Cervical cancer pathological cell classification system of microscope pathological photograph

Abstract

The invention relates to a cervical cancer pathological cell classification system of a microscope pathological photograph. The system comprises a detection network model and a classification networkmodel. The detection network model is trained by using incompletely classified cell labeling data, and is used for detecting cells in the microscope photograph after the training is completed. The classification network model is trained according to the incomplete cell classification annotation data, and is used for classifying the detected cells and predicting the specific types of the cells after the training is completed. According to the system, the pathological cell detection / classification model can be obtained by training incompletely labeled samples.

Description

technical field [0001] The invention relates to the technical field of cervical cancer lesion cell classification, in particular to a cervical cancer lesion cell classification system based on microscope pathological photos. Background technique [0002] Cervical cancer is the fourth leading cause of cancer death in women. Therefore, cervical cancer screening is very important for the treatment of cervical cancer. A Pap smear is a physical examination technique that is widely used to prevent cervical cancer and to find cells that are potentially cancerous. However, the diagnosis process relies heavily on the experience of doctors, which is very time-consuming on the one hand and prone to human misjudgment on the other hand. Due to the latest development of computer technology, computer-aided cell detection technology can play an important role in cytology screening. It can not only screen accurately, reduce labor intensity and workload, but also eliminate the psychological...

AI Technical Summary

Problems solved by technology

Despite recent advances in cell segmentation algorithms, accurate segmentation remains a challenge due to irregularities, overlaps, and uneven staining of cells in microscopic images.

In addition, the performance of various algorithms is also limited by feature design and selection, and the performance of various algorithms is not good.

Images