Cervical Histopathology Image Clustering Using Graph Based Unsupervised Learning

Abstract

In order to apply the important topological information to solve a Cervical Histopathology Image Clustering (CHIC) problem, a Graph Based Unsupervised Learning (GBUL) approach is proposed in this paper. First, the GBUL method applies color features and k-means clustering for a first-stage “coarse” clustering. Then, a Skeletonization Based Node Generation (SBNG) approach is introduced to approximate the distribution of cervical cell nuclei. Thirdly, based on the SBNG nodes, a minimum spanning tree graph is constructed. Next, graph features and additional geometrical features are extracted based on the constructed graph. Finally, the k-means clustering is applied again for the second-stage clustering. In the experiment, a practical cervical histopathology image dataset with ten whole scanned images is tested, obtaining a promising CHIC result and showing a huge potential in the cancer risk prediction field.