Cell classification method and device, and equipment

Abstract

The invention discloses a cell classification method and device, and equipment, and the method comprises the steps: inputting an image of a to-be-classified cell into a cell feature extraction model,and obtaining the category features and common features of the to-be-classified cell after the image is processed by the cell feature extraction model, wherein the category features comprise unique features of each cell category, and the common features comprise common features of all cell categories; and classifying the to-be-classified cells based on the category features and the common featuresof the to-be-classified cells to obtain a classification result of the to-be-classified cells. The category characteristics can reflect the unique features of each cell category in the to-be-classified cells and the common features can reflect the common features of all the cell categories, so that the to-be-classified cells can be subjected to feature expression more accurately by combining theunique features and the common features, and the classification result obtained after the to-be-classified cells are classified based on the category features and the common features is more accurate.

Description

technical field [0001] The present application relates to the field of data processing, in particular to a cell classification method, device and equipment. Background technique [0002] In the medical field, cell classification is a key step in computer-aided pathology. Due to the complexity and variability of the appearance of the nuclei in the pathological image under the microscope, the morphological similarity between various types of nuclei, and the particularly large number, etc. Factors that cause difficulties in the identification of cell nuclei, resulting in cell classification has always been a difficult problem. [0003] For cell classification, how to extract the characteristics of the cells to be classified is a key factor affecting the accuracy of cell classification results. Specifically, cell feature extraction refers to converting raw data in cells into an expression form that can be efficiently utilized and processed in machine learning. The current cell...

AI Technical Summary

Problems solved by technology

However, under the premise of intelligent feature extraction, the current specific cell feature extraction method only focuses on the features related to each cell type in the cell, that is, only extracts the common features related to each cell type in the cell, which are used as The characteristic expression of the cell obviously cannot describe the characteristics of the cell relatively comprehensively, which leads to the inaccurate results of the cell classification based on the characteristic expression

Images