Center for Functional Data Analysis

The goal is to establish a Center of functional data analysis with a particular focus on online prediction of clinical adverse events. The types of data focused will be continuously collected data such as ECG, EEG, optical imaging data, longitudinal biomarkers, vital sign data, and data collected through wearable devices or other online monitors.

The overarching aim is to use patient-level, continuous data for feature extraction and a) to train prediction models; b) to validate the results prospectively; and c) to design timely interventions to prevent acute and subacute complications of disease. We will collaborate broadly with investigators in the University of Pennsylvania Perelman School of Medicine to analyze these types of data and to develop methodology to meet the new challenges.