1
|
Mendelsohn Z, Pemberton HG, Gray J, Goodkin O, Carrasco FP, Scheel M, Nawabi J, Barkhof F. Commercial volumetric MRI reporting tools in multiple sclerosis: a systematic review of the evidence. Neuroradiology 2023; 65:5-24. [PMID: 36331588 PMCID: PMC9816195 DOI: 10.1007/s00234-022-03074-w] [Citation(s) in RCA: 5] [Impact Index Per Article: 5.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/13/2022] [Accepted: 09/29/2022] [Indexed: 11/06/2022]
Abstract
PURPOSE MRI is integral to the diagnosis of multiple sclerosis (MS) and is important for clinical prognostication. Quantitative volumetric reporting tools (QReports) can improve the accuracy and objectivity of MRI-based assessments. Several QReports are commercially available; however, validation can be difficult to establish and does not currently follow a common pathway. To aid evidence-based clinical decision-making, we performed a systematic review of commercial QReports for use in MS including technical details and published reports of validation and in-use evaluation. METHODS We categorized studies into three types of testing: technical validation, for example, comparison to manual segmentation, clinical validation by clinicians or interpretation of results alongside clinician-rated variables, and in-use evaluation, such as health economic assessment. RESULTS We identified 10 companies, which provide MS lesion and brain segmentation and volume quantification, and 38 relevant publications. Tools received regulatory approval between 2006 and 2020, contextualize results to normative reference populations, ranging from 620 to 8000 subjects, and require T1- and T2-FLAIR-weighted input sequences for longitudinal assessment of whole-brain volume and lesions. In MS, six QReports provided evidence of technical validation, four companies have conducted clinical validation by correlating results with clinical variables, only one has tested their QReport by clinician end-users, and one has performed a simulated in-use socioeconomic evaluation. CONCLUSION We conclude that there is limited evidence in the literature regarding clinical validation and in-use evaluation of commercial MS QReports with a particular lack of clinician end-user testing. Our systematic review provides clinicians and institutions with the available evidence when considering adopting a quantitative reporting tool for MS.
Collapse
Affiliation(s)
- Zoe Mendelsohn
- grid.83440.3b0000000121901201Neuroradiological Academic Unit, UCL Queen Square Institute of Neurology, University College London, London, UK ,grid.83440.3b0000000121901201Department of Medical Physics and Bioengineering, Centre for Medical Image Computing (CMIC), University College London, London, UK ,grid.83440.3b0000000121901201Department of Neuroinflammation, Queen Square Multiple Sclerosis Centre, UCL Institute of Neurology, University College London, London, UK ,grid.6363.00000 0001 2218 4662Department of Neuroradiology, Charité School of Medicine and University Hospital Berlin, Berlin, Germany ,grid.6363.00000 0001 2218 4662Department of Radiology, Charité School of Medicine and University Hospital Berlin, Berlin, Germany
| | - Hugh G. Pemberton
- grid.83440.3b0000000121901201Neuroradiological Academic Unit, UCL Queen Square Institute of Neurology, University College London, London, UK ,grid.83440.3b0000000121901201Department of Medical Physics and Bioengineering, Centre for Medical Image Computing (CMIC), University College London, London, UK ,grid.420685.d0000 0001 1940 6527GE Healthcare, Amersham, UK
| | - James Gray
- grid.416626.10000 0004 0391 2793Stepping Hill Hospital, NHS Foundation Trust, Stockport, UK
| | - Olivia Goodkin
- grid.83440.3b0000000121901201Neuroradiological Academic Unit, UCL Queen Square Institute of Neurology, University College London, London, UK ,grid.83440.3b0000000121901201Department of Medical Physics and Bioengineering, Centre for Medical Image Computing (CMIC), University College London, London, UK ,grid.83440.3b0000000121901201Department of Neuroinflammation, Queen Square Multiple Sclerosis Centre, UCL Institute of Neurology, University College London, London, UK
| | - Ferran Prados Carrasco
- grid.83440.3b0000000121901201Department of Medical Physics and Bioengineering, Centre for Medical Image Computing (CMIC), University College London, London, UK ,grid.83440.3b0000000121901201Department of Neuroinflammation, Queen Square Multiple Sclerosis Centre, UCL Institute of Neurology, University College London, London, UK ,grid.36083.3e0000 0001 2171 6620E-Health Centre, Universitat Oberta de Catalunya, Barcelona, Spain
| | - Michael Scheel
- grid.6363.00000 0001 2218 4662Department of Neuroradiology, Charité School of Medicine and University Hospital Berlin, Berlin, Germany
| | - Jawed Nawabi
- grid.6363.00000 0001 2218 4662Department of Radiology, Charité School of Medicine and University Hospital Berlin, Berlin, Germany ,grid.484013.a0000 0004 6879 971XBerlin Institute of Health at Charité – Universitätsmedizin Berlin, BIH Biomedical Innovation Academy, BIH Charité Digital Clinician Scientist Program, Berlin, Germany
| | - Frederik Barkhof
- grid.83440.3b0000000121901201Neuroradiological Academic Unit, UCL Queen Square Institute of Neurology, University College London, London, UK ,grid.83440.3b0000000121901201Department of Medical Physics and Bioengineering, Centre for Medical Image Computing (CMIC), University College London, London, UK ,grid.83440.3b0000000121901201Department of Neuroinflammation, Queen Square Multiple Sclerosis Centre, UCL Institute of Neurology, University College London, London, UK ,grid.12380.380000 0004 1754 9227Radiology & Nuclear Medicine, Amsterdam University Medical Centers, Vrije Universiteit, Amsterdam, The Netherlands
| |
Collapse
|
2
|
Gaubert M, Dell'Orco A, Lange C, Garnier-Crussard A, Zimmermann I, Dyrba M, Duering M, Ziegler G, Peters O, Preis L, Priller J, Spruth EJ, Schneider A, Fliessbach K, Wiltfang J, Schott BH, Maier F, Glanz W, Buerger K, Janowitz D, Perneczky R, Rauchmann BS, Teipel S, Kilimann I, Laske C, Munk MH, Spottke A, Roy N, Dobisch L, Ewers M, Dechent P, Haynes JD, Scheffler K, Düzel E, Jessen F, Wirth M. Performance evaluation of automated white matter hyperintensity segmentation algorithms in a multicenter cohort on cognitive impairment and dementia. Front Psychiatry 2022; 13:1010273. [PMID: 36713907 PMCID: PMC9877422 DOI: 10.3389/fpsyt.2022.1010273] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 08/02/2022] [Accepted: 12/07/2022] [Indexed: 01/15/2023] Open
Abstract
BACKGROUND White matter hyperintensities (WMH), a biomarker of small vessel disease, are often found in Alzheimer's disease (AD) and their advanced detection and quantification can be beneficial for research and clinical applications. To investigate WMH in large-scale multicenter studies on cognitive impairment and AD, appropriate automated WMH segmentation algorithms are required. This study aimed to compare the performance of segmentation tools and provide information on their application in multicenter research. METHODS We used a pseudo-randomly selected dataset (n = 50) from the DZNE-multicenter observational Longitudinal Cognitive Impairment and Dementia Study (DELCODE) that included 3D fluid-attenuated inversion recovery (FLAIR) images from participants across the cognitive continuum. Performances of top-rated algorithms for automated WMH segmentation [Brain Intensity Abnormality Classification Algorithm (BIANCA), lesion segmentation toolbox (LST), lesion growth algorithm (LGA), LST lesion prediction algorithm (LPA), pgs, and sysu_media] were compared to manual reference segmentation (RS). RESULTS Across tools, segmentation performance was moderate for global WMH volume and number of detected lesions. After retraining on a DELCODE subset, the deep learning algorithm sysu_media showed the highest performances with an average Dice's coefficient of 0.702 (±0.109 SD) for volume and a mean F1-score of 0.642 (±0.109 SD) for the number of lesions. The intra-class correlation was excellent for all algorithms (>0.9) but BIANCA (0.835). Performance improved with high WMH burden and varied across brain regions. CONCLUSION To conclude, the deep learning algorithm, when retrained, performed well in the multicenter context. Nevertheless, the performance was close to traditional methods. We provide methodological recommendations for future studies using automated WMH segmentation to quantify and assess WMH along the continuum of cognitive impairment and AD dementia.
Collapse
Affiliation(s)
- Malo Gaubert
- German Center for Neurodegenerative Diseases, Dresden, Germany.,Department of Neuroradiology, Rennes University Hospital (CHU), Rennes, France
| | - Andrea Dell'Orco
- German Center for Neurodegenerative Diseases, Dresden, Germany.,Department of Neuroradiology, Charité - Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin and Humboldt-Universität zu Berlin, Berlin, Germany
| | - Catharina Lange
- German Center for Neurodegenerative Diseases, Dresden, Germany.,Department of Nuclear Medicine, Charité - Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin and Humboldt-Universität zu Berlin, Berlin, Germany
| | - Antoine Garnier-Crussard
- Clinical and Research Memory Center of Lyon, Lyon Institute for Elderly, Hospices Civils de Lyon, Lyon, France.,Normandie Univ, UNICAEN, INSERM, U1237, PhIND "Physiopathology and Imaging of Neurological Disorders," Institut Blood and Brain @ Caen-Normandie, Caen, France.,Neuroscience Research Centre of Lyon, INSERM 1048, CNRS 5292, Lyon, France
| | | | - Martin Dyrba
- German Center for Neurodegenerative Diseases, Rostock, Germany
| | - Marco Duering
- Department of Biomedical Engineering, Medical Image Analysis Center (MIAC) and qbig, University of Basel, Basel, Switzerland
| | - Gabriel Ziegler
- German Center for Neurodegenerative Diseases, Magdeburg, Germany
| | - Oliver Peters
- German Center for Neurodegenerative Diseases, Berlin, Germany.,Department of Psychiatry, Charité - Universitätsmedizin Berlin, Berlin, Germany
| | - Lukas Preis
- Department of Psychiatry, Charité - Universitätsmedizin Berlin, Berlin, Germany
| | - Josef Priller
- German Center for Neurodegenerative Diseases, Berlin, Germany.,Department of Psychiatry and Psychotherapy, Charité - Universitätsmedizin Berlin, Berlin, Germany.,Centre for Clinical Brain Sciences, University of Edinburgh and UK Dementia Research Institute (DRI), Edinburgh, United Kingdom.,Department of Psychiatry and Psychotherapy, School of Medicine, Technical University of Munich, Munich, Germany
| | - Eike Jakob Spruth
- German Center for Neurodegenerative Diseases, Berlin, Germany.,Department of Psychiatry and Psychotherapy, Charité - Universitätsmedizin Berlin, Berlin, Germany
| | - Anja Schneider
- German Center for Neurodegenerative Diseases, Bonn, Germany.,Department of Neurodegenerative Disease and Geriatric Psychiatry, University of Bonn Medical Center, Bonn, Germany
| | - Klaus Fliessbach
- German Center for Neurodegenerative Diseases, Bonn, Germany.,Department of Neurodegenerative Disease and Geriatric Psychiatry, University of Bonn Medical Center, Bonn, Germany
| | - Jens Wiltfang
- German Center for Neurodegenerative Diseases, Göttingen, Germany.,Department of Psychiatry and Psychotherapy, University Medical Center Göttingen, University of Göttingen, Göttingen, Germany.,Department of Medical Sciences, Neurosciences and Signaling Group, Institute of Biomedicine (iBiMED), University of Aveiro, Aveiro, Portugal
| | - Björn H Schott
- German Center for Neurodegenerative Diseases, Göttingen, Germany.,Department of Psychiatry and Psychotherapy, University Medical Center Göttingen, University of Göttingen, Göttingen, Germany.,Leibniz Institute for Neurobiology, Magdeburg, Germany
| | - Franziska Maier
- Department of Psychiatry, Medical Faculty, University of Cologne, Cologne, Germany
| | - Wenzel Glanz
- German Center for Neurodegenerative Diseases, Magdeburg, Germany
| | - Katharina Buerger
- German Center for Neurodegenerative Diseases, Munich, Germany.,Department of Psychiatry and Psychotherapy, University Hospital, Ludwig Maximilian University of Munich (LMU), Munich, Germany
| | - Daniel Janowitz
- Department of Psychiatry and Psychotherapy, University Hospital, Ludwig Maximilian University of Munich (LMU), Munich, Germany
| | - Robert Perneczky
- German Center for Neurodegenerative Diseases, Munich, Germany.,Department of Psychiatry and Psychotherapy, University Hospital, Ludwig Maximilian University of Munich (LMU), Munich, Germany.,Munich Cluster for Systems Neurology (SyNergy) Munich, Munich, Germany.,Ageing Epidemiology Research Unit (AGE), School of Public Health, Imperial College London, London, United Kingdom.,Sheffield Institute for Translational Neuroscience (SITraN), University of Sheffield, Sheffield, United Kingdom
| | - Boris-Stephan Rauchmann
- Department of Psychiatry and Psychotherapy, University Hospital, Ludwig Maximilian University of Munich (LMU), Munich, Germany
| | - Stefan Teipel
- German Center for Neurodegenerative Diseases, Rostock, Germany.,Department of Psychosomatic Medicine, Rostock University Medical Center, Rostock, Germany
| | - Ingo Kilimann
- German Center for Neurodegenerative Diseases, Rostock, Germany.,Department of Psychosomatic Medicine, Rostock University Medical Center, Rostock, Germany
| | - Christoph Laske
- German Center for Neurodegenerative Diseases, Tübingen, Germany.,Department of Psychiatry and Psychotherapy, University of Tübingen, Tübingen, Germany
| | - Matthias H Munk
- German Center for Neurodegenerative Diseases, Tübingen, Germany.,Department of Psychiatry and Psychotherapy, University of Tübingen, Tübingen, Germany
| | - Annika Spottke
- German Center for Neurodegenerative Diseases, Bonn, Germany.,Department of Neurology, University of Bonn, Bonn, Germany
| | - Nina Roy
- German Center for Neurodegenerative Diseases, Bonn, Germany
| | - Laura Dobisch
- German Center for Neurodegenerative Diseases, Magdeburg, Germany
| | - Michael Ewers
- German Center for Neurodegenerative Diseases, Munich, Germany
| | - Peter Dechent
- MR-Research in Neurosciences, Department of Cognitive Neurology, Georg-August-University of Göttingen, Göttingen, Germany
| | - John Dylan Haynes
- Bernstein Center for Computational Neuroscience, Charité - Universitätsmedizin, Berlin, Germany
| | - Klaus Scheffler
- Department for Biomedical Magnetic Resonance, University of Tübingen, Tübingen, Germany
| | - Emrah Düzel
- German Center for Neurodegenerative Diseases, Magdeburg, Germany.,Institute of Cognitive Neurology and Dementia Research (IKND), Otto-von-Guericke University, Magdeburg, Germany
| | - Frank Jessen
- German Center for Neurodegenerative Diseases, Bonn, Germany.,Department of Psychiatry, Medical Faculty, University of Cologne, Cologne, Germany.,Excellence Cluster on Cellular Stress Responses in Aging-Associated Diseases (CECAD), University of Cologne, Köln, Germany
| | - Miranka Wirth
- German Center for Neurodegenerative Diseases, Dresden, Germany
| | | |
Collapse
|
3
|
Hotz I, Deschwanden PF, Liem F, Mérillat S, Malagurski B, Kollias S, Jäncke L. Performance of three freely available methods for extracting white matter hyperintensities: FreeSurfer, UBO Detector, and BIANCA. Hum Brain Mapp 2021; 43:1481-1500. [PMID: 34873789 PMCID: PMC8886667 DOI: 10.1002/hbm.25739] [Citation(s) in RCA: 9] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/23/2021] [Revised: 11/11/2021] [Accepted: 11/26/2021] [Indexed: 11/07/2022] Open
Abstract
White matter hyperintensities (WMH) of presumed vascular origin are frequently found in MRIs of healthy older adults. WMH are also associated with aging and cognitive decline. Here, we compared and validated three algorithms for WMH extraction: FreeSurfer (T1w), UBO Detector (T1w + FLAIR), and FSL's Brain Intensity AbNormality Classification Algorithm (BIANCA; T1w + FLAIR) using a longitudinal dataset comprising MRI data of cognitively healthy older adults (baseline N = 231, age range 64–87 years). As reference we manually segmented WMH in T1w, three‐dimensional (3D) FLAIR, and two‐dimensional (2D) FLAIR images which were used to assess the segmentation accuracy of the different automated algorithms. Further, we assessed the relationships of WMH volumes provided by the algorithms with Fazekas scores and age. FreeSurfer underestimated the WMH volumes and scored worst in Dice Similarity Coefficient (DSC = 0.434) but its WMH volumes strongly correlated with the Fazekas scores (rs = 0.73). BIANCA accomplished the highest DSC (0.602) in 3D FLAIR images. However, the relations with the Fazekas scores were only moderate, especially in the 2D FLAIR images (rs = 0.41), and many outlier WMH volumes were detected when exploring within‐person trajectories (2D FLAIR: ~30%). UBO Detector performed similarly to BIANCA in DSC with both modalities and reached the best DSC in 2D FLAIR (0.531) without requiring a tailored training dataset. In addition, it achieved very high associations with the Fazekas scores (2D FLAIR: rs = 0.80). In summary, our results emphasize the importance of carefully contemplating the choice of the WMH segmentation algorithm and MR‐modality.
Collapse
Affiliation(s)
- Isabel Hotz
- Division of Neuropsychology, Department of Psychology, University of Zurich, Zurich, Switzerland.,University Research Priority Program (URPP), Dynamics of Healthy Aging, University of Zurich, Zurich, Switzerland
| | | | - Franziskus Liem
- University Research Priority Program (URPP), Dynamics of Healthy Aging, University of Zurich, Zurich, Switzerland
| | - Susan Mérillat
- University Research Priority Program (URPP), Dynamics of Healthy Aging, University of Zurich, Zurich, Switzerland
| | - Brigitta Malagurski
- University Research Priority Program (URPP), Dynamics of Healthy Aging, University of Zurich, Zurich, Switzerland
| | - Spyros Kollias
- Department of Neuroradiology, University Hospital Zurich, Zurich, Switzerland
| | - Lutz Jäncke
- Division of Neuropsychology, Department of Psychology, University of Zurich, Zurich, Switzerland.,University Research Priority Program (URPP), Dynamics of Healthy Aging, University of Zurich, Zurich, Switzerland
| |
Collapse
|
4
|
Sundaresan V, Zamboni G, Rothwell PM, Jenkinson M, Griffanti L. Triplanar ensemble U-Net model for white matter hyperintensities segmentation on MR images. Med Image Anal 2021; 73:102184. [PMID: 34325148 PMCID: PMC8505759 DOI: 10.1016/j.media.2021.102184] [Citation(s) in RCA: 18] [Impact Index Per Article: 6.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/29/2020] [Revised: 03/10/2021] [Accepted: 07/16/2021] [Indexed: 01/05/2023]
Abstract
White matter hyperintensities (WMHs) have been associated with various cerebrovascular and neurodegenerative diseases. Reliable quantification of WMHs is essential for understanding their clinical impact in normal and pathological populations. Automated segmentation of WMHs is highly challenging due to heterogeneity in WMH characteristics between deep and periventricular white matter, presence of artefacts and differences in the pathology and demographics of populations. In this work, we propose an ensemble triplanar network that combines the predictions from three different planes of brain MR images to provide an accurate WMH segmentation. In the loss functions the network uses anatomical information regarding WMH spatial distribution in loss functions, to improve the efficiency of segmentation and to overcome the contrast variations between deep and periventricular WMHs. We evaluated our method on 5 datasets, of which 3 are part of a publicly available dataset (training data for MICCAI WMH Segmentation Challenge 2017 - MWSC 2017) consisting of subjects from three different cohorts, and we also submitted our method to MWSC 2017 to be evaluated on the unseen test datasets. On evaluating our method separately in deep and periventricular regions, we observed robust and comparable performance in both regions. Our method performed better than most of the existing methods, including FSL BIANCA, and on par with the top ranking deep learning methods of MWSC 2017.
Collapse
Affiliation(s)
- Vaanathi Sundaresan
- Wellcome Centre for Integrative Neuroimaging, Oxford Centre for Functional MRI of the Brain, Nuffield Department of Clinical Neurosciences, University of Oxford, UK
- Oxford-Nottingham Centre for Doctoral Training in Biomedical Imaging, University of Oxford, UK
- Oxford India Centre for Sustainable Development, Somerville College, University of Oxford, UK
| | - Giovanna Zamboni
- Wellcome Centre for Integrative Neuroimaging, Oxford Centre for Functional MRI of the Brain, Nuffield Department of Clinical Neurosciences, University of Oxford, UK
- Centre for Prevention of Stroke and Dementia, Nuffield Department of Clinical Neurosciences, University of Oxford, UK
- Dipartimento di Scienze Biomediche, Metaboliche e Neuroscienze, Universitá di Modena e Reggio Emilia, Italy
| | - Peter M. Rothwell
- Centre for Prevention of Stroke and Dementia, Nuffield Department of Clinical Neurosciences, University of Oxford, UK
| | - Mark Jenkinson
- Wellcome Centre for Integrative Neuroimaging, Oxford Centre for Functional MRI of the Brain, Nuffield Department of Clinical Neurosciences, University of Oxford, UK
- Australian Institute for Machine Learning (AIML), School of Computer Science, The University of Adelaide, Adelaide, Australia
| | - Ludovica Griffanti
- Wellcome Centre for Integrative Neuroimaging, Oxford Centre for Functional MRI of the Brain, Nuffield Department of Clinical Neurosciences, University of Oxford, UK
- Wellcome Centre for Integrative Neuroimaging, Oxford Centre for Human Brain Activity, Department of Psychiatry, University of Oxford, Oxford, UK
| |
Collapse
|
5
|
Comparison of Multispectral Image-Processing Methods for Brain Tissue Classification in BrainWeb Synthetic Data and Real MR Images. BIOMED RESEARCH INTERNATIONAL 2021; 2021:9820145. [PMID: 33748284 PMCID: PMC7959972 DOI: 10.1155/2021/9820145] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 07/27/2020] [Revised: 01/28/2021] [Accepted: 02/08/2021] [Indexed: 11/30/2022]
Abstract
Accurate quantification of brain tissue is a fundamental and challenging task in neuroimaging. Over the past two decades, statistical parametric mapping (SPM) and FMRIB's Automated Segmentation Tool (FAST) have been widely used to estimate gray matter (GM) and white matter (WM) volumes. However, they cannot reliably estimate cerebrospinal fluid (CSF) volumes. To address this problem, we developed the TRIO algorithm (TRIOA), a new magnetic resonance (MR) multispectral classification method. SPM8, SPM12, FAST, and the TRIOA were evaluated using the BrainWeb database and real magnetic resonance imaging (MRI) data. In this paper, the MR brain images of 140 healthy volunteers (51.5 ± 15.8 y/o) were obtained using a whole-body 1.5 T MRI system (Aera, Siemens, Erlangen, Germany). Before classification, several preprocessing steps were performed, including skull stripping and motion and inhomogeneity correction. After extensive experimentation, the TRIOA was shown to be more effective than SPM and FAST. For real data, all test methods revealed that the participants aged 20–83 years exhibited an age-associated decline in GM and WM volume fractions. However, for CSF volume estimation, SPM8-s and SPM12-m both produced different results, which were also different compared with those obtained by FAST and the TRIOA. Furthermore, the TRIOA performed consistently better than both SPM and FAST for GM, WM, and CSF volume estimation. Compared with SPM and FAST, the proposed TRIOA showed more advantages by providing more accurate MR brain tissue classification and volume measurements, specifically in CSF volume estimation.
Collapse
|
6
|
Gryska E, Schneiderman J, Björkman-Burtscher I, Heckemann RA. Automatic brain lesion segmentation on standard magnetic resonance images: a scoping review. BMJ Open 2021; 11:e042660. [PMID: 33514580 PMCID: PMC7849889 DOI: 10.1136/bmjopen-2020-042660] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 07/11/2020] [Revised: 01/09/2021] [Accepted: 01/12/2021] [Indexed: 12/11/2022] Open
Abstract
OBJECTIVES Medical image analysis practices face challenges that can potentially be addressed with algorithm-based segmentation tools. In this study, we map the field of automatic MR brain lesion segmentation to understand the clinical applicability of prevalent methods and study designs, as well as challenges and limitations in the field. DESIGN Scoping review. SETTING Three databases (PubMed, IEEE Xplore and Scopus) were searched with tailored queries. Studies were included based on predefined criteria. Emerging themes during consecutive title, abstract, methods and whole-text screening were identified. The full-text analysis focused on materials, preprocessing, performance evaluation and comparison. RESULTS Out of 2990 unique articles identified through the search, 441 articles met the eligibility criteria, with an estimated growth rate of 10% per year. We present a general overview and trends in the field with regard to publication sources, segmentation principles used and types of lesions. Algorithms are predominantly evaluated by measuring the agreement of segmentation results with a trusted reference. Few articles describe measures of clinical validity. CONCLUSIONS The observed reporting practices leave room for improvement with a view to studying replication, method comparison and clinical applicability. To promote this improvement, we propose a list of recommendations for future studies in the field.
Collapse
Affiliation(s)
- Emilia Gryska
- Medical Radiation Sciences, Goteborgs universitet Institutionen for kliniska vetenskaper, Goteborg, Sweden
| | - Justin Schneiderman
- Sektionen för klinisk neurovetenskap, Goteborgs Universitet Institutionen for Neurovetenskap och fysiologi, Goteborg, Sweden
| | | | - Rolf A Heckemann
- Medical Radiation Sciences, Goteborgs universitet Institutionen for kliniska vetenskaper, Goteborg, Sweden
| |
Collapse
|
7
|
Brune S, Høgestøl EA, Cengija V, Berg-Hansen P, Sowa P, Nygaard GO, Harbo HF, Beyer MK. LesionQuant for Assessment of MRI in Multiple Sclerosis-A Promising Supplement to the Visual Scan Inspection. Front Neurol 2020; 11:546744. [PMID: 33362682 PMCID: PMC7759639 DOI: 10.3389/fneur.2020.546744] [Citation(s) in RCA: 4] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/29/2020] [Accepted: 11/23/2020] [Indexed: 11/17/2022] Open
Abstract
Background and Goals: Multiple sclerosis (MS) is a central nervous system inflammatory disease where magnetic resonance imaging (MRI) is an important tool for diagnosis and disease monitoring. Quantitative measurements of lesion volume, lesion count, distribution of lesions, and brain atrophy have a potentially significant value for evaluating disease progression. We hypothesize that utilizing software designed for evaluating MRI data in MS will provide more accurate and detailed analyses compared to the visual neuro-radiological evaluation. Methods: A group of 56 MS patients (mean age 35 years, 70% females and 96% relapsing-remitting MS) was examined with brain MRI one and 5 years after diagnosis. The T1 and FLAIR brain MRI sequences for all patients were analyzed using the LesionQuant (LQ) software. These data were compared with data from structured visual evaluations of the MRI scans performed by neuro-radiologists, including assessments of atrophy, and lesion count. The data from LQ were also compared with data from other validated research methods for brain segmentation, including assessments of whole brain volume and lesion volume. Correlations with clinical tests like the timed 25-foot walk test (T25FT) were performed to explore additional value of LQ analyses. Results: Lesion count assessments by LQ and by the neuro-radiologist were significantly correlated one year (cor = 0.92, p = 2.2 × 10−16) and 5 years (cor = 0.84, p = 2.7 × 10−16) after diagnosis. Analyzes of the intra- and interrater variability also correlated significantly (cor = 0.96, p < 0.001, cor = 0.97, p < 0.001). Significant positive correlation was found between lesion volume measured by LQ and by the software Cascade (cor = 0.7, p < 0.001. LQ detected a reduction in whole brain percentile >10 in 10 patients across the time-points, whereas the neuro-radiologist assessment identified six of these. The neuro-radiologist additionally identified five patients with increased atrophy in the follow-up period, all of them displayed decreasing low whole brain percentiles (median 11, range 8–28) in the LQ analysis. Significant positive correlation was identified between lesion volume measured by LQ and test performance on the T25FT both at 1 and 5 years after diagnosis. Conclusion: For the number of MS lesions at both time-points, we demonstrated strong correlations between the assessments done by LQ and the neuro-radiologist. Lesion volume evaluated with LQ correlated with T25FT performance. LQ-analyses classified more patients to have brain atrophy than the visual neuro-radiological evaluation. In conclusion, LQ seems like a promising supplement to the evaluation performed by neuro-radiologists, providing an automated tool for evaluating lesions in MS patients and also detecting early signs of atrophy in both a longitudinal and cross-sectional setting.
Collapse
Affiliation(s)
- Synne Brune
- Institute of Clinical Medicine, University of Oslo, Oslo, Norway.,Department of Neurology, Oslo University Hospital, Oslo, Norway
| | - Einar A Høgestøl
- Institute of Clinical Medicine, University of Oslo, Oslo, Norway.,Department of Neurology, Oslo University Hospital, Oslo, Norway
| | - Vanja Cengija
- Division of Radiology and Nuclear Medicine, Oslo University Hospital, Oslo, Norway
| | - Pål Berg-Hansen
- Institute of Clinical Medicine, University of Oslo, Oslo, Norway.,Department of Neurology, Oslo University Hospital, Oslo, Norway
| | - Piotr Sowa
- Division of Radiology and Nuclear Medicine, Oslo University Hospital, Oslo, Norway
| | - Gro O Nygaard
- Department of Neurology, Oslo University Hospital, Oslo, Norway
| | - Hanne F Harbo
- Institute of Clinical Medicine, University of Oslo, Oslo, Norway.,Department of Neurology, Oslo University Hospital, Oslo, Norway
| | - Mona K Beyer
- Institute of Clinical Medicine, University of Oslo, Oslo, Norway.,Division of Radiology and Nuclear Medicine, Oslo University Hospital, Oslo, Norway
| |
Collapse
|
8
|
Ribaldi F, Altomare D, Jovicich J, Ferrari C, Picco A, Pizzini FB, Soricelli A, Mega A, Ferretti A, Drevelegas A, Bosch B, Müller BW, Marra C, Cavaliere C, Bartrés-Faz D, Nobili F, Alessandrini F, Barkhof F, Gros-Dagnac H, Ranjeva JP, Wiltfang J, Kuijer J, Sein J, Hoffmann KT, Roccatagliata L, Parnetti L, Tsolaki M, Constantinidis M, Aiello M, Salvatore M, Montalti M, Caulo M, Didic M, Bargallo N, Blin O, Rossini PM, Schonknecht P, Floridi P, Payoux P, Visser PJ, Bordet R, Lopes R, Tarducci R, Bombois S, Hensch T, Fiedler U, Richardson JC, Frisoni GB, Marizzoni M. Accuracy and reproducibility of automated white matter hyperintensities segmentation with lesion segmentation tool: A European multi-site 3T study. Magn Reson Imaging 2020; 76:108-115. [PMID: 33220450 DOI: 10.1016/j.mri.2020.11.008] [Citation(s) in RCA: 20] [Impact Index Per Article: 5.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/22/2020] [Revised: 10/02/2020] [Accepted: 11/14/2020] [Indexed: 01/18/2023]
Abstract
Brain vascular damage accumulate in aging and often manifest as white matter hyperintensities (WMHs) on MRI. Despite increased interest in automated methods to segment WMHs, a gold standard has not been achieved and their longitudinal reproducibility has been poorly investigated. The aim of present work is to evaluate accuracy and reproducibility of two freely available segmentation algorithms. A harmonized MRI protocol was implemented in 3T-scanners across 13 European sites, each scanning five volunteers twice (test-retest) using 2D-FLAIR. Automated segmentation was performed using Lesion segmentation tool algorithms (LST): the Lesion growth algorithm (LGA) in SPM8 and 12 and the Lesion prediction algorithm (LPA). To assess reproducibility, we applied the LST longitudinal pipeline to the LGA and LPA outputs for both the test and retest scans. We evaluated volumetric and spatial accuracy comparing LGA and LPA with manual tracing, and for reproducibility the test versus retest. Median volume difference between automated WMH and manual segmentations (mL) was -0.22[IQR = 0.50] for LGA-SPM8, -0.12[0.57] for LGA-SPM12, -0.09[0.53] for LPA, while the spatial accuracy (Dice Coefficient) was 0.29[0.31], 0.33[0.26] and 0.41[0.23], respectively. The reproducibility analysis showed a median reproducibility error of 20%[IQR = 41] for LGA-SPM8, 14% [31] for LGA-SPM12 and 10% [27] with the LPA cross-sectional pipeline. Applying the LST longitudinal pipeline, the reproducibility errors were considerably reduced (LGA: 0%[IQR = 0], p < 0.001; LPA: 0% [3], p < 0.001) compared to those derived using the cross-sectional algorithms. The DC using the longitudinal pipeline was excellent (median = 1) for LGA [IQR = 0] and LPA [0.02]. LST algorithms showed moderate accuracy and good reproducibility. Therefore, it can be used as a reliable cross-sectional and longitudinal tool in multi-site studies.
Collapse
Affiliation(s)
- Federica Ribaldi
- Laboratory of Alzheimer's Neuroimaging and Alzheimer's Epidemiology, IRCCS Istituto Centro San Giovanni di Dio Fatebenefratelli, Brescia, Italy; Department of Molecular and Translational Medicine, University of Brescia, Brescia, Italy; Laboratory of Neuroimaging of Aging (LANVIE), University of Geneva, Geneva, Switzerland; Memory Clinic, Geneva University Hospitals, Geneva, Switzerland.
| | - Daniele Altomare
- Laboratory of Neuroimaging of Aging (LANVIE), University of Geneva, Geneva, Switzerland; Memory Clinic, Geneva University Hospitals, Geneva, Switzerland
| | - Jorge Jovicich
- Center for Mind/Brain Sciences (CIMEC), University of Trento, Rovereto, Italy
| | - Clarissa Ferrari
- Unit of Statistics, IRCCS Istituto Centro San Giovanni di Dio Fatebenefratelli, Brescia, Italy
| | - Agnese Picco
- Department of Neuroscience, Ophthalmology, Genetics and Mother-Child Health (DINOGMI), University of Genoa, Genoa, Italy
| | | | | | - Anna Mega
- Laboratory of Alzheimer's Neuroimaging and Alzheimer's Epidemiology, IRCCS Istituto Centro San Giovanni di Dio Fatebenefratelli, Brescia, Italy
| | - Antonio Ferretti
- Department of Neuroscience Imaging and Clinical Sciences, University "G. d'Annunzio" of Chieti, Italy; Institute for Advanced Biomedical Technologies (ITAB), University "G. d'Annunzio" of Chieti, Italy
| | - Antonios Drevelegas
- Interbalkan Medical Center of Thessaloniki, Thessaloniki, Greece; Department of Radiology, Aristotle University of Thessaloniki, Thessaloniki, Greece
| | - Beatriz Bosch
- Department of Psychiatry and Clinical Psychobiology, Universitat de Barcelona and IDIBAPS, Barcelona, Spain
| | - Bernhard W Müller
- LVR-Clinic for Psychiatry and Psychotherapy, Institutes and Clinics of the University Duisburg-Essen, Essen, Germany
| | - Camillo Marra
- Center for Neuropsychological Research, Catholic University, Rome, Italy
| | | | - David Bartrés-Faz
- Department of Psychiatry and Clinical Psychobiology, Universitat de Barcelona and IDIBAPS, Barcelona, Spain
| | - Flavio Nobili
- Dept. of Neuroscience (DINOGMI), University of Genoa, Italy; IRCCS Ospedale Policlinico San Martino Genova, Italy
| | - Franco Alessandrini
- Radiology, Dept. of Diagnostic and Public Health, Verona University, Verona, Italy
| | - Frederik Barkhof
- Centre for Medical Image Computing, Department of Medical Physics and Biomedical Engineering, University College London, London, UK; Department of Radiology and Nuclear Medicine, Amsterdam University Medical Centers, Amsterdam, the Netherlands
| | - Helene Gros-Dagnac
- ToNIC, Toulouse NeuroImaging Center, Université de Toulouse, Inserm, UPS, France; Université Toulouse 3 Paul Sabatier, UMR 825 Imagerie Cérébrale et Handicaps Neurologiques, F-31024 Toulouse, France
| | - Jean-Philippe Ranjeva
- Institut de Neurosciences de la Timone (INT), Aix-Marseille Université, CNRS, UMR 7289, 13005 Marseille, France
| | - Jens Wiltfang
- Department of Psychiatry and Psychotherapy, University Medical Center (UMG), Georg-August University, Göttingen, Germany
| | - Joost Kuijer
- Department of Radiology and Nuclear Medicine, Amsterdam University Medical Centers, Amsterdam, the Netherlands
| | - Julien Sein
- Institut de Neurosciences de la Timone (INT), Aix-Marseille Université, CNRS, UMR 7289, 13005 Marseille, France
| | | | - Luca Roccatagliata
- IRCCS Ospedale Policlinico San Martino Genova, Italy; Dept. of Health Sciences (DISSAL), University of Genoa, Italy
| | - Lucilla Parnetti
- Section of Neurology, Centre for Memory Disturbances, University of Perugia, Perugia, Italy
| | - Magda Tsolaki
- 1st Department of Neurology, Aristotle University of Thessaloniki, Makedonia, Greece
| | | | | | | | - Martina Montalti
- Laboratory of Alzheimer's Neuroimaging and Alzheimer's Epidemiology, IRCCS Istituto Centro San Giovanni di Dio Fatebenefratelli, Brescia, Italy
| | - Massimo Caulo
- Department of Neuroscience Imaging and Clinical Sciences, University "G. d'Annunzio" of Chieti, Italy; Institute for Advanced Biomedical Technologies (ITAB), University "G. d'Annunzio" of Chieti, Italy
| | - Mira Didic
- APHM, Timone, Service de Neurologie et Neuropsychologie, APHM Hôpital Timone Adultes, Marseille, France; Aix Marseille Univ, INSERM, INS, Inst Neurosci Syst, Marseille, France
| | - Núria Bargallo
- Department of Neuroradiology and Magnetic Resonance Image Core Facility, Hospital Clínic de Barcelona, IDIBAPS, Barcelona, Spain
| | - Olivier Blin
- Aix Marseille University, UMR-INSERM 1106, Service de Pharmacologie Clinique, AP-HM, Marseille, France
| | - Paolo M Rossini
- Dept. Neuroscience & Neurorehabilitation, IRCCS-San Raffaele-Pisana, Rome, Italy
| | - Peter Schonknecht
- Department of Psychiatry and Psychotherapy, University of Leipzig Medical Center, Leipzig, Germany
| | - Piero Floridi
- Neuroradiology Unit, Perugia General Hospital, Perugia, Italy
| | - Pierre Payoux
- ToNIC, Toulouse NeuroImaging Center, Université de Toulouse, Inserm, UPS, France
| | - Pieter Jelle Visser
- Alzheimer Center Amsterdam, Department of Neurology, Amsterdam Neuroscience, Vrije Universiteit Amsterdam, Amsterdam UMC, Amsterdam, The Netherlands
| | - Régis Bordet
- Univ. Lille, INSERM, CHU Lille, Lille Neuroscience & Cognition - Degenerative and Vascular Cognitive Disorders-U1172. F-59000 Lille, France
| | - Renaud Lopes
- Univ. Lille, INSERM, CHU Lille, Lille Neuroscience & Cognition - Degenerative and Vascular Cognitive Disorders-U1172. F-59000 Lille, France
| | | | - Stephanie Bombois
- Univ. Lille, INSERM, CHU Lille, Lille Neuroscience & Cognition - Degenerative and Vascular Cognitive Disorders-U1172. F-59000 Lille, France
| | - Tilman Hensch
- Department of Psychiatry and Psychotherapy, University of Leipzig Medical Center, Leipzig, Germany
| | - Ute Fiedler
- LVR-Clinic for Psychiatry and Psychotherapy, Institutes and Clinics of the University Duisburg-Essen, Essen, Germany
| | - Jill C Richardson
- Neurosciences Therapeutic Area, GlaxoSmithKline R&D, Gunnels Wood Road, Stevenage, United Kingdom
| | - Giovanni B Frisoni
- Laboratory of Neuroimaging of Aging (LANVIE), University of Geneva, Geneva, Switzerland; Memory Clinic, Geneva University Hospitals, Geneva, Switzerland
| | - Moira Marizzoni
- Laboratory of Alzheimer's Neuroimaging and Alzheimer's Epidemiology, IRCCS Istituto Centro San Giovanni di Dio Fatebenefratelli, Brescia, Italy
| |
Collapse
|
9
|
Heinen R, Steenwijk MD, Barkhof F, Biesbroek JM, van der Flier WM, Kuijf HJ, Prins ND, Vrenken H, Biessels GJ, de Bresser J. Performance of five automated white matter hyperintensity segmentation methods in a multicenter dataset. Sci Rep 2019; 9:16742. [PMID: 31727919 PMCID: PMC6856351 DOI: 10.1038/s41598-019-52966-0] [Citation(s) in RCA: 31] [Impact Index Per Article: 6.2] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/31/2019] [Accepted: 10/22/2019] [Indexed: 11/23/2022] Open
Abstract
White matter hyperintensities (WMHs) are a common manifestation of cerebral small vessel disease, that is increasingly studied with large, pooled multicenter datasets. This data pooling increases statistical power, but poses challenges for automated WMH segmentation. Although there is extensive literature on the evaluation of automated WMH segmentation methods, such evaluations in a multicenter setting are lacking. We performed WMH segmentations in sixty patients scanned on six different magnetic resonance imaging (MRI) scanners (10 patients per scanner) using five freely available and fully-automated WMH segmentation methods (Cascade, kNN-TTP, Lesion-TOADS, LST-LGA and LST-LPA). Different MRI scanner vendors and field strengths were included. We compared these automated WMH segmentations with manual WMH segmentations as a reference. Performance of each method both within and across scanners was assessed using spatial and volumetric correspondence with the reference segmentations by Dice's similarity coefficient (DSC) and intra-class correlation coefficient (ICC) respectively. We found the best performance, both within and across scanners, for kNN-TTP, followed by LST-LPA and LST-LGA, with worse performance for Lesion-TOADS and Cascade. Our findings can serve as a guide for choosing a method and also highlight the importance to further improve and evaluate consistency of methods in a multicenter setting.
Collapse
Affiliation(s)
- Rutger Heinen
- Department of Neurology and Neurosurgery, UMC Utrecht Brain Center, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands.
| | - Martijn D Steenwijk
- Department of Anatomy and Neurosciences, Amsterdam Neuroscience, Vrije Universiteit Amsterdam, Amsterdam UMC, Amsterdam, The Netherlands
- Department of Radiology and Nuclear Medicine, Amsterdam Neuroscience, Vrije Universiteit Amsterdam, Amsterdam UMC, Amsterdam, The Netherlands
| | - Frederik Barkhof
- Department of Radiology and Nuclear Medicine, Amsterdam Neuroscience, Vrije Universiteit Amsterdam, Amsterdam UMC, Amsterdam, The Netherlands
- Institutes of Neurology & Healthcare Engineering, University College London (UCL), London, United Kingdom
| | - J Matthijs Biesbroek
- Department of Neurology and Neurosurgery, UMC Utrecht Brain Center, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands
| | - Wiesje M van der Flier
- Alzheimer Center & Department of Neurology, Vrije Universiteit Amsterdam, Amsterdam UMC, Amsterdam, The Netherlands
- Department of Epidemiology and Biostatistics, Vrije Universiteit Amsterdam, Amsterdam UMC, Amsterdam, The Netherlands
| | - Hugo J Kuijf
- Image Sciences Institute, University Medical Center Utrecht, Utrecht, The Netherlands
| | - Niels D Prins
- Alzheimer Center & Department of Neurology, Vrije Universiteit Amsterdam, Amsterdam UMC, Amsterdam, The Netherlands
| | - Hugo Vrenken
- Department of Anatomy and Neurosciences, Amsterdam Neuroscience, Vrije Universiteit Amsterdam, Amsterdam UMC, Amsterdam, The Netherlands
- Department of Radiology and Nuclear Medicine, Amsterdam Neuroscience, Vrije Universiteit Amsterdam, Amsterdam UMC, Amsterdam, The Netherlands
| | - Geert Jan Biessels
- Department of Neurology and Neurosurgery, UMC Utrecht Brain Center, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands
| | - Jeroen de Bresser
- Department of Radiology, Leiden University Medical Center, Leiden, The Netherlands
| |
Collapse
|
10
|
Høgestøl EA, Kaufmann T, Nygaard GO, Beyer MK, Sowa P, Nordvik JE, Kolskår K, Richard G, Andreassen OA, Harbo HF, Westlye LT. Cross-Sectional and Longitudinal MRI Brain Scans Reveal Accelerated Brain Aging in Multiple Sclerosis. Front Neurol 2019; 10:450. [PMID: 31114541 PMCID: PMC6503038 DOI: 10.3389/fneur.2019.00450] [Citation(s) in RCA: 41] [Impact Index Per Article: 8.2] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/20/2018] [Accepted: 04/12/2019] [Indexed: 11/13/2022] Open
Abstract
Multiple sclerosis (MS) is an inflammatory disorder of the central nervous system. By combining longitudinal MRI-based brain morphometry and brain age estimation using machine learning, we tested the hypothesis that MS patients have higher brain age relative to chronological age than healthy controls (HC) and that longitudinal rate of brain aging in MS patients is associated with clinical course and severity. Seventy-six MS patients [71% females, mean age 34.8 years (range 21-49) at inclusion] were examined with brain MRI at three time points with a mean total follow up period of 4.4 years (±0.4 years). We used additional cross-sectional MRI data from 235 HC for case-control comparison. We applied a machine learning model trained on an independent set of 3,208 HC to estimate individual brain age and to calculate the difference between estimated and chronological age, termed brain age gap (BAG). We also assessed the longitudinal change rate in BAG in individuals with MS. MS patients showed significantly higher BAG (4.4 ± 6.6 years) compared to HC (Cohen's D = 0.69, p = 4.0 × 10-6). Longitudinal estimates of BAG in MS patients showed high reliability and suggested an accelerated rate of brain aging corresponding to an annual increase of 0.41 (SE = 0.15) years compared to chronological aging (p = 0.008). Multiple regression analyses revealed higher rate of brain aging in patients with more brain atrophy (Cohen's D = 0.86, p = 4.3 × 10-15) and increased white matter lesion load (WMLL) (Cohen's D = 0.55, p = 0.015). On average, patients with MS had significantly higher BAG compared to HC. Progressive brain aging in patients with MS was related to brain atrophy and increased WMLL. No significant clinical associations were found in our sample, future studies are warranted on this matter. Brain age estimation is a promising method for evaluation of subtle brain changes in MS, which is important for predicting clinical outcome and guide choice of intervention.
Collapse
Affiliation(s)
| | - Tobias Kaufmann
- NORMENT, Division of Mental Health and Addiction, Oslo University Hospital & Institute of Clinical Medicine, University of Oslo, Oslo, Norway
| | - Gro O. Nygaard
- Department of Neurology, Oslo University Hospital, Oslo, Norway
| | - Mona K. Beyer
- Institute of Clinical Medicine, University of Oslo, Oslo, Norway
- Division of Radiology and Nuclear Medicine, Oslo University Hospital, Oslo, Norway
| | - Piotr Sowa
- Division of Radiology and Nuclear Medicine, Oslo University Hospital, Oslo, Norway
| | | | - Knut Kolskår
- NORMENT, Division of Mental Health and Addiction, Oslo University Hospital & Institute of Clinical Medicine, University of Oslo, Oslo, Norway
- Sunnaas Rehabilitation Hospital HT, Nesodden, Norway
- Department of Psychology, University of Oslo, Oslo, Norway
| | - Geneviève Richard
- NORMENT, Division of Mental Health and Addiction, Oslo University Hospital & Institute of Clinical Medicine, University of Oslo, Oslo, Norway
- Sunnaas Rehabilitation Hospital HT, Nesodden, Norway
- Department of Psychology, University of Oslo, Oslo, Norway
| | - Ole A. Andreassen
- NORMENT, Division of Mental Health and Addiction, Oslo University Hospital & Institute of Clinical Medicine, University of Oslo, Oslo, Norway
| | - Hanne F. Harbo
- Institute of Clinical Medicine, University of Oslo, Oslo, Norway
- Department of Neurology, Oslo University Hospital, Oslo, Norway
| | - Lars T. Westlye
- NORMENT, Division of Mental Health and Addiction, Oslo University Hospital & Institute of Clinical Medicine, University of Oslo, Oslo, Norway
- Department of Psychology, University of Oslo, Oslo, Norway
| |
Collapse
|
11
|
Symptoms of fatigue and depression is reflected in altered default mode network connectivity in multiple sclerosis. PLoS One 2019; 14:e0210375. [PMID: 30933977 PMCID: PMC6443168 DOI: 10.1371/journal.pone.0210375] [Citation(s) in RCA: 17] [Impact Index Per Article: 3.4] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/19/2018] [Accepted: 03/19/2019] [Indexed: 01/08/2023] Open
Abstract
Background Fatigue and depression are frequent and often co-occurring symptoms in multiple sclerosis (MS). Resting-state functional magnetic resonance imaging (rs-fMRI) represents a promising tool for disentangling differential associations between depression and fatigue and brain network function and connectivity. In this study we tested for associations between symptoms of fatigue and depression and DMN connectivity in patients with MS. Materials and methods Seventy-four MS patients were included on average 14 months after diagnosis. They underwent MRI scanning of the brain including rs-fMRI, and symptoms of fatigue and depression were assessed with Fatigue Severity Scale (FSS) and Beck Depression Inventory II (BDI). A principal component analysis (PCA) on FSS and BDI scores was performed, and the component scores were analysed using linear regression models to test for associations with default mode network (DMN) connectivity. Results We observed higher DMN connectivity with higher scores on the primary principal component reflecting common symptom burden for fatigue and depression (Cohen’s f2 = 0.075, t = 2.17, p = 0.03). The secondary principal component reflecting a pattern of low fatigue scores with high scores of depression was associated with lower DMN connectivity (Cohen’s f2 = 0.067, t = -2.1, p = 0.04). Using continuous mean scores of FSS we also observed higher DMN connectivity with higher symptom burden (t = 3.1, p = 0.003), but no significant associations between continuous sum scores of BDI and DMN connectivity (t = 0.8, p = 0.4). Conclusion Multivariate decomposition of FSS and BDI data supported both overlapping and unique manifestation of fatigue and depression in MS patients. Rs-fMRI analyses showed that symptoms of fatigue and depression were reflected in altered DMN connectivity, and that higher DMN activity was seen in MS patients with fatigue even with low depression scores.
Collapse
|
12
|
Validation and Optimization of BIANCA for the Segmentation of Extensive White Matter Hyperintensities. Neuroinformatics 2019; 16:269-281. [PMID: 29594711 DOI: 10.1007/s12021-018-9372-2] [Citation(s) in RCA: 14] [Impact Index Per Article: 2.8] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/17/2022]
Abstract
White matter hyperintensities (WMH) are a hallmark of small vessel diseases (SVD). Yet, no automated segmentation method is readily and widely used, especially in patients with extensive WMH where lesions are close to the cerebral cortex. BIANCA (Brain Intensity AbNormality Classification Algorithm) is a new fully automated, supervised method for WMH segmentation. In this study, we optimized and compared BIANCA against a reference method with manual editing in a cohort of patients with extensive WMH. This was achieved in two datasets: a clinical protocol with 90 patients having 2-dimensional FLAIR and an advanced protocol with 66 patients having 3-dimensional FLAIR. We first determined simultaneously which input modalities (FLAIR alone or FLAIR + T1) and which training sets were better compared to the reference. Three strategies for the selection of the threshold that is applied to the probabilistic output of BIANCA were then evaluated: chosen at the group level, based on Fazekas score or determined individually. Accuracy of the segmentation was assessed through measures of spatial agreement and volumetric correspondence with respect to reference segmentation. Based on all our tests, we identified multimodal inputs (FLAIR + T1), mixed WMH load training set and individual threshold selection as the best conditions to automatically segment WMH in our cohort. A median Dice similarity index of 0.80 (0.80) and an intraclass correlation coefficient of 0.97 (0.98) were obtained for the clinical (advanced) protocol. However, Bland-Altman plots identified a difference with the reference method that was linearly related to the total burden of WMH. Our results suggest that BIANCA is a reliable and fast segmentation method to extract masks of WMH in patients with extensive lesions.
Collapse
|
13
|
Bartlett E, Shaw M, Schwarz C, Feinberg C, DeLorenzo C, Krupp LB, Charvet LE. Brief Computer-Based Information Processing Measures are Linked to White Matter Integrity in Pediatric-Onset Multiple Sclerosis. J Neuroimaging 2018; 29:140-150. [DOI: 10.1111/jon.12566] [Citation(s) in RCA: 7] [Impact Index Per Article: 1.2] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/06/2018] [Revised: 09/18/2018] [Accepted: 09/19/2018] [Indexed: 12/13/2022] Open
Affiliation(s)
- Elizabeth Bartlett
- Department of Biomedical Engineering; Stony Brook University; Stony Brook NY
| | - Michael Shaw
- Department of Neurology, New York University Langone Medical Center; NYU Langone Health; New York NY
| | - Colleen Schwarz
- Department of Nursing; Stony Brook University; Stony Brook NY
| | - Charles Feinberg
- Department of Neurology, New York University Langone Medical Center; NYU Langone Health; New York NY
| | - Christine DeLorenzo
- Department of Biomedical Engineering; Stony Brook University; Stony Brook NY
- Department of Psychiatry; Stony Brook University; Stony Brook NY
| | - Lauren B. Krupp
- Department of Neurology, New York University Langone Medical Center; NYU Langone Health; New York NY
| | - Leigh E. Charvet
- Department of Neurology, New York University Langone Medical Center; NYU Langone Health; New York NY
| |
Collapse
|
14
|
Nylander R, Kilander L, Ahlström H, Lind L, Larsson EM. Small Vessel Disease on Neuroimaging in a 75-Year-Old Cohort (PIVUS): Comparison With Cognitive and Executive Tests. Front Aging Neurosci 2018; 10:217. [PMID: 30061827 PMCID: PMC6054972 DOI: 10.3389/fnagi.2018.00217] [Citation(s) in RCA: 6] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/10/2018] [Accepted: 06/26/2018] [Indexed: 12/28/2022] Open
Abstract
Background and Purpose: Signs of small vessel disease (SVD) are commonly seen on magnetic resonance imaging (MRI) of the brain in cognitively healthy elderly individuals, and the clinical relevance of these are often unclear. We have previously described three different MRI manifestations of SVD as well as cerebral perfusion in a longitudinal study of non-demented 75-year-old subjects. The purpose of the present study was to evaluate the relationship of these findings to cognition and executive function at age 75 and changes after 5 years. Methods: In all, 406 subjects from the Prospective Investigation of the Vasculature in Uppsala Seniors (PIVUS) study were examined with MRI of the brain at age 75 years. Two-hundred and fifty of the subjects were re-examined 5 years later. White matter hyperintensities (WMHs) and lacunar infarcts (LIs) were assessed on both occasions, but microbleeds (MBs) and perfusion only at age 75. Cognitive function was screened by the Mini Mental State Examination (MMSE). Trail Making Test A and B (TMT-A and TMT-B) were performed at baseline and at follow-up at age 80. Results: At baseline, 93% performed >27 points in the MMSE. The TMT-B at age 75 was significantly related to WMH visual scoring after adjustment for sex, education and cerebrovascular disease risk factors (+80 s (95% CI 0.3-161 s), P < 0.05 for grade 2-3 vs. grade 0). Neither MMSE nor TMT-A was significantly related to WMH scoring. There was no relation between any test performance and WMH volume, white matter volume, number of MBs or brain perfusion at age 75. Subjects who had sustained a new LI (n = 26) showed a greater increase of the time to perform TMT-A at the 5-year follow-up (+25 s vs. +4 s in LI-free subjects, P = 0.003). Changes in MMSE or TMT-A and -B test performance between ages 75 and 80 were not related to changes in WMH scoring or volume during the 5 years follow-up, or to brain perfusion at age 75. Conclusion: In this cognitively healthy community-based population, moderate-severe WMHs and incident LIs on brain MRI in individuals aged 75-80 years were associated with a mild impairment of processing speed and executive function.
Collapse
Affiliation(s)
- Ruta Nylander
- Department of Surgical Sciences, Radiology, Uppsala University, Uppsala, Sweden
| | - Lena Kilander
- Department of Public Health and Caring Sciences, Geriatrics, Uppsala University, Uppsala, Sweden
| | - Håkan Ahlström
- Department of Surgical Sciences, Radiology, Uppsala University, Uppsala, Sweden
| | - Lars Lind
- Department of Medical Sciences, Uppsala University, Uppsala, Sweden
| | - Elna-Marie Larsson
- Department of Surgical Sciences, Radiology, Uppsala University, Uppsala, Sweden
| |
Collapse
|
15
|
Jiang J, Liu T, Zhu W, Koncz R, Liu H, Lee T, Sachdev PS, Wen W. UBO Detector – A cluster-based, fully automated pipeline for extracting white matter hyperintensities. Neuroimage 2018; 174:539-549. [DOI: 10.1016/j.neuroimage.2018.03.050] [Citation(s) in RCA: 40] [Impact Index Per Article: 6.7] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/09/2018] [Revised: 03/16/2018] [Accepted: 03/21/2018] [Indexed: 11/27/2022] Open
|
16
|
Nylander R, Fahlström M, Rostrup E, Kullberg J, Damangir S, Ahlström H, Lind L, Larsson EM. Quantitative and qualitative MRI evaluation of cerebral small vessel disease in an elderly population: a longitudinal study. Acta Radiol 2018; 59:612-618. [PMID: 28814098 DOI: 10.1177/0284185117727567] [Citation(s) in RCA: 25] [Impact Index Per Article: 4.2] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/20/2022]
Abstract
Background Cerebral white matter hyperintensities (WMHs), lacunes, and microbleeds are seen on magnetic resonance imaging (MRI) in small vessel disease (SVD). Purpose To assess SVD on MRI and its evolution over five years in an elderly population and to investigate whether relative cerebral blood flow (rCBF) at baseline was related to the progression of white matter (WM) lesions. Material and Methods In a population-based study, 406 participants aged 75 years underwent morphological MRI of the brain and 252 of them again at age 80 years. At age 75 years, a perfusion scan was also done. WMHs were evaluated qualitatively (visual scoring) and quantitatively (CASCADE software). Lacunes and microbleeds were counted. Results A significant progression of the WMH score and WMH volume occurred over five years ( P < 0.0001). New lacunes were seen in 10%. Participants with new lacunes at age 80 years showed a more pronounced increase in WMHs (P < 0.0001). Microbleeds were present in 14% at age 75 years. The visual WMH score was significantly associated with the presence of microbleeds ( P < 0.0001). There was no relationship between total WM rCBF and WMH volume at age 75 years, and no significant associations between regional or total rCBF at age 75 years and changes in WMH volume over five years. The total WM and GM volume decreased significantly between the ages of 75 and 80 years ( P < 0.0001). Conclusion MRI manifestations of SVD progressed over five years in an elderly population (age range = 75-80 years). rCBF was not associated with WMH volume or progression of WMH volume.
Collapse
Affiliation(s)
- Ruta Nylander
- Department of Surgical Sciences, Radiology, Uppsala University, Sweden
| | - Markus Fahlström
- Department of Surgical Sciences, Radiology, Uppsala University, Sweden
| | - Egill Rostrup
- Functional Imaging Unit, Department of Diagnostics, Glostrup hospital, University of Copenhagen, Denmark
| | - Joel Kullberg
- Department of Surgical Sciences, Radiology, Uppsala University, Sweden
| | - Soheil Damangir
- Department of Neurobiology, Care Sciences and Society, Karolinska Institutet, Stockholm, Sweden
| | - Håkan Ahlström
- Department of Surgical Sciences, Radiology, Uppsala University, Sweden
| | - Lars Lind
- Department of Medical Sciences, Uppsala University, Sweden
| | | |
Collapse
|
17
|
Rincón M, Díaz-López E, Selnes P, Vegge K, Altmann M, Fladby T, Bjørnerud A. Improved Automatic Segmentation of White Matter Hyperintensities in MRI Based on Multilevel Lesion Features. Neuroinformatics 2018; 15:231-245. [PMID: 28378263 DOI: 10.1007/s12021-017-9328-y] [Citation(s) in RCA: 12] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/25/2022]
Abstract
Brain white matter hyperintensities (WMHs) are linked to increased risk of cerebrovascular and neurodegenerative diseases among the elderly. Consequently, detection and characterization of WMHs are of significant clinical importance. We propose a novel approach for WMH segmentation from multi-contrast MRI where both voxel-based and lesion-based information are used to improve overall performance in both volume-oriented and object-oriented metrics. Our segmentation method (AMOS-2D) consists of four stages following a "generate-and-test" approach: pre-processing, Gaussian white matter (WM) modelling, hierarchical multi-threshold WMH segmentation and object-based WMH filtering using support vector machines. Data from 28 subjects was used in this study covering a wide range of lesion loads. Volumetric T1-weighted images and 2D fluid attenuated inversion recovery (FLAIR) images were used as basis for the WM model and lesion masks defined manually in each subject by experts were used for training and evaluating the proposed method. The method obtained an average agreement (in terms of the Dice similarity coefficient, DSC) with experts equivalent to inter-expert agreement both in terms of WMH number (DSC = 0.637 vs. 0.651) and volume (DSC = 0.743 vs. 0.781). It allowed higher accuracy in detecting WMH compared to alternative methods tested and was further found to be insensitive to WMH lesion burden. Good agreement with expert annotations combined with stable performance largely independent of lesion burden suggests that AMOS-2D will be a valuable tool for fully automated WMH segmentation in patients with cerebrovascular and neurodegenerative pathologies.
Collapse
Affiliation(s)
- M Rincón
- Department of Artificial Intelligence, UNED, Madrid, Spain.
| | - E Díaz-López
- Department of Artificial Intelligence, UNED, Madrid, Spain
| | - P Selnes
- Department of Neurology, Akershus University Hospital, Oslo, Norway
| | - K Vegge
- Department of Radiology, Akershus University Hospital, Oslo, Norway
| | - M Altmann
- Department of Neurology, Akershus University Hospital, Oslo, Norway
| | - T Fladby
- Department of Neurology, Akershus University Hospital, Oslo, Norway
| | - A Bjørnerud
- The Intervention Centre, Oslo University Hospital, Oslo, Norway
- Department of Physics, University of Oslo, Oslo, Norway
| |
Collapse
|
18
|
Guerrero R, Qin C, Oktay O, Bowles C, Chen L, Joules R, Wolz R, Valdés-Hernández MC, Dickie DA, Wardlaw J, Rueckert D. White matter hyperintensity and stroke lesion segmentation and differentiation using convolutional neural networks. NEUROIMAGE-CLINICAL 2017. [PMID: 29527496 PMCID: PMC5842732 DOI: 10.1016/j.nicl.2017.12.022] [Citation(s) in RCA: 104] [Impact Index Per Article: 14.9] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Indexed: 12/21/2022]
Abstract
White matter hyperintensities (WMH) are a feature of sporadic small vessel disease also frequently observed in magnetic resonance images (MRI) of healthy elderly subjects. The accurate assessment of WMH burden is of crucial importance for epidemiological studies to determine association between WMHs, cognitive and clinical data; their causes, and the effects of new treatments in randomized trials. The manual delineation of WMHs is a very tedious, costly and time consuming process, that needs to be carried out by an expert annotator (e.g. a trained image analyst or radiologist). The problem of WMH delineation is further complicated by the fact that other pathological features (i.e. stroke lesions) often also appear as hyperintense regions. Recently, several automated methods aiming to tackle the challenges of WMH segmentation have been proposed. Most of these methods have been specifically developed to segment WMH in MRI but cannot differentiate between WMHs and strokes. Other methods, capable of distinguishing between different pathologies in brain MRI, are not designed with simultaneous WMH and stroke segmentation in mind. Therefore, a task specific, reliable, fully automated method that can segment and differentiate between these two pathological manifestations on MRI has not yet been fully identified. In this work we propose to use a convolutional neural network (CNN) that is able to segment hyperintensities and differentiate between WMHs and stroke lesions. Specifically, we aim to distinguish between WMH pathologies from those caused by stroke lesions due to either cortical, large or small subcortical infarcts. The proposed fully convolutional CNN architecture, called uResNet, that comprised an analysis path, that gradually learns low and high level features, followed by a synthesis path, that gradually combines and up-samples the low and high level features into a class likelihood semantic segmentation. Quantitatively, the proposed CNN architecture is shown to outperform other well established and state-of-the-art algorithms in terms of overlap with manual expert annotations. Clinically, the extracted WMH volumes were found to correlate better with the Fazekas visual rating score than competing methods or the expert-annotated volumes. Additionally, a comparison of the associations found between clinical risk-factors and the WMH volumes generated by the proposed method, was found to be in line with the associations found with the expert-annotated volumes. Robust, fully automatic white matter hyperintensity and stroke lesion segmentation and differentiation A novel patch sampling strategy used during CNN training that avoids the introduction of erroneous locality assumptions Improved segmentation accuracy in terms of Dice scores when compared to well established state-of-the-art methods
Collapse
Affiliation(s)
- R Guerrero
- Department of Computing, Imperial College London, UK.
| | - C Qin
- Department of Computing, Imperial College London, UK
| | - O Oktay
- Department of Computing, Imperial College London, UK
| | - C Bowles
- Department of Computing, Imperial College London, UK
| | - L Chen
- Department of Computing, Imperial College London, UK
| | | | - R Wolz
- IXICO plc., UK; Department of Computing, Imperial College London, UK
| | - M C Valdés-Hernández
- UK Dementia Research Institute at The University of Edinburgh, Edinburgh Medical School, 47 Little France Crescent, Edinburgh EH16 4TJ, UK
| | - D A Dickie
- UK Dementia Research Institute at The University of Edinburgh, Edinburgh Medical School, 47 Little France Crescent, Edinburgh EH16 4TJ, UK
| | - J Wardlaw
- UK Dementia Research Institute at The University of Edinburgh, Edinburgh Medical School, 47 Little France Crescent, Edinburgh EH16 4TJ, UK
| | - D Rueckert
- Department of Computing, Imperial College London, UK
| |
Collapse
|
19
|
Sindi S, Kåreholt I, Spulber G, Soininen H, Kivipelto M, Solomon A. Midlife Work-Related Stress is Associated with Late-Life Gray Matter Volume Atrophy. J Alzheimers Dis Rep 2017; 1:219-227. [PMID: 30480239 PMCID: PMC6159714 DOI: 10.3233/adr-170035] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/30/2022] Open
Abstract
Background: Work-related stress has been associated with an increased dementia risk. However, less is known about the mechanisms that underlie these associations. Objective: The goal is to examine associations between midlife work-related stress and late-life structural brain alterations. Methods: The Cardiovascular Risk Factors, Aging, and Dementia (CAIDE) study participants were randomly selected from independent population-based surveys (mean age 50) in Finland. MRI measurements included gray matter (GM) volume, white matter lesions (WML) and medial temporal atrophy (MTA) (1st re-examination, n = 102); and GM volume, hippocampal volume, WML volume, cortical thickness, and MTA (2nd re-examination, n = 64). Work-related stress comprised a score from two questions administered in midlife. Results: Higher levels of midlife work-related stress were associated with lower GM volume (β= –0.077, p = 0.033) at the first re-examination, even after adjusting for several confounders. No significant associations were found with MTA, WML, or MRI measurements at the second re-examination. Conclusion: Previously shown associations of midlife work-related stress with dementia risk may be at least partly explained by associations with lower GM volume in late-life. The lack of associations at the second re-examination may indicate a critical time window for the effects of midlife work-related stress, and/or selective survival/participation.
Collapse
Affiliation(s)
- Shireen Sindi
- Aging Research Center, Karolinska Institutet and Stockholm University, Stockholm, Sweden.,Karolinska Institutet Center for Alzheimer Research, Stockholm, Sweden.,Neuroepidemiology and Ageing Research Unit, School of Public Health, Imperial College London, London, UK
| | - Ingemar Kåreholt
- Aging Research Center, Karolinska Institutet and Stockholm University, Stockholm, Sweden.,Institute of Gerontology, School of Health and Welfare, Jönköping University, Jönköping, Sweden
| | - Gabriela Spulber
- Karolinska Institutet Center for Alzheimer Research, Stockholm, Sweden
| | - Hilkka Soininen
- Department of Neurology, Institute of Clinical Medicine, University of Eastern Finland, Kuopio, Finland.,Department of Neurology, University of Eastern Finland, and Kuopio University Hospital, Kuopio, Finland
| | - Miia Kivipelto
- Karolinska Institutet Center for Alzheimer Research, Stockholm, Sweden.,Neuroepidemiology and Ageing Research Unit, School of Public Health, Imperial College London, London, UK.,Department of Neurology, Institute of Clinical Medicine, University of Eastern Finland, Kuopio, Finland.,Department of Chronic Disease Prevention, National Institute for Health and Welfare, Helsinki, Finland.,Department of Geriatrics, Karolinska University Hospital, Stockholm, Sweden
| | - Alina Solomon
- Aging Research Center, Karolinska Institutet and Stockholm University, Stockholm, Sweden.,Karolinska Institutet Center for Alzheimer Research, Stockholm, Sweden.,Department of Neurology, Institute of Clinical Medicine, University of Eastern Finland, Kuopio, Finland.,Department of Geriatrics, Karolinska University Hospital, Stockholm, Sweden
| |
Collapse
|
20
|
de Sitter A, Steenwijk MD, Ruet A, Versteeg A, Liu Y, van Schijndel RA, Pouwels PJW, Kilsdonk ID, Cover KS, van Dijk BW, Ropele S, Rocca MA, Yiannakas M, Wattjes MP, Damangir S, Frisoni GB, Sastre-Garriga J, Rovira A, Enzinger C, Filippi M, Frederiksen J, Ciccarelli O, Kappos L, Barkhof F, Vrenken H. Performance of five research-domain automated WM lesion segmentation methods in a multi-center MS study. Neuroimage 2017; 163:106-114. [PMID: 28899746 DOI: 10.1016/j.neuroimage.2017.09.011] [Citation(s) in RCA: 24] [Impact Index Per Article: 3.4] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/24/2017] [Revised: 08/31/2017] [Accepted: 09/06/2017] [Indexed: 11/17/2022] Open
Abstract
BACKGROUND AND PURPOSE In vivoidentification of white matter lesions plays a key-role in evaluation of patients with multiple sclerosis (MS). Automated lesion segmentation methods have been developed to substitute manual outlining, but evidence of their performance in multi-center investigations is lacking. In this work, five research-domain automated segmentation methods were evaluated using a multi-center MS dataset. METHODS 70 MS patients (median EDSS of 2.0 [range 0.0-6.5]) were included from a six-center dataset of the MAGNIMS Study Group (www.magnims.eu) which included 2D FLAIR and 3D T1 images with manual lesion segmentation as a reference. Automated lesion segmentations were produced using five algorithms: Cascade; Lesion Segmentation Toolbox (LST) with both the Lesion growth algorithm (LGA) and the Lesion prediction algorithm (LPA); Lesion-Topology preserving Anatomical Segmentation (Lesion-TOADS); and k-Nearest Neighbor with Tissue Type Priors (kNN-TTP). Main software parameters were optimized using a training set (N = 18), and formal testing was performed on the remaining patients (N = 52). To evaluate volumetric agreement with the reference segmentations, intraclass correlation coefficient (ICC) as well as mean difference in lesion volumes between the automated and reference segmentations were calculated. The Similarity Index (SI), False Positive (FP) volumes and False Negative (FN) volumes were used to examine spatial agreement. All analyses were repeated using a leave-one-center-out design to exclude the center of interest from the training phase to evaluate the performance of the method on 'unseen' center. RESULTS Compared to the reference mean lesion volume (4.85 ± 7.29 mL), the methods displayed a mean difference of 1.60 ± 4.83 (Cascade), 2.31 ± 7.66 (LGA), 0.44 ± 4.68 (LPA), 1.76 ± 4.17 (Lesion-TOADS) and -1.39 ± 4.10 mL (kNN-TTP). The ICCs were 0.755, 0.713, 0.851, 0.806 and 0.723, respectively. Spatial agreement with reference segmentations was higher for LPA (SI = 0.37 ± 0.23), Lesion-TOADS (SI = 0.35 ± 0.18) and kNN-TTP (SI = 0.44 ± 0.14) than for Cascade (SI = 0.26 ± 0.17) or LGA (SI = 0.31 ± 0.23). All methods showed highly similar results when used on data from a center not used in software parameter optimization. CONCLUSION The performance of the methods in this multi-center MS dataset was moderate, but appeared to be robust even with new datasets from centers not included in training the automated methods.
Collapse
Affiliation(s)
- Alexandra de Sitter
- Department of Radiology and Nuclear Medicine, VUmc, Amsterdam, The Netherlands.
| | | | - Aurélie Ruet
- Department of Neurology, CHU-Bordeaux, Bordeaux, France; University of Bordeaux, Bordeaux, France; Inserm U-1215 Magendie Neurocenter-Pathophysiology of Neural Plasticity, CHU-Bordeaux, Bordeaux, France
| | - Adriaan Versteeg
- Department of Radiology and Nuclear Medicine, VUmc, Amsterdam, The Netherlands
| | - Yaou Liu
- Department of Radiology and Nuclear Medicine, VUmc, Amsterdam, The Netherlands
| | | | - Petra J W Pouwels
- Department of Radiology and Nuclear Medicine, VUmc, Amsterdam, The Netherlands
| | - Iris D Kilsdonk
- Department of Radiology and Nuclear Medicine, VUmc, Amsterdam, The Netherlands
| | - Keith S Cover
- Department of Radiology and Nuclear Medicine, VUmc, Amsterdam, The Netherlands
| | - Bob W van Dijk
- Department of Anatomy and Neuroscience, VUmc, Amsterdam, The Netherlands
| | - Stefan Ropele
- Department of Neurology, Medical University of Graz, Graz, Austria
| | - Maria A Rocca
- Neuroimaging Research Unit, Institute of Experimental Neurology, Division of Neuroscience, San Raffaele Scientific Institute, UniSR, Milan, Italy
| | - Marios Yiannakas
- Department of Neuroinflammation, Institute of Neurology, UCL, London, UK
| | - Mike P Wattjes
- Department of Radiology and Nuclear Medicine, VUmc, Amsterdam, The Netherlands
| | - Soheil Damangir
- Department of Neurobiology, Care Sciences and Society, Karolinska Institute, Stockholm, Sweden
| | - Giovanni B Frisoni
- Laboratory of Epidemiology, Neuroimaging and Telemedicine, IRCCS Centro "S. Giovanni di Dio-F.B.F.", Brescia, Italy; Memory Clinic and LANVIE - Laboratory of Neuroimaging of Aging, HUG, Geneva, Switzerland
| | - Jaume Sastre-Garriga
- Centre d'Esclerosi Múltiple de Catalunya (Cemcat), Department of Neurology/Neuroimmunology, VHIR, Barcelona, Spain
| | - Alex Rovira
- Magnetic Resonance Unit, Department of Radiology (IDI), VHIR, Barcelona, Spain
| | - Christian Enzinger
- Department of Neurology, Medical University of Graz, Graz, Austria; Division of Neuroradiology, Vascular and Interventional Radiology, Department of Radiology, Medical University of Graz, Graz, Austria
| | - Massimo Filippi
- Neuroimaging Research Unit, Institute of Experimental Neurology, Division of Neuroscience, San Raffaele Scientific Institute, UniSR, Milan, Italy
| | - Jette Frederiksen
- Department of Neurology, Glostrup University Hospital, Copenhagen, Denmark
| | - Olga Ciccarelli
- UK/NIHR UCL-UCLH Biomedical Research Centre, Institute of Neurology, UCL, London, UK
| | - Ludwig Kappos
- Neurologic Clinic and Policlinic, University Hospital, University of Basel, Switzerland
| | - Frederik Barkhof
- Department of Radiology and Nuclear Medicine, VUmc, Amsterdam, The Netherlands; Institutes of Neurology & Healthcare Engineering, UCL, London, UK
| | - Hugo Vrenken
- Department of Radiology and Nuclear Medicine, VUmc, Amsterdam, The Netherlands; Department of Anatomy and Neuroscience, VUmc, Amsterdam, The Netherlands
| |
Collapse
|
21
|
Ferreira D, Machado A, Molina Y, Nieto A, Correia R, Westman E, Barroso J. Cognitive Variability during Middle-Age: Possible Association with Neurodegeneration and Cognitive Reserve. Front Aging Neurosci 2017. [PMID: 28649200 PMCID: PMC5465264 DOI: 10.3389/fnagi.2017.00188] [Citation(s) in RCA: 45] [Impact Index Per Article: 6.4] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/14/2023] Open
Abstract
Objective: Increased variability in cognition with age has been argued as an indication of pathological processes. Focusing on early detection of neurodegenerative disorders, we investigated variability in cognition in healthy middle-aged adults. In order to understand possible determinants of this variability, we also investigated associations with cognitive reserve, neuroimaging markers, subjective memory complaints, depressive symptomatology, and gender. Method: Thirty-one 50 ± 2 years old individuals were investigated as target group and deviation was studied in comparison to a reference younger group of 30 individuals 40 ± 2 years old. Comprehensive neuropsychological and structural imaging protocols were collected. Brain regional volumes and cortical thickness were calculated with FreeSurfer, white matter hyperintensities with CASCADE, and mean diffusivity with FSL. Results: Across-individuals variability showed greater dispersion in lexical access, processing speed, executive functions, and memory. Variability in global cognition correlated with, reduced cortical thickness in the right parietal-temporal-occipital association cortex, and increased mean diffusivity in the cingulum bundle and right inferior fronto-occipital fasciculus. A trend was also observed for the correlation between global cognition and hippocampal volume and female gender. All these associations were influenced by cognitive reserve. No correlations were found with subjective memory complaints, white matter hyperintensities and depressive symptomatology. Across-domains and across-tasks variability was greater in several executive components and cognitive processing speed. Conclusion: Variability in cognition during middle-age is associated with neurodegeneration in the parietal–temporal–occipital association cortex and white matter tracts connecting this to the prefrontal dorsolateral cortex and the hippocampus. Moreover, this effect is influenced by cognitive reserve. Studying variability offers valuable information showing that differences do not occur in the same magnitude and direction across individuals, cognitive domains and tasks. These findings may have important implications for early detection of subtle cognitive impairment and clinical interpretation of deviation from normality.
Collapse
Affiliation(s)
- Daniel Ferreira
- Division of Clinical Geriatrics-Center for Alzheimer Research, Department of Neurobiology, Care Sciences and Society, Karolinska InstitutetStockholm, Sweden.,Faculty of Psychology, University of La LagunaLa Laguna, Spain
| | - Alejandra Machado
- Division of Clinical Geriatrics-Center for Alzheimer Research, Department of Neurobiology, Care Sciences and Society, Karolinska InstitutetStockholm, Sweden.,Faculty of Psychology, University of La LagunaLa Laguna, Spain
| | - Yaiza Molina
- Faculty of Psychology, University of La LagunaLa Laguna, Spain.,Faculty of Health Sciences, University Fernando Pessoa CanariasLas Palmas, Spain
| | - Antonieta Nieto
- Faculty of Psychology, University of La LagunaLa Laguna, Spain
| | - Rut Correia
- Faculty of Psychology, University of La LagunaLa Laguna, Spain.,Facultad de Educación, Universidad Diego PortalesSantiago, Chile
| | - Eric Westman
- Division of Clinical Geriatrics-Center for Alzheimer Research, Department of Neurobiology, Care Sciences and Society, Karolinska InstitutetStockholm, Sweden
| | - José Barroso
- Faculty of Psychology, University of La LagunaLa Laguna, Spain
| |
Collapse
|
22
|
Damangir S, Westman E, Simmons A, Vrenken H, Wahlund LO, Spulber G. Reproducible segmentation of white matter hyperintensities using a new statistical definition. MAGMA (NEW YORK, N.Y.) 2017; 30:227-237. [PMID: 27943055 PMCID: PMC5440501 DOI: 10.1007/s10334-016-0599-3] [Citation(s) in RCA: 7] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 09/14/2016] [Revised: 11/17/2016] [Accepted: 11/19/2016] [Indexed: 12/25/2022]
Abstract
OBJECTIVES We present a method based on a proposed statistical definition of white matter hyperintensities (WMH), which can work with any combination of conventional magnetic resonance (MR) sequences without depending on manually delineated samples. MATERIALS AND METHODS T1-weighted, T2-weighted, FLAIR, and PD sequences acquired at 1.5 Tesla from 119 subjects from the Kings Health Partners-Dementia Case Register (healthy controls, mild cognitive impairment, Alzheimer's disease) were used. The segmentation was performed using a proposed definition for WMH based on the one-tailed Kolmogorov-Smirnov test. RESULTS The presented method was verified, given all possible combinations of input sequences, against manual segmentations and a high similarity (Dice 0.85-0.91) was observed. Comparing segmentations with different input sequences to one another also yielded a high similarity (Dice 0.83-0.94) that exceeded intra-rater similarity (Dice 0.75-0.91). We compared the results with those of other available methods and showed that the segmentation based on the proposed definition has better accuracy and reproducibility in the test dataset used. CONCLUSION Overall, the presented definition is shown to produce accurate results with higher reproducibility than manual delineation. This approach can be an alternative to other manual or automatic methods not only because of its accuracy, but also due to its good reproducibility.
Collapse
Affiliation(s)
- Soheil Damangir
- Department of Neurobiology, Care Sciences and Society, Karolinska Institutet, Hälsovägen 7, Huddinge, 14157, Stockholm, Sweden.
| | - Eric Westman
- Department of Neurobiology, Care Sciences and Society, Karolinska Institutet, Hälsovägen 7, Huddinge, 14157, Stockholm, Sweden
| | - Andrew Simmons
- Department of Neurobiology, Care Sciences and Society, Karolinska Institutet, Hälsovägen 7, Huddinge, 14157, Stockholm, Sweden
- Institute of Psychiatry, King's College London, Box P089, De Crespigny Park, London, SE5 8AF, UK
| | - Hugo Vrenken
- Department of Physics and Medical Technology, VU University Medical Center, De Boelelaan 1118, 1081HZ, Amsterdam, The Netherlands
- Department of Radiology and Nuclear Medicine, VU University Medical Center, De Boelelaan 1118, 1081HZ, Amsterdam, The Netherlands
| | - Lars-Olof Wahlund
- Department of Neurobiology, Care Sciences and Society, Karolinska Institutet, Hälsovägen 7, Huddinge, 14157, Stockholm, Sweden
| | - Gabriela Spulber
- Department of Neurobiology, Care Sciences and Society, Karolinska Institutet, Hälsovägen 7, Huddinge, 14157, Stockholm, Sweden
| |
Collapse
|
23
|
Griffanti L, Zamboni G, Khan A, Li L, Bonifacio G, Sundaresan V, Schulz UG, Kuker W, Battaglini M, Rothwell PM, Jenkinson M. BIANCA (Brain Intensity AbNormality Classification Algorithm): A new tool for automated segmentation of white matter hyperintensities. Neuroimage 2016; 141:191-205. [PMID: 27402600 PMCID: PMC5035138 DOI: 10.1016/j.neuroimage.2016.07.018] [Citation(s) in RCA: 239] [Impact Index Per Article: 29.9] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/10/2016] [Revised: 07/06/2016] [Accepted: 07/07/2016] [Indexed: 12/21/2022] Open
Abstract
Reliable quantification of white matter hyperintensities of presumed vascular origin (WMHs) is increasingly needed, given the presence of these MRI findings in patients with several neurological and vascular disorders, as well as in elderly healthy subjects. We present BIANCA (Brain Intensity AbNormality Classification Algorithm), a fully automated, supervised method for WMH detection, based on the k-nearest neighbour (k-NN) algorithm. Relative to previous k-NN based segmentation methods, BIANCA offers different options for weighting the spatial information, local spatial intensity averaging, and different options for the choice of the number and location of the training points. BIANCA is multimodal and highly flexible so that the user can adapt the tool to their protocol and specific needs. We optimised and validated BIANCA on two datasets with different MRI protocols and patient populations (a "predominantly neurodegenerative" and a "predominantly vascular" cohort). BIANCA was first optimised on a subset of images for each dataset in terms of overlap and volumetric agreement with a manually segmented WMH mask. The correlation between the volumes extracted with BIANCA (using the optimised set of options), the volumes extracted from the manual masks and visual ratings showed that BIANCA is a valid alternative to manual segmentation. The optimised set of options was then applied to the whole cohorts and the resulting WMH volume estimates showed good correlations with visual ratings and with age. Finally, we performed a reproducibility test, to evaluate the robustness of BIANCA, and compared BIANCA performance against existing methods. Our findings suggest that BIANCA, which will be freely available as part of the FSL package, is a reliable method for automated WMH segmentation in large cross-sectional cohort studies.
Collapse
Affiliation(s)
- Ludovica Griffanti
- Centre for the Functional MRI of the Brain (FMRIB), Nuffield Department of Clinical Neurosciences, University of Oxford, UK.
| | - Giovanna Zamboni
- Centre for the Functional MRI of the Brain (FMRIB), Nuffield Department of Clinical Neurosciences, University of Oxford, UK; Centre for Prevention of Stroke and Dementia, Nuffield Department of Clinical Neurosciences, University of Oxford, UK
| | - Aamira Khan
- Centre for the Functional MRI of the Brain (FMRIB), Nuffield Department of Clinical Neurosciences, University of Oxford, UK
| | - Linxin Li
- Centre for Prevention of Stroke and Dementia, Nuffield Department of Clinical Neurosciences, University of Oxford, UK
| | - Guendalina Bonifacio
- Centre for the Functional MRI of the Brain (FMRIB), Nuffield Department of Clinical Neurosciences, University of Oxford, UK
| | - Vaanathi Sundaresan
- Centre for the Functional MRI of the Brain (FMRIB), Nuffield Department of Clinical Neurosciences, University of Oxford, UK
| | - Ursula G Schulz
- Centre for Prevention of Stroke and Dementia, Nuffield Department of Clinical Neurosciences, University of Oxford, UK
| | - Wilhelm Kuker
- Centre for Prevention of Stroke and Dementia, Nuffield Department of Clinical Neurosciences, University of Oxford, UK
| | - Marco Battaglini
- Department of Medicine, Surgery and Neuroscience, University of Siena, Siena, Italy
| | - Peter M Rothwell
- Centre for Prevention of Stroke and Dementia, Nuffield Department of Clinical Neurosciences, University of Oxford, UK
| | - Mark Jenkinson
- Centre for the Functional MRI of the Brain (FMRIB), Nuffield Department of Clinical Neurosciences, University of Oxford, UK
| |
Collapse
|
24
|
Sowa P, Bjørnerud A, Nygaard GO, Damangir S, Spulber G, Celius EG, Due-Tønnessen P, Harbo HF, Beyer MK. Reduced perfusion in white matter lesions in multiple sclerosis. Eur J Radiol 2015; 84:2605-12. [PMID: 26391230 DOI: 10.1016/j.ejrad.2015.09.007] [Citation(s) in RCA: 11] [Impact Index Per Article: 1.2] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/28/2015] [Revised: 08/14/2015] [Accepted: 09/08/2015] [Indexed: 10/23/2022]
Abstract
OBJECTIVE To investigate dynamic susceptibility contrast (DSC) perfusion weighted imaging (PWI) in white matter lesions (WML) in patients with multiple sclerosis (MS), using automatically generated binary masks of brain tissue. BACKGROUND WML in MS have in some studies demonstrated perfusion abnormalities compared to normal appearing white matter (NAWM), however perfusion changes in WML in MS have in general not been well documented. METHODS DSC PWI was performed at 1.5 Tesla in 69 newly diagnosed MS patients. Parametric perfusion maps representing cerebral blood volume (CBV), cerebral blood flow (CBF) and mean transit time (MTT) were obtained. Binary masks of WML, white matter (WM) and grey matter (GM) were automatically generated and co-registered to the perfusion maps. The WML mask was manually edited and modified to correct for errors in the automatic lesion detection. Perfusion parameters were derived both from WML and NAWM using the manually modified WML mask, and using the original non-modified WML mask (with and without GM exclusion mask). Differences in perfusion measures between WML and NAWM were analyzed. RESULTS CBF was significantly lower (p<0.001) and MTT significantly higher (p<0.001) in WML compared to NAWM. CBV did not show significant difference between WML and NAWM. The non-modified WML mask gave similar results as manually modified WML mask if the GM exclusion mask was used in the analysis. CONCLUSIONS DSC PWI revealed lower CBF and higher MTT, consistent with reduced perfusion, in WML compared to NAWM in patients with early MS. Automatically generated binary masks are a promising tool in perfusion analysis of WML.
Collapse
Affiliation(s)
- Piotr Sowa
- Department of Radiology and Nuclear Medicine, Oslo University Hospital, Oslo, Norway; Institute of Clinical Medicine, University of Oslo, Oslo, Norway.
| | - Atle Bjørnerud
- Intervention Center, Oslo University Hospital, Oslo, Norway; Department of Physics, University of Oslo, Oslo, Norway.
| | - Gro O Nygaard
- Institute of Clinical Medicine, University of Oslo, Oslo, Norway; Department of Neurology, Oslo University Hospital, Oslo, Norway.
| | - Soheil Damangir
- Department of Neurobiology, Care Sciences and Society, Karolinska Institute, Stockholm, Sweden.
| | - Gabriela Spulber
- Department of Neurobiology, Care Sciences and Society, Karolinska Institute, Stockholm, Sweden.
| | - Elisabeth G Celius
- Institute of Clinical Medicine, University of Oslo, Oslo, Norway; Department of Neurology, Oslo University Hospital, Oslo, Norway.
| | - Paulina Due-Tønnessen
- Department of Radiology and Nuclear Medicine, Oslo University Hospital, Oslo, Norway; Institute of Clinical Medicine, University of Oslo, Oslo, Norway.
| | - Hanne F Harbo
- Institute of Clinical Medicine, University of Oslo, Oslo, Norway; Department of Neurology, Oslo University Hospital, Oslo, Norway.
| | - Mona K Beyer
- Department of Radiology and Nuclear Medicine, Oslo University Hospital, Oslo, Norway; Department of Life Sciences and Health, Oslo and Akershus University College of Applied Sciences, Oslo, Norway.
| |
Collapse
|
25
|
Eye and hand motor interactions with the Symbol Digit Modalities Test in early multiple sclerosis. Mult Scler Relat Disord 2015; 4:585-9. [PMID: 26590666 DOI: 10.1016/j.msard.2015.08.003] [Citation(s) in RCA: 20] [Impact Index Per Article: 2.2] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/19/2015] [Revised: 08/11/2015] [Accepted: 08/14/2015] [Indexed: 11/24/2022]
Abstract
PURPOSE Eye and hand motor dysfunction may be present early in the disease course of relapsing-remitting multiple sclerosis (RRMS), and can affect the results on visual and written cognitive tests. We aimed to test for differences in saccadic initiation time (SI time) between RRMS patients and healthy controls, and whether SI time and hand motor speed interacted with the written version of the Symbol Digit Modalities Test (wSDMT). METHODS Patients with RRMS (N = 44, age 35.1 ± 7.3 years), time since diagnosis < 3 years and matched controls (N = 41, age 33.2 ± 6.8 years) were examined with ophthalmological, neurological and neuropsychological tests, as well as structural MRI (white matter lesion load (WMLL) and brainstem lesions), visual evoked potentials (VEP) and eye-tracker examinations of saccades. RESULTS SI time was longer in RRMS than controls (p < 0.05). SI time was not related to the Paced Auditory Serial Addition Test (PASAT), WMLL or to the presence of brainstem lesions. 9 hole peg test (9HP) correlated significantly with WMLL (r = 0.58, p < 0.01). Both SI time and 9HP correlated negatively with the results of wSDMT (r = -0.32, p < 0.05, r = -0.47, p < 0.01), but none correlated with the results of PASAT. CONCLUSIONS RRMS patients have an increased SI time compared to controls. Cognitive tests results, exemplified by the wSDMT, may be confounded by eye and hand motor function.
Collapse
|
26
|
Supervised segmentation of MRI brain images using combination of multiple classifiers. AUSTRALASIAN PHYSICAL & ENGINEERING SCIENCES IN MEDICINE 2015; 38:241-53. [DOI: 10.1007/s13246-015-0352-7] [Citation(s) in RCA: 12] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Track Full Text] [Subscribe] [Scholar Register] [Received: 01/27/2015] [Accepted: 05/21/2015] [Indexed: 10/23/2022]
|
27
|
Vuorinen M, Damangir S, Niskanen E, Miralbell J, Rusanen M, Spulber G, Soininen H, Kivipelto M, Solomon A. Coronary heart disease and cortical thickness, gray matter and white matter lesion volumes on MRI. PLoS One 2014; 9:e109250. [PMID: 25302686 PMCID: PMC4193798 DOI: 10.1371/journal.pone.0109250] [Citation(s) in RCA: 11] [Impact Index Per Article: 1.1] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/09/2014] [Accepted: 09/10/2014] [Indexed: 11/19/2022] Open
Abstract
Coronary heart disease (CHD) has been linked with cognitive decline and dementia in several studies. CHD is strongly associated with blood pressure, but it is not clear how blood pressure levels or changes in blood pressure over time affect the relation between CHD and dementia-related pathology. The aim of this study was to investigate relations between CHD and cortical thickness, gray matter volume and white matter lesion (WML) volume on MRI, considering CHD duration and blood pressure levels from midlife to three decades later. The study population included 69 elderly at risk of dementia who participated in the Cardiovascular Risk Factors, Aging and Dementia (CAIDE) study. CAIDE participants were examined in midlife, re-examined 21 years later, and then after additionally 7 years (in total up to 30 years follow-up). MRIs from the second re-examination were used to calculate cortical thickness, gray matter and WML volume. CHD diagnoses were obtained from the Finnish Hospital Discharge Register. Linear regression analyses were adjusted for age, sex, follow-up time and scanner type, and additionally total intracranial volume in GM volume analyses. Adding diabetes, cholesterol or smoking to the models did not influence the results. CHD was associated with lower thickness in multiple regions, and lower total gray matter volume, particularly in people with longer disease duration (>10 years). Associations between CHD, cortical thickness and gray matter volume were strongest in people with CHD and hypertension in midlife, and those with CHD and declining blood pressure after midlife. No association was found between CHD and WML volumes. Based on these results, long-term CHD seems to have detrimental effects on brain gray matter tissue, and these effects are influenced by blood pressure levels and their changes over time.
Collapse
Affiliation(s)
- Miika Vuorinen
- Department of Neurology, School of Medicine, University of Eastern Finland, Kuopio, Finland
- * E-mail:
| | - Soheil Damangir
- Division of Clinical Geriatrics, NVS, Karolinska Institute, Novum, Stockholm, Sweden
| | - Eini Niskanen
- Department of Applied Physics, University of Eastern Finland, Kuopio, Finland
| | - Julia Miralbell
- Department of Psychiatry and Clinical Psychobiology, University of Barcelona, Barcelona, Spain
| | - Minna Rusanen
- Department of Neurology, School of Medicine, University of Eastern Finland, Kuopio, Finland
| | - Gabriela Spulber
- Division of Clinical Geriatrics, NVS, Karolinska Institute, Novum, Stockholm, Sweden
| | - Hilkka Soininen
- Department of Neurology, School of Medicine, University of Eastern Finland, Kuopio, Finland
| | - Miia Kivipelto
- Department of Neurology, School of Medicine, University of Eastern Finland, Kuopio, Finland
- Department of Neurobiology, Care Sciences and Society, KI- Alzheimer Disease Research Center (KI-ADRC), Karolinska Institute, Stockholm, Sweden
| | - Alina Solomon
- Department of Neurology, School of Medicine, University of Eastern Finland, Kuopio, Finland
- Department of Neurobiology, Care Sciences and Society, KI- Alzheimer Disease Research Center (KI-ADRC), Karolinska Institute, Stockholm, Sweden
| |
Collapse
|
28
|
Nygaard GO, Walhovd KB, Sowa P, Chepkoech JL, Bjørnerud A, Due-Tønnessen P, Landrø NI, Damangir S, Spulber G, Storsve AB, Beyer MK, Fjell AM, Celius EG, Harbo HF. Cortical thickness and surface area relate to specific symptoms in early relapsing–remitting multiple sclerosis. Mult Scler 2014; 21:402-14. [DOI: 10.1177/1352458514543811] [Citation(s) in RCA: 63] [Impact Index Per Article: 6.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/16/2022]
Abstract
Background: Cortical atrophy is common in early relapsing–remitting multiple sclerosis (RRMS). Whether this atrophy is caused by changes in cortical thickness or cortical surface area is not known, nor is their separate contributions to clinical symptoms. Objectives: To investigate the difference in cortical surface area, thickness and volume between early RRMS patients and healthy controls; and the relationship between these measures and neurological disability, cognitive decline, fatigue and depression. Methods: RRMS patients ( n = 61) underwent magnetic resonance imaging (MRI), neurological and neuropsychological examinations. We estimated cortical surface area, thickness and volume and compared them with matched healthy controls ( n = 61). We estimated the correlations between clinical symptoms and cortical measures within the patient group. Results: We found no differences in cortical surface area, but widespread differences in cortical thickness and volume between the groups. Neurological disability was related to regionally smaller cortical thickness and volume. Better verbal memory was related to regionally larger surface area; and better visuo-spatial memory, to regionally larger cortical volume. Higher depression scores and fatigue were associated with regionally smaller cortical surface area and volume. Conclusions: We found that cortical thickness, but not cortical surface area, is affected in early RRMS. We identified specific structural correlates to the main clinical symptoms in early RRMS.
Collapse
Affiliation(s)
- Gro O Nygaard
- Oslo University Hospital, Norway/University of Oslo, Norway
| | | | - Piotr Sowa
- Oslo University Hospital, Norway/University of Oslo, Norway
| | | | - Atle Bjørnerud
- Oslo University Hospital, Norway/University of Oslo, Norway
| | | | | | | | | | | | - Mona K Beyer
- Oslo University Hospital, Norway/University of Oslo, Norway
| | - Anders M Fjell
- Oslo University Hospital, Norway/University of Oslo, Norway
| | | | - Hanne F Harbo
- Oslo University Hospital, Norway/University of Oslo, Norway
| |
Collapse
|
29
|
Steenwijk MD, Pouwels PJW, Daams M, van Dalen JW, Caan MWA, Richard E, Barkhof F, Vrenken H. Accurate white matter lesion segmentation by k nearest neighbor classification with tissue type priors (kNN-TTPs). NEUROIMAGE-CLINICAL 2013; 3:462-9. [PMID: 24273728 PMCID: PMC3830067 DOI: 10.1016/j.nicl.2013.10.003] [Citation(s) in RCA: 130] [Impact Index Per Article: 11.8] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 09/20/2013] [Revised: 10/01/2013] [Accepted: 10/03/2013] [Indexed: 11/25/2022]
Abstract
Introduction The segmentation and volumetric quantification of white matter (WM) lesions play an important role in monitoring and studying neurological diseases such as multiple sclerosis (MS) or cerebrovascular disease. This is often interactively done using 2D magnetic resonance images. Recent developments in acquisition techniques allow for 3D imaging with much thinner sections, but the large number of images per subject makes manual lesion outlining infeasible. This warrants the need for a reliable automated approach. Here we aimed to improve k nearest neighbor (kNN) classification of WM lesions by optimizing intensity normalization and using spatial tissue type priors (TTPs). Methods The kNN-TTP method used kNN classification with 3.0 T 3DFLAIR and 3DT1 intensities as well as MNI-normalized spatial coordinates as features. Additionally, TTPs were computed by nonlinear registration of data from healthy controls. Intensity features were normalized using variance scaling, robust range normalization or histogram matching. The algorithm was then trained and evaluated using a leave-one-out experiment among 20 patients with MS against a reference segmentation that was created completely manually. The performance of each normalization method was evaluated both with and without TTPs in the feature set. Volumetric agreement was evaluated using intra-class coefficient (ICC), and voxelwise spatial agreement was evaluated using Dice similarity index (SI). Finally, the robustness of the method across different scanners and patient populations was evaluated using an independent sample of elderly subjects with hypertension. Results The intensity normalization method had a large influence on the segmentation performance, with average SI values ranging from 0.66 to 0.72 when no TTPs were used. Independent of the normalization method, the inclusion of TTPs as features increased performance particularly by reducing the lesion detection error. Best performance was achieved using variance scaled intensity features and including TTPs in the feature set: this yielded ICC = 0.93 and average SI = 0.75 ± 0.08. Validation of the method in an independent sample of elderly subjects with hypertension, yielded even higher ICC = 0.96 and SI = 0.84 ± 0.14. Conclusion Adding TTPs increases the performance of kNN based MS lesion segmentation methods. Best performance was achieved using variance scaling for intensity normalization and including TTPs in the feature set, showing excellent agreement with the reference segmentations across a wide range of lesion severity, irrespective of the scanner used or the pathological substrate of the lesions. Intensity normalization has a large influence on lesion segmentation performance. Inclusion of tissue type priors as features increases segmentation performance. Best performance was achieved using variance scaling and tissue type priors.
Collapse
Affiliation(s)
- Martijn D Steenwijk
- Department of Radiology and Nuclear Medicine, Neuroscience Campus Amsterdam, VU University Medical Center, Amsterdam, The Netherlands
| | | | | | | | | | | | | | | |
Collapse
|