) and cerebrovascular disease are common, co-existing pathologies in older adults. Whether the effects of cerebrovascular disease and AD biomarkers on cognition are additive or synergistic remains unclear.

OBJECTIVE

To examine whether white matter hyperintensity (WMH) volume moderates the independent association between each AD biomarker and cognition.

METHODS

In 586 older adults without dementia, linear regressions tested the interaction between amyloid-β (Aβ) positron emission tomography (PET) and WMH volume on cognition, independent of tau-PET. We also tested the interaction between tau-PET and WMH volume on cognition, independent of Aβ-PET.

RESULTS

Adjusting for tau-PET, the quadratic effect of WMH interacted with Aβ-PET to impact memory. There was no interaction between either the linear or quadratic effect of WMH and Aβ-PET on executive function. There was no interaction between WMH volume and tau-PET on either cognitive measure.

CONCLUSION

Results suggest that cerebrovascular lesions act synergistically with Aβ to affect memory, independent of tau, highlighting the importance of incorporating vascular pathology into biomarker assessment of AD.

Lower hypothalamic volume with lower body mass index is associated with shorter survival in patients with amyotrophic lateral sclerosis

BACKGROUND AND PURPOSE

Weight loss in patients with amyotrophic lateral sclerosis (ALS) is associated with faster disease progression and shorter survival. Decreased hypothalamic volume is proposed to contribute to weight loss due to loss of appetite and/or hypermetabolism. We aimed to investigate the relationship between hypothalamic volume and body mass index (BMI) in ALS and Alzheimer's disease (AD), and the associations of hypothalamic volume with weight loss, appetite, metabolism and survival in patients with ALS.

METHODS

We compared hypothalamic volumes from magnetic resonance imaging scans with BMI for patients with ALS (n = 42), patients with AD (n = 167) and non-neurodegenerative disease controls (n = 527). Hypothalamic volumes from patients with ALS were correlated with measures of appetite and metabolism, and change in anthropomorphic measures and disease outcomes.

RESULTS

Lower hypothalamic volume was associated with lower and higher BMI in ALS (quadratic association; probability of direction = 0.96). This was not observed in AD patients or controls. Hypothalamic volume was not associated with loss of appetite (p = 0.58) or hypermetabolism (p = 0.49). Patients with lower BMI and lower hypothalamic volume tended to lose weight (p = 0.08) and fat mass (p = 0.06) over the course of their disease, and presented with an increased risk of earlier death (hazard ratio [HR] 3.16, p = 0.03). Lower hypothalamic volume alone trended for greater risk of earlier death (HR 2.61, p = 0.07).

CONCLUSION

These observations suggest that lower hypothalamic volume in ALS contributes to positive and negative energy balance, and  is not universally associated with loss of appetite or hypermetabolism. Critically, lower hypothalamic volume with lower BMI was associated with weight loss and earlier death.

Deep label fusion: A generalizable hybrid multi-atlas and deep convolutional neural network for medical image segmentation

Deep convolutional neural networks (DCNN) achieve very high accuracy in segmenting various anatomical structures in medical images but often suffer from relatively poor generalizability. Multi-atlas segmentation (MAS), while less accurate than DCNN in many applications, tends to generalize well to unseen datasets with different characteristics from the training dataset. Several groups have attempted to integrate the power of DCNN to learn complex data representations and the robustness of MAS to changes in image characteristics. However, these studies primarily focused on replacing individual components of MAS with DCNN models and reported marginal improvements in accuracy. In this study we describe and evaluate a 3D end-to-end hybrid MAS and DCNN segmentation pipeline, called Deep Label Fusion (DLF). The DLF pipeline consists of two main components with learnable weights, including a weighted voting subnet that mimics the MAS algorithm and a fine-tuning subnet that corrects residual segmentation errors to improve final segmentation accuracy. We evaluate DLF on five datasets that represent a diversity of anatomical structures (medial temporal lobe subregions and lumbar vertebrae) and imaging modalities (multi-modality, multi-field-strength MRI and Computational Tomography). These experiments show that DLF achieves comparable segmentation accuracy to nnU-Net (Isensee et al., 2020), the state-of-the-art DCNN pipeline, when evaluated on a dataset with similar characteristics to the training datasets, while outperforming nnU-Net on tasks that involve generalization to datasets with different characteristics (different MRI field strength or different patient population). DLF is also shown to consistently improve upon conventional MAS methods. In addition, a modality augmentation strategy tailored for multimodal imaging is proposed and demonstrated to be beneficial in improving the segmentation accuracy of learning-based methods, including DLF and DCNN, in missing data scenarios in test time as well as increasing the interpretability of the contribution of each individual modality.

Cot-side imaging of functional connectivity in the developing brain during sleep using wearable high-density diffuse optical tomography

Studies of cortical function in newborn infants in clinical settings are extremely challenging to undertake with traditional neuroimaging approaches. Partly in response to this challenge, functional near-infrared spectroscopy (fNIRS) has become an increasingly common clinical research tool but has significant limitations including a low spatial resolution and poor depth specificity. Moreover, the bulky optical fibres required in traditional fNIRS approaches present significant mechanical challenges, particularly for the study of vulnerable newborn infants. A new generation of wearable, modular, high-density diffuse optical tomography (HD-DOT) technologies has recently emerged that overcomes many of the limitations of traditional, fibre-based and low-density fNIRS measurements. Driven by the development of this new technology, we have undertaken the first cot-side study of newborn infants using wearable HD-DOT in a clinical setting. We use this technology to study functional brain connectivity (FC) in newborn infants during sleep and assess the effect of neonatal sleep states, active sleep (AS) and quiet sleep (QS), on resting state FC. Our results demonstrate that it is now possible to obtain high-quality functional images of the neonatal brain in the clinical setting with few constraints. Our results also suggest that sleep states differentially affect FC in the neonatal brain, consistent with prior reports.

MsVRL: Self-Supervised Multiscale Visual Representation Learning via Cross-Level Consistency for Medical Image Segmentation

Automated medical image segmentation for organs or lesions plays an essential role in clinical diagnoses and treatment plannings. However, training an accurate and robust segmentation model is still a long-standing challenge due to the time-consuming and expertise-intensive annotations for training data, especially 3-D medical images. Recently, self-supervised learning emerges as a promising approach for unsupervised visual representation learning, showing great potential to alleviate the expertise annotations for medical images. Although global representation learning has attained remarkable results on iconic datasets, such as ImageNet, it can not be applied directly to medical image segmentation, because the segmentation task is non-iconic, and the targets always vary in physical scales. To address these problems, we propose a Multi-scale Visual Representation self-supervised Learning (MsVRL) model, to perform finer-grained representation and deal with different target scales. Specifically, a multi-scale representation conception, a canvas matching method, an embedding pre-sampling module, a center-ness branch, and a cross-level consistent loss are introduced to improve the performance. After pre-trained on unlabeled datasets (RibFrac and part of MSD), MsVRL performs downstream segmentation tasks on labeled datasets (BCV, spleen of MSD, and KiTS). Results of the experiments show that MsVRL outperforms other state-of-the-art works on these medical image segmentation tasks.

Pathological functional impairment: Neuropsychological correlates of the shared variance between everyday functioning and brain volumetrics

Objective

Given that several non-cognitive factors can contribute to difficulties with everyday functioning, examining the extent to which cognition is associated with brain-related changes in everyday functioning is critical to accurate characterization of cognitive disorders. In this study, we examined neuropsychological correlates of the shared variance between everyday functioning and pathological indicators of cognitive aging using MRI brain volumetrics.

Participants and methods

Participants were 600 adults aged 55 and older without dementia [432 cognitively normal; 168 mild cognitive impairment (MCI)] from the National Alzheimer's Coordinating Center cohort who underwent neuropsychological testing, informant-rated everyday functioning, and brain MRI scanning at baseline. The shared variance between everyday functioning and brain volumetrics (i.e., hippocampal volume, white matter hyperintensity volume) was extracted using the predicted value from multiple regression. The shared variance was used as an indicator of pathological everyday functional impairment. The residual variance from the regression analysis was used to examine functional reserve.

Results

Larger white matter hyperintensity volumes (p = 0.002) and smaller hippocampal volumes (p < 0.001) were significantly correlated with worse informant-rated everyday functioning. Among individuals with MCI, worse performances on delayed recall (p = 0.013) and category fluency (p = 0.012) were significantly correlated with pathological functional impairment in multiple regression analysis. In the cognitively normal group, only worse auditory working memory (i.e., digit span backward; p = 0.025) significantly correlated with pathological functioning. Functional reserve was inversely related to anxiety (p < 0.001) in the MCI group and was associated with depressive symptoms (p = 0.003) and apathy (p < 0.001) in the cognitively normal group.

Conclusion

Subtle brain-related everyday functioning difficulties are evident in MCI and track with expected preclinical Alzheimer's disease cognitive phenotypes in this largely amnestic sample. Our findings indicate that functional changes occur early in the disease process and that interventions to target neuropsychiatric symptoms may help to bolster functional reserve in those at risk.

Local Spatiotemporal Representation Learning for Longitudinally-consistent Neuroimage Analysis

Recent self-supervised advances in medical computer vision exploit the global and local anatomical self-similarity for pretraining prior to downstream tasks such as segmentation. However, current methods assume i.i.d. image acquisition, which is invalid in clinical study designs where follow-up longitudinal scans track subject-specific temporal changes. Further, existing self-supervised methods for medically-relevant image-to-image architectures exploit only spatial or temporal self-similarity and do so via a loss applied only at a single image-scale, with naive multi-scale spatiotemporal extensions collapsing to degenerate solutions. To these ends, this paper makes two contributions: (1) It presents a local and multi-scale spatiotemporal representation learning method for image-to-image architectures trained on longitudinal images. It exploits the spatiotemporal self-similarity of learned multi-scale intra-subject image features for pretraining and develops several feature-wise regularizations that avoid degenerate representations; (2) During finetuning, it proposes a surprisingly simple self-supervised segmentation consistency regularization to exploit intra-subject correlation. Benchmarked across various segmentation tasks, the proposed framework outperforms both well-tuned randomly-initialized baselines and current self-supervised techniques designed for both i.i.d. and longitudinal datasets. These improvements are demonstrated across both longitudinal neurodegenerative adult MRI and developing infant brain MRI and yield both higher performance and longitudinal consistency.

Deep learning empowered volume delineation of whole-body organs-at-risk for accelerated radiotherapy

In radiotherapy for cancer patients, an indispensable process is to delineate organs-at-risk (OARs) and tumors. However, it is the most time-consuming step as manual delineation is always required from radiation oncologists. Herein, we propose a lightweight deep learning framework for radiotherapy treatment planning (RTP), named RTP-Net, to promote an automatic, rapid, and precise initialization of whole-body OARs and tumors. Briefly, the framework implements a cascade coarse-to-fine segmentation, with adaptive module for both small and large organs, and attention mechanisms for organs and boundaries. Our experiments show three merits: 1) Extensively evaluates on 67 delineation tasks on a large-scale dataset of 28,581 cases; 2) Demonstrates comparable or superior accuracy with an average Dice of 0.95; 3) Achieves near real-time delineation in most tasks with <2 s. This framework could be utilized to accelerate the contouring process in the All-in-One radiotherapy scheme, and thus greatly shorten the turnaround time of patients.

High-resolution atlasing and segmentation of the subcortex: Review and perspective on challenges and opportunities created by machine learning

This paper reviews almost three decades of work on atlasing and segmentation methods for subcortical structures in human brain MRI. In writing this survey, we have three distinct aims. First, to document the evolution of digital subcortical atlases of the human brain, from the early MRI templates published in the nineties, to the complex multi-modal atlases at the subregion level that are available today. Second, to provide a detailed record of related efforts in the automated segmentation front, from earlier atlas-based methods to modern machine learning approaches. And third, to present a perspective on the future of high-resolution atlasing and segmentation of subcortical structures in in vivo human brain MRI, including open challenges and opportunities created by recent developments in machine learning.

Multi-atlas segmentation and quantification of muscle, bone and subcutaneous adipose tissue in the lower leg using peripheral quantitative computed tomography

Accurate and reproducible tissue identification is essential for understanding structural and functional changes that may occur naturally with aging, or because of a chronic disease, or in response to intervention therapies. Peripheral quantitative computed tomography (pQCT) is regularly employed for body composition studies, especially for the structural and material properties of the bone. Furthermore, pQCT acquisition requires low radiation dose and the scanner is compact and portable. However, pQCT scans have limited spatial resolution and moderate SNR. pQCT image quality is frequently degraded by involuntary subject movement during image acquisition. These limitations may often compromise the accuracy of tissue quantification, and emphasize the need for automated and robust quantification methods. We propose a tissue identification and quantification methodology that addresses image quality limitations and artifacts, with increased interest in subject movement. We introduce a multi-atlas image segmentation (MAIS) framework for semantic segmentation of hard and soft tissues in pQCT scans at multiple levels of the lower leg. We describe the stages of statistical atlas generation, deformable registration and multi-tissue classifier fusion. We evaluated the performance of our methodology using multiple deformable registration approaches against reference tissue masks. We also evaluated the performance of conventional model-based segmentation against the same reference data to facilitate comparisons. We studied the effect of subject movement on tissue segmentation quality. We also applied the top performing method to a larger out-of-sample dataset and report the quantification results. The results show that multi-atlas image segmentation with diffeomorphic deformation and probabilistic label fusion produces very good quality over all tissues, even for scans with significant quality degradation. The application of our technique to the larger dataset reveals trends of age-related body composition changes that are consistent with the literature. Because of its robustness to subject motion artifacts, our MAIS methodology enables analysis of larger number of scans than conventional state-of-the-art methods. Automated analysis of both soft and hard tissues in pQCT is another contribution of this work.

Harmonized Z-Scores Calculated from a Large-Scale Normal MRI Database to Evaluate Brain Atrophy in Neurodegenerative Disorders

Alzheimer’s disease (AD), the most common type of dementia in elderly individuals, slowly and progressively diminishes the cognitive function. Mild cognitive impairment (MCI) is also a significant risk factor for the onset of AD. Magnetic resonance imaging (MRI) is widely used for the detection and understanding of the natural progression of AD and other neurodegenerative disorders. For proper assessment of these diseases, a reliable database of images from cognitively healthy participants is important. However, differences in magnetic field strength or the sex and age of participants between a normal database and an evaluation data set can affect the accuracy of the detection and evaluation of neurodegenerative disorders. We developed a brain segmentation procedure, based on 30 Japanese brain atlases, and suggest a harmonized Z-score to correct the differences in field strength and sex and age from a large data set (1235 cognitively healthy participants), including 1.5 T and 3 T T1-weighted brain images. We evaluated our harmonized Z-score for AD discriminative power and classification accuracy between stable MCI and progressive MCI. Our procedure can perform brain segmentation in approximately 30 min. The harmonized Z-score of the hippocampus achieved high accuracy (AUC = 0.96) for AD detection and moderate accuracy (AUC = 0.70) to classify stable or progressive MCI. These results show that our method can detect AD with high accuracy and high generalization capability. Moreover, it may discriminate between stable and progressive MCI. Our study has some limitations: the age groups in the 1.5 T data set and 3 T data set are significantly different. In this study, we focused on AD, which is primarily a disease of elderly patients. For other diseases in different age groups, the harmonized Z-score needs to be recalculated using different data sets.

[18F]FE-PE2I PET is a feasible alternative to [123I]FP-CIT SPECT for dopamine transporter imaging in clinically uncertain parkinsonism

BACKGROUND

Dopamine transporter (DAT) imaging of striatum is clinically used in Parkinson's disease (PD) and neurodegenerative parkinsonian syndromes (PS) especially in the early disease stages. The aim of the present study was to evaluate the diagnostic performance of the recently developed tracer for DAT imaging [¹⁸F]FE-PE2I PET/CT to the reference standard [¹²³I]FP-CIT SPECT.

METHODS

Ninety-eight unselected patients referred for DAT imaging were included prospectively and consecutively and evaluated with [¹⁸F]FE-PE2I PET/CT and [¹²³I]FP-CIT SPECT on two separate days. PET and SPECT scans were categorized independently by two blinded expert readers as either normal, vascular changes, or mixed. Semiquantitative values were obtained for each modality and compared regarding effect size using Glass' delta.

RESULTS

Fifty-six of the [¹²³I]FP-CIT SPECT scans were considered abnormal (52 caused by PS, 4 by infarctions). Using [¹⁸F]FE-PE2I PET/CT, 95 of the 98 patients were categorized identically to SPECT as PS or non-PS with a sensitivity of 0.94 [0.84-0.99] and a specificity of 1.00 [0.92-1.00]. Inter-reader agreement for [¹⁸F]FE-PE2I PET with a kappa of 0.97 [0.89-1.00] was comparable to the agreement for [¹²³I]FP-CIT SPECT of 0.96 [0.76-1.00]. Semiquantitative values for short 10-min reconstructions of [¹⁸F]FE-PE2I PET/CT were comparable to longer reconstructions. The effect size for putamen/caudate nucleus ratio was significantly increased using PET compared to SPECT.

CONCLUSIONS

The high correspondence of [¹⁸F]FE-PE2I PET compared to reference standard [¹²³I]FP-CIT SPECT establishes [¹⁸F]FE-PE2I PET as a feasible PET tracer for clinical use with favourable scan logistics.

A joint ventricle and WMH segmentation from MRI for evaluation of healthy and pathological changes in the aging brain

Age-related changes in brain structure include atrophy of the brain parenchyma and white matter changes of presumed vascular origin. Enlargement of the ventricles may occur due to atrophy or impaired cerebrospinal fluid (CSF) circulation. The co-occurrence of these changes in neurodegenerative diseases and in aging brains often requires investigators to take both into account when studying the brain, however, automated segmentation of enlarged ventricles and white matter hyperintensities (WMHs) can be a challenging task. Here, we present a hybrid multi-atlas segmentation and convolutional autoencoder approach for joint ventricle parcellation and WMH segmentation from magnetic resonance images (MRIs). Our fully automated approach uses a convolutional autoencoder to generate a standardized image of grey matter, white matter, CSF, and WMHs, which, in conjunction with labels generated by a multi-atlas segmentation approach, is then fed into a convolutional neural network to parcellate the ventricular system. Hence, our approach does not depend on manually delineated training data for new data sets. The segmentation pipeline was validated on both healthy elderly subjects and subjects with normal pressure hydrocephalus using ground truth manual labels and compared with state-of-the-art segmentation methods. We then applied the method to a cohort of 2401 elderly brains to investigate associations of ventricle volume and WMH load with various demographics and clinical biomarkers, using a multiple regression model. Our results indicate that the ventricle volume and WMH load are both highly variable in a cohort of elderly subjects and there is an independent association between the two, which highlights the importance of taking both the possibility of enlarged ventricles and WMHs into account when studying the aging brain.

Quantitative susceptibility atlas construction in Montreal Neurological Institute space: towards histological-consistent iron-rich deep brain nucleus subregion identification

Iron-rich deep brain nuclei (DBN) of the human brain are involved in various motoric, emotional and cognitive brain functions. The abnormal iron alterations in the DBN are closely associated with multiple neurological and psychiatric diseases. Quantitative susceptibility mapping (QSM) provides the spatial distribution of the magnetic susceptibility of human brain tissues. Compared to traditional structural imaging, QSM provides superiority for imaging the iron-rich DBN owing to the susceptibility difference existing between brain tissues. In this study, we constructed a Montreal Neurological Institute (MNI) space unbiased QSM human brain atlas via group-wise registration from 100 healthy subjects aged 19-29 years. The atlas construction process was guided by hybrid images that were fused from multi-modal magnetic resonance images (MRI). We named it as Multi-modal-fused magnetic Susceptibility (MuSus-100) atlas. The high-quality susceptibility atlas provides extraordinary image contrast between iron-rich DBN with their surroundings. Parcellation maps of DBN and their subregions that are highly related to neurological and psychiatric pathology were then manually labeled based on the atlas set with the assistance of an image border-enhancement process. Especially, the bilateral thalamus was delineated into 64 detailed subregions referring to the Schaltenbrand-Wahren stereotactic atlas. To our best knowledge, the histological-consistent thalamic nucleus parcellation map is well defined for the first time in the MNI space. Compared with existing atlases that emphasizing DBN parcellation, the newly proposed atlas outperforms on the task of atlas-guided individual brain image DBN segmentation both in accuracy and robustness. Moreover, we applied the proposed DBN parcellation map to conduct detailed identification of the pathology-related iron content alterations in subcortical nuclei for Parkinson's Disease (PD) patients. We envision that the MuSus-100 atlas can play a crucial role in improving the accuracy of DBN segmentation for the research of neurological and psychiatric disease progress and also be helpful for target planning in deep brain stimulation surgery.

Integrated 3d flow-based multi-atlas brain structure segmentation

MRI brain structure segmentation plays an important role in neuroimaging studies. Existing methods either spend much CPU time, require considerable annotated data, or fail in segmenting volumes with large deformation. In this paper, we develop a novel multi-atlas-based algorithm for 3D MRI brain structure segmentation. It consists of three modules: registration, atlas selection and label fusion. Both registration and label fusion leverage an integrated flow based on grayscale and SIFT features. We introduce an effective and efficient strategy for atlas selection by employing the accompanying energy generated in the registration step. A 3D sequential belief propagation method and a 3D coarse-to-fine flow matching approach are developed in both registration and label fusion modules. The proposed method is evaluated on five public datasets. The results show that it has the best performance in almost all the settings compared to competitive methods such as ANTs, Elastix, Learning to Rank and Joint Label Fusion. Moreover, our registration method is more than 7 times as efficient as that of ANTs SyN, while our label transfer method is 18 times faster than Joint Label Fusion in CPU time. The results on the ADNI dataset demonstrate that our method is applicable to image pairs that require a significant transformation in registration. The performance on a composite dataset suggests that our method succeeds in a cross-modality manner. The results of this study show that the integrated 3D flow-based method is effective and efficient for brain structure segmentation. It also demonstrates the power of SIFT features, multi-atlas segmentation and classical machine learning algorithms for a medical image analysis task. The experimental results on public datasets show the proposed method's potential for general applicability in various brain structures and settings.

Leveraging non-expert crowdsourcing to segment the optic cup and disc of multicolor fundus images

Multicolor scanning laser imaging (MCI) images have broad application potential in the diagnosis of fundus diseases such as glaucoma. However, the performance level of automatic aided diagnosis systems based on MCI images is limited by the lack of high-quality annotations of numerous images. Producing annotations for vast amounts of MCI images will be a prolonged process if we only employ experts. Therefore, we consider non-expert crowdsourcing, which is an alternative approach to produce useful annotations efficiently and low cost. In this work, we aim to explore the effectiveness of non-expert crowdsourcing on the segmentation of the optic cup (OC) and optic disc (OD), which is an upstream task for glaucoma diagnosis, using MCI images. To this end, desensitized MCI images are independently annotated by four non-expert annotators, constructing a crowdsourcing dataset. To profit from crowdsourcing, we propose a model consisting of coupled regularization network and segmentation network. The regularization network generates learnable pixel-wise confusion matrices (CMs) that reflects preferences of each annotator. During training, the CMs and segmentation network are simultaneously optimized to enable dynamic trade-offs for non-expert annotations and generate reliable predictions. Crowdsourcing learning using our method have an average Mean Intersection Over Union ( M ) of 91.34%, while the average M of model trained by expert annotations is 91.72%. In addition, comparative experiments show that in our segmentation task non-expert crowdsourcing can be on a par with the expert who annotates 90% of data. Our work suggests that crowdsourcing in the segmentation of OC and OD using MCI images has the potential to be a substitute to expert annotation, which will accelerate the construction of large datasets to facilitate the application of deep learning in clinical diagnosis using MCI images.

Development of a New Index to Distinguish Hepatic Encephalopathy through Automated Quantification of Globus Pallidal Signal Intensity Using MRI

Hyperintensities within the bilateral globus pallidus on T1-weighted magnetic resonance images were present in some liver cirrhosis patients with hepatic encephalopathy. The symptoms of covert hepatic encephalopathy are similar to those of mild dementia. We aimed to develop a new diagnostic index in which to distinguish hepatic encephalopathy from dementia. The globus pallidus signal hyperintensity was quantified using three-dimensional images. In addition, the new index value distribution was evaluated in a cohort of dementia patients. Signal intensity of globus pallidus significantly increased in liver cirrhosis patients with hepatic encephalopathy compared to those without hepatic encephalopathy (p < 0.05), healthy subjects (p < 0.05) or dementia patients (p < 0.001). Only 12.5% of liver cirrhosis patients without hepatic encephalopathy and 2% of dementia patients exceeded the new index cut-off value of 0.994, which predicts hepatic encephalopathy. One dementia patient in our evaluation had a history of liver cancer treatment and was assumed to have concomitant hepatic encephalopathy. The automatic assessment of signal intensity in globus pallidus is useful for distinguishing liver cirrhosis patients with hepatic encephalopathy from healthy subjects and liver cirrhosis patients without hepatic encephalopathy. Our image analyses exclude possible cases of hepatic encephalopathy from patients with neurocognitive impairment, including dementia.

LGMSU-Net: Local Features, Global Features, and Multi-Scale Features Fused the U-Shaped Network for Brain Tumor Segmentation

Brain tumors are one of the deadliest cancers in the world. Researchers have conducted a lot of research work on brain tumor segmentation with good performance due to the rapid development of deep learning for assisting doctors in diagnosis and treatment. However, most of these methods cannot fully combine multiple feature information and their performances need to be improved. This study developed a novel network fusing local features representing detailed information, global features representing global information, and multi-scale features enhancing the model’s robustness to fully extract the features of brain tumors and proposed a novel axial-deformable attention module for modeling global information to improve the performance of brain tumor segmentation to assist clinicians in the automatic segmentation of brain tumors. Moreover, positional embeddings were used to make the network training faster and improve the method’s performance. Six metrics were used to evaluate the proposed method on the BraTS2018 dataset. Outstanding performance was obtained with Dice score, mean Intersection over Union, precision, recall, params, and inference time of 0.8735, 0.7756, 0.9477, 0.8769, 69.02 M, and 15.66 millisecond, respectively, for the whole tumor. Extensive experiments demonstrated that the proposed network obtained excellent performance and was helpful in providing supplementary advice to the clinicians.

Simulated Attack Reveals How Lesions Affect Network Properties in Poststroke Aphasia

Aphasia is a prevalent cognitive syndrome caused by stroke. The rarity of premorbid imaging and heterogeneity of lesion obscures the links between the local effects of the lesion, global anatomic network organization, and aphasia symptoms. We applied a simulated attack approach in humans to examine the effects of 39 stroke lesions (16 females) on anatomic network topology by simulating their effects in a control sample of 36 healthy (15 females) brain networks. We focused on measures of global network organization thought to support overall brain function and resilience in the whole brain and within the left hemisphere. After removing lesion volume from the network topology measures and behavioral scores [the Western Aphasia Battery Aphasia Quotient (WAB-AQ), four behavioral factor scores obtained from a neuropsychological battery, and a factor sum], we compared the behavioral variance accounted for by simulated poststroke connectomes to that observed in the randomly permuted data. Global measures of anatomic network topology in the whole brain and left hemisphere accounted for 10% variance or more of the WAB-AQ and the lexical factor score beyond lesion volume and null permutations. Streamline networks provided more reliable point estimates than FA networks. Edge weights and network efficiency were weighted most highly in predicting the WAB-AQ for FA networks. Overall, our results suggest that global network measures provide modest statistical value beyond lesion volume when predicting overall aphasia severity, but less value in predicting specific behaviors. Variability in estimates could be induced by premorbid ability, deafferentation and diaschisis, and neuroplasticity following stroke.SIGNIFICANCE STATEMENT Poststroke, the remaining neuroanatomy maintains cognition and supports recovery. However, studies often use small, cross-sectional samples that cannot fully model the interactions between lesions and other variables that affect networks in stroke. Alternate methods are required to account for these effects. "Simulated attack" models are computational approaches that apply virtual damage to the brain and measure their putative consequences. Using a simulated attack model, we estimated how simulated damage to anatomic networks could account for language performance. Overall, our results reveal that global network measures can provide modest statistical value predicting overall aphasia severity, but less value in predicting specific behaviors. These findings suggest that more theoretically precise network models could be necessary to robustly predict individual outcomes in aphasia.