Auto-segmentations by convolutional neural network in cervical and anorectal cancer with clinical structure sets as the ground truth
(2020) In Clinical and Translational Radiation Oncology 25. p.37-45- Abstract
Background: It is time-consuming for oncologists to delineate volumes for radiotherapy treatment in computer tomography (CT) images. Automatic delineation based on image processing exists, but with varied accuracy and moderate time savings. Using convolutional neural network (CNN), delineations of volumes are faster and more accurate. We have used CTs with the annotated structure sets to train and evaluate a CNN. Material and methods: The CNN is a standard segmentation network modified to minimize memory usage. We used CTs and structure sets from 75 cervical cancers and 191 anorectal cancers receiving radiation therapy at Skåne University Hospital 2014-2018. Five structures were investigated: left/right femoral heads, bladder, bowel... (More)
Background: It is time-consuming for oncologists to delineate volumes for radiotherapy treatment in computer tomography (CT) images. Automatic delineation based on image processing exists, but with varied accuracy and moderate time savings. Using convolutional neural network (CNN), delineations of volumes are faster and more accurate. We have used CTs with the annotated structure sets to train and evaluate a CNN. Material and methods: The CNN is a standard segmentation network modified to minimize memory usage. We used CTs and structure sets from 75 cervical cancers and 191 anorectal cancers receiving radiation therapy at Skåne University Hospital 2014-2018. Five structures were investigated: left/right femoral heads, bladder, bowel bag, and clinical target volume of lymph nodes (CTVNs). Dice score and mean surface distance (MSD) (mm) evaluated accuracy, and one oncologist qualitatively evaluated auto-segmentations. Results: Median Dice/MSD scores for anorectal cancer: 0.91–0.92/1.93–1.86 femoral heads, 0.94/2.07 bladder, and 0.83/6.80 bowel bag. Median Dice scores for cervical cancer were 0.93–0.94/1.42–1.49 femoral heads, 0.84/3.51 bladder, 0.88/5.80 bowel bag, and 0.82/3.89 CTVNs. With qualitative evaluation, performance on femoral heads and bladder auto-segmentations was mostly excellent, but CTVN auto-segmentations were not acceptable to a larger extent. Discussion: It is possible to train a CNN with high overlap using structure sets as ground truth. Manually delineated pelvic volumes from structure sets do not always strictly follow volume boundaries and are sometimes inaccurately defined, which leads to similar inaccuracies in the CNN output. More data that is consistently annotated is needed to achieve higher CNN accuracy and to enable future clinical implementation.
(Less)
- author
- Sartor, Hanna LU ; Minarik, David LU ; Enqvist, Olof LU ; Ulén, Johannes LU ; Wittrup, Anders LU ; Bjurberg, Maria LU and Trägårdh, Elin LU
- organization
- publishing date
- 2020
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- Automatic segmentation, Cervical cancer radiotherapy, Clinical Target Volume, Convolutional neural network, Organs-at-risk
- in
- Clinical and Translational Radiation Oncology
- volume
- 25
- pages
- 9 pages
- publisher
- Elsevier
- external identifiers
-
- scopus:85091487337
- pmid:33005756
- ISSN
- 2405-6308
- DOI
- 10.1016/j.ctro.2020.09.004
- language
- English
- LU publication?
- yes
- id
- 6288f7ed-3efe-4a00-b0f0-c3ef0b709583
- date added to LUP
- 2020-10-22 15:31:58
- date last changed
- 2024-11-28 16:29:56
@article{6288f7ed-3efe-4a00-b0f0-c3ef0b709583, abstract = {{<p>Background: It is time-consuming for oncologists to delineate volumes for radiotherapy treatment in computer tomography (CT) images. Automatic delineation based on image processing exists, but with varied accuracy and moderate time savings. Using convolutional neural network (CNN), delineations of volumes are faster and more accurate. We have used CTs with the annotated structure sets to train and evaluate a CNN. Material and methods: The CNN is a standard segmentation network modified to minimize memory usage. We used CTs and structure sets from 75 cervical cancers and 191 anorectal cancers receiving radiation therapy at Skåne University Hospital 2014-2018. Five structures were investigated: left/right femoral heads, bladder, bowel bag, and clinical target volume of lymph nodes (CTVNs). Dice score and mean surface distance (MSD) (mm) evaluated accuracy, and one oncologist qualitatively evaluated auto-segmentations. Results: Median Dice/MSD scores for anorectal cancer: 0.91–0.92/1.93–1.86 femoral heads, 0.94/2.07 bladder, and 0.83/6.80 bowel bag. Median Dice scores for cervical cancer were 0.93–0.94/1.42–1.49 femoral heads, 0.84/3.51 bladder, 0.88/5.80 bowel bag, and 0.82/3.89 CTVNs. With qualitative evaluation, performance on femoral heads and bladder auto-segmentations was mostly excellent, but CTVN auto-segmentations were not acceptable to a larger extent. Discussion: It is possible to train a CNN with high overlap using structure sets as ground truth. Manually delineated pelvic volumes from structure sets do not always strictly follow volume boundaries and are sometimes inaccurately defined, which leads to similar inaccuracies in the CNN output. More data that is consistently annotated is needed to achieve higher CNN accuracy and to enable future clinical implementation.</p>}}, author = {{Sartor, Hanna and Minarik, David and Enqvist, Olof and Ulén, Johannes and Wittrup, Anders and Bjurberg, Maria and Trägårdh, Elin}}, issn = {{2405-6308}}, keywords = {{Automatic segmentation; Cervical cancer radiotherapy; Clinical Target Volume; Convolutional neural network; Organs-at-risk}}, language = {{eng}}, pages = {{37--45}}, publisher = {{Elsevier}}, series = {{Clinical and Translational Radiation Oncology}}, title = {{Auto-segmentations by convolutional neural network in cervical and anorectal cancer with clinical structure sets as the ground truth}}, url = {{http://dx.doi.org/10.1016/j.ctro.2020.09.004}}, doi = {{10.1016/j.ctro.2020.09.004}}, volume = {{25}}, year = {{2020}}, }