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.

From the NeuroPoly Laboratory Institute of Biomedical Engineering Polytechnique Montréal 2500 Chemin de Polytechnique Montréal Québec Canada H3T 1J4

medRxiv. 2024 Apr 21:2024.01.03.24300794. doi: 10.1101/2024.01.03.24300794. PubMed

Zobrazit více v PubMed

Ahuja CS , Wilson JR , Nori S , et al. . Traumatic spinal cord injury . Nat Rev Dis Primers 2017. ; 3 ( 1 ): 17018 . PubMed

Freund P , Seif M , Weiskopf N , et al. . MRI in traumatic spinal cord injury: from clinical assessment to neuroimaging biomarkers . Lancet Neurol 2019. ; 18 ( 12 ): 1123 – 1135 . PubMed

David G , Mohammadi S , Martin AR , et al. . Traumatic and nontraumatic spinal cord injury: pathological insights from neuroimaging . Nat Rev Neurol 2019. ; 15 ( 12 ): 718 – 731 . PubMed

Scivoletto G , Torre M , Mammone A , et al. . Acute Traumatic and Ischemic Spinal Cord Injuries Have a Comparable Course of Recovery . Neurorehabil Neural Repair 2020. ; 34 ( 8 ): 723 – 732 . PubMed

New PW , Cripps RA , Bonne Lee B . Global maps of non-traumatic spinal cord injury epidemiology: towards a living data repository . Spinal Cord 2014. ; 52 ( 2 ): 97 – 109 . [Published correction appears in Spinal Cord 2014;52(5):417.] PubMed

Iseli E , Cavigelli A , Dietz V , Curt A . Prognosis and recovery in ischaemic and traumatic spinal cord injury: clinical and electrophysiological evaluation . J Neurol Neurosurg Psychiatry 1999. ; 67 ( 5 ): 567 – 571 . PubMed PMC

Pfyffer D , Huber E , Sutter R , Curt A , Freund P . Tissue bridges predict recovery after traumatic and ischemic thoracic spinal cord injury . Neurology 2019. ; 93 ( 16 ): e1550 – e1560 . PubMed PMC

Miyanji F , Furlan JC , Aarabi B , Arnold PM , Fehlings MG . Acute cervical traumatic spinal cord injury: MR imaging findings correlated with neurologic outcome--prospective study with 100 consecutive patients . Radiology 2007. ; 243 ( 3 ): 820 – 827 . PubMed

Dobran M , Aiudi D , Liverotti V , et al. . Prognostic MRI parameters in acute traumatic cervical spinal cord injury . Eur Spine J 2023. ; 32 ( 5 ): 1584 – 1590 . PubMed

Huber E , Lachappelle P , Sutter R , Curt A , Freund P . Are midsagittal tissue bridges predictive of outcome after cervical spinal cord injury? Ann Neurol 2017. ; 81 ( 5 ): 740 – 748 . PubMed

Smith AC , Albin SR , O’Dell DR , et al. . Axial MRI biomarkers of spinal cord damage to predict future walking and motor function: a retrospective study . Spinal Cord 2021. ; 59 ( 6 ): 693 – 699 . PubMed PMC

Kurpad S , Martin AR , Tetreault LA , et al. . Impact of Baseline Magnetic Resonance Imaging on Neurologic, Functional, and Safety Outcomes in Patients With Acute Traumatic Spinal Cord Injury . Global Spine J 2017. ; 7 ( 3 Suppl ): 151S – 174S . PubMed PMC

Martin AR , De Leener B , Cohen-Adad J , et al. . A novel MRI biomarker of spinal cord white matter injury: T2*-weighted white matter to gray matter signal intensity ratio . AJNR Am J Neuroradiol 2017. ; 38 ( 6 ): 1266 – 1273 . PubMed PMC

Badhiwala JH , Ahuja CS , Akbar MA , et al. . Degenerative cervical myelopathy - update and future directions . Nat Rev Neurol 2020. ; 16 ( 2 ): 108 – 124 . PubMed

De Leener B , Lévy S , Dupont SM , et al. . SCT: Spinal Cord Toolbox, an open-source software for processing spinal cord MRI data . Neuroimage 2017. ; 145 ( Pt A ): 24 – 43 . PubMed

Bischof A , Papinutto N , Keshavan A , et al. . Spinal Cord Atrophy Predicts Progressive Disease in Relapsing Multiple Sclerosis . Ann Neurol 2022. ; 91 ( 2 ): 268 – 281 . PubMed PMC

Smith AC , Weber KA 2nd , O’Dell DR , Parrish TB , Wasielewski M , Elliott JM . Lateral Corticospinal Tract Damage Correlates With Motor Output in Incomplete Spinal Cord Injury . Arch Phys Med Rehabil 2018. ; 99 ( 4 ): 660 – 666 . PubMed PMC

Mummaneni N , Burke JF , DiGiorgio AM , et al. . Injury volume extracted from MRI predicts neurologic outcome in acute spinal cord injury: A prospective TRACK-SCI pilot study . J Clin Neurosci 2020. ; 82 ( Pt B ): 231 – 236 . PubMed

Vallotton K , Huber E , Sutter R , Curt A , Hupp M , Freund P . Width and neurophysiologic properties of tissue bridges predict recovery after cervical injury . Neurology 2019. ; 92 ( 24 ): e2793 – e2802 . PubMed PMC

Pfyffer D , Vallotton K , Curt A , Freund P . Predictive Value of Midsagittal Tissue Bridges on Functional Recovery After Spinal Cord Injury . Neurorehabil Neural Repair 2021. ; 35 ( 1 ): 33 – 43 . PubMed PMC

Khan O , Badhiwala JH , Wilson JRF , Jiang F , Martin AR , Fehlings MG . Predictive Modeling of Outcomes After Traumatic and Nontraumatic Spinal Cord Injury Using Machine Learning: Review of Current Progress and Future Directions . Neurospine 2019. ; 16 ( 4 ): 678 – 685 . PubMed PMC

Dietz N , Vaitheesh Jaganathan , Alkin V , Mettille J , Boakye M , Drazin D . Machine learning in clinical diagnosis, prognostication, and management of acute traumatic spinal cord injury (SCI): A systematic review . J Clin Orthop Trauma 2022. ; 35 : 102046 . PubMed PMC

Asgari Taghanaki S , Abhishek K , Cohen JP , Cohen-Adad J , Hamarneh G . Deep semantic segmentation of natural and medical images: a review . Artif Intell Rev 2021. ; 54 ( 1 ): 137 – 178 .

McCoy DB , Dupont SM , Gros C , et al. . Convolutional Neural Network-Based Automated Segmentation of the Spinal Cord and Contusion Injury: Deep Learning Biomarker Correlates of Motor Impairment in Acute Spinal Cord Injury . AJNR Am J Neuroradiol 2019. ; 40 ( 4 ): 737 – 744 . PubMed PMC

Rupp R , Biering-Sørensen F , Burns SP , et al. . International Standards for Neurological Classification of Spinal Cord Injury: Revised 2019 . Top Spinal Cord Inj Rehabil 2021. ; 27 ( 2 ): 1 – 22 . PubMed PMC

Smith AC , O’Dell DR , Thornton WA , et al. . Spinal Cord Tissue Bridges Validation Study: Predictive Relationships With Sensory Scores Following Cervical Spinal Cord Injury . Top Spinal Cord Inj Rehabil 2022. ; 28 ( 2 ): 111 – 115 . PubMed PMC

Cohen-Adad J , El Mendili MM , Lehéricy S , et al. . Demyelination and degeneration in the injured human spinal cord detected with diffusion and magnetization transfer MRI . Neuroimage 2011. ; 55 ( 3 ): 1024 – 1033 . PubMed

Gorgolewski KJ , Auer T , Calhoun VD , et al. . The brain imaging data structure, a format for organizing and describing outputs of neuroimaging experiments . Sci Data 2016. ; 3 ( 1 ): 160044 . PubMed PMC

Gros C , De Leener B , Badji A , et al. . Automatic segmentation of the spinal cord and intramedullary multiple sclerosis lesions with convolutional neural networks . Neuroimage 2019. ; 184 : 901 – 915 . PubMed PMC

Bédard S , Enamundram NK , Tsagkas C , et al. . Towards contrast-agnostic soft segmentation of the spinal cord . ArXiv 2310.15402 [preprint] https://arxiv.org/abs/2310.15402. Posted October 23, 2023. Accessed November 2023 . PubMed

Isensee F , Jaeger PF , Kohl SAA , Petersen J , Maier-Hein KH . nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation . Nat Methods 2021. ; 18 ( 2 ): 203 – 211 . PubMed

Budd S , Robinson EC , Kainz B . A survey on active learning and human-in-the-loop deep learning for medical image analysis . Med Image Anal 2021. ; 71 : 102062 . PubMed

De Leener B , Kadoury S , Cohen-Adad J . Robust, accurate and fast automatic segmentation of the spinal cord . Neuroimage 2014. ; 98 : 528 – 536 . PubMed

Commowick O , Istace A , Kain M , et al. . Objective Evaluation of Multiple Sclerosis Lesion Segmentation using a Data Management and Processing Infrastructure . Sci Rep 2018. ; 8 ( 1 ): 13650 . PubMed PMC

Virtanen P , Gommers R , Oliphant TE , et al. . SciPy 1.0: fundamental algorithms for scientific computing in Python . Nat Methods 2020. ; 17 ( 3 ): 261 – 272 . [Published correction appears in Nat Methods 2020;17(3):352.] PubMed PMC

Blanc C , Shahrampour S , Mohamed FB , de Leener B . Combining PropSeg and a convolutional neural network for automatic spinal cord segmentation in pediatric populations and patients with spinal cord injury . Int J Imaging Syst Technol 2023. ; 33 ( 4 ): 1396 – 1405 .

Nouri A , Martin AR , Kato S , Reihani-Kermani H , Riehm LE , Fehlings MG . The Relationship Between MRI Signal Intensity Changes, Clinical Presentation, and Surgical Outcome in Degenerative Cervical Myelopathy: Analysis of a Global Cohort . Spine 2017. ; 42 ( 24 ): 1851 – 1858 . PubMed

Martin AR , Tadokoro N , Tetreault L , et al. . Imaging Evaluation of Degenerative Cervical Myelopathy: Current State of the Art and Future Directions . Neurosurg Clin N Am 2018. ; 29 ( 1 ): 33 – 45 . PubMed

Scheuren PS , David G , Kramer JLK , et al. . Combined Neurophysiologic and Neuroimaging Approach to Reveal the Structure-Function Paradox in Cervical Myelopathy . Neurology 2021. ; 97 ( 15 ): e1512 – e1522 . PubMed

Smith AC , Weber KA , Parrish TB , et al. . Ambulatory function in motor incomplete spinal cord injury: a magnetic resonance imaging study of spinal cord edema and lower extremity muscle morphometry . Spinal Cord 2017. ; 55 ( 7 ): 672 – 678 . PubMed PMC

Cummins DP , Connor JR , Heller KA , et al. . Establishing the inter-rater reliability of spinal cord damage manual measurement using magnetic resonance imaging . Spinal Cord Ser Cases 2019. ; 5 ( 1 ): 20 . PubMed PMC