Pathological image classification method and system based on deep learning and machine learning

Abstract

The invention discloses a pathological image classification method and system based on deep learning and machine learning, and the method mainly comprises the following steps: constructing a target detection network model based on the deep learning technology, obtaining the positions of lymphocytes in a pathological image, and carrying out the statistics of the infiltration focus number of the lymphocytes; extracting global spatial arrangement and distribution characteristics of lymphocytes on the basis, and quantifying the infiltration degree of the lymphocytes in the pathological image; andtraining a machine learning classifier to obtain a pathological image classification result. According to the method, an artificial intelligence technology is utilized, and accurate and automatic classification of pathological images with different lymphocyte infiltration degrees is realized through automatic detection of lymphocyte infiltration lesions and numeralization feature extraction of infiltration degrees.

Description

technical field [0001] The present disclosure relates to the technical fields of image processing and artificial intelligence, in particular, to a pathological image classification method and system based on deep learning and machine learning. Background technique [0002] Pathological examination is still the gold standard for clinical disease diagnosis. In clinical practice, doctors usually need to make pathological sections of the diseased tissues of the patient's body with histopathological methods, and after staining and sealing, observe the morphological and structural changes of cells and tissues under a high-power microscope to determine the nature of the disease. Make a pathological diagnosis. The commonly used staining method for pathological sections is Hematoxylin-Eosin (HE) staining. Among them, hematoxylin is a basic dye that can dye basophilic substances in tissues blue, such as chromatin in the nucleus, etc.; eosin is an acid dye that can dye eosinophilic s...

AI Technical Summary

Problems solved by technology

[0006] In order to solve the problem of pathological image classification in the prior art, the embodiments of the present disclosure provide a pathological image classification method and system based on deep learning and machine learning, combined with deep learning technology and machine learning algorithm, through the automatic detection of lymphocyte infiltration Detection and numerical feature extraction of the degree of infiltration can realize accurate and automatic classification of pathological images with different degrees of lymphocyte infiltration, and provide a basis for the establishment of a computer-aided analysis system based on pathological images, thereby reducing the workload of pathologists and improving diagnosis. efficiency and accuracy

Images