Deep learning-based classification of organs at risk and delineation guideline in pelvic cancer radiation therapy
(2023) In Journal of Applied Clinical Medical Physics 24(9).- Abstract
Deep learning (DL) models for radiation therapy (RT) image segmentation require accurately annotated training data. Multiple organ delineation guidelines exist; however, information on the used guideline is not provided with the delineation. Extraction of training data with coherent guidelines can therefore be challenging. We present a supervised classification method for pelvis structure delineations where bowel cavity, femoral heads, bladder, and rectum data, with two guidelines, were classified. The impact on DL-based segmentation quality using mixed guideline training data was also demonstrated. Bowel cavity was manually delineated on CT images for anal cancer patients (n = 170) according to guidelines Devisetty and RTOG. The DL... (More)
Deep learning (DL) models for radiation therapy (RT) image segmentation require accurately annotated training data. Multiple organ delineation guidelines exist; however, information on the used guideline is not provided with the delineation. Extraction of training data with coherent guidelines can therefore be challenging. We present a supervised classification method for pelvis structure delineations where bowel cavity, femoral heads, bladder, and rectum data, with two guidelines, were classified. The impact on DL-based segmentation quality using mixed guideline training data was also demonstrated. Bowel cavity was manually delineated on CT images for anal cancer patients (n = 170) according to guidelines Devisetty and RTOG. The DL segmentation quality from using training data with coherent or mixed guidelines was investigated. A supervised 3D squeeze-and-excite SENet-154 model was trained to classify two bowel cavity delineation guidelines. In addition, a pelvis CT dataset with manual delineations from prostate cancer patients (n = 1854) was used where data with an alternative guideline for femoral heads, rectum, and bladder were generated using commercial software. The model was evaluated on internal (n = 200) and external test data (n = 99). By using mixed, compared to coherent, delineation guideline training data mean DICE score decreased 3% units, mean Hausdorff distance (95%) increased 5 mm and mean surface distance (MSD) increased 1 mm. The classification of bowel cavity test data achieved 99.8% unweighted classification accuracy, 99.9% macro average precision, 97.2% macro average recall, and 98.5% macro average F1. Corresponding metrics for the pelvis internal test data were all 99% or above and for the external pelvis test data they were 96.3%, 96.6%, 93.3%, and 94.6%. Impaired segmentation performance was observed for training data with mixed guidelines. The DL delineation classification models achieved excellent results on internal and external test data. This can facilitate automated guideline-specific data extraction while avoiding the need for consistent and correct structure labels.
(Less)
- author
- Lempart, Michael LU ; Scherman, Jonas ; Nilsson, Martin P. LU and Jamtheim Gustafsson, Christian LU
- organization
- publishing date
- 2023
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- deep learning, guideline, organs at risk, radiation therapy, structure classification
- in
- Journal of Applied Clinical Medical Physics
- volume
- 24
- issue
- 9
- publisher
- American College of Medical Physics
- external identifiers
-
- pmid:37177830
- scopus:85159122674
- ISSN
- 1526-9914
- DOI
- 10.1002/acm2.14022
- language
- English
- LU publication?
- yes
- id
- 3b8f2191-89fc-4649-9078-715d4a28afa4
- date added to LUP
- 2023-08-15 10:09:18
- date last changed
- 2024-06-29 06:54:14
@article{3b8f2191-89fc-4649-9078-715d4a28afa4, abstract = {{<p>Deep learning (DL) models for radiation therapy (RT) image segmentation require accurately annotated training data. Multiple organ delineation guidelines exist; however, information on the used guideline is not provided with the delineation. Extraction of training data with coherent guidelines can therefore be challenging. We present a supervised classification method for pelvis structure delineations where bowel cavity, femoral heads, bladder, and rectum data, with two guidelines, were classified. The impact on DL-based segmentation quality using mixed guideline training data was also demonstrated. Bowel cavity was manually delineated on CT images for anal cancer patients (n = 170) according to guidelines Devisetty and RTOG. The DL segmentation quality from using training data with coherent or mixed guidelines was investigated. A supervised 3D squeeze-and-excite SENet-154 model was trained to classify two bowel cavity delineation guidelines. In addition, a pelvis CT dataset with manual delineations from prostate cancer patients (n = 1854) was used where data with an alternative guideline for femoral heads, rectum, and bladder were generated using commercial software. The model was evaluated on internal (n = 200) and external test data (n = 99). By using mixed, compared to coherent, delineation guideline training data mean DICE score decreased 3% units, mean Hausdorff distance (95%) increased 5 mm and mean surface distance (MSD) increased 1 mm. The classification of bowel cavity test data achieved 99.8% unweighted classification accuracy, 99.9% macro average precision, 97.2% macro average recall, and 98.5% macro average F1. Corresponding metrics for the pelvis internal test data were all 99% or above and for the external pelvis test data they were 96.3%, 96.6%, 93.3%, and 94.6%. Impaired segmentation performance was observed for training data with mixed guidelines. The DL delineation classification models achieved excellent results on internal and external test data. This can facilitate automated guideline-specific data extraction while avoiding the need for consistent and correct structure labels.</p>}}, author = {{Lempart, Michael and Scherman, Jonas and Nilsson, Martin P. and Jamtheim Gustafsson, Christian}}, issn = {{1526-9914}}, keywords = {{deep learning; guideline; organs at risk; radiation therapy; structure classification}}, language = {{eng}}, number = {{9}}, publisher = {{American College of Medical Physics}}, series = {{Journal of Applied Clinical Medical Physics}}, title = {{Deep learning-based classification of organs at risk and delineation guideline in pelvic cancer radiation therapy}}, url = {{http://dx.doi.org/10.1002/acm2.14022}}, doi = {{10.1002/acm2.14022}}, volume = {{24}}, year = {{2023}}, }