A Novel Approach to Breast Ultrasound Image Segmentation Based on the Characteristics of Breast Tissue and Particle Swarm Optimization

Abstract

Breast cancer occurs to over 8% women during their lifetime, and is a leading cause of death among women. Sonography is superior to mammography in its ability to detect focal abnormalities in the dense breasts and has no side-effect. In this paper, we proposed a novel automatic segmentation algorithm based on the characteristics of breast tissue and the eliminating particle swarm optimization (EPSO) clustering analysis. The characteristics of mammary gland in breast ultrasound (BUS) images are analyzed and utilized, and a method based on step-down threshold technique is employed to locate the mammary gland area. The EPSO clustering algorithm employes the idea of "survival of the superior and weeding out the inferior". The experimental results demonstrate that the proposed approach can segment BUS image with high accuracy and low computational time. The EPSO clustering method reduces the computational time by 32.75% compared with the standard PSO clustering algorithm. The proposed approach would find wide applications in automatic lesion classification and computer aided diagnosis (CAD) systems of breast cancer.

PSO Aided Neuro Fuzzy Inference System for Ultrasound Image Segmentation

“…The proposed segmentation method was tested on breast ultrasound images database, experimental results are presented to demonstrate the performance of the proposed method. In Figures 3(a) Original US image, in which most of the bright areas are the breast and muscle tissues, and the suspicious tumor areas are corrupted by speckle noise [18]. In Figures 3(b) through 3(f) are the segmentation results by the proposed expert system.…”

“…After segmentation, lesions can be detected and classified easier and better. The match rate (MR) between the manually determined areas [18] and the automatically located lesions by the proposed algorithm is used to quantitatively evaluate the performance of the proposed algorithm. The MR is defined as:…”

“…The advantages of PSO are that PSO is easy to implement and there are few parameters to adjust [10]. PSO has been successfully applied in many areas: function optimization, artificial neural network training [18], fuzzy system control, and other areas where GA can be applied.…”

Automated Segmentation of Ultrasonic Breast Lesions Using Statistical Texture Classification and Active Contour Based on Probability Distance

“…In this paper, based on radiologist experience, the two thresholds are selected as: D SM 564 and D MM 53=4H ori initially, where H ori is the height of the image. And a novel and fast method (Guo et al 2008) is used for adjusting the two selected thresholds. The method focuses on finding the subcutaneous fat margin and muscle boarders in BUS images according to the fact that they have high gray levels.…”