Click here to


Are you sure ?

Yes, do it No, cancel

BEST IN PHYSICS (THERAPY): Deep Learning-Assisted Algorithm for Catheter Reconstruction During MR-Only Gynecological Interstitial Brachytherapy

A Shaaer1*, M Paudel2,3,M Smith4,F Tonolete4,E Leung 2,5,A Ravi2,3 (1) Department of Physics, Ryerson University, Toronto, ON, Canada, (2) Department of Radiation Oncology, University of Toronto, Toronto, ON, Canada, (3) Department of Medical Physics, Sunnybrook Health Sciences Centre, Toronto, ON, Canada, (4) Department of Radiation Therapy, Sunnybrook Health Sciences Centre, Toronto, ON, Canada, (5) Department of Radiation Oncology, Sunnybrook Health Sciences Centre, Toronto, ON, Canada


(Thursday, 7/16/2020) 10:30 AM - 11:30 AM [Eastern Time (GMT-4)]

Room: Track 3

Purpose: resonance imaging (MRI) offers excellent soft-tissue contrast enabling the contouring of targets and OARs during gynecological interstitial brachytherapy procedure. Despite its advantage, one of the main obstacles preventing a transition to an MRI-only workflow is that the implanted catheters are not reliably visualized on MR images.This study aims to evaluate the feasibility of a deep-learning based algorithm for semi-automatic reconstruction of interstitial catheters during an MR-only workflow.

Methods: MR images of 10 gynecological patients were used in this study. 187 catheters were reconstructed using T1- and T2-weighted images by 5 experienced brachytherapy planners. The mean of the 5 reconstructed paths were considered as “ground truth” for training (103 catheters), validation (15 catheters), and testing/evaluation (69 catheters). To automatically identify and localize the catheters, U-Net was used to find their approximate location in each slice. Once localized, blob detection and threshholding were applied to those regions to find the extrema, as catheters appear as bright and dark regions in T1- and T2-weighted, respectively. The localized dwell positions of the proposed algorithm were compared to the ground truth reconstruction. Reconstruction time was also evaluated.

Results: total of 19,604 catheter dwell positions were evaluated between algorithm and all planners to estimate the reconstruction variability. The average variation was 2 ± 1 mm. The average reconstruction time for this approach was 13 ± 1 minutes, compared with 51 ± 10 minutes for the expert planner.

Conclusion: study suggests that the proposed deep learning, MR-based framework has potential to replace the conventional manual catheter reconstruction. The adoption of this approach in the brachytherapy workflow is expected to improve treatment efficiency while reducing planning time, resources, and errors.


HDR, Interstitial Brachytherapy, MRI


Not Applicable / None Entered.

Contact Email