Abstract
Myasthenia gravis (MG) is a rare autoimmune neuromuscular disease. Clinical trials with rigorously collected data, especially for rare diseases, provide opportunities for mathematical modeling of patient outcomes over time; however, building a larger data set from multiple trials faces the challenge of harmonization of outcome measures. To accurately model MG and predict individual patient trajectories, one requires integrating three primary data types: (i) Laboratory and medication data, (ii) Electronic Health Record (EHR) data (e.g., age, sex, years since diagnosis, BMI), (iii) Disease severity scores. Among these, MG severity scores are crucial for measuring disease progression from the patient's and clinical evaluator's perspectives. However, clinical studies often employ various scoring systems (e.g., ADL, QMG, MG-CE, MGQOL-15), making it challenging to determine the most reliable measure. In this study, we investigate the relationships among clinical outcome measures across multiple clinical studies. Our objective is to develop a robust "Myasthenia Gravis Portrait" that can be applied across diverse clinical studies. This standardized portrait will facilitate the creation of a virtual population of digital twins, enabling the application of machine learning techniques to a larger patient population.