Method and system for automatically identifying patient orientation through identification features

Abstract

The invention provides a method and system for automatically identifying patient orientation through identification features, and the method comprises the steps: placing an identification according toa set rule, and obtaining an X-ray digital image of the patient with the identification through a CB machine; filtering the digital image to reduce noise interference and obtain a filtered image, averaging pixel gray values of the filtered image, screening according to an average value, and enhancing a picture identification area of the filtered image; carrying out binarization processing on thefiltered image, searching a connected domain according to a pixel with a pixel value of 0 on the filtered image, finding a pixel set with the largest connected domain, and marking the pixel set as anidentification pixel; obtaining the identification direction by comparing pixel features in the identification pixels, and obtaining the orientation of the patient on the digital image map by corresponding identification. According to the method, the markers are placed according to a certain rule, and the imaging structure of the special markers in the X-ray image of the patient is analyzed, so that the problem of how to automatically and quickly determine the direction of the X-ray image of the patient is solved.

Description

technical field [0001] The invention relates to the technical field of image recognition, in particular to a method and a system for automatically recognizing a patient's position through identification features. Background technique [0002] At present, in the surgical navigation technology, it is necessary to use intraoperative 2D X-ray images and preoperative CT / MR 3D image registration. Since the patient's orientation and posture in the preoperative 3D images are known, if the intraoperative 2D images can be determined The orientation will greatly speed up the registration efficiency. In the method of how to determine the patient’s X-ray image, through manual observation and then manual input, the observer needs to have certain experience in medical image reading, and the efficiency is relatively low; through the method of deep learning, a large amount of training is required in the early stage; by placing the logo The method of manually observing and manually inputting...

AI Technical Summary

Problems solved by technology

In the method of how to determine the patient’s X-ray image, through manual observation and then manual input, the observer needs to have certain experience in medical image reading, and the efficiency is relatively low; through the method of deep learning, a large amount of training is required in the early stage; by placing the logo The method of manual observation and manual input of objects is relatively inefficient

Images