MAG-Net: Multi-task Attention Guided Network for Brain Tumor Segmentation and Classification

Brain tumor is the most common and deadliest disease that can be found in all age groups. Generally, MRI modality adopted for identifying diagnosing tumors by radiologists. The correct identification of regions its type aid to diagnose with followup treatment plans. However, any radiologist analysing such scans a complex time-consuming task. Motivated deep learning based computer-aided-diagnosis systems, this paper proposes multi-task attention guided encoder-decoder network (MAG-Net) classify segment brain using images. MAG-Net trained evaluated on Figshare dataset includes coronal, axial, sagittal views 3 types meningioma, glioma, pituitary tumor. With exhaustive experimental trials model achieved promising results as compared existing state-of-the-art models, while having least number training parameters among other models .