Related Experiment Video
Updated: Jun 14, 2025
Diffusion Tensor Magnetic Resonance Imaging in Chronic Spinal Cord Compression
Published on: May 7, 2019
SCIseg: Automatic Segmentation of Intramedullary Lesions in Spinal Cord Injury on T2-weighted MRI Scans
Enamundram Naga Karthik1, Jan Valošek1, Andrew C Smith1
1From the NeuroPoly Laboratory, Institute of Biomedical Engineering, Polytechnique Montréal, 2500 Chemin de Polytechnique, Montréal, Québec, Canada H3T 1J4 (E.N.K., J.V., J.C.A.); Mila-Quebec AI Institute, Montréal, Québec, Canada (E.N.K., J.V., J.C.A.); Department of Neurosurgery and Department of Neurology, Faculty of Medicine and Dentistry, Palacký University Olomouc, Olomouc, Czechia (J.V.); Department of Physical Medicine and Rehabilitation Physical Therapy Program, University of Colorado School of Medicine, Aurora, Colo (A.C.S.); Spinal Cord Injury Center, Balgrist University Hospital, University of Zürich, Zürich, Switzerland (D.P., S.S.S., L.F., P.F.); Department of Anesthesiology, Perioperative and Pain Medicine, Stanford University School of Medicine, Stanford, Calif (D.P., K.A.W.); Department of Neurophysics, Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany (P.F.); and Functional Neuroimaging Unit, CRIUGM and Centre de Recherche du CHU Sainte-Justine, Université de Montréal, Montréal, Québec, Canada (J.C.A.).
Abstract:
Purpose To develop a deep learning tool for the automatic segmentation of the spinal cord and intramedullary lesions in spinal cord injury (SCI) on T2-weighted MRI scans. Materials and Methods This retrospective study included MRI data acquired between July 2002 and February 2023. The data consisted of T2-weighted MRI scans acquired using different scanner manufacturers with various image resolutions (isotropic and anisotropic) and orientations (axial and sagittal). Patients had different lesion etiologies (traumatic, ischemic, and hemorrhagic) and lesion locations across the cervical, thoracic, and lumbar spine. A deep learning model, SCIseg (which is open source and accessible through the Spinal Cord Toolbox, version 6.2 and above), was trained in a three-phase process involving active learning for the automatic segmentation of intramedullary SCI lesions and the spinal cord. The segmentations from the proposed model were visually and quantitatively compared with those from three other open-source methods (PropSeg, DeepSeg, and contrast-agnostic, all part of the Spinal Cord Toolbox). The Wilcoxon signed rank test was used to compare quantitative MRI biomarkers of SCI (lesion volume, lesion length, and maximal axial damage ratio) derived from the manual reference standard lesion masks and biomarkers obtained automatically with SCIseg segmentations. Results The study included 191 patients with SCI (mean age, 48.1 years ± 17.9 [SD]; 142 [74%] male patients). SCIseg achieved a mean Dice score of 0.92 ± 0.07 and 0.61 ± 0.27 for spinal cord and SCI lesion segmentation, respectively. There was no evidence of a difference between lesion length (P = .42) and maximal axial damage ratio (P = .16) computed from manually annotated lesions and the lesion segmentations obtained using SCIseg. Conclusion SCIseg accurately segmented intramedullary lesions on a diverse dataset of T2-weighted MRI scans and automatically extracted clinically relevant lesion characteristics. Keywords: Spinal Cord, Trauma, Segmentation, MR Imaging, Supervised Learning, Convolutional Neural Network (CNN) Published under a CC BY 4.0 license.
More Related Videos
12:50Lesion Explorer: A Video-guided, Standardized Protocol for Accurate and Reliable MRI-derived Volumetrics in Alzheimer's Disease and Normal Elderly
Published on: April 14, 2014
10:25Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
Published on: September 25, 2019