Diffusion Tensor Imaging in Amyotrophic Lateral Sclerosis: Machine Learning for Biomarker Development

Abstract

Diffusion tensor imaging (DTI) allows the in vivo imaging of pathological white matter alterations, either with unbiased voxel-wise or hypothesis-guided tract-based analysis. Alterations of diffusion metrics are indicative of the cerebral status of patients with amyotrophic lateral sclerosis (ALS) at the individual level. Using machine learning (ML) models to analyze complex and high-dimensional neuroimaging data sets, new opportunities for DTI-based biomarkers in ALS arise. This review aims to summarize how different ML models based on DTI parameters can be used for supervised diagnostic classifications and to provide individualized patient stratification with unsupervised approaches in ALS. To capture the whole spectrum of neuropathological signatures, DTI might be combined with additional modalities, such as structural T1w 3-D MRI in ML models. To further improve the power of ML in ALS and enable the application of deep learning models, standardized DTI protocols and multi-center collaborations are needed to validate multimodal DTI biomarkers. The application of ML models to multiparametric MRI/multimodal DTI-based data sets will enable a detailed assessment of neuropathological signatures in patients with ALS and the development of novel neuroimaging biomarkers that could be used in the clinical workup.

Toward diffusion tensor imaging as a biomarker in neurodegenerative diseases: technical considerations to optimize recordings and data processing

“…Furthermore, ML was applied to the differentiation of parkinsonian syndromes ( Haller et al, 2012 ; Du et al, 2017 ; Chougar et al, 2021 ; Talai et al, 2021 ). Since first applications of ML methods to ALS ( Welsh et al, 2013 ; Sarica et al, 2017 ), ML was used to improve diagnostic accuracy ( Kocar et al, 2021 ; Behler et al, 2023 ) and clinical associations ( Li et al, 2021 ). For the evaluation of imaging biomarkers in HD, ML methods have been applied for many years ( Klöppel et al, 2008 ; Rizk-Jackson et al, 2011 ; Hu B. et al, 2023 ).…”

Machine learning classificatory as a tool in the diagnosis of amyotrophic lateral sclerosis using diffusion tensor imaging parameters collected with 1.5T MRI scanner: A case study

“…Currently, ML is widely used in ALS research, for example, to find and analyse biomedical signals inherent in this disease, identify and predict clinical subgroups, evaluate the results of muscle ultrasound in the early stages of ALS [27][28][29][30] etc. In addition, there are several publications devoted to using ML to analyse images of the brain affected by the disease obtained by neuroimaging and, in particular, diffusion tensor images [24,[31][32][33][34][35]. This particular MRI method can detect changes in the white matter that occur in ALS by measuring differences in restrictions on water diffusion in the brain.…”

“…This particular MRI method can detect changes in the white matter that occur in ALS by measuring differences in restrictions on water diffusion in the brain. Summarizing the results of a systematic review of using suiML algorithms to analyse data obtained by DTI-based neuroimaging, A. Behler et al [31] note their enormous academic and clinical potential in developing biomarkers for this disease, which can be useful both at the group and individual level, for example, to improve individual differential diagnosis or serve as endpoints in clinical trials.…”

Fluid biomarkers for amyotrophic lateral sclerosis: a review

“…These endeavors have benefited from the continuous growth of our knowledge of ALS pathomechanisms and advances in technologies and bioinformatics. Over the course of 20 + years, several fluid biomarker candidates for ALS have emerged, as have numerous non-fluid biomarkers ranging from clinical to radiographic markers [ 27 – 32 ]. Here, we provide an overview of the more promising fluid biomarkers for ALS, highlight exciting new strategies towards testing the utility of TDP-43—a protein intimately involved in ALS pathogenesis—as a reliable biomarker, and discuss present limitations and future avenues being explored for ALS biomarker discovery.…”