All Projects
Computer VisionMedical AIFeatured
Brain Tumor Detection — Medical Computer Vision
A Deep Learning medical imaging system for automated brain tumor classification from MRI scans. Reached over 95% accuracy in 4-class multi-class classification, featuring data preprocessing, augmentation, and Grad-CAM explainability heatmaps.
Tech Stack
TensorFlowKerasOpenCVPythonGrad-CAM
Keywords
Deep LearningCNNTensorFlowMedical ImagingOpenCVExplainable AI
Visualizations1 Charts
Deep Dive
Brain Tumor Detection is an explainable deep learning system designed for the automatic classification of brain tumors from MRI scans.
Key Technical Details:
- ▸High Classification Accuracy: Reached over 95% accuracy on 4 different classes of MRI scans (glioma, meningioma, pituitary tumor, and no tumor).
- ▸Computer Vision Pipeline: Custom preprocessing, data augmentation (OpenCV, Albumentations), model training (TensorFlow/Keras), and validation.
- ▸Grad-CAM Explainability: Generates visual heatmaps to highlight the exact regions the CNN model focused on for classification, supporting clinicians.