A brain tumor is a collection, or mass, of abnormal cells in your brain. Your skull, which encloses your brain, is very rigid. Any growth inside such a restricted space can cause problems. Brain tumors can be cancerous (malignant) or noncancerous (benign). When benign or malignant tumors grow, they can cause the pressure inside your skull to increase. This can cause brain damage, and it can be life-threatening.
Early detection and classification of brain tumors is an important research domain in the field of medical imaging and accordingly helps in selecting the most convenient treatment method to save patients life therefore.
| Package Name | Version |
|---|---|
python |
3.7.12 |
tensorflow |
2.6.0 |
keras |
2.6.0 |
keras-preprocessing |
1.1.2 |
matplotlib |
3.0.2 |
opencv |
4.1.2 |
scikit-learn |
0.22.2 |
The dataset was taken from here.
This dataset contains 7022 images of human brain MRI images which are classified into 4 classes:
- glioma
- meningioma
- no tumor
- pituitary
About 22% of the images are intended for model testing and the rest for model training. Pay attention that The size of the images in this dataset is different. You can resize images to the desired size after pre-processing and removing the extra margins.
Crop the part of the image that contains only the brain (which is the most important part of the image): The cropping technique is used to find the extreme top, bottom, left and right points of the brain using OpenCV. You can do this with Preprocessing.py
A pre-trained model is a model that was trained on a large benchmark dataset to solve a problem similar to the one that we want to solve. Accordingly, due to the computational cost of training such models, it is common practice to import and use models from published literature (e.g. VGG, Inception, ResNet50). For this project, I decided to use ResNet50 model to perform image classification for brain tumor MRI images.Resnet50 Article
You can see more details about training steps and testing results inside Brain_Tumor_Classification.ipynb