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
. 2025 Aug 25;25(1):347.
doi: 10.1186/s12880-025-01894-9.

Reducing radiomics errors in nasopharyngeal cancer via deep learning-based synthetic CT generation from CBCT

Affiliations

Reducing radiomics errors in nasopharyngeal cancer via deep learning-based synthetic CT generation from CBCT

Ying Xiao et al. BMC Med Imaging. .

Abstract

Purpose: This study investigates the impact of cone beam computed tomography (CBCT) image quality on radiomic analysis and evaluates the potential of deep learning-based enhancement to improve radiomic feature accuracy in nasopharyngeal cancer (NPC).

Methods: The CBAMRegGAN model was trained on 114 paired CT and CBCT datasets from 114 nasopharyngeal cancer patients to enhance CBCT images, with CT images as ground truth. The dataset was split into 82 patients for training, 12 for validation, and 20 for testing. The radiomic features in 6 different categories, including first-order, gray-level co-occurrence matrix (GLCM), gray-level run-length matrix (GLRLM), gray-level size-zone matrix(GLSZM), neighbouring gray tone difference matrix (NGTDM), and gray-level dependence matrix (GLDM), were extracted from the gross tumor volume (GTV) of original CBCT, enhanced CBCT, and CT. Comparing feature errors between original and enhanced CBCT showed that deep learning-based enhancement improves radiomic feature accuracy.

Results: The CBAMRegGAN model achieved improved image quality with a peak signal-to-noise ratio (PSNR) of 29.52 ± 2.28 dB, normalized mean absolute error (NMAE) of 0.0129 ± 0.004, and structural similarity index (SSIM) of 0.910 ± 0.025 for enhanced CBCT images. This led to reduced errors in most radiomic features, with average reductions across 20 patients of 19.0%, 24.0%, 3.0%, 19%, 15.0%, and 5.0% for first-order, GLCM, GLRLM, GLSZM, NGTDM, and GLDM features.

Conclusion: This study demonstrates that CBCT image quality significantly influences radiomic analysis, and deep learning-based enhancement techniques can effectively improve both image quality and the accuracy of radiomic features in NPC.

Keywords: CBCT; Deep learning; Generative adversarial network; Image enhancement; Radiomics.

PubMed Disclaimer

Conflict of interest statement

Declarations. Ethics approval and consent to participate: We declare that all of us obey the principles of the Declaration of Helsinki. In other words, all experiments and methods in this paper are in accordance with these principles.The studies involving humans were approved by Scientific Research Ethical Review of Ganzhou Cancer Hospital. The studies were conducted in accordance with the local legislation and institutional requirements. And the requirement for informed consent was waived for this study by Scientific Research Ethical Review of Ganzhou Cancer Hospital because of the anonymous nature of the data. Consent for publication: Not applicable for this paper. Competing interests: The authors declare no competing interests.

Figures

Fig. 1
Fig. 1
(a) is the overall architecture of CBAMRegGAN; (b), (c), and (d) represent the operation flowcharts of the channel and spatial attention mechanisms, respectively
Fig. 2
Fig. 2
Illustration of CBAMRegGAN. The model begins with a convolutional layer for input image processing, followed by CBAM
Fig. 3
Fig. 3
Comprehensive analysis of PSNR, SSIM, and NMAE violin plots for each model predicted
Fig. 4
Fig. 4
Comparison of deformed CT(dCT) and enhance CT images generated by different models. The first and second columns show CBCT images and dCT. And the third, fourth columns, fifth column, and the sixth column respectively display enhance CT images generated by the Pix2Pix model, the CycleGAN, RegGAN, and CBAMRegGAN
Fig. 5
Fig. 5
GTV and whole body intensity histogram of test
Fig. 6
Fig. 6
Comparison of original CBCT, CycleGAN, Pix2Pix, RegGAN and CBAMRegGAN for reducing the radiomic feature errors

References

    1. Gillies RJ, Kinahan PE, Hricak H. Radiomics: images are more than pictures, they are data. Radiology. 2016;278:563–77. - PMC - PubMed
    1. Scapicchio C, Gabelloni M, Barucci A, Cioni D, Saba L, Neri E. A deep look into radiomics. Radiol Med. 2021;1–16. - PMC - PubMed
    1. Lambin P, Rios-Velazquez E, Leijenaar R, Carvalho S, Aerts HJWL. Radiomics: extracting more information from medical images using advanced feature analysis. Eur J Cancer. 2012;43:441–6. - PMC - PubMed
    1. Virendra, Kumar, Yuhua, Gu, Satrajit B et al. Radiomics: the process and the challenge. Magn Reson Imaging. 2012. - PMC - PubMed
    1. Haider SP, Burtness B, Yarbrough WG, Payabvash S. Applications of radiomics in precision diagnosis, prognostication and treatment planning of head and neck squamous cell carcinomas. Cancers Head Neck. 2020;5. - PMC - PubMed

LinkOut - more resources