Automated Deep Learning-based Segmentation of the Dentate Nucleus Using Quantitative Susceptibility Mapping MRI.
Authors
Affiliations (26)
Affiliations (26)
- Department of Neurology, School of Medical Sciences, University of Campinas (Unicamp), Rua Vital Brasil, 89-99, Cidade Universitária "Zeferino Vaz", Campinas, São Paulo, Brazil 13083-888.
- School of Psychological Sciences, The Turner Institute for Brain and Mental Health, Monash University, Clayton, Australia.
- Department of Neuroscience, School of Translational Medicine, Monash University, Melbourne, Australia.
- Center for Magnetic Resonance Research, Department of Radiology, University of Minnesota, Minneapolis, Minn.
- Department of Advanced Biomedical Sciences, University of Naples Federico II, Naples, Italy.
- Bruce Lefroy Centre for Genetic Health Research, Murdoch Children's Research Institute, Parkville, Australia.
- Department of Paediatrics, University of Melbourne, Parkville, Australia.
- University Clinic and Outpatient Clinic for Radiology, Department for Radiation Medicine, University Hospital Halle (Saale), University Medicine Halle, Halle (Saale), Germany.
- Halle MR Imaging Core Facility, Medical Faculty, Martin-Luther-University Halle-Wittenberg, Halle (Saale), Germany.
- Department of Neurology, RWTH Aachen University, Aachen, Germany.
- JARA-BRAIN Institute Molecular Neuroscience and Neuroimaging, Forschungszentrum Jülich GmbH and RWTH Aachen University, Aachen, Germany.
- Department of Radiology, Children's Hospital of Philadelphia, Philadelphia, Pa.
- Department of Neuroradiology, Fondazione IRCCS Istituto Neurologico Carlo Besta, Milan, Italy.
- Department of Neurology, Medical University of Graz, Graz, Austria.
- Division of Biostatistics and Health Data Science, School of Public Health, University of Minnesota, Minneapolis, Minn.
- Department of Neurology, Children's Hospital of Philadelphia, Philadelphia, Pa.
- Department of Biochemistry and Molecular Biology, University of Florida, Gainesville, Fla.
- Institute of Biostructures and Bioimaging, Italian National Research Council, Naples, Italy.
- Department of Neurology and Neurosurgery, McGill University, Montreal, Canada.
- McConnell Brain Imaging Centre, Montreal Neurological Institute and Hospital, Montreal, Canada.
- Department of Biomedical Engineering, McGill University, Montreal, Canada.
- Department of Neurology and the Fixel Institute for Neurological Diseases, University of Florida, Gainesville, Fla.
- Department of Neurology and Center for Translational and Behavioral Neuroscience (C-TNBS), Essen University Hospital, University of Duisburg-Essen, Essen, Germany.
- QIMR Berghofer Medical Research Institute, Brisbane, Australia.
- School of Translational Medicine, Monash University, Melbourne, Australia.
- Paulo Gontijo Institute, São Paulo, Brazil.
Abstract
<i>"Just Accepted" papers have undergone full peer review and have been accepted for publication in <i>Radiology: Artificial Intelligence</i>. This article will undergo copyediting, layout, and proof review before it is published in its final version. Please note that during production of the final copyedited article, errors may be discovered which could affect the content.</i> Purpose To develop a dentate nucleus (DN) segmentation tool using deep learning (DL) applied to brain MRI-based quantitative susceptibility mapping (QSM) images. Materials and Methods Brain QSM images from healthy controls and individuals with cerebellar ataxia or multiple sclerosis were collected from nine different datasets (2016-2023) worldwide for this retrospective study (ClinicalTrials.gov Identifier: NCT04349514). Manual delineation of the DN was performed by experienced raters. Automated segmentation performance was evaluated against manual reference segmentations following training with several DL architectures. A two-step approach was used, consisting of a localization model followed by DN segmentation. Performance metrics included intraclass correlation coefficient (ICC), Dice score, and Pearson correlation coefficient. Results The training and testing datasets comprised 328 individuals (age range, 11-64 years; 171 female), including 141 healthy individuals and 187 with cerebellar ataxia or multiple sclerosis. The manual tracing protocol produced reference standards with high intrarater (average ICC 0.91) and interrater reliability (average ICC 0.78). Initial DL architecture exploration indicated that the nnU-Net framework performed best. The two-step localization plus segmentation pipeline achieved a Dice score of 0.90 ± 0.03 and 0.89 ± 0.04 for left and right DN segmentation, respectively. In external testing, the proposed algorithm outperformed the current leading automated tool (mean Dice scores for left and right DN: 0.86 ± 0.04 vs 0.57 ± 0.22, <i>P</i> < .001; 0.84 ± 0.07 vs 0.58 ± 0.24, <i>P</i> < .001). The model demonstrated generalizability across datasets unseen during the training step, with automated segmentations showing high correlation with manual annotations (left DN: r = 0.74; <i>P</i> < .001; right DN: r = 0.48; <i>P</i> = .03). Conclusion The proposed model accurately and efficiently segmented the DN from brain QSM images. The model is publicly available (https://github.com/art2mri/DentateSeg). ©RSNA, 2025.