Dataset Information

Deciphering multiple sclerosis disability with deep learning attention maps on clinical MRI.

ABSTRACT: The application of convolutional neural networks (CNNs) to MRI data has emerged as a promising approach to achieving unprecedented levels of accuracy when predicting the course of neurological conditions, including multiple sclerosis, by means of extracting image features not detectable through conventional methods. Additionally, the study of CNN-derived attention maps, which indicate the most relevant anatomical features for CNN-based decisions, has the potential to uncover key disease mechanisms leading to disability accumulation. From a cohort of patients prospectively followed up after a first demyelinating attack, we selected those with T1-weighted and T2-FLAIR brain MRI sequences available for image analysis and a clinical assessment performed within the following six months (N = 319). Patients were divided into two groups according to expanded disability status scale (EDSS) score: ≥3.0 and < 3.0. A 3D-CNN model predicted the class using whole-brain MRI scans as input. A comparison with a logistic regression (LR) model using volumetric measurements as explanatory variables and a validation of the CNN model on an independent dataset with similar characteristics (N = 440) were also performed. The layer-wise relevance propagation method was used to obtain individual attention maps. The CNN model achieved a mean accuracy of 79% and proved to be superior to the equivalent LR-model (77%). Additionally, the model was successfully validated in the independent external cohort without any re-training (accuracy = 71%). Attention-map analyses revealed the predominant role of frontotemporal cortex and cerebellum for CNN decisions, suggesting that the mechanisms leading to disability accrual exceed the mere presence of brain lesions or atrophy and probably involve how damage is distributed in the central nervous system.

SUBMITTER: Coll L

PROVIDER: S-EPMC10034138 | biostudies-literature | 2023 Mar

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Deciphering multiple sclerosis disability with deep learning attention maps on clinical MRI.

Coll Llucia L Pareto Deborah D Carbonell-Mirabent Pere P Cobo-Calvo Álvaro Á Arrambide Georgina G Vidal-Jordana Ángela Á Comabella Manuel M Castilló Joaquín J Rodríguez-Acevedo Breogán B Zabalza Ana A Galán Ingrid I Midaglia Luciana L Nos Carlos C Salerno Annalaura A Auger Cristina C Alberich Manel M Río Jordi J Sastre-Garriga Jaume J Oliver Arnau A Montalban Xavier X Rovira Àlex À Tintoré Mar M Lladó Xavier X Tur Carmen C

NeuroImage. Clinical 20230315

The application of convolutional neural networks (CNNs) to MRI data has emerged as a promising approach to achieving unprecedented levels of accuracy when predicting the course of neurological conditions, including multiple sclerosis, by means of extracting image features not detectable through conventional methods. Additionally, the study of CNN-derived attention maps, which indicate the most relevant anatomical features for CNN-based decisions, has the potential to uncover key disease mechanis ...[more]

PMID: 36940621

Similar Datasets

Project description:Manually segmenting multiple sclerosis (MS) cortical lesions (CLs) is extremely time consuming, and past studies have shown only moderate inter-rater reliability. To accelerate this task, we developed a deep-learning-based framework (CLAIMS: Cortical Lesion AI-Based Assessment in Multiple Sclerosis) for the automated detection and classification of MS CLs with 7 T MRI. Two 7 T datasets, acquired at different sites, were considered. The first consisted of 60 scans that include 0.5 mm isotropic MP2RAGE acquired four times (MP2RAGE×4), 0.7 mm MP2RAGE, 0.5 mm T2 *-weighted GRE, and 0.5 mm T2 *-weighted EPI. The second dataset consisted of 20 scans including only 0.75 × 0.75 × 0.9 mm3 MP2RAGE. CLAIMS was first evaluated using sixfold cross-validation with single and multi-contrast 0.5 mm MRI input. Second, the performance of the model was tested on 0.7 mm MP2RAGE images after training with either 0.5 mm MP2RAGE×4, 0.7 mm MP2RAGE, or alternating the two. Third, its generalizability was evaluated on the second external dataset and compared with a state-of-the-art technique based on partial volume estimation and topological constraints (MSLAST). CLAIMS trained only with MP2RAGE×4 achieved results comparable to those of the multi-contrast model, reaching a CL true positive rate of 74% with a false positive rate of 30%. Detection rate was excellent for leukocortical and subpial lesions (83%, and 70%, respectively), whereas it reached 53% for intracortical lesions. The correlation between disability measures and CL count was similar for manual and CLAIMS lesion counts. Applying a domain-scanner adaptation approach and testing CLAIMS on the second dataset, the performance was superior to MSLAST when considering a minimum lesion volume of 6 μL (lesion-wise detection rate of 71% versus 48%). The proposed framework outperforms previous state-of-the-art methods for automated CL detection across scanners and protocols. In the future, CLAIMS may be useful to support clinical decisions at 7 T MRI, especially in the field of diagnosis and differential diagnosis of MS patients.

Project description:Next to the disseminated clinical symptoms, cognitive dysfunctions are common features of multiple sclerosis (MS). Over the recent years several different MRI measures became available representing the various features of the pathology, but the contribution to various clinical and cognitive functions is not yet fully understood. In this multiparametric MRI study we set out to identify the set of parameters that best predict the clinical and cognitive disability in MS. High resolution T1 weighted structural and high angular resolution diffusion MRI images were measured in 53 patients with relapsing remitting MS and 53 healthy controls. Clinical disability was inflicted by EDSS and cognitive functions were evaluated with the BICAMS tests. The contribution of lesion load, partial brain, white matter, gray matter and subcortical volumes as well as the diffusion parameters in the area of the lesions and the normal appearing white matter were examined by model free, partial least square (PLS) approach. Significance of the predictors was tested with Variable Importance in the Projection (VIP) score and 1 was used for threshold of significance. The PLS analysis indicated that the axial diffusivity of the NAWM contributed the most to the clinical disability (VIP score: 1.979). For the visuo-spatial working memory the most critical contributor was the size of the bilateral hippocampi (VIP scores: 1.183 and 1.2 left and right respectively). For the verbal memory the best predictors were the size of the right hippocampus (VIP score: 1.972), lesion load (VIP score: 1.274) and the partial brain volume (VIP score: 1.119). In case of the information processing speed the most significant contribution was from the diffusion parameters (fractional anisotropy, mean and radial diffusivity, VIP scores: 1.615, 1.321 respectively) of the normal appearing white matter. Our results indicate that various MRI measurable factors of MS pathology contribute differently to clinical and cognitive disability. These results point out the importance of the volumetry of the subcortical structures and the diffusion measures of the white matter in understanding the disability progression.

Dataset Information

Deciphering multiple sclerosis disability with deep learning attention maps on clinical MRI.

Publications

Deciphering multiple sclerosis disability with deep learning attention maps on clinical MRI.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets