Computer Vision to detect brain tumor

This project applies computer vision and deep learning techniques to analyze Brain MRI scans for medical diagnostics. It focuses on detecting anomalies such as brain tumors by training convolutional neural networks (CNNs) on a labeled MRI dataset.

The workflow includes:

Preprocessing: Image resizing, grayscale normalization, and data augmentation.

Modeling: A CNN-based architecture (e.g., VGG, ResNet, or a custom model) is trained to classify MRIs into normal or abnormal categories.

Evaluation: Performance is measured using accuracy, precision, recall, and F1-score, with visualization tools such as confusion matrices and Grad-CAM for interpretability.

This system aids radiologists by automating preliminary diagnosis, ensuring faster and more consistent assessments of brain MRI scans.