This method makes it possible to simulate the clinical course of patients with Amyotrophic Lateral Sclerosis (ALS) starting from the variables measured during a single visit (e.g., the first one).
It allows prediction of disease evolution over time, showing the probability of losing functional independence in four specific domains (breathing, swallowing, communication, and walking/self-care), along with survival.
Based on Dynamic Bayesian Networks (DBNs), the method has been implemented on a cohort of more than 4,000 patients from 6 different international clinical centres and validated on an independent test set.
Enabling the generation of in silico population with specific features, the present invention was developed for the purpose of supporting clinical trial as an alternative to placebo cohorts. In addition, it can be used in clinical settings to support medical decision-making.