Abnormal degree quantification method of gastric mucosa dyeing magnified image microstructure

Abstract

The invention discloses an abnormal degree quantification method of gastric mucosa dyeing magnified image microstructure. The method comprises an image segmentation method, a density method, a unit area method, a centroid eccentricity method and a distribution symmetry method. The image segmentation method is used for extracting a clear area and a microstructure whole image in the gastroscope image; the density method is used for calculating the microstructure whole graph density; the unit area method is used for calculating the area occupied by each microstructure; the centroid eccentricity method is used for calculating the offset distance of the equivalent centroid of the microstructure whole graph relative to the centroid of the clear area; the distribution symmetry method is used for quantifying the four-quadrant symmetry of the microstructure. And finally, the microstructure density, the microstructure unit area, the microstructure centroid eccentricity and the microstructure distribution symmetry are weighted to obtain a microstructure anomaly degree coefficient, and thus judging the microstructure anomaly degree grade according to the microstructure anomaly degree coefficient.

Description

technical field [0001] The invention relates to the technical field of image processing in the medical field, in particular to a method for quantifying the abnormality degree of the microstructure of the enlarged image of gastric mucosa dyeing. Background technique [0002] Gastric cancer is one of the most common digestive tract malignancies and the third deadliest malignancy in the world. In 2015, the number of gastric cancer cases in my country reached 400,000, and the death toll was close to 300,000, seriously endangering people's lives and health. The root cause of gastric cancer endangering human health is that it cannot be detected early. The advent of chromoendoscopy established the endoscopic diagnosis method for early gastric cancer. The combined application of magnifying gastroscopy and electronic chromoendoscopy can observe the tiny blood vessel structure and the fine structure of mucosal surface that cannot be observed by ordinary gastroscopy, which provides c...

AI Technical Summary

Problems solved by technology

However, whether the shape of the microstructure is uniform, whether the distribution is symmetrical, etc., are all qualitative descriptions or experience summaries based on the experience of early cancer diagnosis by endoscopists. Quantitative calculation methods and corresponding conclusions are not given for each description, which has great significance. Strong subjectivity, learning difficulties

Images