Search

Background Multiparametric MRI localizes cancer in the prostate, allowing for MRI guided biopsy (MRI-GB) 43 alongside transrectal ultrasound-guided systematic biopsy (TRUS-GB). Three MRI-GB approaches exist; visual estimation (COG-TB); fusion software-assisted (FUS-TB) and MRI ‘in-bore’ biopsy (IB-TB). It is unknown whether any of these are superior. We conducted a systematic review and meta-analysis to address three questions. First, whether MRI-GB is superior to TRUS-GB at detecting clinically significant PCa (csPCa). Second, whether MRI-GB is superior to TRUS-GB at avoiding detection of insignificant PCa. Third, whether any MRI-GB strategy is superior at detecting csPCa. Methods A systematic literature review from 2015 to 2019 was performed in accordance with the START recommendations. Studies reporting PCa detection rates, employing MRI-GB and TRUS-GB were included and evaluated using the QUADAS-2 checklist. 1553 studies were found, of which 43 were included in the meta-analysis. Results For csPCa, MRI-GB was superior in detection to TRUS-GB (0.83 vs. 0.63 [p = 0.02]). MRI-GB was superior in detection to TRUS-GB at avoiding detection of insignificant PCa. No MRI-GB technique was superior at detecting csPCa (IB-TB 0.87; COG TB 0.81; FUS-TB 0.81, [p = 0.55]). There was significant heterogeneity observed between the included studies. Conclusions In patients with suspected PCa on MRI, MRI-GB offers superior rates of csPCa detection and reduces detection of insignificant PCa compared to TRUS-GB. No individual MRI-GB technique was found to be better in csPCa detection. Prospective adequately powered randomized controlled trials are required.

Current methods of accurate soft tissue injections in small animals are prone to many sources of error. Although efforts have been made to improve the accuracy of needle deliveries, none of the efforts have provided accurate soft tissue references. An MR image-guided robot was designed to function inside the bore of a 9.4T MR scanner to accurately deliver needles to locations within the mouse brain. The robot was designed to have no noticeable negative effects on the image quality and was localized in the MR images through the use of an MR image visible fiducial. The robot was mechanically calibrated and subsequently validated in an image-guided phantom experiment, where the mean needle targeting accuracy and needle trajectory accuracy were calculated to be 178 ± 54µm and 0.27 ± 0.65º, respectively. Finally, the device successfully demonstrated an image-guided needle targeting procedure in situ.