Room: Room 207
Purpose: Current treatment planning for high dose-rate Brachytherapy (HDRBT) of cervical cancer is performed in a manual fashion, suffering from problems such as low efficiency, being prone to human error, and plan quality variation between planners. In addition, recent ICRU-89 guideline recommends documenting tens of variables per plan for a research-oriented institution, which is infeasible under the manual planning process with clinical time constraints. To overcome these problems, we have initiated efforts to develop an AutoBrachy system with a streamlined planning, quality assurance (QA), and documentation process. The system initially supports HDRBT with a tandem-and-ovoid (T/O) applicator.
Methods: AutoBrachy contains a web-based interface and a backend computation server. It performs the following steps. 1) CT images are imported via DICOM interface. After the applicator is automatically segmented using spatial clustering analysis, dynamic thresholding, and weighted least squares polynomial fitting, the dwell positions are automatically marked. Points of interest (points A and B, bladder point, rectum point, and posterior-inferior border of symphysis) are identified using geometrical relationships and analysis of the CT images. Dwell times are determined using inverse optimization. 2) Once the plan is approved by physician, a QA module is launched to comprehensively evaluate geometric and dosimetric accuracy with potential problems highlighted. 3) Once the plan passes QA, a pdf report is generated. At the same time, all ICRU-89 recommended data are recorded in AutoBrachy database.
Results: The AutoBrachy system has been implemented at our institution for routine clinical use. It takes ~2min to generate a plan and ~40 sec for QA. Among 20 cases tested, applicators are successfully digitized in all of them and points are accurately drawn in 16 cases. AutoBrachy accurately records data in each plan.
Conclusion: AutoBrachy system enabled a streamlined treatment planning, QA, and documentation process for HDRBT treated with a T/O applicator.
Brachytherapy, HDR, Treatment Planning