Auxiliary diagnosis method and device based on semi-supervised learning and multi-scale feature fusion

Abstract

The invention provides an auxiliary diagnosis method and device based on semi-supervised learning and multi-scale feature fusion, and the method comprises the steps: inputting a to-be-diagnosed CT image into a multi-scale feature fusion module of a classification model, and obtaining the multi-scale fusion features of the to-be-diagnosed CT image; inputting the multi-scale fusion features into a classification module of the classification model to obtain an auxiliary medical diagnosis result of the to-be-diagnosed CT image; wherein the classification model is obtained by performing semi-supervised learning training based on a first training data set; the first training data set comprises a first sample CT image with a label and a second sample CT image without a label. According to the method, the classification model with higher generalization and accuracy can be obtained when the effective annotation data volume is small, and then the auxiliary medical diagnosis result can be rapidly and accurately obtained.

Description

technical field [0001] The invention relates to the technical field of auxiliary medical diagnosis, in particular to an auxiliary diagnosis method and device based on semi-supervised learning and multi-scale feature fusion. Background technique [0002] Infectious diseases spread rapidly around the world, seriously threatening human life, health and safety. To combat infectious diseases, effective screening of patients suspected of infection is critical. Compared with reverse transcription polymerase chain reaction, CT (Computed Tomography, computer tomography) images have the advantages of high accuracy and wide popularity, and have become an effective means to diagnose whether patients are infected with infectious diseases. [0003] In the prior art, a doctor examines a large number of CT images to determine whether a patient corresponding to the CT image is infected with an infectious disease, which results in the doctor needing a lot of energy to repeatedly check and di...

AI Technical Summary

Problems solved by technology

[0005] The present invention provides an auxiliary diagnosis method and device based on semi-supervised learning and multi-scale feature fusion, which is used to solve the shortage of effectively labeled CT image samples in the prior art, and it is difficult to train a high-performance deep learning model, which leads to auxiliary medical diagnosis. The defect of poor effect is to realize the rapid and accurate acquisition of auxiliary medical diagnosis results in the absence of effectively labeled CT image samples

Images