Method and system for detecting microcalcifications in mammogram

Abstract

The invention discloses a method and a system for detecting microcalcifications in a mammogram, wherein the system comprises carrying out a processing of high-pass filtering on a mammogram to be identified by adopting a high-pass filter; carrying out a normalized processing on the filtered mammogram; extracting a feature vector of the normalized mammogram, and inputting the feature vector into an adaptive relevance vector machine classifier; and identifying microcalcifications in the mammogram according to the feature vector by using the adaptive relevance vector machine classifier. The method and the system of the invention apply an adaptive relevance vector machine classifier to the detection of microcalcifications in a mammogram. An optimization of incremental amount of model parameters is carried out directly by using parameter adaptive optimization ability of the adaptive relevance vector machine classifier, and identification accuracy and efficiency are improved.

Description

technical field [0001] The present application relates to the technical field of automatic analysis and processing of medical images, in particular to a method and system for detecting microcalcification points in breast images. Background technique [0002] Breast cancer is one of the most common malignant tumors in women and the second biggest killer threatening women's health. In recent years, the incidence of breast cancer in my country has increased rapidly. Accurate diagnosis of the nature of lesions, early detection, and early diagnosis play a decisive role in reducing the mortality of breast cancer and improving the survival rate. The advancement and popularization of modern imaging examination technology has laid the foundation for this. Among them, mammography is one of the most accurate and widely used imaging examination methods recognized by breast experts at home and abroad for diagnosing breast cancer. However, because diagnosticians may be affected by subjec...

AI Technical Summary

Problems solved by technology

However, the model parameter selection process in this method requires manual cross-validation to achieve optimization through repeated trials, and the detection efficiency is low, which is not conducive to clinical application.

Images