Computer-aided detection method for breast carcinoma calcification point based on local binary pattern and support vector machine

Abstract

The invention discloses a computer-aided detection method for a breast carcinoma calcification point based on a local binary pattern and a support vector machine. Specific implementation of the computer-aided detection method comprises the steps of 1, performing image enhancement on an original X-ray image by using a filter group; 2, coding enhanced and blocked X-ray image blocks by using the local binary pattern, and performing classification on acquired binary codes so as to acquire texture information represented by statistical distribution of the binary codes; and 3, training according to the extracted texture information and a known mapping relation for indicating whether a calcified tissue exists or not by using the support vector machine, and acquiring a detector which is capable of automatically judging whether a calcification point exists or not in the X-ray image. According to the computer-aided detection method, texture features of the breast carcinoma calcification point based on the X-ray image are extracted efficiently and accurately, a function of assisting doctors to detect the breast carcinoma calcification point is realized, and a detection effect for the breast carcinoma calcification point based on the method provided by the invention has excellent consistence with subjective judgment of the doctors.

Description

technical field [0001] The invention belongs to the field of computer-aided diagnosis of medical images, and relates to a breast cancer calcification point detection method based on image processing and machine learning. Background technique [0002] Computer-aided diagnosis of medical images refers to the means of assisting in the detection of lesions and improving the accuracy of diagnosis and the degree of automation through imaging, medical image processing technology, and physiological, biochemical and other methods, combined with computer analysis and calculation. The wide application of computer-aided diagnosis system helps to improve the sensitivity and specificity of doctors' diagnosis. [0003] Early breast cancer calcifications have defects such as too small calcifications, difficult to observe, and time-consuming and laborious. Therefore, the accuracy of purely manual judgment is low and depends on the clinical experience of doctors. Computer-aided detection met...

AI Technical Summary

Problems solved by technology

[0003] Early breast cancer calcifications have defects such as too small calcifications, difficult to observe, time-consuming and labor-intensive, so the accuracy of purely manual judgment is low, and it depends on the clinical experience of doctors

Images