Deep Learning Corpus Callosum Segmentation as a Neurodegenerative Marker in Multiple Sclerosis

Abstract

Background and Purpose Corpus callosum atrophy is a sensitive biomarker of multiple sclerosis (MS) neurodegeneration but typically requires manual 2D or volumetric 3D‐based segmentations. We developed a supervised machine learning algorithm, DeepnCCA, for corpus callosum segmentation and relate callosal morphology to clinical disability using conventional MRI scans collected in clinical routine. Methods In a prospective study of 553 MS patients with 704 acquisitions, 200 unique 2D T2‐weighted MRI scans were delineated to develop, train, and validate DeepnCCA. Comparative FreeSurfer segmentations were obtained in 504 3D T1‐weighted scans. Both FreeSurfer and DeepnCCA outputs were correlated with clinical disability. Using principal component analysis of the DeepnCCA output, the morphological changes were explored in relation to clinical disease burden. Results DeepnCCA and manual segmentations had high similarity (Dice coefficients 98.1±.11%, 89.3±.76%, for intracranial and corpus callosum area, respectively through 10‐fold cross‐validation). DeepnCCA had numerically stronger correlations with cognitive and physical disability as compared to FreeSurfer: Expanded disability status scale (EDSS) ±6 months (r = –.22 P = .002; r = –.17, P = .013), future EDSS (r = –.26, P<.001; r = –.17, P = .012), and future symbol digit modalities test (r = .26, P = .001; r = .24, P = .003). The corpus callosum became thinner with increasing cognitive and physical disability. Increasing physical disability, additionally, significantly correlated with a more angled corpus callosum. Conclusions DeepnCCA (https://github.com/plattenmichael/DeepnCCA/) is an openly available tool that can provide fast and accurate corpus callosum measurements applicable to large MS cohorts, potentially suitable for monitoring disease progression and therapy response.

Automatic deep learning multicontrast corpus callosum segmentation in multiple sclerosis

“…The same approach showed high and consistent performance across Compared to a previous study by our group, 14 the main technical novelties of the present work are the introduction of an automatic midslice selection, the possibility of running multiple MRI sequences, and an adaptation for modern 3-dimensional volumetric sequences.…”

“…Following the same strategy presented in previous literature, 14 Thereafter, the intraclass correlation coefficient (ICC) was calculated as a metric of precision, using the ICC(A,1) model. 26 We also aimed to investigate if the automatic midslice selection (midCNN) introduced bias into the CC-Net and IC-Net segmentations.…”

“…12 Previous studies have applied CNNs to segment the CC in 2 dimensions in healthy volunteers as well as in MS but on older sagittal two-dimensional T2-weighted (T2w) and T1-weighted (T1w) images. 13,14 Meanwhile, the current MAGNIMS-CMSC-NAIMS consensus guidelines recommend that 3-dimensional T2w fluid-attenuated inversion recovery (FLAIR) and T1w images are acquired for diagnosis and monitoring in MS. 15,16 Therefore, we developed a fully automatic, computationally efficient deep learning-based segmentation tool tailored for modern 3dimensional FLAIR and T1w scans in MS. The aim was to provide a biomarker of neurodegeneration in MS feasible to be run on large datasets.…”

“…In small sample sizes, a U‐net architecture is popular and it is often combined with effective image augmentation for improved classification 12 . Previous studies have applied CNNs to segment the CC in 2 dimensions in healthy volunteers as well as in MS but on older sagittal two‐dimensional T2‐weighted (T2w) and T1‐weighted (T1w) images 13,14 . Meanwhile, the current MAGNIMS‐CMSC‐NAIMS consensus guidelines recommend that 3‐dimensional T2w fluid‐attenuated inversion recovery (FLAIR) and T1w images are acquired for diagnosis and monitoring in MS 15,16 …”

Machine Learning Approaches in Study of Multiple Sclerosis Disease Through Magnetic Resonance Images

“…In addition, the MS course in patients, even from the early stages, is characterized by a slow progression of disabilities independent of relapses (58). Automatic segmentation of MS lesions, including gadolinium-enhancing, new T2 or enlarging T2, are essential biomarkers for the progression of the disease as we as the treatment options and allow to explore the morphological changes in relation to clinical disease burden (48,59). Our review shows that three studies in this section applied ML to predict the progression of MS by measuring Expanded Disability Status Scale (EDSS) in the first years of disease evolution (38,46,49).…”