Prostate cancer probability maps based on ultrasound RF time series and SVM classifiers

Abstract

We describe a very efficient method based on ultrasound RF time series analysis and support vector machine classification for generating probabilistic prostate cancer colormaps to augment the biopsy process. To form the RF time series, we continuously record ultrasound RF echoes backscattered from tissue while the imaging probe and the tissue are stationary in position. In an in-vitro study involving 30 prostate specimens, we show that the features extracted from RF time series are significantly more accurate and sensitive compared to two other established categories of ultrasound-based tissue typing methods. The method results in an area under ROC curve of 0.95 in 10-fold cross-validation. © 2008 Springer-Verlag Berlin Heidelberg.