Ultrasonic breast mass grading detection method based on shearlet features and hierarchical binary tree svm classifier

Abstract

The present invention provides a method for grading and detecting ultrasonic breast masses based on Shearlet features and a hierarchical binary tree SVM classifier. By inputting breast ultrasonic RF data, extracting Shearlet features and reducing dimensionality, breast mass feature extraction is realized based on Shearlet transform, and Shearlet features based on LBP Dimensionality reduction; Hierarchical binary tree SVM classifier was used for grading detection of breast masses. This method is based on the Shearlet transform for breast mass feature extraction, which can accurately describe the feature differences of different grades of breast mass. At the same time, the dimensionality reduction algorithm based on LBP coding can neither lose the feature information of breast mass nor reduce the computational complexity of the algorithm. , which is conducive to improving the effectiveness of the algorithm; through the hierarchical binary tree SVM classifier, breast masses can be effectively graded; the accuracy of film reading can be improved, the influence of subjective factors of doctors can be reduced, and the accuracy of doctor diagnosis can be improved.

Description

technical field [0001] The invention relates to a method for grading and detecting ultrasonic mammary gland masses based on Shearlet features and a hierarchical binary tree SVM classifier. Background technique [0002] According to statistics published by the American Cancer Society in 2013, breast cancer is one of the most common types of cancer in women, accounting for 29% of all cancer cases. According to the World Health Organization, about 1.38 million women are diagnosed with breast cancer every year in the world, which accounts for 23% of all cancer cases. Early detection and timely treatment play an important role in reducing the number of breast cancer deaths. At present, ultrasound imaging technology is one of the commonly used methods for early screening of breast cancer. It is non-invasive, simple, and repeatable. Ultrasound doctors perform graded diagnosis on ultrasound breast images according to the BI-RADS (Breast Imaging Reporting And Data System) standard. ...

AI Technical Summary

Problems solved by technology

[0013] The purpose of the present invention is to provide a kind of ultrasonic mammary gland tumor classification detection method based on Shearlet feature and hierarchical binary tree SVM classifier, solve the existing in the prior art and adopt scoring method or single machine learning algorithm to carry out classification detection to mammary gland tumor, its detection Performance still needs to be further improved

Images