Abstract:
|
Amyotrophic Lateral Sclerosis (ALS), also known as Lou Gehrig's disease, is a rare disease with extreme between-subject variability, especially with respect to rate of disease progression. This combined with a typically small number of longitudinal measurements per subject makes modelling a subject's disease progression, which is measured by the ALS Functional Rating Scale (ALSFRS), very difficult. Consider the problem of predicting a subject's ALSFRS score at 9 or 12 months after a given time-point. Here we investigate how leveraging an additional artificial data-point using time of disease onset can improve this prediction. The predictive mean-square-error using this onset-anchor is drastically reduced when compared to the MSE provided by either a Bayesian hierarchical linear model (similar to a mixed-effects model) or a Bayesian mixture model.
|