New developments in tissue-type imaging (TTI) for guiding prostate biopsies and for planning and monitoring treatment of prostate cancer

摘要

Our research is aimed at ultrasonically characterizing cancerous prostate tissue so that we can imaging it effectively and thereby provide improved means of detecting, treating, and monitoring prostate cancer. We base our characterization methods on spectrum analysis of RF echo signals combined with clinical variables such as prostate-specific antigen (PSA). These parameters are classified using artificial neural networks (ANNs), and classification efficacy is measured using relative-operating-characteristic (ROC) methods. These methods produced ROC-curve areas of 0.80 compared to 0.64 for conventional methods. We then used our optimal classifiers to generate lookup tables (LUTs) that translate spectral parameters and clinical variables to pixel values in tissue-type images (TTIs). TTIs show cancerous regions in 2D or 3D, and may prove to be particularly useful clinically in combination with other ultrasonic and non-ultrasonic methods, e.g., magnetic-resonance methods.