{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"bioRxiv"}],"indexed":{"date-parts":[[2026,11,10]],"date-time":"2026-11-10T01:28:25Z","timestamp":1794274105073,"version":"4.9.0"},"posted":{"date-parts":[[2025,3,11]]},"group-title":"Neuroscience","reference-count":54,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2025,3,11]],"date-time":"2025-03-11T00:00:00Z","timestamp":1741651200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2025,3,11]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:p>\n                  The fine-grained segmentation of cerebellar structures is an essential step towards supplying increasingly accurate anatomically informed analyses, including, for example, white matter diffusion magnetic resonance imaging (MRI) tractography. Cerebellar tissue segmentation is typically performed on structural magnetic resonance imaging data, such as T1-weighted data, while connectivity between segmented regions is mapped using diffusion MRI tractography data. Small deviations in structural to diffusion MRI data co-registration may negatively impact connectivity analyses. Reliable segmentation of brain tissue performed directly on diffusion MRI data helps to circumvent such inaccuracies. Diffusion MRI enables the computation of many image contrasts, including a variety of tissue microstructure maps. While multiple methods have been proposed for the segmentation of cerebellar structures using diffusion MRI, little attention has been paid to the systematic evaluation of the performance of different available input image contrasts for the segmentation task. In this work, we evaluate and compare the segmentation performance of diffusion MRI-derived contrasts on the cerebellar segmentation task. Specifically, we include spherical mean (diffusion-weighted image average) and b0 (non-diffusion-weighted image average) contrasts, local signal parameterization contrasts (diffusion tensor and kurtosis fit maps), and the structural T1-weighted MRI contrast that is most commonly employed for the task. We train a popular deep-learning architecture using a publicly available dataset (HCP-YA), leveraging cerebellar region labels from the atlas-based SUIT cerebellar segmentation pipeline. By training and testing using many diffusion-MRI-derived image inputs, we find that the spherical mean image computed from b=1000 s\/mm\n                  <jats:sup>2<\/jats:sup>\n                  shell data provides stable performance across different metrics and significantly outperforms the tissue microstructure contrasts that are traditionally used in machine learning segmentation methods for diffusion MRI.\n                <\/jats:p>\n                <jats:sec>\n                  <jats:title>Key points<\/jats:title>\n                  <jats:list list-type=\"bullet\">\n                    <jats:list-item>\n                      <jats:p>We provide evidence about the performance of different dMRI contrasts for cerebellar structure segmentation using a deep learning neural network.<\/jats:p>\n                    <\/jats:list-item>\n                    <jats:list-item>\n                      <jats:p>The diffusion MRI spherical mean provides improved and stable cerebellar structure segmentation performance.<\/jats:p>\n                    <\/jats:list-item>\n                    <jats:list-item>\n                      <jats:p>The spherical mean is easy to compute and can be used for cerebellar structure segmentation on retrospective clinical diffusion MRI data.<\/jats:p>\n                    <\/jats:list-item>\n                  <\/jats:list>\n                <\/jats:sec>","DOI":"10.1101\/2025.03.10.642452","type":"posted-content","created":{"date-parts":[[2025,3,11]],"date-time":"2025-03-11T08:10:23Z","timestamp":1741680623000},"source":"Crossref","is-referenced-by-count":2,"title":["Towards an informed choice of diffusion MRI image contrasts for cerebellar segmentation"],"prefix":"10.64898","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9661-1396","authenticated-orcid":false,"given":"Jon Haitz","family":"Legarreta","sequence":"first","affiliation":[{"name":"Department of Radiology, Brigham and Women\u2019s Hospital, Mass General Brigham\/Harvard Medical School, Boston MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhou","family":"Lan","sequence":"additional","affiliation":[{"name":"Department of Radiology, Brigham and Women\u2019s Hospital, Mass General Brigham\/Harvard Medical School, Boston MA, USA"},{"name":"Center for Clinical Investigation, Brigham and Women\u2019s Hospital, Mass General Brigham, Boston MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuqian","family":"Chen","sequence":"additional","affiliation":[{"name":"Department of Radiology, Brigham and Women\u2019s Hospital, Mass General Brigham\/Harvard Medical School, Boston MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fan","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Edward","family":"Yeterian","sequence":"additional","affiliation":[{"name":"Department of Anatomy and Neurobiology, Boston University Chobanian & Avedisian School of Medicine, Boston MA, USA"},{"name":"Department of Psychology, Colby College, Waterville ME, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nikos","family":"Makris","sequence":"additional","affiliation":[{"name":"Department of Psychiatry, Brigham and Women\u2019s Hospital, Mass General Brigham\/Harvard Medical School, Boston MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jarrett","family":"Rushmore","sequence":"additional","affiliation":[{"name":"Department of Anatomy and Neurobiology, Boston University Chobanian & Avedisian School of Medicine, Boston MA, USA"},{"name":"Department of Psychiatry, Brigham and Women\u2019s Hospital, Mass General Brigham\/Harvard Medical School, Boston MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yogesh","family":"Rathi","sequence":"additional","affiliation":[{"name":"Department of Psychiatry, Brigham and Women\u2019s Hospital, Mass General Brigham\/Harvard Medical School, Boston MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lauren J.","family":"O\u2019Donnell","sequence":"additional","affiliation":[{"name":"Department of Radiology, Brigham and Women\u2019s Hospital, Mass General Brigham\/Harvard Medical School, Boston MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"54368","reference":[{"key":"2026041112563185000_2025.03.10.642452v1.1","unstructured":"Altman, J. , & Bayer, S. A . (1996). Development of the cerebellar system. CRC Press."},{"key":"2026041112563185000_2025.03.10.642452v1.2","doi-asserted-by":"publisher","DOI":"10.1002\/cne.21974"},{"key":"2026041112563185000_2025.03.10.642452v1.3","doi-asserted-by":"publisher","DOI":"10.1093\/cercor\/bhab184"},{"key":"2026041112563185000_2025.03.10.642452v1.4","doi-asserted-by":"publisher","DOI":"10.1007\/s12311-014-0627-7"},{"key":"2026041112563185000_2025.03.10.642452v1.5","doi-asserted-by":"publisher","DOI":"10.1002\/hbm.25604"},{"key":"2026041112563185000_2025.03.10.642452v1.6","doi-asserted-by":"crossref","unstructured":"Boerner, T. J. , Deems, S. , Furlani, T. R. , Knuth, S. L. , & Towns, J . (2023). ACCESS: Advancing Innovation: NSF\u2019s Advanced Cyberinfrastructure Coordination Ecosystem: Services & Support. Practice and Experience in Advanced Research Computing, 173\u2013176.","DOI":"10.1145\/3569951.3597559"},{"key":"2026041112563185000_2025.03.10.642452v1.7","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuron.2013.10.044"},{"key":"2026041112563185000_2025.03.10.642452v1.8","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2018.08.003"},{"key":"2026041112563185000_2025.03.10.642452v1.9","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.24736"},{"key":"2026041112563185000_2025.03.10.642452v1.10","doi-asserted-by":"publisher","DOI":"10.1016\/j.mri.2020.02.010"},{"key":"2026041112563185000_2025.03.10.642452v1.11","doi-asserted-by":"publisher","DOI":"10.1002\/nbm.3931"},{"key":"2026041112563185000_2025.03.10.642452v1.12","doi-asserted-by":"publisher","DOI":"10.5281\/zenodo.8436376"},{"key":"2026041112563185000_2025.03.10.642452v1.13","doi-asserted-by":"publisher","DOI":"10.1002\/047134608x.w8258"},{"key":"2026041112563185000_2025.03.10.642452v1.14","doi-asserted-by":"publisher","DOI":"10.2307\/1932409"},{"issue":"1","key":"2026041112563185000_2025.03.10.642452v1.15","first-page":"127","volume":"33","year":"2006","journal-title":"A spatially unbiased atlas template of the human cerebellum. NeuroImage"},{"key":"2026041112563185000_2025.03.10.642452v1.16","first-page":"119703","volume":"264","author":"ESMI MRI Study Group, & Reuter, M","year":"2022","journal-title":"CerebNet: A fast and reliable deep-learning pipeline for detailed cerebellum sub-segmentation. NeuroImage"},{"key":"2026041112563185000_2025.03.10.642452v1.17","first-page":"8","article-title":"Dipy, a library for the analysis of diffusion MRI data","volume":"8","year":"2014","journal-title":"Frontiers in Neuroinformatics"},{"key":"2026041112563185000_2025.03.10.642452v1.18","doi-asserted-by":"crossref","unstructured":"Gaviraghi, M. , Savini, G. , Castellazzi, G. , Palesi, F. , Rolandi, N. , Sacco, S. , Pichiecchio, A. , Mariani, V. , Tartara, E. , Tassi, L. , Vitali, P. , D\u2019Angelo, E. , & Wheeler-Kingshott, C. A. M. G . (2021). Automatic Segmentation of Dentate Nuclei for Microstructure Assessment: Example of Application to Temporal Lobe Epilepsy Patients. Computational Diffusion MRI, 263\u2013278.","DOI":"10.1007\/978-3-030-73018-5_21"},{"key":"2026041112563185000_2025.03.10.642452v1.19","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2013.04.127"},{"key":"2026041112563185000_2025.03.10.642452v1.20","doi-asserted-by":"crossref","unstructured":"Hagmann, P. , Jonasson, L. , Maeder, P. , Thiran, J.-P. , Wedeen, V. J. , & Meuli, R . (2006). Understanding Diffusion MR Imaging Techniques: From Scalar Diffusion-weighted Imaging to Diffusion Tensor Imaging and Beyond. RadioGraphics, 26(suppl_1), S205\u2013S223.","DOI":"10.1148\/rg.26si065510"},{"key":"2026041112563185000_2025.03.10.642452v1.21","doi-asserted-by":"crossref","unstructured":"Hancock, D. Y. , Fischer, J. , Lowe, J. M. , Snapp-Childs, W. , Pierce, M. , Marru, S. , Coulter, J. E. , Vaughn, M. , Beck, B. , Merchant, N. , Skidmore, E. , & Jacobs, G . (2021). Jetstream2: Accelerating cloud computing via Jetstream. Practice and Experience in Advanced Research Computing , Article Article 11.","DOI":"10.1145\/3437359.3465565"},{"key":"2026041112563185000_2025.03.10.642452v1.22","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2020.116819"},{"key":"2026041112563185000_2025.03.10.642452v1.23","first-page":"31","article-title":"The human brain in numbers: a linearly scaled-up primate brain","volume":"3","year":"2009","journal-title":"Frontiers in Human Neuroscience"},{"key":"2026041112563185000_2025.03.10.642452v1.24","doi-asserted-by":"publisher","DOI":"10.1002\/nbm.1518"},{"key":"2026041112563185000_2025.03.10.642452v1.25","doi-asserted-by":"crossref","unstructured":"Jia, J. , Staring, M. , & Stoel, B. C. (2024). Seg-metrics: a Python package to compute segmentation metrics. In arXiv [cs.CV]. arXiv. http:\/\/arxiv.org\/abs\/2403.07884","DOI":"10.1101\/2024.02.22.24303215"},{"key":"2026041112563185000_2025.03.10.642452v1.26","first-page":"12336","volume":"2021","year":"2021","journal-title":"Learning calibrated medical image segmentation via multi-rater agreement modeling"},{"key":"2026041112563185000_2025.03.10.642452v1.27","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.20033"},{"key":"2026041112563185000_2025.03.10.642452v1.28","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.20283"},{"key":"2026041112563185000_2025.03.10.642452v1.29","doi-asserted-by":"publisher","DOI":"10.1145\/304181.304219"},{"key":"2026041112563185000_2025.03.10.642452v1.30","first-page":"101550","article-title":"Deep Cerebellar Nuclei Segmentation via Semi-Supervised Deep Context-Aware Learning from 7T Diffusion MRI. IEEE Access : Practical Innovations","volume":"8","year":"2020","journal-title":"Open Solutions"},{"key":"2026041112563185000_2025.03.10.642452v1.31","unstructured":"Kingma, D. P. , & Ba, J . (2015). Adam: A Method for Stochastic Optimization. In Y. Bengio & Y. LeCun (Eds.), 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, May 7-9, 2015, Conference Track Proceedings. http:\/\/arxiv.org\/abs\/1412.6980"},{"key":"2026041112563185000_2025.03.10.642452v1.32","doi-asserted-by":"publisher","DOI":"10.1016\/j.cobeha.2024.101353"},{"key":"2026041112563185000_2025.03.10.642452v1.33","doi-asserted-by":"crossref","unstructured":"Magin, R. L . (2016). Models of diffusion signal decay in magnetic resonance imaging: Capturing complexity. Concepts in Magnetic Resonance. Part A, Bridging Education and Research, 45A(4), e21401.","DOI":"10.1002\/cmr.a.21401"},{"key":"2026041112563185000_2025.03.10.642452v1.34","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-023-02151-z"},{"key":"2026041112563185000_2025.03.10.642452v1.35","doi-asserted-by":"crossref","unstructured":"Morell-Ortega, S. , Ruiz-Perez, M. , Gadea, M. , Vivo-Hernando, R. , Rubio, G. , Aparici, F. , de la Iglesia-Vaya, M. , Catheline, G. , Coup\u00e9, P. , & Manj\u00f3n, J. V. (2024). DeepCERES: A Deep learning method for cerebellar lobule segmentation using ultra-high resolution multimodal MRI. In arXiv [eess.IV]. arXiv. http:\/\/arxiv.org\/abs\/2401.12074","DOI":"10.1016\/j.neuroimage.2025.121063"},{"key":"2026041112563185000_2025.03.10.642452v1.36","doi-asserted-by":"publisher","DOI":"10.1016\/j.cortex.2009.07.018"},{"key":"2026041112563185000_2025.03.10.642452v1.37","doi-asserted-by":"crossref","unstructured":"Myronenko, A . (2019). 3D MRI Brain Tumor Segmentation Using Autoencoder Regularization. Brainlesion: Glioma, Multiple Sclerosis , Stroke and Traumatic Brain Injuries, 311\u2013320.","DOI":"10.1007\/978-3-030-11726-9_28"},{"issue":"4","key":"2026041112563185000_2025.03.10.642452v1.38","first-page":"044007","article-title":"Using deep learning for a diffusion-based segmentation of the dentate nucleus and its benefits over atlas-based methods","volume":"6","year":"2019","journal-title":"The Journal of Medical Investigation: JMI"},{"key":"2026041112563185000_2025.03.10.642452v1.39","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2015.05.008"},{"key":"2026041112563185000_2025.03.10.642452v1.40","first-page":"383","article-title":"Thalamic Nuclei Segmentation Using Tractography","volume":"2018","year":"2018","journal-title":"Population-Specific Priors and Local Fibre Orientation. Medical Image Computing and Computer Assisted Intervention \u2013 MICCAI"},{"key":"2026041112563185000_2025.03.10.642452v1.41","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2013.05.057"},{"key":"2026041112563185000_2025.03.10.642452v1.42","doi-asserted-by":"publisher","DOI":"10.1146\/annurev.neuro.31.060407.125606"},{"key":"2026041112563185000_2025.03.10.642452v1.43","doi-asserted-by":"crossref","unstructured":"Sun, P. , Wu, Y. , Chen, G. , Wu, J. , Shen, D. , & Yap, P.-T . (2019). Tissue Segmentation Using Sparse Non-negative Matrix Factorization of Spherical Mean Diffusion MRI Data. Computational Diffusion MRI, 69\u201376.","DOI":"10.1007\/978-3-030-05831-9_6"},{"key":"2026041112563185000_2025.03.10.642452v1.44","doi-asserted-by":"publisher","DOI":"10.1016\/j.compmedimag.2019.04.005"},{"key":"2026041112563185000_2025.03.10.642452v1.45","doi-asserted-by":"publisher","DOI":"10.3389\/fnimg.2022.917806"},{"key":"2026041112563185000_2025.03.10.642452v1.46","doi-asserted-by":"publisher","DOI":"10.1002\/nbm.3017"},{"key":"2026041112563185000_2025.03.10.642452v1.47","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2013.05.041"},{"issue":"3","key":"2026041112563185000_2025.03.10.642452v1.48","first-page":"261","volume":"17","author":"SciPy 1.0 Contributors","year":"2020","journal-title":"SciPy 1.0: fundamental algorithms for scientific computing in Python. Nature Methods"},{"key":"2026041112563185000_2025.03.10.642452v1.49","first-page":"34","article-title":"Aleatoric uncertainty estimation with test-time augmentation for medical image segmentation with convolutional neural networks","volume":"335","year":"2019","journal-title":"Neurocomputing"},{"key":"2026041112563185000_2025.03.10.642452v1.50","doi-asserted-by":"publisher","DOI":"10.1016\/j.jneumeth.2020.108828"},{"key":"2026041112563185000_2025.03.10.642452v1.51","doi-asserted-by":"publisher","DOI":"10.1109\/iceic51217.2021.9369797"},{"key":"2026041112563185000_2025.03.10.642452v1.52","doi-asserted-by":"publisher","DOI":"10.1016\/J.NEUROIMAGE.2021.117934"},{"key":"2026041112563185000_2025.03.10.642452v1.53","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2023.3331691"},{"key":"2026041112563185000_2025.03.10.642452v1.54","doi-asserted-by":"crossref","unstructured":"Zhang, L. , Tanno, R. , Xu, M. , Huang, Y. , Bronik, K. , Jin, C. , Jacob, J. , Zheng, Y. , Shao, L. , Ciccarelli, O. , Barkhof, F. , & Alexander, D. C . (2023). Learning from multiple annotators for medical image segmentation. Pattern Recognition, 138(109400), None. [dataset] Human Connectome Project, WU-Minn Consortium; 2017; 1200 Subjects Data Release (S1200); https:\/\/db.humanconnectome.org\/","DOI":"10.1016\/j.patcog.2023.109400"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1101\/2025.03.10.642452","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,11]],"date-time":"2026-04-11T19:56:41Z","timestamp":1775937401000},"score":1,"resource":{"primary":{"URL":"http:\/\/biorxiv.org\/lookup\/doi\/10.1101\/2025.03.10.642452"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,3,11]]},"references-count":54,"URL":"https:\/\/doi.org\/10.1101\/2025.03.10.642452","relation":{},"subject":[],"published":{"date-parts":[[2025,3,11]]},"subtype":"preprint"}}