Hairong Wang Institute: Georgia Institute of Technology Date: April 10, 2025 Title: Advancing precision medicine: Knowledge-informed methods with data efficiency Abstract: In recent decades, machine learning (ML) has emerged as a promising tool for analyzing complex patterns from large datasets. The computational power and versatility of ML has enabled in-depth analysis of medical imaging, clinical, and molecular data, significantly enhancing diagnosis, prognosis, and treatment planning in healthcare. However, an intrinsic bottleneck exists in healthcare data acquisition, limited by the invasiveness or high expense of sample collection, the need for highly-specialized experts to create accurate labels, the rarity of some diseases in the population, and the difficulty in patient recruitment. In this talk, I will discuss my recent development on enhancing data efficiency in the context of precision medicine (PM). Motivated by the practical limitations, we developed knowledge-informed, data efficient algorithms to address the challenges of labeled sample size and implicit hierarchical knowledge integration, aiming for practical solutions in PM. These works have been applied in real-world contexts in collaboration with Columbia University Medical Center and Mayo Clinic, demonstrating considerable potential in boosting the accuracy, robustness, and interpretability of model outcomes.
