Skip to main content

RPA-Head-and-Neck-Lymph-Nodes

The Cancer Imaging Archive

RPA-Head-and-Neck-Lymph-Nodes | The Head and Neck Lymph Nodes Dataset, Definitive, Non-contrast

DOI: 10.7937/fe3t-za58 | Data Citation Required | Image Collection

Location Species Subjects Data Types Cancer Types Size Status Updated
Head-Neck Human 46 CT, RTSTRUCT, Other Head and Neck Cancer 5GB Limited, Complete 2024/10/23

Summary

This dataset was generated to train models for research in the Radiation Planning Assistant (RPA), aimed at auto-contouring cervical lymph node levels in the head and neck.

  • Patients: Training = 32 (31 unique); Test = 15
  • Acquisition Protocol: See accompanying spreadsheet (TCIA_RPA_HN_LNs_Aquisition_Protocols) for more details. All patients were imaged supine with head holder and thermoplastic mask specified in our institution’s simulation protocol for radiotherapy head and neck.
  • Scanner Details: Philips/Brilliance Big Bore (89.4%) | SIEMENS/SOMATOM Definition Edge (10.6%)
  • Patient Inclusion Criteria: CT scans of 46 unique patients who retrospectively received head and neck radiotherapy at MD Anderson Cancer Center from April 2019 to June 2021 were curated. All data was gathered under an approved institutional review board protocol. These patients received definitive radiotherapy to the oropharynx with no nodal dissection.
  • Segmentation Generation: First, five head and neck sub-specialized radiation oncologists manually contoured seven lymph node levels (IA, IB, II, III, IV, V, RP (Retropharyngeal or VIIA)) on 3 patients each, totaling 15 patients. This resulted in a collection of 105 lymph node level contours (5 physicians x 3 patients x 7 contours). Left and right contours for each nodal level were combined into one volume to prevent misclassification from left-right flipping augmentations. Contours were anatomically drawn without margin according to institutional practice. This dataset was comprised of images and contours of definitive Oropharynx Cancer (OPX) patients only and did not contain images and contours of those who had nodal level dissections at the time of radiotherapy. These 15 patients’ contours were used to train an initial model. This initial model was then used to generate lymph node level contours on 32 additional CT scans (31 unique patients), which were subsequently edited and reviewed for quality by a radiation oncology resident to be training data for the final model. The 15-patient cohort was then used as test data for the final model. The contours were split into left and right contours in post-processing.
    • Structures of interest: LN_Neck_IA_L, LN_Neck_IA_R, LN_Neck_IB_L, LN_Neck_IB_R, LN_Neck_II_L, LN_Neck_II_R, LN_Neck_III_L, LN_Neck_III_R, LN_Neck_IV_L, LN_Neck_IV_R, LN_Neck_V_L, LN_Neck_V_R, LN_Neck_RP_L, LN_Neck_RP_R
  • Note: In processing the image data, a single patient was accidentally anonymized more than once. The dataset for this model used a duplicate of one of those patients as 2 training subjects. The duplication was discovered after the journal publication. The duplicate anonymized patient IDs are RPA-HN-Lymph-Nodes-006 and RPA-HN-Lymph-Nodes-017. The duplicate was removed from the TCIA dataset; RPA-HN-Lymph-Nodes-006 was removed. Patient ID RPA-HN-Lymph-Nodes-017 may be used twice if the user wishes to duplicate the findings in the journal publication.

Data Access

Version 1: Updated 2024/10/23

Title Data Type Format Access Points Subjects Studies Series Images License
Images and Radiation Therapy Structures CT, RTSTRUCT DICOM
Download requires NBIA Data Retriever
46 46 92 8,788 TCIA Restricted
Train and Test Patient Ids Other XLSX 46 CC BY 4.0
Acquisition Protocol Other XLSX 46 CC BY 4.0
Related Datasets
No related Analysis Results found: Submit your proposal! No related Collections found
Legend: Analysis Results| Collections

Citations & Data Usage Policy

Data Citation Required: Users must abide by the TCIA Data Usage Policy and Restrictions. Attribution must include the following citation, including the Digital Object Identifier:

Data Citation

Maroongroge, S., Mohamed, A. S. R., Nguyen, C., Guma De la Vega, J., Frank, S. J., Garden, A. S., Gunn, G. B., Lee, A., Mayo, L., Moreno, A., Morrison, W. H., Phan, J., Spiotto, M. T., Court, L. E., Fuller, C. D., Rosenthal, D. I., & Netherton, T. J. (2024). The Head and Neck Lymph Nodes Dataset, Definitive, Non-contrast (RPA-Head-and-Neck-Lymph-Nodes) (Version 1) [Data set]. The Cancer Imaging Archive. https://doi.org/10.7937/FE3T-ZA58

Acknowledgements

Data was curated by The Radiation Planning Assistant Team and Head and Neck Radiation Oncology Group at the Unversity of Texas MD Anderson Cancer Center.

Related Publications

Publications by the Dataset Authors

The authors recommended this paper as the best source of additional information about this dataset:

  • Maroongroge, S., Mohamed, A. SR., Nguyen, C., Guma De la Vega, J., Frank, S. J., Garden, A. S., Gunn, B. G., Lee, A., Mayo, L., Moreno, A., Morrison, W. H., Phan, J., Spiotto, M. T., Court, L. E., Fuller, C. D., Rosenthal, D. I., & Netherton, T. J. (2024). Clinical acceptability of automatically generated lymph node levels and structures of deglutition and mastication for head and neck radiation therapy. In Physics and Imaging in Radiation Oncology (Vol. 29, p. 100540). Elsevier BV. https://doi.org/10.1016/j.phro.2024.100540

.

Publication Citation

Maroongroge, S., Mohamed, A. SR., Nguyen, C., Guma De la Vega, J., Frank, S. J., Garden, A. S., Gunn, B. G., Lee, A., Mayo, L., Moreno, A., Morrison, W. H., Phan, J., Spiotto, M. T., Court, L. E., Fuller, C. D., Rosenthal, D. I., & Netherton, T. J. (2024). Clinical acceptability of automatically generated lymph node levels and structures of deglutition and mastication for head and neck radiation therapy. In Physics and Imaging in Radiation Oncology (Vol. 29, p. 100540). Elsevier BV. https://doi.org/10.1016/j.phro.2024.100540

Research Community Publications

TCIA maintains a list of publications that leveraged this dataset. If you have a manuscript you’d like to add please contact TCIA’s Helpdesk.

Other Publications Using this Data

TCIA maintains a list of publications that leveraged this dataset. If you have a manuscript you’d like to add please contact TCIA’s Helpdesk.