Longitudinal Data Analysis

Fillmore has employed longitudinal data analysis to identify early biomarkers and predictive patterns for neurodegenerative and oncologic diseases. His work utilizes deep sequential modeling of electronic health records to predict pancreatic cancer risk years before clinical diagnosis and uses long-term cohort data to establish the association between systemic inflammatory markers and the subsequent development of dementia. These studies demonstrate how the analysis of temporal health data can refine risk stratification and improve early detection for high-mortality conditions.