Meeting Banner
Abstract #0696

Automatic MS lesion segmentation in the spinal cord using deep learning

Charley Gros1, Atef Badji1,2, Josefina Maranzano3, Ren Zhuoquiong4, Yaou Liu4,5, Elise Bannier6,7, Anne Kerbrat7,8, Gilles Edan8, Pierre Labauge9, Virginie Callot10,11, Jean Pelletier11,12, Bernard Audoin11,12, Henitsoa Rasoanandrianina10,11, Paola Valsasina13, Massimo Filippi13, Rohit Bakshi14, Shahamat Tauhid14, Ferran Prados15, Marios Yiannakas15, Hugh Kearney15, Olga Ciccarelli15, Sridar Narayanan3, and Julien Cohen-Adad1,16

1NeuroPoly Lab, Polytechnique Montreal, Montréal, QC, Canada, 2Faculty of Medicine, University of Montreal, Montreal, QC, Canada, 3Montreal Neurological Institute, Montreal, QC, Canada, 4Department of Radiology, Beijing Tiantan Hospital, Capital Medical University, Beijing, China, 5Department of Radiology, Xuanwu Hospital, Capital Medical University, Beijing, China, 6Department of Radiology, University Hospital of Rennes, Rennes, France, 7Univ Rennes, Inria, CNRS, Inserm, IRISA UMR 6074, VISAGES ERL U-1228, F-35000, Rennes, France, 8Department of Neurology, University Hospital of Rennes, Rennes, France, 9University Hospital of Montpellier, Montpellier, France, 10CRMBM, CNRS, Aix-Marseille University, Marseille, France, 11CEMEREM, Hôpital de la Timone, AP-HM, Marseille, France, 12Department of Neurology, CHU Timone, APHM, Marseille, France, 13Neuroimaging Research Unit, INSPE, Division of Neuroscience, San Raffaele Scientific Institute, Vita-Salute San Raffaele University, Milan, Italy, 14Brigham and Women's Hospital, Harvard Medical School, Boston, MA, United States, 15Queen Square MS Centre, UCL Institute of Neurology, Faculty of Brain Sciences, University College London, London, United Kingdom, 16Functional Neuroimaging Unit, CRIUGM, Université de Montréal, Montreal, QC, Canada

Detection of multiple sclerosis (MS) lesions in the spinal cord is clinically important for diagnosis and disease progression assessment. Although several automatic segmentation methods have been proposed for brain lesions, these methods cannot be directly applied to spinal lesions. We propose a fully automatic pipeline based on deep learning to segment the spinal cord and spinal MS lesions, and validate it against a dataset of T2-w images (265 patients from 5 centers). The proposed cord segmentation achieved better results than the current state-of-the-art, and lesion segmentation yielded a median Dice of 63.4%. The pipeline will be available as open-source.

This abstract and the presentation materials are available to members only; a login is required.

Join Here