Integrating support vector machine and graph cuts for medical image segmentation

• Published in: Journal of visual communication and image representation 2018-08, Vol.55, p.157-165
• Main Authors: Zheng, Qiang, Li, Honglun, Fan, Baode, Wu, Shuanhu, Xu, Jindong
• Format: Article
• Language: English
• Subjects: Graph cuts; Medical image segmentation; Support vector machine
• ISSN: 1047-3203; 1095-9076
• DOI: 10.1016/j.jvcir.2018.06.005

•Medical image segmentation suffers from intensity inhomogeneity.•Novel method integrating support vector machine and graph cuts.•Localized training scheme is proposed to train the classifier for each pixel.•Graph cuts are for post-processing based on constraint of classification result. Medical image segmentation remains a challenged problem because of intensity inhomogeneity and surrounding complex background. In this paper, we propose a novel method for medical image segmentation by integrating support vector machine and graph cuts. Particularly, a novel localized training scheme is proposed to train a classifier for each pixel based on the target image information, and then a novel graph cuts-based segmentation method that combines the constraint information of machine learning result, the edge information, the local information, and the remote-local information is proposed for post-processing. Instead of delineating an initialized curve around the object boundary, we directly draw a narrowband mask for the initialization in the paper. Experiments on synthetic and medical images demonstrate that the proposed method can achieve better performance than the state-of-the-art.