Semi-supervised X-ray image automatic labeling based on generative adversarial network

Abstract

The invention provides a semi-supervised X-ray automatic labeling method based on a generative adversarial network. A traditional training method is improved on the basis of an existing generative adversarial network method, and a semi-supervised training method combining supervised loss and unsupervised loss is used for carrying out image classification recognition based on a small number of labeled samples. The problem of data scarcity annotation of the X-ray image is studied. The method comprises: firstly, on the basis of a traditional unsupervised generative adversarial network, using a softmax for replacing a final output layer; expanding the X-ray image into a semi-supervised generative adversarial network, defining additional category label guide training for the generated sample, optimizing network parameters by adopting the semi-supervised training, and finally, automatically labeling the X-ray image by adopting a trained discriminant network. Compared with traditional supervised learning and other semi-supervised learning algorithms, the method has the advantage that in the aspect of medical X-ray image automatic labeling, the performance is improved.

Description

[0001] Technical field: [0002] The invention is a semi-supervised image automatic labeling method, and in particular relates to the problem of scarcity of medical X-ray image label samples. Specifically, an automated method for semi-supervised medical X-ray images based on generative adversarial networks. [0003] Background technique: [0004] In recent years, with the breakthrough progress of deep learning technology in computer vision, the performance of natural image computer vision tasks such as image classification, object detection and instance segmentation has been significantly improved. Therefore, computer vision methods based on deep learning have also begun to be widely used in early disease detection and image diagnosis. Among them, medical X-ray images are the most intuitive and accurate means of disease detection and diagnosis. Therefore, the development of X-ray image classification and recognition Significant. However, the performance of X-ray image classif...

AI Technical Summary

Problems solved by technology

However, in practical problems, it is often difficult to accurately predict and assume the model in advance, so that the final effect of the algorithm is greatly affected by the model setting, which limits its application range

Images