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

• DOI: 10.3174/ajnr.a6020
• Publication Date: 2019-03-28T17:55:16Z
• Authors: D.B. McCoy; S.M. Dupont; C. Gros; J. Cohen-Adad; R.J. Huie; A. Ferguson; X. Duong-Fernandez; L.H. Thomas; V. Singh; J. Narvid; L. Pascual; N. Kyritsis; M.S. Beattie; J.C. Bresnahan; S. Dhall; W. Whetstone; J.F. Talbott
• Keywords: Male; Physical Injury - Accidents and Adverse Effects; Image Processing; Contusions; Clinical Sciences; Motor Disorders; Neurodegenerative; 03 medical and health sciences; Computer-Assisted; Deep Learning; 0302 clinical medicine; Image Interpretation, Computer-Assisted; Image Processing, Computer-Assisted; Humans; Spinal Cord Injury; Image Interpretation; Traumatic Head and Spine Injury; Spinal Cord Injuries; Neurosciences; Magnetic Resonance Imaging; Nuclear Medicine & Medical Imaging; Good Health and Well Being; Neurological; Biomedical Imaging; Female; TRACK-SCI Investigators

Our aim was to use 2D convolutional neural networks for automatic segmentation of the spinal cord and traumatic contusion injury from axial T2-weighted MR imaging in a cohort of patients with acute spinal cord injury.Forty-seven patients who underwent 3T MR imaging within 24 hours of spinal cord injury were included. We developed an image-analysis pipeline integrating 2D convolutional neural networks for whole spinal cord and intramedullary spinal cord lesion segmentation. Linear mixed modeling was used to compare test segmentation results between our spinal cord injury convolutional neural network (Brain and Spinal Cord Injury Center segmentation) and current state-of-the-art methods. Volumes of segmented lesions were then used in a linear regression analysis to determine associations with motor scores.Compared with manual labeling, the average test set Dice coefficient for the Brain and Spinal Cord Injury Center segmentation model was 0.93 for spinal cord segmentation versus 0.80 for PropSeg and 0.90 for DeepSeg (both components of the Spinal Cord Toolbox). Linear mixed modeling showed a significant difference between Brain and Spinal Cord Injury Center segmentation compared with PropSeg (P < .001) and DeepSeg (P < .05). Brain and Spinal Cord Injury Center segmentation showed significantly better adaptability to damaged areas compared with PropSeg (P < .001) and DeepSeg (P < .02). The contusion injury volumes based on automated segmentation were significantly associated with motor scores at admission (P = .002) and discharge (P = .009).Brain and Spinal Cord Injury Center segmentation of the spinal cord compares favorably with available segmentation tools in a population with acute spinal cord injury. Volumes of injury derived from automated lesion segmentation with Brain and Spinal Cord Injury Center segmentation correlate with measures of motor impairment in the acute phase. Targeted convolutional neural network training in acute spinal cord injury enhances algorithm performance for this patient population and provides clinically relevant metrics of cord injury.