EXPLORE PUBLICATIONS BY COUNTRIES

Xinjiang High Morbidity of Kazak, Esophageal Cancer, Comprehensive Feature, Feature Extraction, Image Classification.

Image feature extraction technology has been widely applied in image data mining, pattern recognition and classification. Esophageal cancer is a common digestive malignant tumor, China is one of the world's highest incidence and mortality rates of esophageal cancer among the countries, Xinjiang Uygur Autonomous Region is a high incidence area of esophageal cancer, and the kazak is the esophageal cancer high-risk groups. In this paper, we selected 60 advanced esophageal X-ray barium images, half of them are constricted esophagus and the rest are ulcerous esophagus. Firstly, image preprocessing approaches were used to preprocess images. Secondly, extracting the gray-scale histogram features and the GLCM features of the images, then composing the two features into comprehensive feature. Finally, using Bayes discriminant analysis to verify the classification ability of the comprehensive feature. The classification accuracy for constricted esophagus was 86.7%, for ulcerous esophagus was 93.3%.