Surgical navigation method and medium based on ultrasonic image feature plane detection

Abstract

The invention relates to a technical scheme of a surgical navigation method and a medium based on ultrasonic image feature plane detection, including: creating a deep convolutional neural network, preprocessing preoperative enhanced three-dimensional CT image data, The preprocessed preoperative enhanced 3D CT image data is used for multi-point localization to obtain a multi-point localization model; a detection model is constructed based on the DQN framework and the multi-point localization model. The ultrasound image is processed to obtain a three-dimensional space transformation matrix for aligning the preoperative three-dimensional CT image and the intraoperative two-dimensional ultrasound image, and the surgical navigation is performed through the three-dimensional space transformation matrix. The beneficial effects of the invention are as follows: the intelligent positioning method of the ultrasound feature standard section in the preoperative three-dimensional CT data is realized, and the manual, time-consuming and laborious positioning of the key point positions during the operation is avoided.

Description

technical field [0001] The invention relates to the field of computers, in particular to a surgical navigation method and medium based on ultrasonic image feature plane detection. Background technique [0002] In clinical practice, ultrasound or CT images are usually used to navigate the puncture path. There are four types of image navigation techniques generally used in the navigation system of interventional surgery: [0003] 1. CT guidance: The enhanced CT image can clearly display the trend of blood vessels, the size, location and surrounding tissue structure of the tumor area. However, it is impossible to detect the position of surgical instruments in the patient's body in real time. The entire interventional operation depends on the experience of the interventional physician. During the operation, multiple CT scans are required to continuously adjust the position of the puncture needle. Generally, the operation time is longer, and complications such as bleeding are p...

AI Technical Summary

Problems solved by technology

However, it is impossible to detect the position of the surgical instrument in the patient's body in real time. The entire interventional operation depends on the experience of the interventional surgeon. During the operation, multiple CT scans are required to continuously adjust the position of the puncture needle.

Generally, the operation time is long, and complications such as bleeding are prone to occur;

[0004] 2. MRI guidance: clearly display tumors, but for patients with magnetic substances on their bodies, surgical instruments must have electromagnetic compatibility and are expensive;

[0005] 3. Ultrasound guidance: It can detect the position of surgical instruments in real time and the price is low, but the imaging quality is checked, and generally two-dimensional imaging;

[0008] The existing technology has the following disadvantages: it is very dependent on the experience of interventional physicians, and this process is laborious and time-consuming, which seriously hinders the implementation and promotion of interventional techniques; The position is moved, it needs to be marked again, and the process of image scanning is repeated, which is cumbersome

Images