Mismatch between clinically defined classification of ALS stage and the burden of cerebral pathology.

This study aimed to investigate the clinical stratification of amyotrophic lateral sclerosis (ALS) patients in relation to in vivo cerebral degeneration. One hundred forty-nine ALS patients and one hundred forty-four healthy controls (HCs) were recruited from the Canadian ALS Neuroimaging Consortium (CALSNIC). Texture analysis was performed on T1-weighted scans to extract the texture feature "autocorrelation" (autoc), an imaging biomarker of cerebral degeneration. Patients were stratified at baseline into early and advanced disease stages based on criteria adapted from ALS clinical trials and the King's College staging system, as well as into slow and fast progressors (disease progression rates, DPR). Patients had increased autoc in the internal capsule. These changes extended beyond the internal capsule in early-stage patients (clinical trial-based criteria), fast progressors, and in advanced-stage patients (King's staging criteria). Longitudinal increases in autoc were observed in the postcentral gyrus, corticospinal tract, posterior cingulate cortex, and putamen; whereas decreases were observed in corpus callosum, caudate, central opercular cortex, and frontotemporal areas. Both longitudinal increases and decreases of autoc were observed in non-overlapping regions within insula and precentral gyrus. Within-criteria comparisons of autoc revealed more pronounced changes at baseline and longitudinally in early- (clinical trial-based criteria) and advanced-stage (King's staging criteria) patients and fast progressors. In summary, comparative patterns of baseline and longitudinal progression in cerebral degeneration are dependent on sub-group selection criteria, with clinical trial-based stratification insufficiently characterizing disease stage based on pathological cerebral burden.

Accurate personalized survival prediction for amyotrophic lateral sclerosis patients.

Amyotrophic Lateral Sclerosis (ALS) is a rapidly progressive neurodegenerative disease. Accurately predicting the survival time for ALS patients can help patients and clinicians to plan for future treatment and care. We describe the application of a machine-learned tool that incorporates clinical features and cortical thickness from brain magnetic resonance (MR) images to estimate the time until a composite respiratory failure event for ALS patients, and presents the prediction as individual survival distributions (ISDs). These ISDs provide the probability of survival (none of the respiratory failures) at multiple future time points, for each individual patient. Our learner considers several survival prediction models, and selects the best model to provide predictions. We evaluate our learned model using the mean absolute error margin (MAE-margin), a modified version of mean absolute error that handles data with censored outcomes. We show that our tool can provide helpful information for patients and clinicians in planning future treatment.

Effects of MRI scanner manufacturers in classification tasks with deep learning models.

Deep learning has become a leading subset of machine learning and has been successfully employed in diverse areas, ranging from natural language processing to medical image analysis. In medical imaging, researchers have progressively turned towards multi-center neuroimaging studies to address complex questions in neuroscience, leveraging larger sample sizes and aiming to enhance the accuracy of deep learning models. However, variations in image pixel/voxel characteristics can arise between centers due to factors including differences in magnetic resonance imaging scanners. Such variations create challenges, particularly inconsistent performance in machine learning-based approaches, often referred to as domain shift, where the trained models fail to achieve satisfactory or improved results when confronted with dissimilar test data. This study analyzes the performance of multiple disease classification tasks using multi-center MRI data obtained from three widely used scanner manufacturers (GE, Philips, and Siemens) across several deep learning-based networks. Furthermore, we investigate the efficacy of mitigating scanner vendor effects using ComBat-based harmonization techniques when applied to multi-center datasets of 3D structural MR images. Our experimental results reveal a substantial decline in classification performance when models trained on one type of scanner manufacturer are tested with data from different manufacturers. Moreover, despite applying ComBat-based harmonization, the harmonized images do not demonstrate any noticeable performance enhancement for disease classification tasks.