Ultrasonic section image quality control method and device and computer equipment

Abstract

The invention relates to an ultrasonic section image quality control method and device and computer equipment. The method comprises the following steps: acquiring an ultrasonic multi-section image data set, performing section classification on the ultrasonic multi-section image data set to obtain a category label corresponding to each section image, and calling a corresponding trained target detection network to predict each section image according to the category label corresponding to each section image to obtain a prediction result, and according to the prediction result and a preset tangent plane scoring evaluation rule, obtaining a quality score and an evaluation result of each tangent plane image. The scheme is different from the traditional ultrasonic section image for single section identification, multi-section identification is used as a base point for processing, and the target detection network for predicting each section image is obtained by training based on the ultrasonic multi-section image data set carrying the target structure annotation information, so that the reliability of the image prediction result can be ensured, and the accuracy of the image prediction result is improved. Therefore, the consistency of image quality scores can be ensured, and the quality control efficiency is improved.

Description

technical field [0001] The present application relates to the technical field of image processing, in particular to a method, device, computer equipment and storage medium for controlling the quality of ultrasonic section images. Background technique [0002] With the development of medical level, medical ultrasonography plays an important role in medical diagnosis. Medical sonography (sonography, ultrasonography) is an ultrasound-based medical imaging diagnostic technique that enables the visualization of muscles and internal organs including their size, structure and pathological lesions. [0003] Due to factors such as the state of the patient, the doctor's operation technique, and the physical characteristics of ultrasound during the ultrasound examination, the image clarity and standardization of the ultrasound-based slice images often vary greatly, which in turn affects the accuracy of the examination results. For example, fetal ultrasonography is the first choice for...

AI Technical Summary

Problems solved by technology

However, through the doctor's subjective evaluation method, this method consumes a lot of manpower and material resources, and it is difficult to be widely used in actual clinical practice, and different experts or doctors have different cognitions on the standard of ultrasound slice images (such as fetal ultrasound slice images). All the same results in inconsistencies in the quality control results; to identify a single section through deep learning, it is necessary to extract the diagram of the section from all kinds of diagrams and put them into the system in turn for automatic quality control. Poor operability and readability of quality control results

In summary, the existing quality control of ultrasound slice images has the problem of low quality control efficiency

Images