Esophageal cancer B3 type blood vessel recognition method based on variable coefficient method

Abstract

The invention relates to the technical field of image recognition in the medical field, in particular to an esophageal cancer B3 type blood vessel recognition method based on a variable coefficient method, and the method comprises an image segmentation method, a clustering algorithm and the variable coefficient method. The image segmentation method is used for extracting a whole blood vessel image in the esophagus image; the clustering algorithm is used for performing data processing on all blood vessel diameters in the whole blood vessel graph and dividing the blood vessel into a plurality of classes according to the blood vessel diameters; the variable coefficient method is used for calculating the variable coefficient of each class; and obtaining a blood vessel diameter dispersion degree coefficient through the maximum class and minimum class variable coefficients, and further judging whether the B3 type blood vessel is contained or not according to the blood vessel diameter dispersion degree coefficient. According to the method, a clustering algorithm and a variable coefficient method are used to quantify the dispersion degree of the diameter of the capillary vessel, the dispersion degree of the diameter of the capillary vessel is quantified, whether the esophageal endoscopic image contains the B3-type blood vessel is determined, and an endoscopic physician can be assisted to improve the reliability and accuracy of esophageal cancer analysis and diagnosis.

Description

technical field [0001] The invention relates to the technical field of image recognition in the medical field, in particular to a method for recognizing blood vessels of type B3 esophageal cancer based on a coefficient of variation method. Background technique [0002] The depth of invasion of esophageal squamous cell carcinoma is crucial for determining the exact indication for endoscopic resection. Studies have shown that the vascular pattern observed by narrow-band magnifying endoscopy is closely related to the invasion depth of superficial cancer, and the lymph node metastasis rate is directly proportional to the cancer invasion depth. Therefore, the depth of invasion of superficial esophageal squamous cell carcinoma is predicted. The Japanese Esophagus Society (JES) proposed a classification method based on the degree of irregularity of epithelial intrapapillary capillary loops (IPCL) under magnifying endoscopy to evaluate the invasion depth of superficial esophageal s...

AI Technical Summary

Problems solved by technology

However, the classification effect of this method is not ideal on the validation set.

Images