A method for the preparation of pleural effusion monomer cancer cells identified by AI for non-diagnostic purposes

Abstract

The present invention provides a method for preparing pleural effusion single cancer cells identified by AI for non-diagnostic purposes, comprising the following steps: S1, preparing lung cancer single cell separation liquid; S2, preparing liquid-based tablet; S3, detecting based on an AI system The cell areas in the pathological scan images of the liquid-based preparations of lung adenocarcinoma cells were obtained, and the VGG 16 deep convolutional neural network model was used, combined with the transfer learning technology, to identify the pleural effusion single cancer cells in the detected cell areas. In the present invention, most cancer cells are dispersed into monomers, and then liquid-based cell preparation is performed, which not only retains its integrity and morphological characteristics, but also facilitates further detection. It also has good application prospects in biological research (such as cell culture, invasion assay, single cell analysis, etc.), clinical testing, drug sensitivity testing, and efficacy analysis.

Description

technical field [0001] The invention belongs to the field of clinical medicine, and in particular relates to a preparation method of a pleural effusion single cancer cell used for AI identification for non-diagnostic purposes. Background technique [0002] Pathological diagnosis is the cornerstone of modern medicine, which determines the direction and prognosis of clinical diagnosis and treatment. Pathological section analysis is the gold standard in cancer diagnosis, but even for experienced pathologists, it is very difficult to read pathological sections. complex process. [0003] Lung cancer ranks first among the top ten malignant tumors, and has three high phenomena: high incidence, high mortality, and high rate of increase. It is a major disease affecting human health and one of the major medical problems in the 21st century. The 5-year survival rate of lung cancer in the population at home and abroad is only 5%-15%, and pleural effusion often occurs in advanced lung c...

AI Technical Summary

Problems solved by technology

At present, there is an obvious problem in the application of liquid-based thin-layer cell preparation of pleural effusion exfoliated cells to the development of artificial intelligence cytopathology diagnosis module based on deep learning, that is, the phenomenon of pleural effusion cells clustering and clustering is more obvious in liquid-based preparation (like figure 1 ), it is not easy to focus under the microscope, and requires high diagnostic experience and level of pathologists. At the same time, it increases the difficulty of machine learning and slows down the operation speed of the module, which is not conducive to the promotion and application of the module in daily high-throughput pathological diagnosis.

Images