Automated Cervical Spinal Cord Segmentation in Real-World MRI of Multiple Sclerosis Patients by Optimized Hybrid Residual Attention-Aware Convolutional Neural Networks.

Magnetic resonance (MR) imaging is the most sensitive clinical tool in the diagnosis and monitoring of multiple sclerosis (MS) alterations. Spinal cord evaluation has gained interest in this clinical scenario in recent years, but, unlike the brain, there is a more limited choice of algorithms to assist spinal cord segmentation. Our goal was to investigate and develop an automatic MR cervical cord segmentation method, enabling automated and seamless spinal cord atrophy assessment and setting the stage for the development of an aggregated algorithm for the extraction of lesion-related imaging biomarkers. The algorithm was developed using a real-world MR imaging dataset of 121 MS patients (96 cases used as a training dataset and 25 cases as a validation dataset). Transversal, 3D T1-weighted gradient echo MR images (TE/TR/FA = 1.7-2.7 ms/5.6-8.2 ms/12°) were acquired in a 3 T system (Signa HD, GEHC) as standard of care in our clinical practice. Experienced radiologists supervised the manual labelling, which was considered the ground-truth. The 2D convolutional neural network consisted of a hybrid residual attention-aware segmentation method trained to delineate the cervical spinal cord. The training was conducted using a focal loss function, based on the Tversky index to address label imbalance, and an automatic optimal learning rate finder. Our automated model provided an accurate segmentation, achieving a validation DICE coefficient of 0.904 ± 0.101 compared with the manual delineation. An automatic method for cervical spinal cord segmentation on T1-weighted MR images was successfully implemented. It will have direct implications serving as the first step for accelerating the process for MS staging and follow-up through imaging biomarkers.

Oligoclonal M bands and cervical spinal cord lesions predict early secondary progressive multiple sclerosis.

Objective: To determine baseline cerebrospinal fluid and magnetic resonance imaging (MRI) variables at the onset of a clinically isolated syndrome (CIS) suggestive of multiple sclerosis (MS) that predict evolution to secondary progressive MS (SPMS). Methods: 276 CIS patients with a minimum follow-up of 10 years were studied. Baseline presence of oligoclonal IgG and IgM bands (OCGB and OCMB respectively); number of brain T2 lesions (B-T2L), brain gadolinium enhancement lesions (brain-GEL), cervical spinal cord T2 lesions (cSC-T2L); and fulfillment of 2017 McDonald criteria among other variables were collected. Results: 14 patients ended up with a non-MS condition. 138/276 CIS patients fulfilled 2017 McDonald criteria. Mean age was 32.4 years, 185 female. 227 received treatment, 95 as CIS. After a mean follow-up of 12 years, 36 patients developed SPMS. Conversion to SPMS was associated with OCGB (p = 0.02), OCMB (p = 0.0001); ≥ 9 B-T2L (p = 0.03), brain-GEL (p = 0.03), and cSC-T2L (p = 0.03). However, after adjusting for sex, age, BT2L, brain-GEL, SC-T2, and OCMB status, only OCMB (HR 4.4, 1.9-10.6) and cSC-T2L (HR 2.2, 1.0-6.2) suggested an independent association with risk of conversion to SPMS. Patients with both risk factors had a HR of 6.12 (2.8-12.9). Discussion: OCMB and SC-T2 lesions are potential independent predictors of conversion to SPMS.