Deep Convolutional Neural Network with Segmentation Techniques for Chest X-Ray Analysis

Abstract

The deep ConvNets is suitable for learning the mapping between CXR gradients. This paper proposes an example segmentation algorithm based on deep learning applied to xray medical image automatic segmentation annotation. The basic convolutional neural network is used to extract the feature map of the image, and the corresponding branch structure: classification, regression, and mask that can complete the automatic analysis of the image's infrastructure. Our method is evaluated on a dataset that consisted of 180 cases of real two-exposure dual-energy subtraction chest radiographs. Meanwhile, we design histogram averaging and data augmentation to enhance the low contrast image. Finally, we visualize the image and the good results have been achieved in the segmentation and labeling of clavicle and rib. We hope that our research will provide a good prospect for the application of deep learning in automatic segmentation and labeling of medical images.