Brain abnormality detection using neural network

Web application

Inspiration

Artificial intelligence is thriving in the health sector and current research have been exploring its abilities to process medical imaging and improve detection. In particular, the importance of a precise detection of brain abnormalities can allow better further treatment for the patients.

What it does

The web application analyses medical images and detects, with a certain precision score, the possible abnormalities in the brain. For now, it analyze for brain tumors, Alzheimer's disease, and Parkinson's disease.

How we built it

We used public databases containing MRI images of brains affected with brain cancer, Alzheimer's, and Parkinson's, as well as MRI images of healthy brains. We then trained a neural network to be able to detect and classify future images, giving a diagnosis with a certain level of accuracy.

Challenges we ran into

The biggest challenge was training the neural network in a limited amout of time.

Accomplishments that we're proud of

It's our first time participating in a hackathon and also our first time using a neural network.

What we learned

What's next for Brain abnormality detection using neural network

Our project has a lot of possible future application in the health sector. Our goal is to be able to train the model so that it is able to track patient's medical imagies over many years, predict the apparition of a disease and make an early diagnosis. Theoretically, it can be adapted for other parts of the body. This program could be implemented directly in medical machines, MRIs, or scan, particularly in remote areas where a radiologist is not present.

An amelioration would be to standardize the images that are giving as an input to be able to detect more diseases using one image only.