Inspiration
Many of our group members had common experiences with people they knew receiving a brain cancer diagnosis This led us to explore how diagnosis occurs and we found x-rays to be an invaluable tool for doctors From this, we wanted to be able to help doctors identify tumors quicker in less evident x-ray scans to diagnose brain cancer at an earlier stage
What it does
First, we had to find datasets containing multiple x-ray images of brains so that our Machine Learning Model could be trained We read the files into Python and scanned each image We divided the image into clusters and collected the average gray value of each cluster. We calculated the standard deviation and range of the average gray values of the whole image. The standard deviation and range of the gray values was used for our multivariate classification algorithm to calculate the weights, and determine the accuracy of the model with 5 folds.
How we built it
First, we found a dataset of x-ray images of brains with and without tumors. Then, we divided each image up in clusters and found the average grey value of each cluster. We found the standard deviation and range of the average grey values for each image and gave it to our machine learning algorithm to train it.
Challenges we ran into
We ran into challenges with integrating the various libraries together.
Accomplishments that we're proud of
We are proud of our accuracy and recall value for the machine learning algorithm being over 70 percent
What we learned
We learned how to collaborate as a team with different languages and libraries
What's next for Brain Vision
We next want to fully integrate this into an iOS app and make it available on the App Store
Log in or sign up for Devpost to join the conversation.