ePoster Forums
Purpose: Individual template guided interstitial brachytherapy (ITG-IHT) has the advantages of personalized needle path and high needle placement accuracy. However, there is no integrated planning method for both the needle path and dwell time, and manual setup of needle path is still required. This work proposes an integrated optimization method of needle path and dwell time for ITG-IHT.
Methods: Data of ten patient who received ITG-IB was collected. The CT images were scanned with vaginal cylinder template setup. First, a candidate needle set was generated according to the patient anatomy and the vaginal cylinder. Each candidate needle path was determined by the needle entry point penetrating the vaginal cylinder and a target point on the surface of the target. The entry points were limited to the proximal end of the vaginal cylinder, which encouraged the needles to lie in the cylinder as much as possible. Then, a nested optimization mechanism of simulated annealing and quasi-Newton method was used for plan optimization. While simulated annealing method was adopted to optimize needle paths, the quasi-Newton method to optimize the dwell time for a specific configuration of needle paths.
Results: The HR-CTV volume was 78.2±40.9cc. With the HR-CTV D90 normalized to 600cGy, the D2cc for bladder, rectum, and urethra were decreased from 471.1±48.6cGy, 434.0±64.0cGy, 197.7±49.3cGy for clinical plans to 454.1±46.2cGy, 428.6±62.5cGy, 164.9±40.5cGy with p values of 0.001, 0.376, 0.001 respectively. The dwell time of the automatic plans (344±107s) was similar with the clinical plans (353±105s). The generated needle path lay more inside the cylinder than the clinical plans, which has the advantage of reducing needle puncture damage.
Conclusion: The proposed method was effective for integrated optimization of needle path and dwell time for ITG-IHT.
Funding Support, Disclosures, and Conflict of Interest: This work was supported by the National Key Research and Development Program of China under Grant 2019YFB1311300 and 2019YFB1311301.
Brachytherapy, Optimization, Treatment Planning
TH- Brachytherapy: Treatment planning using machine learning/automation