Intelligent auxiliary identification equipment and method for autonomously evaluating benign and malignant thyroid nodules

Abstract

The invention provides intelligent auxiliary identification equipment and method for autonomously evaluating benign and malignant thyroid nodules. The intelligent auxiliary identification equipment comprises a simulation ultrasonic probe, a simulation skin platform, a control module, a field execution mechanism, a DICOM workstation, a benign and malignant nodule classification module, a nodule region calculation module, a visual operation module and report output equipment. An ultrasonic scanning method comprises the following steps: acquiring the pose of the simulation ultrasonic probe, the positive pressure between the simulation ultrasonic probe and the simulation skin platform and two-dimensional coordinates of the simulation ultrasonic probe on the surface of the simulation skin platform according to the movement of the simulation ultrasonic probe on the simulation skin platform so as to control a field manipulator to perform ultrasonic scanning. A benign and malignant thyroid nodule classification method comprises the step of completing the classification task of thyroid nodule ultrasonic images by a machine learning method. The invention provides an auxiliary workstation which is convenient to operate. The intelligent auxiliary equipment increases the telepresence of an operator greatly, can obtain identification results in time, and provides reference for further diagnosis by doctors.

Description

technical field [0001] The present invention relates to a medical image analysis, machine learning and compliance control technology in the field of robot-human-computer cooperation technology, and specifically relates to an intelligent auxiliary identification device and method for autonomously evaluating benign and malignant thyroid nodules. Background technique [0002] Thyroid nodules refer to lumps in the thyroid gland, which can move up and down with the thyroid gland when swallowing. It is a common clinical disease and can be caused by various etiologies. Thyroid nodules are divided into two types, benign and malignant. According to the analysis of the incidence of the two types of thyroid nodules at the present stage, benign accounts for about 95%. Therefore, timely, accurate and effective preoperative diagnosis is conducive to reducing overtreatment Damage to patients, reduce the loss of medical resources, and reduce the incidence of thyroid cancer. [0003] There ...

AI Technical Summary

Problems solved by technology

[0013] In the Chinese invention patent application with publication number CN109288540A, the operator uses a circular handle to perform simulated scanning operations, which cannot reflect the operator's hand feeling to the actual scanning in real time.

[0014] In the Chinese invention patent application with the publication number CN109998590A, the phantom that can be reconstructed in three dimensions is used to increase the operator's sense of presence, but the three-dimensional scanning before the operation is time-consuming, the hardware cost is high, and the real-time reconstruction requires real-time data transmission It is too high. In addition, there is no force sensor installed on the mechanical arm, so it is difficult to accurately reproduce the operator's hand feeling in the actual scanning, and it does not have auxiliary diagnostic functions.

[0015] The Chinese invention patent application with the publication number CN111127391A discloses a benign and malignant evaluation method based on thyroid ultrasound video stream to identify benign and malignant thyroid nodules. The processing method for ultrasound video is complex, the model structure is complicated, and the training is difficult. time, and there is no obvious advantage in recognition accuracy

[0016] The Chinese invention patent application with publication number CN106056595A discloses another benign and malignant evaluation method, but this method requires a large sample size, heavy interception and labeling preparation work in the early stage, takes a long time, and does not study the automatic update of the model

[0017] At present, there is no description or report of the similar technology of the present invention, and no similar data at home and abroad have been collected yet.

Images