A method and system for intelligent diagnosis of diffuse thyroid disease based on ultrasound images

Abstract

The present invention relates to the field of ultrasonic image analysis, in particular to a method for intelligent diagnosis of diffuse thyroid disease based on ultrasonic images, comprising the following steps: intercepting the region of interest of the ultrasonic image based on the Faster R-CNN model, and highlighting the region of interest through image enhancement processing Region; based on the DSSD_Inception_V3 model, the detection and identification of the diffuse thyroid region in the region of interest is realized, and the diffuse thyroid region is intercepted; the size of the diffuse thyroid region is measured based on the aspect ratio of the connected component circumscribed rectangle, and the measurement results are output; according to the position of the diffuse thyroid region The parameter, the number parameter, and the corresponding size parameter realize the diagnosis of diffuse thyroid disease. The invention can automatically detect the type and probability of the diffuse thyroid disease, and at the same time can accurately identify the size, location and coverage of the diffuse thyroid disease.

Description

Technical field [0001] The present invention relates to the field of ultrasound image analysis, specifically to a method and system for intelligent diagnosis of diffuse thyroid diseases based on ultrasound images. Background [0002] Diffuse thyroid disease is a common thyroid disorder, the main common of which is Hashimoto's disease and Graves' disease. Ultrasonography is currently one of the most widely used thyroid disease examination methods. It is of great clinical significance to conduct research on diffuse thyroid disease based on the texture characteristics of ultrasound images and to diagnose patients with diffuse thyroid disease in a timely manner. [0003] At present, in the field of computer-aided diagnosis, deep learning and imaging omics and other methods are commonly used to study thyroid nodular disease based on ultrasound images, but there are few radiomics research data on diffuse thyroid diseases, and only some researchers use imaging omics methods to distingui...

AI Technical Summary

Problems solved by technology

[0003] At present, in the field of computer-aided diagnosis, it is common to use methods such as deep learning and radiomics to study thyroid nodular diseases based on ultrasound images, but there are few radiomics research data on diffuse thyroid diseases, and only some researchers use imaging Omics method, by extracting the texture features of the image, the distinction between Hashimoto's disease and normal thyroid, or the distinction between hyperthyroidism and addition and subtraction

However, the classification of ultrasound images of a variety of common diffuse thyroid diseases, such as Graves' disease, Hashimoto's disease, and normal thyroid tissue, is still blank, and no research has specifically proposed corresponding texture features for the characteristics of thyroid diseases to improve diffuse Disease diagnostic accuracy

Images