SSCLNet : Based Brain MRI Classification

Abstract

Brain tumor detection using MRI imaging is a critical task in modern medical diagnosis. Traditional deep learning models rely heavily on large annotated datasets, which are often limited in availability due to the time-consuming and costly nature of medical image labeling. This project introduces SSCLNet (Self-Supervised Contrastive Learning Network), a novel framework that reduces reliance on labeled data by learning meaningful features from unlabeled brain MRI scans using contrastive learning techniques. By combining self-supervised pre-training with supervised fine-tuning on a smaller labeled dataset, SSCLNet achieves high classification accuracy and robust performance, demonstrating its effectiveness for real-world medical imaging applications.