Inspiration
Physician shortages are expected to grow between 54,100 and 139,000 by 2033. Along with the rapidly growing population, demand for physicians (both primary and specialty care) will continue to increase. But how can we meet our citizens' needs with a shortage of physicians? To solve this concerning issue, we used our programming skills to develop a web application that retrieves live-feed audio input of a patient's symptoms and compares it to a vast database of medical conditions and case histories in order to provide the patient with an accurate diagnoses.
What it does
PatientInsight is a web application that makes use of medical knowledge for patients in need. The user simply records audio through a microphone to their computer and receives personalized insights regarding their diagnosis. A more specific walkthrough is described below: Diagnosis Screen: Users click on the microphone button and describes their symptoms orally. After around 30 seconds to a minute, PatientInsight retrieves the audio data as a file and processes it using OpenAI data models to store and return information for patients and doctors such as notes, questions, and diagnosis .
How we built it
- Figma for UI/UX design
- React and Next.js for frontend
- Serverless function of next js for backend
- Typescript for efficient development
- OpenAI for models
Challenges we ran into
We ran into many challenges and technical bugs, but that is what programming is all about. Our main challenge was using OpenAI tools for the first time, as there are so many different models to choose from, such as gpt4, 3.5, 3.5-turbo, etc. Getting down the responsive design using extended css was also challenging as well.
Accomplishments that we're proud of
We are proud of so many things. We made use of this project to the best of our abilities in the allotted time. We got to dabble with OpenAI tools for the first time, and even have a relatively decent grasp of it now for future use. Additionally, we combined all of our skills together through teamwork and communication to pull off our product in the given timeframe. We never would have thought we could accomplish this much in such a small period of time.
What we learned
- OpenAI for models
- The efficiency of planning out a UI in figma first before developing
- Storage/retrieval of messages
What's next for PatientInsight
Expanding PatientInsight to an abundance of other languages and even incorporating a webcam for those who communicate using sign language.
Built With
- nextjs
- openai
- react


Log in or sign up for Devpost to join the conversation.