TY - JOUR
T1 - Deep Learning for Automated Elective Lymph Node Level Segmentation for Head and Neck Cancer Radiotherapy
AU - Strijbis, Victor I. J.
AU - Dahele, Max
AU - Gurney-Champion, Oliver J.
AU - Blom, Gerrit J.
AU - Vergeer, Marije R.
AU - Slotman, Berend J.
AU - Verbakel, Wilko F. A. R.
N1 - Funding Information: This work was funded by a research grant from Varian Medical Systems. Publisher Copyright: © 2022 by the authors.
PY - 2022/11/1
Y1 - 2022/11/1
N2 - Depending on the clinical situation, different combinations of lymph node (LN) levels define the elective LN target volume in head-and-neck cancer (HNC) radiotherapy. The accurate auto-contouring of individual LN levels could reduce the burden and variability of manual segmentation and be used regardless of the primary tumor location. We evaluated three deep learning approaches for the segmenting individual LN levels I–V, which were manually contoured on CT scans from 70 HNC patients. The networks were trained and evaluated using five-fold cross-validation and ensemble learning for 60 patients with (1) 3D patch-based UNets, (2) multi-view (MV) voxel classification networks and (3) sequential UNet+MV. The performances were evaluated using Dice similarity coefficients (DSC) for automated and manual segmentations for individual levels, and the planning target volumes were extrapolated from the combined levels I–V and II–IV, both for the cross-validation and for an independent test set of 10 patients. The median DSC were 0.80, 0.66 and 0.82 for UNet, MV and UNet+MV, respectively. Overall, UNet+MV significantly (p < 0.0001) outperformed other arrangements and yielded DSC = 0.87, 0.85, 0.86, 0.82, 0.77, 0.77 for the combined and individual level I–V structures, respectively. Both PTVs were also significantly (p < 0.0001) more accurate with UNet+MV, with DSC = 0.91 and 0.90, respectively. The accurate segmentation of individual LN levels I–V can be achieved using an ensemble of UNets. UNet+MV can further refine this result.
AB - Depending on the clinical situation, different combinations of lymph node (LN) levels define the elective LN target volume in head-and-neck cancer (HNC) radiotherapy. The accurate auto-contouring of individual LN levels could reduce the burden and variability of manual segmentation and be used regardless of the primary tumor location. We evaluated three deep learning approaches for the segmenting individual LN levels I–V, which were manually contoured on CT scans from 70 HNC patients. The networks were trained and evaluated using five-fold cross-validation and ensemble learning for 60 patients with (1) 3D patch-based UNets, (2) multi-view (MV) voxel classification networks and (3) sequential UNet+MV. The performances were evaluated using Dice similarity coefficients (DSC) for automated and manual segmentations for individual levels, and the planning target volumes were extrapolated from the combined levels I–V and II–IV, both for the cross-validation and for an independent test set of 10 patients. The median DSC were 0.80, 0.66 and 0.82 for UNet, MV and UNet+MV, respectively. Overall, UNet+MV significantly (p < 0.0001) outperformed other arrangements and yielded DSC = 0.87, 0.85, 0.86, 0.82, 0.77, 0.77 for the combined and individual level I–V structures, respectively. Both PTVs were also significantly (p < 0.0001) more accurate with UNet+MV, with DSC = 0.91 and 0.90, respectively. The accurate segmentation of individual LN levels I–V can be achieved using an ensemble of UNets. UNet+MV can further refine this result.
KW - auto-contouring
KW - computed tomography
KW - deep learning
KW - head-and-neck cancer
KW - lymph nodes
KW - radiation oncology
UR - http://www.scopus.com/inward/record.url?scp=85142519955&partnerID=8YFLogxK
UR - https://pure.amc.nl/ws/files/27326203/cancers_14_05501.pdf
U2 - https://doi.org/10.3390/cancers14225501
DO - https://doi.org/10.3390/cancers14225501
M3 - Article
C2 - 36428593
SN - 2072-6694
VL - 14
JO - Cancers
JF - Cancers
IS - 22
M1 - 5501
ER -