Automated plan generation for prostate radiotherapy patients using deep learning and scripted optimization
Background and Purpose: Treatment planning is a time-intensive task that could be automated. We aimed to develop a “single-click” workflow, fully deployed within a commercial treatment planning system (TPS), for autoplanning prostate radiotherapy treatment plans using predictions from a deep learnin...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-10-01
|
| Series: | Physics and Imaging in Radiation Oncology |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2405631624001118 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846115981893566464 |
|---|---|
| author | Cody Church Michelle Yap Mohamed Bessrour Michael Lamey Dal Granville |
| author_facet | Cody Church Michelle Yap Mohamed Bessrour Michael Lamey Dal Granville |
| author_sort | Cody Church |
| collection | DOAJ |
| description | Background and Purpose: Treatment planning is a time-intensive task that could be automated. We aimed to develop a “single-click” workflow, fully deployed within a commercial treatment planning system (TPS), for autoplanning prostate radiotherapy treatment plans using predictions from a deep learning model (DLM). Materials and Methods: Automatically generated treatment plans were created with a single script, executed from within a commercial TPS scripting environment, that performed two stages sequentially. Initially, a 3D dose distribution was predicted with a ResUNet DLM. The DLM was trained and validated using previously treated datasets (n = 120) which used 3D contours as inputs. Following this, dose predictions were converted into treatment plans by extracting dose-volume metrics from the predictions to use as objectives for the inverse optimizer within the TPS. An independent test dataset (n = 20) was used to evaluate the similarity between automated and clinical plans. Results: For planning target volumes, the median percentage difference and interquartile range between the automatically generated plans and clinical plans were 0.4% [0.2-1.1%] for the V100%, −0.5% [(−1.0)-(−0.2)%] for D99% and −0.5% [(−1.0)-(−0.2)%] for D95%. Bladder and rectum volume-at-dose objectives agreed within −6.1% [(−12.5)-0.9%]. The conversion of the DLM prediction into a treatment plan took 15 min [13-16 min]. Conclusions: An automatic plan generation workflow that uses a DL model with scripted optimization was fully deployed in a commercial TPS. Autoplans were compared to previously treated clinical plans and were found to be non-inferior. |
| format | Article |
| id | doaj-art-c8092acbf30a475d8e5f865b4466f092 |
| institution | Kabale University |
| issn | 2405-6316 |
| language | English |
| publishDate | 2024-10-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Physics and Imaging in Radiation Oncology |
| spelling | doaj-art-c8092acbf30a475d8e5f865b4466f0922024-12-19T10:55:36ZengElsevierPhysics and Imaging in Radiation Oncology2405-63162024-10-0132100641Automated plan generation for prostate radiotherapy patients using deep learning and scripted optimizationCody Church0Michelle Yap1Mohamed Bessrour2Michael Lamey3Dal Granville4Department of Medical Physics, The Ottawa Hospital General Campus, Canada; Corresponding author at: Department of Medical Physics, The Ottawa Hospital General Campus, 501 Smyth Rd, Ottawa, ON K1H 8L6, Canada.Department of Medical Physics, The Ottawa Hospital General Campus, CanadaDepartment of Medical Physics, The Ottawa Hospital General Campus, CanadaDepartment of Medical Physics, The Ottawa Hospital General Campus, CanadaDepartment of Radiation Oncology and Department of Physics and Atmospheric Science, Dalhousie University, CanadaBackground and Purpose: Treatment planning is a time-intensive task that could be automated. We aimed to develop a “single-click” workflow, fully deployed within a commercial treatment planning system (TPS), for autoplanning prostate radiotherapy treatment plans using predictions from a deep learning model (DLM). Materials and Methods: Automatically generated treatment plans were created with a single script, executed from within a commercial TPS scripting environment, that performed two stages sequentially. Initially, a 3D dose distribution was predicted with a ResUNet DLM. The DLM was trained and validated using previously treated datasets (n = 120) which used 3D contours as inputs. Following this, dose predictions were converted into treatment plans by extracting dose-volume metrics from the predictions to use as objectives for the inverse optimizer within the TPS. An independent test dataset (n = 20) was used to evaluate the similarity between automated and clinical plans. Results: For planning target volumes, the median percentage difference and interquartile range between the automatically generated plans and clinical plans were 0.4% [0.2-1.1%] for the V100%, −0.5% [(−1.0)-(−0.2)%] for D99% and −0.5% [(−1.0)-(−0.2)%] for D95%. Bladder and rectum volume-at-dose objectives agreed within −6.1% [(−12.5)-0.9%]. The conversion of the DLM prediction into a treatment plan took 15 min [13-16 min]. Conclusions: An automatic plan generation workflow that uses a DL model with scripted optimization was fully deployed in a commercial TPS. Autoplans were compared to previously treated clinical plans and were found to be non-inferior.http://www.sciencedirect.com/science/article/pii/S2405631624001118Deep learningAutoplanningRadiotherapy |
| spellingShingle | Cody Church Michelle Yap Mohamed Bessrour Michael Lamey Dal Granville Automated plan generation for prostate radiotherapy patients using deep learning and scripted optimization Physics and Imaging in Radiation Oncology Deep learning Autoplanning Radiotherapy |
| title | Automated plan generation for prostate radiotherapy patients using deep learning and scripted optimization |
| title_full | Automated plan generation for prostate radiotherapy patients using deep learning and scripted optimization |
| title_fullStr | Automated plan generation for prostate radiotherapy patients using deep learning and scripted optimization |
| title_full_unstemmed | Automated plan generation for prostate radiotherapy patients using deep learning and scripted optimization |
| title_short | Automated plan generation for prostate radiotherapy patients using deep learning and scripted optimization |
| title_sort | automated plan generation for prostate radiotherapy patients using deep learning and scripted optimization |
| topic | Deep learning Autoplanning Radiotherapy |
| url | http://www.sciencedirect.com/science/article/pii/S2405631624001118 |
| work_keys_str_mv | AT codychurch automatedplangenerationforprostateradiotherapypatientsusingdeeplearningandscriptedoptimization AT michelleyap automatedplangenerationforprostateradiotherapypatientsusingdeeplearningandscriptedoptimization AT mohamedbessrour automatedplangenerationforprostateradiotherapypatientsusingdeeplearningandscriptedoptimization AT michaellamey automatedplangenerationforprostateradiotherapypatientsusingdeeplearningandscriptedoptimization AT dalgranville automatedplangenerationforprostateradiotherapypatientsusingdeeplearningandscriptedoptimization |