Course Director: Faculty
3 credits
This course provides an overview of implementation science and examines ethical, methodological, and practical challenges in the use of artificial intelligence (AI) in healthcare. It focuses on the safe and effective development, evaluation, and implementation of predictive models, large language models, and generative AI systems in real-world clinical settings.
The course introduces frameworks for evaluating AI systems using the principles of fairness, appropriateness, validity, effectiveness, and safety (FAVES), along with performance metrics, generalizability, and robustness across populations and healthcare environments. It examines sources of bias, distribution shift, and unintended consequences, and explores ethical issues through the lens of autonomy, beneficence, nonmaleficence, and justice. Students will analyze trade-offs between performance, fairness, privacy, and safety, and consider the implications for different stakeholders, including patients, providers, and health systems.
The course also covers key concepts in data governance, privacy, and human factors, as well as implementation science frameworks such as Diffusion of Innovation, RE-AIM, and PRISM. Emphasis is placed on translating AI models into practice, including workflow integration, monitoring, and real-world deployment challenges. In addition, the course introduces emerging topics such as large language models and agentic AI systems for clinical decision support and healthcare operations.
We will invite experts in the field to provide guest lectures and to lead student workshops within their areas of expertise.