Classification of benign and malignant lesions of breast lesions based on ultrasound radio frequency time series analysis

Chunyi LIN,Bo ZOU,Jianhua ZHOU

DOI: https://doi.org/10.19529/j.cnki.1672-6278.2018.01.05

2018-01-01

Journal of Biomedical Engineering Research

Abstract:Try to provide high-accuracy and noninvasive imaging-assisted diagnostic methods for benign and malignant breast le-sions,a method for classification of benign and malignant breast lesions based on ultrasound radio frequency time series analysis was proposed.For 275 samples of breast lesions,a conventional ultrasonic probe was used to acquire multi-frame ultrasound echo radio frequency(RF)signal.Fractal dimension,frequency domain and time domain characteristics from the ultrasonic RF time series were extracted,and then these breast samples were classified by using machine learning algorithm support vector machine(SVM)and ran-dom forests.The classification results were as follows:area under receiver operating characteristics curve(AUC)of SVM was 0.914, the classification accuracy rate was 85.37%, the AUC of the random forest was 0.937, and the classification accuracy rate was 91.46%.Compared with the previous studies,the classification accuracy of benign and malignant breast lesions is improved,and the benign and malignant breast lesions classification system is developed,which can provide diagnostic reference for doctors.