NEW IMAGE-BASED TECHNIQUES FOR PROSTATE BIOPSY AND TREATMENT

Training Plan: The career goal of the PI is to become a successful researcher and trainer who can advance the field of image-based medical interventions for treatment and diagnosis of prostate cancer. To achieve this goal, the PI needs a supportive environment where he can focus on research goals. The environment of research excellence in the University of British Columbia, BC Cancer Agency, and Vancouver General Hospital, if complemented by the financial support provided through the Prostate Cancer Training Award, will provide the PI with the opportunity to implement the research ideas put forward in this proposal and develop new image-based techniques for prostate cancer care. The mentorship of Professor Tim Salcudean, Canada Research Chair, and the collaboration of the visionary Dr. Larry Goldenberg will be the key elements of the training plan. Research Plan: Recently, two technical innovations in ultrasonic imaging promised improvements to image-based prostate interventions: (1) A new ultrasound-based vibro-elastography technique, which relies on broadband vibrations applied to an endorectal transducer. This method can be used to determine the various mechanical parameters of the tissue and to detect tissue regions with varied stiffness and viscosity indices. (2) An ultrasound-based method for tissue typing, which relies on the spectral and fractal analysis of time series of RF echo signals. The method has resulted in a sensitive approach for detection of prostate tumors in ex-vivo studies. The proposed research will combine the advantages of these two techniques and aim at the following specific goals: (1) To enhance the prostatectomy and brachytherapy procedures by detecting cancer and the extent of extra-capsular extension of prostate cancer. A combination of the information from VE and RF time series will be used. The resulting methodology will take advantage of novel feature selection and pattern classification approaches and, also, visualization techniques to efficiently illustrate the outcome in real-time on a 3D model. (2) To enhance the current treatment and biopsy procedures by combining conventional ultrasound (B-mode, Doppler) with vibro-elastography data for real-time detection of the boundaries of the prostate gland. (3) To optimize the RF time series tissue-typing approach and VE imaging and prepare them for real-time clinical applications. Study Design: (1) The in-vivo validation of tissue typing capacity of VE and RF time series features: This will be achieved through collecting data from patients immediately before prostatectomy. The PI and his mentor have worked on comprehensive data collection procedures to address the needs of the project. A certificate of ethics approval was recently received from the Office of Research Services, University of British Columbia. (2) Validation of the in-vivo validation of visualization of prostate boundaries in VE: MRI and VE data is being collected from patients undergoing brachytherapy. 3D volumes of prostate will be reconstructed and compared from the two modalities with pre-operation MRI as the gold standard. Registration and visualization techniques will be developed. (3) Automatic segmentation methods based on VE will be developed. (4) Methods will be developed to optimize the data collection process, extract the tissue-typing features in real-time, implement classification techniques for cancer detection, and visualize the combined anatomic and pathologic results. (5) The sources of tissue typing information in RF time series will be investigated based on ultrasound particle motion tracking techniques. Impact: If successful, the combination of the VE and RF time series techniques will drastically improve image-based treatment guidance. In particular, image biopsy guidance with tools will become more accurate, surgical resection planes could be defined to avoid damaging critical anatomical structures, and prostate radiation treatment will be planned more accurately and monitored more easily. Focus Area (Imaging): New techniques for pathologic and anatomic 'imaging' of prostate cancer will improve the diagnosis and treatment of prostate cancer.