NMR Image Segmentation Based on Unsupervised Extreme Learning Machine

Abstract

NMR image is often used in medical diagnosis. And image segmentation is one of the most important steps in the NMR image analysis, which is valuable for the computer-aided detection (CADe) and computer-aided diagnosis (CADx). As traditional image segmentation methods based on supervised learning required a lot of manual intervention. Thus, segmentation methods based on unsupervised learning have been received much concern, and unsupervised extreme learning machine (US-ELM)’s performance is particularly outstanding among the unsupervised learning methods. Therefore, in this paper, we proposed a NMR image segmentation method based on US-ELM, named NS-UE. Firstly, a NMR image feature model is established for the input NMR image; Secondly, the clustering based on US-ELM is proposed to separate the various regions of NMR image. Finally, a large number of experimental evaluation results demonstrated the effectiveness and efficiency of the proposed algorithms for the NMR image segmentation.