Abstract
This work presents a computational model to predict the progression of Amyotrophic Lateral Sclerosis (ALS) by integrating longitudinal clinical data, blood biomarkers, and immune cell profiles collected during a clinical study. Using machine learning techniques, the researchers identified the variables most strongly associated with changes in the ALS Functional Rating Scale–Revised (ALSFRS-R), a widely used measure of disease progression.
The model confirms the prognostic value of established biomarkers such as neurofilament light and heavy chains, while also highlighting the role of specific regulatory T-cell populations in disease progression. Although further validation on larger patient cohorts is needed, the results demonstrate the potential of computational approaches to improve the prediction of ALS progression and support more personalized disease monitoring and clinical decision-making.