Cervical biopsy area identification method and device based on channel information multimodal network

Abstract

The invention discloses a cervical biopsy area identification method and a device based on a channel information multimodal network. The device comprises an image acquisition unit, a data processing unit and a display unit, wherein the image acquisition unit collects the physiological saline image, the acetic acid image and the iodine image of a cervix; the data processing unit includes a trainedcervical biopsy area identification model, analyzes and processes the physiological saline image, the acetic acid image and the iodine image and outputs a probability label for indicating that the cervix has a biopsy area; the cervical biopsy area identification model includes a detection network layer and a characteristic combination prediction network layer; the detection network layer is used for extracting characteristic graphs and positional information of cervical surface areas in the physiological saline image, the acetic acid image and the iodine image; the three characteristic graphsand the positional information are jointed on a channel dimension, and through characteristic fusion and identification, the probability label for indicating that the cervix has the biopsy area is output; and the display unit acquires and displays the probability label. The cervical biopsy area identification device can assist a doctor to make an accurate judgment on whether the cervix of a patient needs further biopsy.

Description

technical field [0001] The invention relates to the field of medical image processing, in particular to a method and device for identifying cervical biopsy regions based on channel information multimodal networks. Background technique [0002] Cervical cancer is the most common gynecological malignancy. The high-incidence age of carcinoma in situ is 30-35 years old, and that of invasive carcinoma is 45-55 years old. In recent years, the onset tends to be younger. [0003] The inspection steps of cervical lesions are mainly divided into three steps: (1) cervical cytology examination, the most commonly used is the Pap smear method; (2) colposcopy, if the cytology results are abnormal, colposcopy examination is required, observation Changes in cervical epithelial color, blood vessels, etc.; (3) Cervical tissue biopsy. If there is any doubt in colposcopy, the doctor will take a small amount of cervical tissue from the suspicious lesion for biopsy under the positioning of the co...

AI Technical Summary

Problems solved by technology

At present, many machine learning and deep learning methods have been applied to colposcopy detection images, including the detection of cervical os, the detection of acetabular areas, and the prediction of the degree of cervical lesions. Although these methods can play a certain role in assisting function, but cannot assist doctors to make more accurate judgments from the root cause

Moreover, most of these methods only use colposcopic cervical images with 3%-5% acetic acid solution, which is combined with the characteristic changes of doctors through 0.9% normal saline, 3%-5% acetic acid solution, compound iodine solution and other images to determine whether there is cervical cancer. Inconsistent medical procedures in the area of ​​the biopsy

Images