Image recognition method, device, electronic equipment and storage medium

Abstract

The present application provides an image recognition method, device, electronic equipment, and storage medium; the method first acquires an initial image of a medical scan corresponding to a target site, and then performs preprocessing on the initial image to obtain a post-processed image of blood vessels in the target site and At least one of the abnormal areas in the blood vessel centerline parameters, and then perform focus identification based on the initial image and the post-processing image of the blood vessel to obtain the focus area, and then determine and identify the focus image in the initial image according to the focus area and the abnormal area, and obtain Lesion identification results. This application adopts multiple ways to identify the lesion on the initial image, which makes up for the omission of manual identification and improves the accuracy of the identification result.

Description

technical field [0001] The present application relates to the technical field of medical image processing, and in particular to an image recognition method, device, electronic equipment and storage medium. Background technique [0002] At present, when diagnosing lesions that appear in human blood vessels, manual recognition is often used or a detector is used to perform global detection on the image to be recognized. However, there are often omissions in the manual recognition method, and the recognition results are mostly judged by the subjective experience of doctors, which is prone to judgment errors, resulting in low accuracy of the recognition results. , because the global detection mainly uses the trained model to complete the detection work, and the model trained by the neural network usually has errors, so the problem of missed detection of lesions is likely to occur, resulting in low accuracy of the recognition results. [0003] Therefore, it is necessary to provi...

AI Technical Summary

Problems solved by technology

However, there are often omissions in manual recognition methods, and the recognition results are mostly judged by the subjective experience of doctors, which is prone to judgment errors, resulting in low accuracy of the recognition results. , because the global detection mainly uses the trained model to complete the detection work, and the model trained by the neural network usually has errors, so the problem of missed detection of lesions is likely to occur, resulting in low accuracy of the recognition results

[0003] Therefore, it is necessary to provide an image recognition method to alleviate technical problems such as low accuracy of recognition results in current image recognition methods

Images