Inspiration
As we can tell by an enormously growing infection rate, it's important that we diagnose potential symptoms of COVID-19 and communicate with our local health departments.
What it does
Trachio is a web app that allows you to upload spectrograms of a patient’s lung sounds in order to diagnose and report potential symptoms of COVID-19 to one’s local health department.
How I built it
I created a Flask backend that communicates with a dynamic HTML/CSS frontend, which allows you to upload audio spectrograms. A classification is made through the Google AutoML Vision API and returned to display potential symptoms on the frontend. There is also a feature that allows you to quickly contact local health authorities via email, which is plain HTML.
Challenges I ran into
Of course, getting Google Cloud authentication to work properly was a hassle. Also, I was a bit slow to develop because I had never used Flask before.
Accomplishments that I'm proud of
But regardless, I'm really proud of how many tools I learned to use in the past 24 hours. I dared myself to go solo and I feel like it paid off for my own learning.
What I learned
Basically the entirety of Flask, and servers in general.
What's next for Trachio
Improve the model's accuracy, and make the frontend prettier.


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