Peter MacDonald
Assistant Professor, University of Waterloo
Room: M3 3127
Patient Similarity in Personalized Predictive Modelling: A Weighted Cosine Similarity Approach and Comparative Evaluation of Metrics
Personalized Predictive Modelling (PPM) has been growing rapidly in recent years, especially with the availability of Electronic Health Records (EHRs). This approach aims to improve a model's predictive performance by fitting a unique model to each individual, where the model is trained on a subset of the training data consisting of individuals who are similar to the individual of interest. In this work, we introduce a weighted cosine similarity metric that extends the standard cosine similarity metric by assigning predictor-specific weights when computing similarity between participants. Results from our simulation study and an analysis of intensive care unit data involving patients with circulatory system disease show that, although the proposed similarity metric leads to a slight deterioration in calibration, it produces substantial gains in discrimination. We conclude with a comprehensive comparison of several similarity metrics to investigate how the choice of similarity metric influences predictive performance in PPM.