Artificial Life for Breast Ultrasound Image Segmentation
Nalan Karunanayake, Stanislav S. Makhanov · 2022
Breast cancer is one of the prominent causes of death among women worldwide. An effective way to reduce the risk of breast cancer is to diagnose it correctly at an early stage of the disease using ultrasound imaging. Breast ultrasound is a common imaging technique used to diagnose breast cancer. Ultrasound images are usually analysed by manual segmentation, which is a time-consuming process that can be inaccurate due to human error. Therefore, computer-assisted diagnosis can be a key tool to improve the accuracy of a breast cancer diagnosis. In this paper, we propose a novel method for US tumour segmentation based on artificial life active particles that interact with the edges of the image. The active particles are able to segment complex-shaped synthetic and real US tumours by grouping the feature edges. The numerical evaluation against five conventional methods shows an advantage of the proposed algorithm.