Workflow Efficiency in Vaginal Cuff High Dose Rate Brachytherapy Using Artificial Intelligence-Based Organ...

Area of Science:

• Medical physics
• Radiation oncology
• Artificial intelligence in healthcare

Background:

• High dose rate brachytherapy (HDR-BT) is a resource-intensive cancer treatment.
• Workflow inefficiencies and inter-user variability are significant challenges in HDR-BT.
• Artificial intelligence (AI) tools offer potential solutions to these challenges.

Purpose of the Study:

• To evaluate the impact of AI-based organ contouring (AC) and applicator modeling (AM) on workflow efficiency and inter-user variability in vaginal cuff HDR-BT.
• To assess the effect of AC and AM on treatment plan quality.

Main Methods:

• A retrospective analysis of 260 HDR-BT fractions was conducted, comparing cases before and after the implementation of AC and AM.
• Treatment planning times (contouring, dose planning, total) were recorded using timestamps.
• Dosimetric data, including organ-at-risk doses (rectum, bladder D2cc) and target coverage (V90-200), were analyzed.
• Statistical analysis using ANOVA and Tukey tests compared differences between medical physicists before and after AI implementation.

Main Results:

• Implementation of AC and AM led to a significant reduction in overall planning time by 13.7 ± 1.7 minutes.
• Inter-user variability in organ and target dose metrics was reduced across most structures.
• Statistically significant differences in rectum D2cc and target V95 between planners were eliminated after AC and AM implementation.

Conclusions:

• AI-based organ contouring and applicator modeling substantially enhance workflow efficiency in HDR-BT.
• These AI tools maintain or improve treatment plan quality while reducing inter-user variability.
• Standardization of workflows through AI can aid in user training and robust outcome assessment for brachytherapy.