Image Analysis Tools for mpMRI Prostate Cancer Diagnosis Using PI-RADS

ABSTRACTProstate cancerPCais one of the most commonly occurring forms of canceraccounting forof all cancer in menMulti parametric magnetic resonance imagingmpMRIhas led to improved capabilities for detectinglocalizingand staging PCaCombined with image guided prostate biopsympMRI has helped to improve diagnosis of clinically significant PCawhich helps to reduce mortality as well as unnecessary biopsies or treatmentsUntil recentlythe diagnostic capabilities of mpMRI were limited by lack of standardization in imaginginterpretationand reporting methodswhich are all subject to high interand intra observer variabilityTo address these problemsthe Prostate Imaging Reporting and Data SystemPI RADSwas designed to standardize the reporting of PCaPI RADS aims to standardize imaging acquisition parameters for mpMRIsimplify radiological reportingand develop assessment categories to stratify levels of PCaA recent metaanalysis reported the diagnostic performance of PI RADS to have a high pooled sensitivity ofand specificity ofunfortunatelythere is still high variability in these resultsThe current clinical practice for interpreting mpMRI has limitations that may contribute to this variationCurrentlyno image registration exists between the imagesand radiologists rely on mental alignment of the images while reading a set of mpMR imageswhich introduces a potential source of variability into PCa diagnosisLocalization and reporting of PCa is specified with respect to the PI RADS sector atlasand this is another source of operator variationAn explicit manual delineation of the prostate into its constituent PI RADS sectors would reduce variationbut this is time consuming and infeasible in the clinical settingThe overarching goal of this proposal is to reduce the interand intra observer variability while interpreting mpMRI images using the PI RADS protocol to improve consistency and accuracy of PCa diagnosisThe primary innovation is creation of a population of PI RADS sector atlases and their application to automatically segment anatomical prostate images with respect to this atlas label protocolThis project is significant in that it has the potential to reduce the variability in PCa interpretation and reporting by providing automated image analysis toolsWhile radiological results are currently communicated in a non standardized formatthe proposed work will facilitate development of automated electronic report generation capabilities to foster data sharing and collaborationsUltimatelyenhancements from this project will create a novel feature for Eigen sthe applicant company sFDAkcleared imaging productProFusethat should improve the diagnosis of PCaIn Aimof this projecta tool to co register and visualize multi parametric prostate MR imaging will be developedIn Aiman image segmentation method to automatically localize the anatomical PI RADS sector map standard within the prostate will be developedBoth aims will utilize a database of existing mpMRI images to develop and validate the algorithms and validate their accuracy PROJECT NARRATIVE Prostate cancer accounts forof all cancer in menThe Prostate Imaging Reporting and Data SystemPIRADSwas designed to standardize the reporting of prostate cancer based on medical imagingbut there is still high variability involved due to interand intra observer variabilityThis project proposes to develop image analysis tools to automate and standardize the interpretation and reporting of radiological prostate cancer diagnosisThis system will be developed by automating components of the PI RADS protocol and integrating these features into an already effective prostate cancer software system that is used for fusion guided biopsies