Abstract
The head and neck (H&N) cancer is the eighth most common cause of cancer death. Radiation therapy is one of the most effective therapies, but it heavily relies on the contouring of tumor volumes on medical images. In this paper, the 3D nnU-Net is first applied to segment H&N tumors in FluoroDeoxyGlucose Positron Emission Tomography (FDG-PET) and Computed Tomography (CT) images. Furthermore we improve upon the 3D nnU-Net by integrating it the spatial and channel ‘squeeze & excitation’ (scSE) blocks, so as to boost those meaningful features while suppressing weak ones. We name the advanced 3D nnU-Net as 3D scSE nnU-Net. Its performance is tested on the HECKTOR 2020 training data by dividing it into training and validation subsets, such as 160 images are in training subset and 41 images are contained in validation subset. The experimental results on the validation images show that the proposed 3D scSE nnU-Net is superior to the original 3D nnU-Net by 1.4% in terms of DSC (Dice Similarity Coefficient) metric on this segmentation task. Our 3D scSE nnU-Net has got the DSC of 0.735 on HECKTOR test data. It has got the third place in this HECKTOR challenge.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
O’rorke, M.A., Ellison, M.V., Murray, L.J., Moran, M., James, J., Anderson, L.A.: Human papillomavirus related head and neck cancer survival: a systematic review and meta-analysis. Oral Oncol. 48(12), 1191–1201 (2012). https://doi.org/10.1016/j.oraloncology.2012.06.019
Gudi, S., et al.: Interobserver variability in the delineation of gross tumor volume and specified organs-at-risk during IMRT for head and neck cancers and the impact of FDG-PET/CT on such variability at the primary site. J. Med. Imaging Radiat. Sci. 48(2), 184–192 (2017). https://doi.org/10.1016/j.jmir.2016.11.003
Andrearczyk, V., et al.: Automatic segmentation of head and neck tumors and nodal metastases in PET-CT scans. In: Medical Imaging with Deep Learning (MIDL) (2020)
Andrearczyk, V., et al.: Overview of the HECKTOR challenge at MICCAI 2020: automatic head and neck tumor segmentation in PET/CT. In: Andrearczyk, V., et al. (eds.) HECKTOR 2020. LNCS, vol. 12603, pp. 1–21. Springer, Cham (2021)
Isensee, F., Petersen, J., Kohl, S.A.A., Jäger, P.F., Maier-Hein, K.H.: nnU-Net: breaking the spell on successful medical image segmentation. arXiv preprint arXiv:1904.08128 (2019)
Roy, A.G., Navab, N., Wachinger, C.: Recalibrating fully convolutional networks with spatial and channel “squeeze and excitation” blocks. IEEE Trans. Med. Imaging 38(2), 540–549 (2018). https://doi.org/10.1109/TMI.2018.2867261
Çiçek, Ö., Abdulkadir, A., Lienkamp, S.S., Brox, T., Ronneberger, O.: 3D U-Net: learning dense volumetric segmentation from sparse annotation. In: Ourselin, S., Joskowicz, L., Sabuncu, M.R., Unal, G., Wells, W. (eds.) MICCAI 2016. LNCS, vol. 9901, pp. 424–432. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-46723-8_49
Wang, Z., Zou, N., Shen, D., Ji, S.: Non-local U-Nets for biomedical image segmentation. In: Proceedings of the 34th AAAI Conference on Artificial Intelligence, California, pp. 6315–6322. AAAI Press (2020). https://doi.org/10.1609/aaai.v34i04.6100
Drozdzal, M., Vorontsov, E., Chartrand, G., Kadoury, S., Pal, C.: The importance of skip connections in biomedical image segmentation. In: Carneiro, G., et al. (eds.) LABELS/DLMIA -2016. LNCS, vol. 10008, pp. 179–187. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-46976-8_19
Acknowledgements
This work is supported in part by the National Natural Science Foundation of China under grant No. 61673251, 62076159 and 12031010, and is also by the National Key Research and Development Program of China under grant No. 2016YFC0901900, and by the Fundamental Research Funds for the Central Universities under grant No. GK201701006 and 2018TS078, the Scientific and Technological Achievements Transformation and Cultivation Funds under grant No. GK201806013, and the Innovation Funds of Graduate Programs at Shaanxi Normal University under grant No. 2015CXS028 and 2016CSY009.
We also acknowledge the HECKTOR2020 challenge organization committee for their providing the competition platform and inviting us submitting this paper for our success wining the third place in this competition.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Xie, J., Peng, Y. (2021). The Head and Neck Tumor Segmentation Using nnU-Net with Spatial and Channel ‘Squeeze & Excitation’ Blocks. In: Andrearczyk, V., Oreiller, V., Depeursinge, A. (eds) Head and Neck Tumor Segmentation. HECKTOR 2020. Lecture Notes in Computer Science(), vol 12603. Springer, Cham. https://doi.org/10.1007/978-3-030-67194-5_3
Download citation
DOI: https://doi.org/10.1007/978-3-030-67194-5_3
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-67193-8
Online ISBN: 978-3-030-67194-5
eBook Packages: Computer ScienceComputer Science (R0)