Engineering FAIR Privacy-preserving Applications that Learn Histories of Disease
This work provides an architectural blueprint for developers building privacy-preserving generative AI applications in healthcare, addressing data privacy concerns for personalized health tasks.
This project explored the engineering challenges of deploying privacy-preserving Generative AI applications in sensitive domains like healthcare, specifically for predicting individual morbidity risk. They successfully deployed a model in-browser using ONNX and a custom JavaScript SDK, demonstrating a secure and high-performance architecture for private generative AI in medicine.
A recent report on "Learning the natural history of human disease with generative transformers" created an opportunity to assess the engineering challenge of delivering user-facing Generative AI applications in privacy-sensitive domains. The application of these models, particularly for personalized healthcare tasks like predicting individual morbidity risk, is typically constrained by data privacy concerns. This project was accordingly designed as an in-browser model deployment exercise (an "App") testing the architectural boundaries of client-side inference generation (no downloads or installations). We relied exclusively on the documentation provided in the reference report to develop the model, specifically testing the "R" component of the FAIR data principles: Findability, Accessibility, Interoperability, and Reusability. The successful model deployment, leveraging ONNX and a custom JavaScript SDK, establishes a secure, high-performance architectural blueprint for the future of private generative AI in medicine.