QuickMed

Inspiration

In 2022, our friend moved from India to California. We shared a strong passion for cricket and found ourselves practicing in the same batch. He didn't really know how to communicate to us in English, often using hand gestures to tell us what he needed. One scorching hot day, during our fitness session, he was holding his stomach and was often leaning over in pain. We knew he wasn't feeling well, but didn't know what was happening. Later that practice, he threw up several times and immediately let for home. We carried guilt of not being able to do anything. However, this hackathon gave us the opportunity to make sure this never happens again. What it does QuickMed gives two ways to express their pain without needing proper english. First you point at where it is hurting. Quickmed will analyze the anatomy/confidence level/visual observations using AI. Next, the patient will describe it in their own words. AI translates foreign words and will provide an emotional stress level from 1-10. The patient then sends their symptoms to the doctor. The doctor will review the symptoms and can mark it as reviewed or flag it if in need of additional instructions. How we built it Front end: React + Tailwind CSS + Framer Motion for a smooth mobile optimized UI We used AI to help us develop a database as well as formatting the doctors view. We used Webspeech API that is set to auto detect any language.

Challenges we ran into

We ran into issues with the camera reliability and the way it tracked the anatomy. We also needed help with mapping the anatomy on the silhouette. Also we came across issues in language auto detection in the voice.

Accomplishments that we're proud of

This is very personal website for us as it touches to something we endured in real life. We hope to work with a local medical center to truly bring QuickMed to life.

What we learned

We learned a lot about video tracking, mediapipe, and voice recognition. We also learned how to integrate different languages into our voice recognition

What's next for QuickMed

As said earlier, we hope to work with a local clinic to bring QuickMed to life. Also we will strive to fine tune this website and eventually make it an IOS app that people can go to.

Built With

• and-ai
• and-we-leveraged-native-browser-apis-?-the-web-speech-api-for-multilingual-voice-and-mediadevices-for-camera-?-so-the-app-is-fast
• auth
• authy
• base44-as-our-backend-as-a-service-for-database
• css
• file-storage
• google-web-speech-api
• html
• javascript
• lightweight
• python
• web