{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,9]],"date-time":"2026-03-09T20:54:15Z","timestamp":1773089655792,"version":"3.50.1"},"reference-count":27,"publisher":"Ovid Technologies (Wolters Kluwer Health)","issue":"3","content-domain":{"domain":["lww.com","ovid.com"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2022,6]]},"abstract":"<jats:sec>\n            <jats:title>Objectives:<\/jats:title>\n            <jats:p>Automated whole brain segmentation from magnetic resonance images is of great interest for the development of clinically relevant volumetric markers for various neurological diseases. Although deep learning methods have demonstrated remarkable potential in this area, they may perform poorly in nonoptimal conditions, such as limited training data availability. Manual whole brain segmentation is an incredibly tedious process, so minimizing the data set size required for training segmentation algorithms may be of wide interest. The purpose of this study was to compare the performance of the prototypical deep learning segmentation architecture (U-Net) with a previously published atlas-free traditional machine learning method, Classification using Derivative-based Features (C-DEF) for whole brain segmentation, in the setting of limited training data.<\/jats:p>\n          <\/jats:sec>\n          <jats:sec>\n            <jats:title>Materials and Methods:<\/jats:title>\n            <jats:p>C-DEF and U-Net models were evaluated after training on manually curated data from 5, 10, and 15 participants in 2 research cohorts: (1) people living with clinically diagnosed HIV infection and (2) relapsing\u2013remitting multiple sclerosis, each acquired at separate institutions, and between 5 and 295 participants' data using a large, publicly available, and annotated data set of glioblastoma and lower grade glioma (brain tumor segmentation). Statistics was performed on the Dice similarity coefficient using repeated-measures analysis of variance and Dunnett\u2013Hsu pairwise comparison.<\/jats:p>\n          <\/jats:sec>\n          <jats:sec>\n            <jats:title>Results:<\/jats:title>\n            <jats:p>C-DEF produced better segmentation than U-Net in lesion (29.2%\u201338.9%) and cerebrospinal fluid (5.3%\u201311.9%) classes when trained with data from 15 or fewer participants. Unlike C-DEF, U-Net showed significant improvement when increasing the size of the training data (24%\u201330% higher than baseline). In the brain tumor segmentation data set, C-DEF produced equivalent or better segmentations than U-Net for enhancing tumor and peritumoral edema regions across all training data sizes explored. However, U-Net was more effective than C-DEF for segmentation of necrotic\/non-enhancing tumor when trained on 10 or more participants, probably because of the inconsistent signal intensity of the tissue class.<\/jats:p>\n          <\/jats:sec>\n          <jats:sec>\n            <jats:title>Conclusions:<\/jats:title>\n            <jats:p>These results demonstrate that classical machine learning methods can produce more accurate brain segmentation than the far more complex deep learning methods when only small or moderate amounts of training data are available (n \u2264 15). The magnitude of this advantage varies by tissue and cohort, while U-Net may be preferable for deep gray matter and necrotic\/non-enhancing tumor segmentation, particularly with larger training data sets (n \u2265 20). Given that segmentation models often need to be retrained for application to novel imaging protocols or pathology, the bottleneck associated with large-scale manual annotation could be avoided with classical machine learning algorithms, such as C-DEF.<\/jats:p>\n          <\/jats:sec>","DOI":"10.1097\/rmr.0000000000000296","type":"journal-article","created":{"date-parts":[[2022,6,29]],"date-time":"2022-06-29T16:01:34Z","timestamp":1656518494000},"page":"31-39","update-policy":"https:\/\/doi.org\/10.1097\/lww.0000000000001000","source":"Crossref","is-referenced-by-count":10,"title":["Logistic Regression\u2013Based Model Is More Efficient Than U-Net Model for Reliable Whole Brain Magnetic Resonance Imaging Segmentation"],"prefix":"10.1097","volume":"31","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-1390-2444","authenticated-orcid":false,"given":"Henry","family":"Dieckhaus","sequence":"first","affiliation":[{"name":"qMRI Core Facility, NINDS, National Institutes of Health, Bethesda, MD;"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Rozanna","family":"Meijboom","sequence":"additional","affiliation":[{"name":"Centre for Clinical Brain Sciences, University of Edinburgh, Edinburgh, United Kingdom;"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Serhat","family":"Okar","sequence":"additional","affiliation":[{"name":"Translational Neuroradiology Section, NINDS, National Institutes of Health, Bethesda, MD;"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Tianxia","family":"Wu","sequence":"additional","affiliation":[{"name":"Clinical Trials Unit, NINDS, National Institutes of Health, Bethesda, MD;"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Prasanna","family":"Parvathaneni","sequence":"additional","affiliation":[{"name":"Translational Neuroradiology Section, NINDS, National Institutes of Health, Bethesda, MD;"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yair","family":"Mina","sequence":"additional","affiliation":[{"name":"Viral Immunology Section, NINDS, National Institutes of Health, Bethesda, MD; and"},{"name":"Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel."}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Siddharthan","family":"Chandran","sequence":"additional","affiliation":[{"name":"Centre for Clinical Brain Sciences, University of Edinburgh, Edinburgh, United Kingdom;"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Adam D.","family":"Waldman","sequence":"additional","affiliation":[{"name":"Centre for Clinical Brain Sciences, University of Edinburgh, Edinburgh, United Kingdom;"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Daniel S.","family":"Reich","sequence":"additional","affiliation":[{"name":"Translational Neuroradiology Section, NINDS, National Institutes of Health, Bethesda, MD;"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Govind","family":"Nair","sequence":"additional","affiliation":[{"name":"qMRI Core Facility, NINDS, National Institutes of Health, Bethesda, MD;"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"276","published-online":{"date-parts":[[2022,6,28]]},"reference":[{"key":"R1-20240824","doi-asserted-by":"crossref","first-page":"21","DOI":"10.1186\/s13024-019-0325-5","article-title":"Neuroimaging biomarkers for Alzheimer's disease","volume":"14","author":"Marquez","year":"2019","journal-title":"Mol Neurodegener"},{"key":"R2-20240824","doi-asserted-by":"crossref","first-page":"75","DOI":"10.1038\/nrneurol.2017.171","article-title":"Assessing treatment outcomes in multiple sclerosis trials and in the clinical setting","volume":"14","author":"Tur","year":"2018","journal-title":"Nat Rev Neurol"},{"key":"R3-20240824","doi-asserted-by":"crossref","first-page":"1756286419859722","DOI":"10.1177\/1756286419859722","article-title":"Advances in brain imaging in multiple sclerosis","volume":"12","author":"Cortese","year":"2019","journal-title":"Ther Adv Neurol Disord"},{"key":"R4-20240824","doi-asserted-by":"crossref","first-page":"552","DOI":"10.1097\/RLI.0b013e31817e90d2","article-title":"White matter lesions in patients with localization-related epilepsy","volume":"43","author":"Jansen","year":"2008","journal-title":"Invest Radiol."},{"key":"R5-20240824","doi-asserted-by":"crossref","first-page":"423","DOI":"10.1097\/01.RLI.0000065426.07178.f1","article-title":"Comparison of multiple sclerosis lesions at 1.5 and 3.0 Tesla","volume":"38","author":"Sicotte","year":"2003","journal-title":"Invest Radiol"},{"key":"R6-20240824","doi-asserted-by":"crossref","first-page":"601","DOI":"10.1097\/RLI.0000000000000666","article-title":"Variability and standardization of quantitative imaging: monoparametric to multiparametric quantification, radiomics, and artificial intelligence","volume":"55","author":"Hagiwara","year":"2020","journal-title":"Invest Radiol"},{"key":"R7-20240824","doi-asserted-by":"crossref","first-page":"e01226","DOI":"10.1016\/j.heliyon.2019.e01226","article-title":"Robust, atlas-free, automatic segmentation of brain MRI in health and disease","volume":"5","author":"Selvaganesan","year":"2019","journal-title":"Heliyon"},{"key":"R8-20240824","doi-asserted-by":"crossref","first-page":"203","DOI":"10.1038\/s41592-020-01008-z","article-title":"nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation","volume":"18","author":"Isensee","year":"2021","journal-title":"Nat Methods"},{"key":"R10-20240824","doi-asserted-by":"crossref","first-page":"449","DOI":"10.1007\/s10278-017-9983-4","article-title":"Deep learning for brain MRI segmentation: state of the art and future directions","volume":"30","author":"Akkus","year":"2017","journal-title":"J Digit Imaging"},{"key":"R11-20240824","doi-asserted-by":"crossref","first-page":"1993","DOI":"10.1109\/TMI.2014.2377694","article-title":"The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS)","volume":"34","author":"Menze","year":"2015","journal-title":"IEEE Trans Med Imaging"},{"key":"R13-20240824","doi-asserted-by":"crossref","first-page":"74","DOI":"10.1016\/j.neunet.2019.08.025","article-title":"MultiResUNet: rethinking the U-Net architecture for multimodal biomedical image segmentation","volume":"121","author":"Ibtehaz","year":"2020","journal-title":"Neural Netw"},{"key":"R14-20240824","doi-asserted-by":"crossref","first-page":"1856","DOI":"10.1109\/TMI.2019.2959609","article-title":"UNet++: redesigning skip connections to exploit multiscale features in image segmentation","volume":"39","author":"Zhou","year":"2020","journal-title":"IEEE Trans Med Imaging"},{"key":"R15-20240824","doi-asserted-by":"crossref","first-page":"197","DOI":"10.1016\/j.media.2019.01.012","article-title":"Attention gated networks: learning to leverage salient regions in medical images","volume":"53","author":"Schlemper","year":"2019","journal-title":"Med Image Anal"},{"key":"R17-20240824","doi-asserted-by":"crossref","first-page":"105","DOI":"10.1016\/j.neuroimage.2019.03.041","article-title":"3D whole brain segmentation using spatially localized atlas network tiles","volume":"194","author":"Huo","year":"2019","journal-title":"NeuroImage"},{"key":"R19-20240824","doi-asserted-by":"crossref","first-page":"e190195","DOI":"10.1148\/ryai.2020190195","article-title":"MRI manufacturer shift and adaptation: increasing the generalizability of deep learning segmentation for MR images acquired with different scanners","volume":"2","author":"Yan","year":"2020","journal-title":"Radiol Artif Intell"},{"key":"R20-20240824","doi-asserted-by":"crossref","first-page":"450341","DOI":"10.1155\/2015\/450341","article-title":"MRI segmentation of the human brain: challenges, methods, and applications","volume":"2015","author":"Despotovi\u0107","year":"2015","journal-title":"Comput Math Methods Med"},{"key":"R21-20240824","doi-asserted-by":"crossref","first-page":"619","DOI":"10.1097\/RLI.0000000000000673","article-title":"Artificial intelligence and machine learning in radiology: current state and considerations for routine clinical implementation","volume":"55","author":"Wichmann","year":"2020","journal-title":"Invest Radiol"},{"key":"R22-20240824","doi-asserted-by":"crossref","first-page":"185012","DOI":"10.1088\/1361-6560\/ac2206","article-title":"The impact of training sample size on deep learning-based organ auto-segmentation for head-and-neck patients","volume":"66","author":"Fang","year":"2021","journal-title":"Phys Med Biol"},{"key":"R23-20240824","doi-asserted-by":"crossref","first-page":"48","DOI":"10.1109\/JBHI.2016.2631401","article-title":"A convolutional neural network for automatic characterization of plaque composition in carotid ultrasound","volume":"21","author":"Lekadir","year":"2017","journal-title":"IEEE J Biomed Health Inform"},{"key":"R24-20240824","doi-asserted-by":"crossref","first-page":"292","DOI":"10.1002\/ana.22366","article-title":"Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria","volume":"69","author":"Polman","year":"2011","journal-title":"Ann Neurol"},{"key":"R25-20240824","doi-asserted-by":"crossref","first-page":"162","DOI":"10.1006\/cbmr.1996.0014","article-title":"AFNI: software for analysis and visualization of functional magnetic resonance neuroimages","volume":"29","author":"Cox","year":"1996","journal-title":"Comput Biomed Res"},{"key":"R26-20240824","doi-asserted-by":"crossref","first-page":"341","DOI":"10.1016\/S0896-6273(02)00569-X","article-title":"Whole brain segmentation: automated labeling of neuroanatomical structures in the human brain","volume":"33","author":"Fischl","year":"2002","journal-title":"Neuron"},{"key":"R29-20240824","doi-asserted-by":"crossref","first-page":"170117","DOI":"10.1038\/sdata.2017.117","article-title":"Advancing The Cancer Genome Atlas glioma MRI collections with expert segmentation labels and radiomic features","volume":"4","author":"Bakas","year":"2017","journal-title":"Sci Data"},{"key":"R30-20240824","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1097\/RLI.0000000000000600","article-title":"A practical guide to artificial intelligence\u2013based image analysis in radiology","volume":"55","author":"Weikert","year":"2020","journal-title":"Invest Radiol"},{"key":"R31-20240824","doi-asserted-by":"crossref","first-page":"2458","DOI":"10.1007\/s00415-012-6762-5","article-title":"Recommendations to improve imaging and analysis of brain lesion load and atrophy in longitudinal studies of multiple sclerosis","volume":"260","author":"Vrenken","year":"2013","journal-title":"J Neurol"},{"key":"R33-20240824","doi-asserted-by":"crossref","first-page":"169","DOI":"10.3390\/diagnostics11020169","article-title":"BrainSeg-Net: brain tumor MR image segmentation via enhanced encoder-decoder Network","volume":"11","author":"Rehman","year":"2021","journal-title":"Diagnostics (Basel)"},{"key":"R34-20240824","doi-asserted-by":"crossref","first-page":"10191","DOI":"10.1038\/s41598-021-89686-3","article-title":"CMM-Net: contextual multi-scale multi-level network for efficient biomedical image segmentation","volume":"11","author":"Al-masni","year":"2021","journal-title":"Sci Rep"}],"container-title":["Topics in Magnetic Resonance Imaging"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.lww.com\/10.1097\/RMR.0000000000000296","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,8,25]],"date-time":"2024-08-25T00:00:38Z","timestamp":1724544038000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.lww.com\/10.1097\/RMR.0000000000000296"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,6]]},"references-count":27,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2022]]}},"URL":"https:\/\/doi.org\/10.1097\/rmr.0000000000000296","relation":{},"ISSN":["1536-1004"],"issn-type":[{"value":"1536-1004","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,6]]},"assertion":[{"value":"2022-04-26","name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2022-05-24","name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}}]}}