Rapid treatment planning for low-dose-rate prostate brachytherapy with TP-GAN

TL;DR

This paper introduces a rapid, AI-based method using conditional GANs to generate consistent, high-quality prostate brachytherapy plans, significantly reducing planning time while maintaining clinical effectiveness.

Contribution

The study presents a novel GAN-based model trained on retrospective data to automate and standardize prostate brachytherapy planning, with optional fine-tuning via simulated annealing.

Findings

Planning time reduced to under 3 seconds without SA and 2.5 minutes with SA.

Achieved 98.9% target volume coverage, comparable to manual plans.

Model demonstrates similar dosimetric quality to expert manual planning.

Abstract

Treatment planning in low-dose-rate prostate brachytherapy (LDR-PB) aims to produce arrangement of implantable radioactive seeds that deliver a minimum prescribed dose to the prostate whilst minimizing toxicity to healthy tissues. There can be multiple seed arrangements that satisfy this dosimetric criterion, not all deemed 'acceptable' for implant from a physician's perspective. This leads to plans that are subjective to the physician's/centre's preference, planning style, and expertise. We propose a method that aims to reduce this variability by training a model to learn from a large pool of successful retrospective LDR-PB data (961 patients) and create consistent plans that mimic the high-quality manual plans. Our model is based on conditional generative adversarial networks that use a novel loss function for penalizing the model on spatial constraints of the seeds. An optional optimizer based on a simulated annealing (SA) algorithm can be used to further fine-tune the plans if necessary (determined by the treating physician). Performance analysis was conducted on 150 test cases demonstrating comparable results to that of the manual prehistorical plans. On average, the clinical target volume covering 100% of the prescribed dose was 98.9% for our method compared to 99.4% for manual plans. Moreover, using our model, the planning time was significantly reduced to an average of 2.5 mins/plan with SA, and less than 3 seconds without SA. Compared to this, manual planning at our centre takes around 20 mins/plan.