Related Experiment Video
Updated: Aug 11, 2025
A Pipeline for 3D Multimodality Image Integration and Computer-assisted Planning in Epilepsy Surgery
Published on: May 20, 2016
Toward real-time automatic treatment planning (RTTP) with a one-step 3D fluence map prediction method and
Jiayuan Peng1, Cui Yang1, Hongbo Guo1
1Department of Radiation Oncology, Fudan University Shanghai Cancer Center, Shanghai, China; Department of Oncology, Shanghai Medical College, Fudan University, Shanghai, China; Shanghai Clinical Research Center for Radiation Oncology, China; Shanghai key laboratory of Radiation Oncology, Shanghai, China.
A new real-time automated treatment planning (RTTP) strategy for rectal cancer radiotherapy significantly improves efficiency and deliverability. This AI-driven approach maintains high plan quality comparable to manual methods, enhancing patient care.
Area of Science:
- Medical Physics
- Radiation Oncology
- Artificial Intelligence in Medicine
Background:
- Rectal cancer radiotherapy requires precise treatment planning for optimal outcomes.
- Current manual planning is time-consuming and can be a bottleneck in clinical workflow.
- Automated planning aims to improve efficiency and consistency in radiotherapy.
Purpose of the Study:
- To develop and evaluate a quasi real-time automated treatment planning (RTTP) strategy for rectal cancer.
- To utilize a one-step 3D fluence map prediction model based on nonorthogonal convolution for automated planning.
- To assess the plan quality, efficiency, and deliverability of the RTTP method compared to manual plans.
Main Methods:
- A 3D deep learning model with nonorthogonal convolution was developed to predict 3D fluence maps directly from CT and anatomical data.
- The model input consists of projections in cone beam space, extracting features along ray-trace paths.
- The predicted fluence maps were converted to Multi-Leaf Collimator (MLC) sequences for treatment plan generation, tested on 314 patients for training and 20 for testing.
Main Results:
- RTTP plans met clinical dose criteria for target coverage and organ-at-risk sparing.
- Planning efficiency dramatically improved, with fluence map generation time reduced from 944s to 39s.
- Deliverability performance showed a 1.91% reduction in total Monitor Units (MU), and 55% of RTTP plans were deemed clinically usable after physician review.
Conclusions:
- The quasi RTTP strategy offers a significant improvement in planning efficiency and deliverability for rectal cancer radiotherapy.
- The method maintains plan quality comparable to optimized manual plans.
- This automated approach shows promise for streamlining radiotherapy workflows and potentially improving patient throughput.
