Inspiration
Chronic diseases (e.g., diabetes, hypertension, cardiovascular diseases) require constant monitoring and personalized treatment plans. The goal of this app is to use predictive AI to assess a patient's risk level for complications based on their medical history, lab results, medication adherence, lifestyle factors, and other relevant data in their electronic health records (EHRs).
What it does
AIrythmia uses a CDS hook when the patient is viewed to prefetch patient data, vital sign observations and cholesterol readings and the data to OpenAI for analysis. AI agents were trained and used for decision making for diagnosing the patient’s blood pressure, cholesterol and ASCVD 10-year risk.
Impacts on healthcare
The impacts on healthcare are through its mitigation of preventable hospitalizations through enabling data-driven, proactive care. In addition, it allows patients to make informed, AI-assisted decisions and promotes patient engagement via insights they can act on to manage their health better.
How we built it
AIrythmia is a smart app hosted on MeldRx. It provides an analysis of a patient's cardiovascular health and evaluation of their overall ASCVD risk.
Challenges we ran into
It was our first delve into the world of OpenAI, training AI agents. While we were excited about the possibilities, we had to quickly learn how to fine-tune models, and optimize our prompts for better results. Debugging AI responses and integrating them effectively into our project required some trial and error, but ultimately, it was an incredible learning experience that pushed us to think creatively!
What's next for AI (AI-driven heart health insights)
Incorporating patients medical history, medications and family history into the ASCVD risk assessment.
Log in or sign up for Devpost to join the conversation.