Inspiration
I choose this problem since it is easy to start off and it is something useful that can be applied to the field of neuroscience.
What it does
Classify the top view of the brain MRI scan to detect tumor.
How I built it
I experiment the CNN model alongside with VGG architecture to properly classify the images. I try with different combination of layers to have a better accuracy.
Challenges I ran into
Constructing convolutional layers that are specifically for mri scan tumour detection is difficult especially if we want to achieve a high accuracy
Accomplishments that I'm proud of
I am most satisfied with my model although the accuracy is not high, but it is my first CNN classification problem and I enjoy a lot when working on the problem.
What I learned
I learned about different architectures of CNN and different metrics for it.
What's next for Brain Tumour Detection
I would definitely add more convolutional layers to improve the model and more data augmentation to avoid overfitting.
Log in or sign up for Devpost to join the conversation.