X-Ray Mammary Image Segmentation Based on Convolutional Neural Network
Shuwan Pan, Jingyi Zhang, Taishen Wang, Lingke Kong · 2019
Medical images segmentation is a challenging research topic in the field of computer vision, especially when applied to breast cancer. X-ray examination is the best method to diagnose breast cancer at present, however, X-ray imaging used to diagnose mass is heavily dependent on the clinician's experience. To address those problems, this paper implements an automated method for revealing mass in a mammary X-ray image based on a convolutional neural network. Traditional deep convolutional neural network and machine learning methods not only lead to poor performance, but also fail to make full use of the long-term dependence between pixels. According to the characteristics of X-ray images, we propose a novel convolutional neural network framework based on one of the most successful medical images segmentation frameworks, U-net. The experimental results show how to improve models by the use of the pre-trained encoder.