QUANTITATIVE ULTRASOUND IMAGE PROCESSING FOR CHARACTERIZATION OF PROSTATE CANCER

点击翻译按钮获取中文摘要

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. Primary support for the subproject and the subproject's principal investigator may have been provided by other sources, including other NIH sources. The Total Cost listed for the subproject likely represents the estimated amount of Center infrastructure utilized by the subproject, not direct funding provided by the NCRR grant to the subproject or subproject staff. We are optimizing trans-rectal ultrasound-coupled optical tomography (TRUST) to improve image guided prostate biopsy by utilizing both the anatomic details of trans-rectal ultrasound (US) and functional information from trans-rectal near infrared (NIR) tomography. Trans-rectal NIR tomography presents a new dimension of information necessary to determining malignancy of lesions suspicious on US and potentially detecting lesions that are otherwise inconspicuous on US. The trans-rectal US not only ensure accurate positioning of the NIR applicator but also can provide spatial prior for guided NIR image reconstruction. In this project, we propose to quantitatively assess TRUST images of prostate in vivo and histological images of prostate cancer. We will focus on important variable in US and NIR imaging, including those associated with optimizing NIR image reconstruction. We will first pre-process US images by the advanced image processing techniques including noise pattern suppression and image segmentation to extract the spatial a priori information to guide NIR image reconstruction. Next, we will quantify the accuracy of optical properties in the US-guided reconstructed with respect to pre-determined true values, and use a perceptual difference computer model (PDM) to determine the visual difference between images reconstructed with and without priori information. Finally, we will establish a relationship between the optical property of prostate cancer and the abnormal cell-morphology using histology images, by converting information such as density and size distribution of cellular or sub-cellular organelles to associated optical properties. This will lead to the understanding of optical scattering spectra of the prostate cancer as a new dimension for detection. These proposed approaches will optimize NIR image reconstruction, and understanding of the optical pathology of prostate cancer. The completion of this research will establish accurate tools for evaluating and optimizing the TRUST detection of prostate cancer. The research outcome may also apply to other multi-modality imaging techniques such as PET/CT, SPECT/CT, and PET/MRI.