
This research paper introduces UnSegMedGAT, a novel unsupervised learning approach for medical image segmentation that leverages pre-trained vision transformers (ViTs) and graph attention networks (GATs) to achieve state-of-the-art performance on benchmark datasets.


coremsg

unsupervised-medical-image-segmentation-using-graph-attention-networks-and-modularity-based-clustering


Unsupervised Medical Image Segmentation Using Graph Attention Networks and Modularity-Based Clustering
