Following his doctoral training in Medical Physics at UCLA and a subsequent fellowship and clinical residency in Therapeutic Medical Physics at the University of Pennsylvania, he embarked on a journey to explore and develop machine learning algorithms and advanced models for their application in medicine. These were constructed with the aim to predict patient outcomes and tailor treatments according to the type and quality of data available, with a particular focus on prostate and thoracic cancer patients. In addition to these interests, his professional journey has led me to develop methods for inspecting and interpreting machine learning models through innovative initiatives like MediBoost, The Additive Tree, Expert Augmented Machine Learning, The Conditional Super Learner, and Representational Gradient Boosting. He has also garnered significant experience in creating and maintaining R and Python packages to meet the unique data analysis needs within oncology, demonstrated by my contributions to EAML, LINAD, and Lockout. In his current role as Director of the "Advanced Machine Learning" graduate course at UCSF, where I also hold joint appointments in the Radiation Oncology and Biostatistics departments, he is committed to developing machine learning algorithms specifically tailored to medicine. More recently, his interests have evolved towards the fascinating convergence of Interpretable Machine Learning and Targeted Learning. This symbiosis has been aimed at delivering transparent and robust Causal Machine Learning inference. His dedication to this field was acknowledged by the NIH NIBIB K08 award, which provided substantial support for his work in crafting precise and interpretable machine learning algorithms across a spectrum of clinical applications.
Machine Learning, Causal Inference, Interpretability, Cancer Research, Algorithm development for Medical Applications
