Inspiration
We were inspired by issues that are faced in the medical field, both for care workers and patients. Sometimes doctors face a lot of trouble determining patient health issues when faced with high-volume traffic, which can lead to misdiagnosis and even no diagnosis, which can be very serious.
What it does
This project automates this process while maintaining validity. It receives a transcript or document of the patient speaking to the doctor, summarizes the file, and emails it to the appropriate departments. The summary will take key points of the transcript, making it easy and fast to read, while protecting the personal information of the patients. Additionally, it provides information on the quality of doctor-patient interactions through a heat map, rated on a scale of 1 to 5.
How we built it
We used NeuralSeek to build 4 agents that use artificial intelligence to execute functions on the documents. We are using the Streamlit tool for the front-end to allow the user to enter the transcript and view analytical data.
Challenges we ran into
We had difficulties with integrating API keys, and it also took us a while to connect all the agents to work with each other. Configuring Agent 4 to send emails to separate departments was also difficult.
Accomplishments that we're proud of
Successfully integrated AI into a website using AI agents while maintaining user security and displaying a heat map on the front-end.
What we learned
We learned how to create and integrate AI agents using NeuralSeek and apply them to solve real-world problems.
What's next for Medalyze
We want to optimize Medalyze to become effective and launch into the real world, making an impact.
Log in or sign up for Devpost to join the conversation.