Purpose: Use of interstitial needles with pose a challenge during MRI-imaging due to patient/organ movement, lack of signal and due to diameter of sheathe being comparable to imaging slice thickness. Due to challenging catheter reconstructions on MRI , despite inferior soft-tissue contrast, CT scans are still used in addition to MRI for manual reconstruction of needles in image-guided brachytherapy. This workflow does not account for change in plastic needle position due to missing stylet during MRI and treatment. Flexi-needles are often in a different position due to changing tissue, organ forces and proximity to ovoids. To address these challenges, MRI Lumen markers were commissioned for clinical use to improve delineation of needles on MRI-scan for treatment planning in HDR gynecological-intracavitary implants with needles.
Methods: MRI lumen markers ™ (Theragenics Inc.) were inserted in the plastic needle sheaths in different media (water, soft-bolus material, in-house poured silicone and on solid-water slabs with soft bolus on top) to assess accuracy of visualization of tip of the needle and delineation of interstitial needles, alongside an empty flexi-needle. Images acquired using clinical protocols on 3T MRI scanner and 0.35T MR-Linac TRUFI (Viewray Inc) scans were compared to CT images of the needles with metal stylets for verification of the tip.
Results: Use of soft-bolus (102 HU) and in-house poured silicone phantom (194 HU) were verified as suitable media for needle visualization study using CT and 0.35T MR linac. Needle marker tips on the TRUFI Images were localized to within +/-1mm of the CT-scans. Needle without the marker was visualized within +/-1.3mm. Results on the 3T MR scanner were inconclusive due to age/resue of markers warranting additional investigation.
Conclusion: Needle markers are visible on 0.35MRI across multiple sequences, and image orientations facilitating treatment planning dosimetry directly on MR-Linac through accurate source path localization and reconstruction.
Not Applicable / None Entered.
Not Applicable / None Entered.