Why we need a vendor neutral specification for delineating prostate cancer with mpMRI

Excerpt

Despite being the most common non-cutaneous cancer and second leading cause of cancer death in American men, the United States Preventive Service Task Force recommends against screening for prostate cancer because of the resultant overtreatment of indolent disease [1]. With the demonstrated ability of multiparametric MRI (mpMRI) to identify cancer and the ability of mpMRI combined with transrectal ultrasound (TRUS) MRI fusion biopsy to increase the yield of significant cancer while decreasing the yield of insignificant cancer, there has been marked growth in volumes and indications for mpMRI of the prostate [2]. TRUS/MRI technology is growing in popularity and acceptance, and it will soon become the standard of care. The National Comprehensive Cancer Network, for example, is expected to suggest its appropriateness for initial biopsy and has recently advocated its use in active surveillance. Building on the demonstration of TRUS/MRI fusion at the National Institutes of Health, we now have nearly a dozen different systems for TRUS/MRI fusion biopsy [3]. This embarrassment of riches has come with an unexpected price—a similar number of commercial contour encoding systems and their associated workstations and image interpretation platforms have been developed for the segmentation of the prostate and suspicious regions within. Two immediate issues arise without vendor neutral specifications for handling contour metadata: the first is related to the problem of multiple referrers using multiple different TRUS/MRI fusion systems for targeted biopsy and its impact on workflow; the second related to archiving, retrieval, and transfer of patient data. …