Quantitative Issues in Cancer Research Working Seminar

Patrick Emedom-NnamdiDoctoral Student, Department of Biostatistics, Harvard University”Determining Surgical Recovery Strategies with Offline Reinforcement Learning”ABSTRACT: mproving patient recovery after a post-operative event continues to be a concern for many medical providers. We propose using offline reinforcement learning algorithms to identify how patients should adapt levels of physical activity and mobility at various post-operative time points in order to improve their recovery process. For this application, we utilize smartphone digital phenotyping data collected from neurosurgical patients with spine disease.