nach oben

Japanese Journal of Radiology

Erschienen in:

05.02.2021 | Original Article

Convolutional neural network in nasopharyngeal carcinoma: how good is automatic delineation for primary tumor on a non-contrast-enhanced fat-suppressed T2-weighted MRI?

verfasst von: Lun M. Wong, Qi Yong H. Ai, Frankie K. F. Mo, Darren M. C. Poon, Ann D. King

Erschienen in: Japanese Journal of Radiology | Ausgabe 6/2021

Einloggen, um Zugang zu erhalten

Abstract

Purpose

Convolutional neural networks (CNNs) show potential for delineating cancers on contrast-enhanced MRI (ce-MRI) but there are clinical scenarios in which administration of contrast is not desirable. We investigated performance of the CNN for delineating primary nasopharyngeal carcinoma (NPC) on non-contrast-enhanced images and compared the performance to that on ce-MRI.

Materials and methods

We retrospectively analyzed primary NPC in 195 patients using a well-established CNN, U-Net, for tumor delineation on the non-contrast-enhanced fat-suppressed (fs)-T2W, ce-T1W and ce-fs-T1W images. The CNN-derived delineations were compared to manual delineations to obtain Dice similarity coefficient (DSC) and average surface distance (ASD). The DSC and ASD on fs-T2W were compared to those on ce-MRI. Primary tumor volumes (PTVs) of CNN-derived delineations were compared to that of manual delineations.

Results

The CNN for NPC delineation on fs-T2W images showed similar DSC (0.71 ± 0.09) and ASD (0.21 ± 0.48 cm) to those on ce-T1W images (0.71 ± 0.09 and 0.17 ± 0.19 cm, respectively) (p > 0.05), and lower DSC but similar ASD to ce-fs-T1W images (0.73 ± 0.09, p < 0.001; and 0.17 ± 0.20 cm, p > 0.05). The CNN overestimated PTVs on all sequences (p < 0.001).

Conclusion

The CNN showed promise for NPC delineation on fs-T2W images in cases where it is desirable to avoid contrast agent injection. The CNN overestimated PTVs on all sequences.

Nur mit Berechtigung zugänglich

King AD. Magnetic resonance imaging staging of nasopharyngeal carcinoma in the head and neck. World J Radiol. 2010;2(5):159. https://doi.org/10.4329/wjr.v2.i5.159.CrossRefPubMedPubMedCentral

Amin MB, Edge S, Greene F, Byrd DR, Brookland RK, Washington MK, et al. AJCC Cancer Staging Manual. 8th ed. Amin MB, Edge S, Greene F, Byrd DR, Brookland RK, Washington MK, et al., editors. Springer International Publishing; 2017.

Li J, Huang S, Jiang X, Ouyang B, Han F, Liu S, et al. Local failure patterns for patients with nasopharyngeal carcinoma after intensity-modulated radiotherapy. Radiat Oncol. 2014;9(1):87. https://doi.org/10.1186/1748-717X-9-87.CrossRefPubMedPubMedCentral

He YX, Wang Y, Cao PF, Shen LF, Zhao YJ, Zhang ZJ, et al. Prognostic value and predictive threshold of tumor volume for patients with locally advanced nasopharyngeal carcinoma receiving intensity-modulated radiotherapy. Chin J Cancer. 2016;35(1):96. https://doi.org/10.1186/s40880-016-0159-2.CrossRefPubMedPubMedCentral

Feng M, Wang W, Fan Z, Fu B, Li J, Zhang S, et al. Tumor volume is an independent prognostic indicator of local control in nasopharyngeal carcinoma patients treated with intensity-modulated radiotherapy. Radiat Oncol. 2013;8(1):208. https://doi.org/10.1186/1748-717X-8-208.CrossRefPubMedPubMedCentral

Linge A, Lohaus F, Löck S, Nowak A, Gudziol V, Valentini C, et al. HPV status, cancer stem cell marker expression, hypoxia gene signatures and tumour volume identify good prognosis subgroups in patients with HNSCC after primary radiochemotherapy: A multicentre retrospective study of the German Cancer Consortium Radiation. Radiother Oncol. 2016;121(3):364–73. https://doi.org/10.1016/j.radonc.2016.11.008.CrossRefPubMed

Işın A, Direkoğlu C, Şah M. Review of MRI-based Brain Tumor Image Segmentation Using Deep Learning Methods. Procedia Comput Sci. 2016;102:317–24. https://doi.org/10.1016/j.procs.2016.09.407.CrossRef

Wang C, Tyagi N, Rimner A, Hu Y-C, Veeraraghavan H, Li G, et al. Segmenting lung tumors on longitudinal imaging studies via a patient-specific adaptive convolutional neural network. Radiother Oncol. 2019;131:101–7. https://doi.org/10.1016/j.radonc.2018.10.037.CrossRefPubMed

Oktay O, Schlemper J, Folgoc L Le, Lee M, Heinrich M, Misawa K, et al. Attention U-Net: Learning Where to Look for the Pancreas [Preprint]. 2018 [cited 2018 Apr 11]. https://arxiv.org/abs/1804.03999

10.

Litjens G, Kooi T, Bejnordi BE, Arindra A, Setio A, Ciompi F, et al. A survey on deep learning in medical image analysis. Med Image Anal. 2017;42:60–88. https://doi.org/10.1016/j.media.2017.07.005.CrossRefPubMed

11.

Lin L, Dou Q, Jin Y-M, Zhou G-Q, Tang Y-Q, Chen W-L, et al. Deep learning for automated contouring of primary tumor volumes by mri for nasopharyngeal carcinoma. Radiology. 2019;3:182012. https://doi.org/10.1148/radiol.2019182012.CrossRef

12.

Chen H, Qi Y, Yin Y, Li T, Liu X, Li X, et al. MMFNet: A multi-modality MRI fusion network for segmentation of nasopharyngeal carcinoma. Neurocomputing. 2020;394:27–40. https://doi.org/10.1016/j.neucom.2020.02.002.CrossRef

13.

Li Q, Xu Y, Chen Z, Liu D, Feng S-T, Law M, et al. Tumor segmentation in contrast-enhanced magnetic resonance imaging for nasopharyngeal carcinoma: deep learning with convolutional neural network. Biomed Res Int. 2018;2018(17):1–7. https://doi.org/10.1155/2018/9128527.CrossRef

14.

Ye Y, Cai Z, Huang B, He Y, Zeng P, Zou G, et al. Fully-automated segmentation of nasopharyngeal carcinoma on dual-sequence MRI using convolutional neural networks. Front Oncol. 2020;19:10. https://doi.org/10.3389/fonc.2020.00166.CrossRef

15.

Ma Z, Wu X, Song Q, Luo Y, Wang Y, Zhou J. Automated nasopharyngeal carcinoma segmentation in magnetic resonance images by combination of convolutional neural networks and graph cut. Exp Ther Med. 2018;16(3):2511–21. https://doi.org/10.3892/etm.2018.6478.CrossRefPubMedPubMedCentral

16.

Thomsen HS, Morcos SK, Almén T, Bellin M-F, Bertolotto M, Bongartz G, et al. Nephrogenic systemic fibrosis and gadolinium-based contrast media: updated ESUR Contrast Medium Safety Committee guidelines. Eur Radiol. 2013;23(2):307–18. https://doi.org/10.1007/s00330-012-2597-9.CrossRefPubMed

17.

Kanda T, Ishii K, Kawaguchi H, Kitajima K, Takenaka D. High signal intensity in the dentate nucleus and globus pallidus on unenhanced T1-weighted MR images: relationship with increasing cumulative dose of a gadolinium-based contrast material. Radiology. 2014;270(3):834–41. https://doi.org/10.1148/radiol.13131669.CrossRefPubMed

18.

Choi JW, Moon W-J. Gadolinium Deposition in the Brain: Current Updates. Korean J Radiol. 2019;20(1):134. https://doi.org/10.3348/kjr.2018.0356.CrossRefPubMed

19.

King AD, Lam WW, Leung SF, Chan YL, Metreweli C. Comparison of T2 weighted fat suppressed turbo spin echo and contrast enhanced T1 weighted spin echo MRI in nasopharyngeal carcinoma. Br J Radiol. 1997;70(840):1208–14. https://doi.org/10.1259/bjr.70.840.9505838.CrossRefPubMed

20.

Tokuda O, Harada Y, Matsunaga N. MRI of soft-tissue tumors: fast STIR sequence as substitute for T1-weighted fat-suppressed contrast-enhanced spin-echo sequence. Am J Roentgenol. 2009;193(6):1607–14. https://doi.org/10.2214/AJR.09.2675.CrossRef

21.

Lau KY. Magnetic resonance for T-staging of nasopharyngeal carcinoma-the most informative pair of sequences. Jpn J Clin Oncol. 2004;34(4):171–5. https://doi.org/10.1093/jjco/hyh033.CrossRefPubMed

22.

Yushkevich PA, Piven J, Hazlett HC, Smith RG, Ho S, Gee JC, et al. User-guided 3D active contour segmentation of anatomical structures: Significantly improved efficiency and reliability. Neuroimage. 2006;31(3):1116–28. https://doi.org/10.1016/j.neuroimage.2006.01.015.CrossRefPubMed

23.

Park SH, Han K. Methodologic guide for evaluating clinical performance and effect of artificial intelligence technology for medical diagnosis and prediction. Radiology. 2018;286(3):800–9. https://doi.org/10.1148/radiol.2017171920.CrossRefPubMed

24.

Ronneberger O, Fischer P, Brox T. U-Net: convolutional networks for biomedical image segmentation. Miccai. 2015. https://doi.org/10.1007/978-3-319-24574-4_28.CrossRef

25.

Paszke A, Gross S, Massa F, Lerer A, Bradbury J, Chanan G, et al. PyTorch: An Imperative Style, High-Performance Deep Learning Library. In: Advances in Neural Information Processing Systems 32. Curran Associates, Inc.; 2019. p. 8024–35.

26.

Taha AA, Hanbury A. Metrics for evaluating 3D medical image segmentation: Analysis, selection, and tool. BMC Med Imaging. 2015. https://doi.org/10.1186/s12880-015-0068-x.CrossRefPubMedPubMedCentral

27.

Giavarina D. Understanding Bland Altman analysis. Biochem Medica. 2015;25(2):141–51. https://doi.org/10.11613/BM.2015.015.CrossRef

28.

Datta D. blandr: a Bland-Altman Method Comparison package for R [Internet]. 2017. [cited 2020 Jan 6] doi: https://doi.org/10.5281/zenodo.824514https://github.com/deepankardatta/blandr

29.

Guan S, Khan A, Sikdar S, Chitnis P. Fully Dense UNet for 2D Sparse Photoacoustic Tomography Artifact Removal. IEEE J Biomed Heal Informatics. 2019. https://doi.org/10.1109/JBHI.2019.2912935.CrossRef

30.

Wang C, Zhao Z, Ren Q, Xu Y, Yu Y. Dense U-net Based on Patch-Based Learning for Retinal Vessel Segmentation. Entropy. 2019;21(2):168. https://doi.org/10.3390/e21020168.CrossRefPubMedCentral

31.

Hua R, Huo Q, Gao Y, Sui H, Zhang B, Sun Y, et al. Segmenting brain tumor using cascaded v-nets in multimodal MR Images. Front Comput Neurosci. 2020. https://doi.org/10.3389/fncom.2020.00009.CrossRefPubMedPubMedCentral

32.

Liang Y, Schott D, Zhang Y, Wang Z, Nasief H, Paulson E, et al. Auto-segmentation of pancreatic tumor in multi-parametric MRI using deep convolutional neural networks. Radiother Oncol. 2020;145:193–200. https://doi.org/10.1016/j.radonc.2020.01.021.CrossRefPubMed

Titel: Convolutional neural network in nasopharyngeal carcinoma: how good is automatic delineation for primary tumor on a non-contrast-enhanced fat-suppressed T2-weighted MRI?
verfasst von: Lun M. Wong
Qi Yong H. Ai
Frankie K. F. Mo
Darren M. C. Poon
Ann D. King
Publikationsdatum: 05.02.2021
Verlag: Springer Singapore
Erschienen in: Japanese Journal of Radiology / Ausgabe 6/2021
Print ISSN: 1867-1071
Elektronische ISSN: 1867-108X
DOI: https://doi.org/10.1007/s11604-021-01092-x

Update Radiologie

Bestellen Sie unseren Fach-Newsletter und bleiben Sie gut informiert.

Newsletter bestellen

Die Highlights vom Kongress des American College of Cardiology 2024

Springer Medizin

Convolutional neural network in nasopharyngeal carcinoma: how good is automatic delineation for primary tumor on a non-contrast-enhanced fat-suppressed T2-weighted MRI?

Abstract

Purpose

Materials and methods

Results

Conclusion

Neu im Fachgebiet Radiologie

„Übersichtlicher Wegweiser“: Lauterbachs umstrittener Klinik-Atlas ist online

Klinikreform soll zehntausende Menschenleben retten

Darf man die Behandlung eines Neonazis ablehnen?

Ein Drittel der jungen Ärztinnen und Ärzte erwägt abzuwandern

Update Radiologie

Die Highlights vom Kongress des American College of Cardiology 2024

Springer Medizin

Abstract

Purpose

Materials and methods

Results

Conclusion

Bitte loggen Sie sich ein, um Zugang zu diesem Inhalt zu erhalten

Weitere Artikel der Ausgabe 6/2021

Pulmonary artery trunk enlargement on admission as a predictor of mortality in in-hospital patients with COVID-19

Spectrum of CT findings in amebic colitis

Correction to: Artificial intelligence: a critical review of current applications in pancreatic imaging

The future of cardiac disease assessment using 18F-FDG PET/CT

Direct aspiration thrombectomy experience with the SOFIA 6F catheter in acute ischemic stroke

Could both intrinsic and extrinsic iodine be successfully suppressed on virtual non-contrast CT images for detecting thyroid calcification?

Neu im Fachgebiet Radiologie

„Übersichtlicher Wegweiser“: Lauterbachs umstrittener Klinik-Atlas ist online

Klinikreform soll zehntausende Menschenleben retten

Darf man die Behandlung eines Neonazis ablehnen?

Ein Drittel der jungen Ärztinnen und Ärzte erwägt abzuwandern

Update Radiologie