Healthcare is a deeply developed real-world case study of this broader problem. My work sits at the intersection of regulatory science, institutional design, and health policy — connecting high-level technology policy with what actually happens when institutions must make consequential decisions about deploying emerging technologies.
My doctoral research at the Johns Hopkins Bloomberg School of Public Health examines the efficacy of regulatory and ethical frameworks in managing patient harms associated with AI/ML-enabled digital-health technologies and clinical decision-support software — asking not just what the rules say, but whether they actually work once technologies reach clinical practice.
Co-created and administered Mayo Clinic's internal Software as a Medical Device Review Board, establishing the institutional process for determining when digital-health and AI technologies constitute regulated medical devices.
A generalizable framework for tiering AI systems by potential for harm, calibrating the level of independent review each tool requires before and after deployment.
Led enterprise-wide alignment on risk-based use of the licensed-practitioner exemption for AI-enabled clinical decision support — a provision written for simple rule-based software now being stretched to cover adaptive machine-learning and LLM-based tools.
Directed development of governance requirements for vended and internally developed non-device AI tools, and established the tracking infrastructure that became Mayo Clinic's enterprise Clinical AI Inventory.
Selected frameworks and tools will be published here as they become publicly shareable.