Related Experiment Video
Updated: Mar 7, 2026
Proton Therapy Delivery and Its Clinical Application in Select Solid Tumor Malignancies
Published on: February 6, 2019
Evaluation of a Machine-Learning Algorithm for Treatment Planning in Prostate Low-Dose-Rate Brachytherapy
Alexandru Nicolae1, Gerard Morton2, Hans Chung2
1Department of Physics, Ryerson University, Toronto, Ontario, Canada; Department of Medical Physics, Odette Cancer Center, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada.
Machine learning (ML) rapidly generates high-quality prostate low-dose-rate (LDR) brachytherapy plans, matching expert quality in significantly less time. This ML approach promises to enhance treatment plan consistency and efficiency in clinical workflows.
Area of Science:
- Medical Physics
- Radiation Oncology
- Artificial Intelligence
Background:
- Prostate low-dose-rate (LDR) brachytherapy is a standard treatment for localized prostate cancer.
- Current treatment planning is time-consuming and relies heavily on expert brachytherapist input.
- There is a need for more efficient and consistent treatment planning methods.
Purpose of the Study:
- To apply a machine learning (ML) algorithm for automatic generation of high-quality prostate LDR brachytherapy treatment plans.
- To evaluate the efficiency, dosimetry, and quality of ML-generated plans compared to expert-created plans.
- To assess the potential of ML to mimic clinically acceptable preoperative treatment plans.
Main Methods:
- A machine learning algorithm was trained on a database of 100 high-quality LDR treatment plans.
- The ML algorithm identifies similar cases to rapidly generate an initial seed distribution, followed by stochastic optimization.
- Preimplantation plans generated by ML were compared to brachytherapist (BT) plans for planning time and dosimetry, with qualitative expert review.
Main Results:
- ML planning time averaged 0.84 minutes, significantly faster than the 17.88 minutes for expert planners (P=.020).
- ML-generated plans were dosimetrically equivalent to BT plans, with a non-clinically significant reduction in prostate V150% (P=.002).
- Expert oncologists rated ML plans as equivalent to BT plans in target coverage, normal tissue avoidance, and overall quality, with difficulty distinguishing between them.
Conclusions:
- Machine learning can rapidly generate prostate LDR preimplantation treatment plans of equivalent quality to those created by brachytherapists.
- Adoption of ML in brachytherapy workflows can improve plan uniformity, reduce planning time, and optimize resource utilization.
- ML offers a promising tool to enhance efficiency and consistency in prostate LDR brachytherapy planning.
More Related Videos
08:25Radiation Planning Assistant - A Streamlined, Fully Automated Radiotherapy Treatment Planning System
Published on: April 11, 2018
07:57Positron Emission Tomography-based Dose Painting Radiation Therapy in a Glioblastoma Rat Model using the Small Animal Radiation Research Platform
Published on: March 24, 2022