Multi-task U-Net for end-to-end image classification and segmentation, ideal for medical imaging and other vision applications.

Deep learning pipeline for breast ultrasound tumor analysis using the BUSI dataset. Implements a multi-task U-Net model for simultaneous tumor segmentation and classification (benign, malignant, normal). Includes preprocessing, augmentation, training, and evaluation scripts.

Breast cancer type classification and segmentation using multi-task u-net model

No description available

Multi-task deep learning for brain tumor MRI: segmentation and classification using U-Net, Attention U-Net, and joint models. Includes data generators, augmentation, training, evaluation, and visualizations. Designed for the BRISC 2025 dataset.

Multi-task deep learning for brain tumor MRI: segmentation and classification using U-Net, Attention U-Net, and joint models. Includes data generators, augmentation, training, evaluation, and visualizations. Designed for the BRISC 2025 dataset.

Deep learning system for automated X-ray analysis. Features a Multi-Task CNN for screening (fracture detection & bone classification) and a U-Net for pixel-level fracture segmentation.

A deep learning project implementing U-Net and Attention U-Net for image segmentation on the BRISC2025 dataset, with an integrated classification head for multi-task learning. The project includes training pipelines, evaluation metrics (mIoU, Dice, pixel accuracy), and comparative analysis between joint and separate training strategies.