A method and device for intelligently identifying image features collected by a colposcope

Abstract

The invention relates to a method and device for intelligently identifying image features collected by a colposcope, which selects and automatically extracts the red feature of the original image of the colposcope in a normal saline test, the blood vessel distribution feature of the original green light image, and the white epithelial distribution feature of an acetic acid test image. Recognition, use an easily recognizable logo shape to locate the center of the cervix, and can automatically recognize the logo shape and position, so as to characterize the four-quadrant distribution of the image, improve the recognizability of the image, and on this basis extract the features Presented in an intuitive and visible form. The invention adopts the method of combining automatic identification and displaying image features, which simplifies the complexity of image interpretation, enriches the amount of information of image interpretation, provides intuitive, visible and easy-to-recognize image features for doctors, and improves image interpretation. And the accuracy, consistency and repeatability of the evaluation method, reducing the dependence on subjective experience.

Description

technical field [0001] The invention relates to the technical field of medical equipment, in particular to a method and device for intelligently identifying image features collected by a colposcope in electronic colposcope detection. Background technique [0002] Electronic colposcopy can detect cervical erosion, cervical polyps, cervical intraepithelial neoplasia, cervical cancer, vaginitis, vulva, vaginal or cervical sarcoidosis, viral infection and subclinical papillomavirus infection. Electronic colposcopy not only has application value in diagnosing early cervical cancer and distinguishing tumor and inflammation, but also has special application value in treatment, especially in the treatment of cervical intraepithelial neoplasia. Because electronic colposcopy can see the position and scope of cervical epithelial changes, the video image or computer video image acquisition and storage of electronic colposcopy is very important for the tracking and observation of cervica...

AI Technical Summary

Problems solved by technology

Image interpretation focused on the epithelial blood vessels after using normal saline, and the epithelial changes after using 5% acetic acid solution and 5% compound iodine solution, but ignored the red feature of cervical epithelium in the images collected by colposcopy after using normal saline. Features include the number of quadrants occupied by red, the direction of the clock, and the degree of offset. In colposcopy, the lack of interpretation of red features will reduce the accuracy of the evaluation results; There are also large deviations in the interpretation of images collected by colposcopy with acetic acid solution, which affects the accuracy, consistency and repeatability of image evaluation

Images