Building AI Models for Healthcare

Explore the complexities of developing AI models for healthcare in this 48-minute Machine Learning Tech Talk. Delve into three common myths surrounding healthcare AI, including the misconception that more data guarantees better models, the belief that model accuracy alone ensures product usefulness, and the assumption that a good product automatically leads to clinical impact. Join Product Manager Lily Peng as she debunks these myths and engages in insightful conversations with Software Engineers Kira Whitehouse and Scott McKinney. Gain valuable insights into the challenges and considerations of applying AI in medical contexts, supported by numerous research resources on topics such as diabetic retinopathy detection, cardiovascular risk assessment, and the human-centered evaluation of AI systems in clinical settings.

- Introduction
- Myth #1: More data is all you need for a better model
- Myth #2: An accurate model is all you need for a useful product
- Myth #3: A good product is sufficient for clinical impact
- Conversation with Kira Whitehouse, Software Engineer
- Conversation with Scott McKinney, Software Engineer