Ultrasound-based Tissue Typing Using RF Time Series: Clinical Feasibility Studies and Applications
RF time series , Prostate cancer , Ablation , Tissue characterization
Prostate cancer (PCa) is the most commonly diagnosed malignancy, and the second leading cancer-related cause of death in North American men. If diagnosed early, PCa can be managed with a 5-year relative survival rate above 95%. The current standard for PCa diagnosis involves ultrasound-guided core needle biopsy; however, the biopsy procedure is not scaled to individual patients due to the lack of sensitivity and specificity of conventional ultrasound images. Previously, a tissue typing approach was proposed that uses ultrasound RF time series acquired from a stationary transducer and tissue position over a few seconds. The goal in this thesis is to present clinical feasibility studies and new applications using RF time series. We pursue this goal in the context of three studies involving: i) prediction of cancer for in vivo radical prostatectomy cases, ii) prediction of cancer following prostate biopsy, and iii) prediction of changes to the tissue following interstitial ablation therapy. Finally, in simulation and controlled laboratory experiments, we explore the rise of tissue temperature as a potential source of tissue typing information in RF time series. In the in vivo clinical study involving prostatectomy cases, PCa likelihood maps are generated using a computer-aided intervention solution that takes advantage of ultrasound RF time series features. In a cross-validation framework, and by using an ultrasound to histology registration, an area under receiver operating characteristic (AUC) of 0.93 is achieved for PCa detection. We also provide cancer likelihood maps and predict the pathology of MRI-identified targets in biopsy cores with an AUC of 0.91. Using a new approach for fusion of RF time series data and without the need for exhaustive search in the feature space, we achieve comparable classification results with those obtained previously using a feature selection approach to characterize PCa in vivo. Moreover, the application of ultrasound RF time series imaging to differentiate ablated and non-ablated tissue results in the cross-validation accuracy of 84.5%, and an AUC of 0.91 for ex vivo data, and an accuracy of 85% for in situ data.