I have made a CNN model to classify whether the MRI brain scan has tumor or not.
The main aim of this project is to use the CNN model and then classify whether a particular MRI scan has a tumor or not.
Since the tumor is very difficult to be seen via naked eyes. I thought building and training a CNN model would be an easy solution to identify if the patient suffers from a brain tumor or not.
I have used the dataset available on kaggle
It has 198 images as training set and 58 images as test sets.
The dataset folder has been divided into training and test set folders which is further divided into yes or no folders.
Single prediction folder contains single images of brain scans that are used to validate whether the model can predict correctly or not.
convolution_neural_network.py contains the main CNN code for classifying the MRI scan,this model has validation accuracy between 77%-85%