Breast cancer detection based on ultrasound B-scan texture analysis and patient age information

摘要

We apply the fractional differencing autoregressive moving average (FARMA) model for ultrasonic breast tissue characterization using RF echo signals. We present estimation techniques to extract the model parameters, namely features, for classification purposes and tissue characterization. Along with the model parameters, we use patient age information as an additional feature to improve ROC results. We evaluate the performance of the proposed method using in vivo ultrasound breast images with benign and malignant tumors.