Detection of Cerebral Carcinoma from MRI-101 Bits N' Bytes-2

Detection of Cerebral Neoplasm using Machine Learning models

Inspiration

It has always been a challenge for doctors and other medical professionals to detect the presence of tumor in the human brain , many of the patients are alerted or are diagnosed only in the last stage of their tumor .Some of them have also gone through chemotherapy even in the absence of a tumor due to difficulty in the identification of tumors. So we have developed a model which can classify the presence or absence of a tumor from the MRI scans of a human brain.

What it does

Our project classifies patient's MRI scans to if it is tumorous or not. We have accomplished this using Image Processing where we have trained our model with both positive and negative examples to get a greater accuracy to predict it in the right class.

How we built it

We have used CNN (Convolutional Neural Network) to implement tumour detection , where we have processed images of MRI scans of patients and have trained our model.

Challenges we ran into

We have faced challenges regarding the selection and accuracy of the model, we have improved our model accuracy by making changes in the activation functions and the dense layer filters.

Accomplishments that we're proud of

We are very proud that after a lot of hard work and time we have managed to increase are validation accuracy to about 86%.

What we learned

We had a great experience learning variety of topics in ML and DL and get out hands dirty in Image Processing.

What's next for Detection of Cerebral Carcinoma from MRI-101 Bits N' Bytes-2

It has a lot of scope in future , once we have classified an MRI as the one with tumor we can further classify it into different types like Gliomas , Meningiomas etc. We can further extend the use of our model not only to brain tumor but also to other cancers in the body , like lungs , breast etc.