作者: Frank Zijlstra , Marinus A. Moerland , Jochem R.N. van der Voort van Zyp , Juus L. Noteboom , Max A. Viergever
DOI: 10.1002/MP.12505
关键词:
摘要: Purpose An MR-only postimplant dosimetry workflow for low dose rate (LDR) brachytherapy could reduce patient burden, improve accuracy, and cost efficiency. However, localization of seeds on MRI scans remains a major challenge this type workflow. In study, we propose validate an seed method, identify remaining challenges. Methods Materials The method was based template matching simulations complex-valued imaging artifacts around metal seeds. The applied to 25 prostate cancer patients who underwent LDR whom performed after four weeks. locations found with the were validated against CT. circumstances in which detection errors made classified gain insight nature errors. Results A total 1490 out 1557 (96%) correctly detected, while 67 false positive made. detected had high spatial accuracy average error 0.8 mm compared A majority positives occurred near other Most negatives either stranded configurations without spacers or seeds. Conclusions The obtained by approach are promising future clinical application dosimetry. most important is robustness regard multiple close vicinity, such as strands spacers. This issue potentially be resolved simulating specific seeds, constraining treatment planning avoid these configurations, make proposed competitive CT-based localization. This article protected copyright. All rights reserved.