A robust approach for subject segmentation of medical Images: Illustration with mammograms and breast magnetic resonance images. (August 2017)

• Record Type: Journal Article
• Authors: Yang, Sheng-Chih
• Is Part Of: Computers & electrical engineering. Volume 62(2017)
• Page Start: 151
• Page End: 165
• Publication Date: 2017-08
• Subjects: Medical image segmentation -- Subject segmentation -- Progressive support-pixel correlation statistical method -- Mammogram -- Breast MRI; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2016.12.022
• Languages: English
• ISSNs: 0045-7906

Abstract

Highlights: A novel, robust technique for medical image segmentation is presented. The proposed method improves on existing methods and preserves their advantages. Real medical image results illustrate the proposed method's excellent performance. Comparison with existing algorithms confirms the proposed method's advantages. Abstract: Medical image segmentation techniques are used to segment subject (suspicious lesions or organs) in medical images in order to provide physicians with accurate information about size, location, or shape characteristics, and therefore are an important technology for clinical diagnosis. However, existing segmentation methods suffer from low accuracy, high complexity, low robustness, and lack of versatility. This paper presents a novel, robust medical image segmentation technique that not only addresses these shortcomings, but also preserves the advantages of existing methods, achieving high image segmentation accuracy. In order to illustrate the excellent results delivered by the proposed progressive support-pixel correlation statistical method (PSCSM) for real medical images, experimental data are categorized as computer-simulated images, actual single-spectral mammograms, and multi-spectral breast magnetic resonance images (MRI). Finally, we compare the experimental results with those of several well-known existing and competitive image segmentation algorithms to confirm the advantages and contributions of the proposed method. Graphical abstract: