Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Feb 24;11(3):214.
doi: 10.3390/bioengineering11030214.

Towards Automation in Radiotherapy Planning: A Deep Learning Approach for the Delineation of Parotid Glands in Head and Neck Cancer

Ioannis Kakkos  1 , Theodoros P Vagenas  1 , Anna Zygogianni  2 , George K Matsopoulos  1
Affiliations

Towards Automation in Radiotherapy Planning: A Deep Learning Approach for the Delineation of Parotid Glands in Head and Neck Cancer

Ioannis Kakkos et al. Bioengineering (Basel). .

Abstract

The delineation of parotid glands in head and neck (HN) carcinoma is critical to assess radiotherapy (RT) planning. Segmentation processes ensure precise target position and treatment precision, facilitate monitoring of anatomical changes, enable plan adaptation, and enhance overall patient safety. In this context, artificial intelligence (AI) and deep learning (DL) have proven exceedingly effective in precisely outlining tumor tissues and, by extension, the organs at risk. This paper introduces a DL framework using the AttentionUNet neural network for automatic parotid gland segmentation in HN cancer. Extensive evaluation of the model is performed in two public and one private dataset, while segmentation accuracy is compared with other state-of-the-art DL segmentation schemas. To assess replanning necessity during treatment, an additional registration method is implemented on the segmentation output, aligning images of different modalities (Computed Tomography (CT) and Cone Beam CT (CBCT)). AttentionUNet outperforms similar DL methods (Dice Similarity Coefficient: 82.65% ± 1.03, Hausdorff Distance: 6.24 mm ± 2.47), confirming its effectiveness. Moreover, the subsequent registration procedure displays increased similarity, providing insights into the effects of RT procedures for treatment planning adaptations. The implementation of the proposed methods indicates the effectiveness of DL not only for automatic delineation of the anatomical structures, but also for the provision of information for adaptive RT support.

Keywords: CT; artificial intelligence; deep learning; head and neck cancer; parotid glands; radiation therapy; registration; segmentation.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflicts of interest.

Figures

Figure 1
Figure 1
The architecture of the AttentionUNet-based method. The attention gate block is presented in the bottom right.
Figure 2
Figure 2
Affine registration flowchart implemented for the pCT to CBCT1 and CBCT1 to CBCT5 steps.
Figure 3
Figure 3
Training and validation loss for 600 epochs.
Figure 4
Figure 4
Segmentation results of the tested models on (a) the public dataset; (b) the private dataset. Each column represents different algorithmic designs. The segmentation masks of the parotid glands are displayed in red.
Figure 5
Figure 5
An example of the registration processes for (a) pCT to CBCT1; (b) CBCT1 to CBCT5.
See this image and copyright information in PMC

References

    1. Argiris A., Karamouzis M.V., Raben D., Ferris R.L. Head and Neck Cancer. Lancet Lond. Engl. 2008;371:1695–1709. doi: 10.1016/S0140-6736(08)60728-X. - DOI - PMC - PubMed
    1. Mnejja W., Daoud H., Fourati N., Sahnoun T., Siala W., Farhat L., Daoud J. Dosimetric Impact on Changes in Target Volumes during Intensity-Modulated Radiotherapy for Nasopharyngeal Carcinoma. Rep. Pract. Oncol. Radiother. J. Gt. Cancer Cent. Poznan Pol. Soc. Radiat. Oncol. 2020;25:41–45. doi: 10.1016/j.rpor.2019.12.012. - DOI - PMC - PubMed
    1. Zhang Z., Zhao T., Gay H., Zhang W., Sun B. ARPM-Net: A Novel CNN-Based Adversarial Method with Markov Random Field Enhancement for Prostate and Organs at Risk Segmentation in Pelvic CT Images. Med. Phys. 2021;48:227–237. doi: 10.1002/mp.14580. - DOI - PubMed
    1. Zhang Z., Zhao T., Gay H., Zhang W., Sun B. Semi-Supervised Semantic Segmentation of Prostate and Organs-at-Risk on 3D Pelvic CT Images. Biomed. Phys. Eng. Express. 2021;7:65023. doi: 10.1088/2057-1976/ac26e8. - DOI - PubMed
    1. Figen M., Çolpan Öksüz D., Duman E., Prestwich R., Dyker K., Cardale K., Ramasamy S., Murray P., Şen M. Radiotherapy for Head and Neck Cancer: Evaluation of Triggered Adaptive Replanning in Routine Practice. Front. Oncol. 2020;10:579917. doi: 10.3389/fonc.2020.579917. - DOI - PMC - PubMed

LinkOut - more resources