Reliable brain morphometry from contrast‐enhanced T1w‐MRI in patients with multiple sclerosis

Abstract

Brain morphometry is usually based on non-enhanced (pre-contrast) T1-weighted MRI. However, such dedicated protocols are sometimes missing in clinical examinations. Instead, an image with a contrast agent is often available. Existing tools such as FreeSurfer yield unreliable results when applied to contrast-enhanced (CE) images.Consequently, these acquisitions are excluded from retrospective morphometry studies, which reduces the sample size. We hypothesize that deep learning (DL)-based morphometry methods can extract morphometric measures also from contrastenhanced MRI. We have extended DL+DiReCT to cope with contrast-enhanced MRI. Training data for our DL-based model were enriched with non-enhanced and CE image pairs from the same session. The segmentations were derived with Free-Surfer from the non-enhanced image and used as ground truth for the coregistered CE image. A longitudinal dataset of patients with multiple sclerosis (MS), comprising relapsing remitting (RRMS) and primary progressive (PPMS) subgroups, was used for the evaluation. Global and regional cortical thickness derived from non-enhanced and CE images were contrasted to results from FreeSurfer. Correlation coefficients of global mean cortical thickness between non-enhanced and CE images were significantly larger with DL+DiReCT (r = 0.92) than with FreeSurfer (r = 0.75). When comparing the longitudinal atrophy rates between the two MS subgroups, the effect sizes between PPMS and RRMS were higher with DL+DiReCT both for non-enhanced (d = À0.304) and CE images (d = À0.169) than for FreeSurfer (non-enhanced d = À0.111, CE d = 0.085). In conclusion, brain morphometry can be derived reliably from contrast-enhanced MRI using DL-based morphometry tools, making additional cases available for analysis and potential future diagnostic morphometry tools.

Multiple sclerosis as a model to investigate SARS‐CoV‐2 effect on brain atrophy

“…MRI data were processed using the freely available software packages FreeSurfer (version 6.0.0) and DL+DiReCT to derive global and regional cortical gray matter (GM) volume and thickness. 11 , 12 , 13 …”

“…Atrophy was quantified on unenhanced, high‐resolution magnetization‐prepared rapid acquisition with gradient echo sequence (MPRAGE) images. MRI data were processed using the freely available software packages FreeSurfer (version 6.0.0) and DL+DiReCT to derive global and regional cortical gray matter (GM) volume and thickness 11–13 …”

“…MRI data were processed using the freely available software packages FreeSurfer (version 6.0.0) and DL+DiReCT to derive global and regional cortical gray matter (GM) volume and thickness. [11][12][13] Patient-specific atrophy rates prior to infection were calculated individually for every patient with at least three pre-SARS-CoV-2 MRI scans that matched with at least one post-SARS-CoV-2 scan with respect to field strength, repetition time, and inversion time (Figure S2), leading to the final cohort of 14 patients (Figure S1). A linear model was fitted to the global and regional GM volumes of the matching pre-SARS-CoV-2 images using patient age as the explanatory variable.…”

Anti‐neurochondrin antibody as a biomarker in primary autoimmune cerebellar ataxia—a case report and review of the literature

“…Quantification of cerebellar volume was performed using DL+DiReCT [8], which allows brain morphometric measures to be derived from high‐resolution, contrast‐enhanced T1‐weighted magnetic resonance imaging (MRI) [9].…”

From basic sciences and engineering to epileptology: A translational approach

“…17 In the context of brain morphometry, CNNs have been used for high-quality segmentation and parcellation of tissue types and brain regions from unenhanced or contrast-enhanced T1-weighted MRI and subsequent cortical-thickness estimation. 18,19 Another aspect that might turn out relevant to the application of brain morphometry in epilepsy is the discovery of a universal and biophysically plausible scaling law of cortical folding. 20 Motivated by the large degree of covariance among variables of cortical morphometry (specifically cortical thickness, and total and exposed surface area), novel and mutually independent variables have been suggested and applied to epilepsy recently.…”