When
1 – 2 p.m., Feb. 14, 2025
Where
Speaker: Nick Henscheid, (PhD alum 2018) Quantitative Medicine Scientist, Critical Path Institute
Title: Classical versus machine learning methods for modeling longitudinal disease data
Abstract: Understanding heterogeneity in disease progression is key to designing better clinical trials and moving towards precision medicine. There are many ways to model heterogeneous longitudinal data, including classical methods such as nonlinear mixed effects and modern machine learning methods like recurrent neural networks and transformers. I’ll talk about why we model disease progression at the Critical Path Institute and discuss some comparisons between classical and AI/ML methods.