BrainTect

Inspiration

Last year, as a freshman in high school, one of my teachers found out he had a brain tumor. This affected his teaching as he experienced hearing loss. This inspired me to create my project around brain tumors. My goal is to improve the efficiency of diagnosis because if the tumor can be detected earlier, the better it is for the patient.

What it does

My application takes in an MRI scan (an image), and detects tumors in that image. How does this help? This program can make parsing through images and identifying tumors more efficient. While combining together two of the biggest industries, this application utilizes machine learning methods to improve the medical industry.

How we built it

I used CVAT to annotate images for training, made a YOLOv5 model to detect the images, Flask to manage the back end, and HTML/CSS for the front end.

Challenges we ran into

Training my tumor model took quite some time as my computer continued to crash. Displaying the final detected image on the screen bizarrely took a long time as well as I was running into many directory-related errors.

Accomplishments that we're proud of

I am proud of the fact that I simply finished my project. With the many errors I ran into, I'm glad I finished. I am also proud of my project in the sense that it helps people in one way or another.

What we learned

Through this project, I learned a lot about brain tumors and how they affect people. Programming-wise, I strengthened my front-end skills as well as learned many new back-end techniques.

What's next for BrainTect

In the future, I plan to extend this project by adding more applications to this website. At the moment, it only detects brain tumors. I plan to train more models so that my application can detect more parts of the brain.