TS-PCNet-based panoramic X-ray film image classification method and system

Abstract

The invention discloses a TS-PCNet-based panoramic X-ray film image classification method and system, and the method comprises the steps: carrying out the data enhancement of an oral panoramic X-ray film, enriching a data set through size transformation, random flipping, image enhancement and other methods, and improving the generalization of a model; carrying out sample equalization processing, and carrying out oversampling on categories with a small number of samples so as to equalize the data set samples; and finally, a two-stage periodontitis classification model (TS-PCNet) provided by the invention is trained on the processed data set, on the basis of the deep residual network, the first stage carries out two-stage classification of illness and health on teeth, the second stage further realizes slight, moderate and severe classification of illness degrees, and finally, accurate diagnosis of periodontitis is realized. The problems that manual film reading is long in time and heavy in burden, and early symptoms are difficult to diagnose are solved. In addition, the TS-PCNet provided by the invention has good generalization, and can be easily applied to other tasks in a transfer learning mode.

Description

technical field [0001] The invention belongs to the technical field of image classification in computer vision, and relates to an image classification method and system, in particular to a panoramic X-ray image classification method and system based on TS-PCNet, which can be applied to the field of periodontitis auxiliary diagnosis. Background technique [0002] Panoramic X-ray film is an important basis for clinical diagnosis of periodontitis. On the one hand, relying on doctors to read images manually requires more time and a heavier burden, further occupying limited medical resources; on the other hand, due to the insignificant early symptoms of periodontitis, manual diagnosis is more difficult. With the development of deep learning, computer vision technology has been widely studied and applied in medical image processing. A large number of studies have shown that this technology can reach or even exceed the level of human beings, and has broad development prospects. ...

AI Technical Summary

Problems solved by technology

[0003] Based on the image classification technology in computer vision, the present invention proposes a new dental panoramic X-ray film image classification method and system, which can be applied in the field of periodontitis auxiliary diagnosis to solve the problem of oral panoramic X-ray film in the prior art. Manual reading time is long, burden is heavy, and it is difficult to diagnose early symptoms

Images