Automatic detection and classification of nasopharyngeal carcinoma on PET/CT with support vector machine

  • Bangxian Wu
  • Pek-Lan Khong
  • Tao Chan
Open Access
Original Article



Positron emission tomography/computed tomography (PET/CT) has established values for imaging of head and neck cancers, including the nasopharyngeal carcinoma (NPC), utilizing both morphologic and functional information. In this paper, we introduce a computerized system for automatic detection of NPC, targeting both the primary tumor and regional nodal metastasis, on PET/CT.


Candidate lesions were extracted based on the features from both PET and CT images and a priori knowledge of anatomical features and subsequently classified by a support vector machine algorithm. The system was validated with 25 PET/CT examinations from 10 patients suffering from NPC. Lesions manually contoured by experienced radiologists were used as the gold standard.


Results showed that the system successfully identified all 53 hypermetabolic lesions larger than 1 cm in size and excluded normal physiological uptake in brown fat, muscles, bone marrow, brain, and salivary glands.


The system combined both imaging features and a priori clinical knowledge for classification between pathological and physiological uptake. Preliminary results showed that the system was highly accurate and promising for adoption in clinical use.


PET/CT Computer automatic diagnosis Support vector machine Nasopharyngeal carcinoma 


Open Access

This article is distributed under the terms of the Creative Commons Attribution Noncommercial License which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.


  1. 1.
    Kapoor V, McCook BM, Torok FS (2004) An introduction to PET-CT imaging. Radiographics 24(2): 523–543PubMedCrossRefGoogle Scholar
  2. 2.
    Townsend DW et al (2004) PET/CT today and tomorrow. J Nucl Med 45(Suppl 1): 4S–14SPubMedGoogle Scholar
  3. 3.
    Weber WA, Figlin R (2007) Monitoring cancer treatment with PET/CT: does it make a difference?. J Nucl Med 48(Suppl 1): 36S–44SPubMedGoogle Scholar
  4. 4.
    Lee SW et al (2008) Prediction of prognosis using standardized uptake value of 2-[(18)F] fluoro-2-deoxy-d-glucose positron emission tomography for nasopharyngeal carcinomas. Radiother Oncol 87(2): 211–216PubMedCrossRefGoogle Scholar
  5. 5.
    Schinagl DA et al (2011) Can FDG PET predict radiation treatment outcome in head and neck cancer? Results of a prospective study. Eur J Nucl Med Mol Imaging 38(8): 1449–1458PubMedCrossRefGoogle Scholar
  6. 6.
    Doi K (2005) Current status and future potential of computer-aided diagnosis in medical imaging. Br J Radiol 78(Spec No 1): S3–S19PubMedCrossRefGoogle Scholar
  7. 7.
    Tozaki YT et al (2003) Computer assisted diagnosis method of whole body cancer using FDG-pet images. In: Proceedings of international conference on image processing, vol 2, pp 1085–1088Google Scholar
  8. 8.
    Opfer R et al (2008) Automatic lesion tracking for a PET/CT based computer aided cancer therapy monitoring system. In: Giger ML, Karssemeijer N (eds) Medical imaging 2008: computer-aided diagnosis. Proceedings of the SPIE, vol 6915, pp 691513.1–691513.10Google Scholar
  9. 9.
    Sharif MS et al (2010) Artificial neural network-based system for PET volume segmentation. Int J Biomed Imaging 2010Google Scholar
  10. 10.
    Gao X et al (2010) Computer-assisted quantitative evaluation of therapeutic responses for lymphoma using serial PET/CT imaging. Acad Radiol 17(4): 479–488PubMedCrossRefGoogle Scholar
  11. 11.
    Cortes C, Vapnik V (1995) Support-vector networks. Mach Learn 20(3): 273–297Google Scholar
  12. 12.
    Suykens JAK, Vandewalle J (1999) Least squares support vector machine classifiers. Neural Process Lett 9(3):293–300Google Scholar
  13. 13.
    Thie JA (2004) Understanding the standardized uptake value, its methods, and implications for usage. J Nucl Med 45(9): 1431–1434PubMedGoogle Scholar
  14. 14.
    Aoki J et al (2001) FDG PET of primary benign and malignant bone tumors: standardized uptake value in 52 lesions. Radiology 219(3): 774–777PubMedGoogle Scholar
  15. 15.
    Kao CH et al (1998) Detection of recurrent or persistent nasopharyngeal carcinomas after radiotherapy with 18-fluoro-2-deoxyglucose positron emission tomography and comparison with computed tomography. J Clin Oncol 16(11): 3550–3555PubMedGoogle Scholar
  16. 16.
    Yen TC et al (2005) Are dual-phase 18F-FDG PET scans necessary in nasopharyngeal carcinoma to assess the primary tumour and loco-regional nodes?. Eur J Nucl Med Mol Imaging 32(5): 541–548PubMedCrossRefGoogle Scholar
  17. 17.
    Allal AS et al (2002) Standardized uptake value of 2-[F-18] fluoro-2-deoxy-D-glucose in predicting outcome in head and neck carcinomas treated by radiotherapy with or without chemotherapy. J Clin Oncol 20(5): 1398–1404PubMedCrossRefGoogle Scholar
  18. 18.
    Pham DL, Xu CY, Prince JL (2000) Current methods in medical image segmentation. Annu Rev Biomed Eng 2:315–337Google Scholar
  19. 19.
    Hill DLG et al (2001) Medical image registration. Phys Med Biol 46(3): R1–R45PubMedCrossRefGoogle Scholar
  20. 20.
    Furey TS et al (2000) Support vector machine classification and validation of cancer tissue samples using microarray expression data. Bioinformatics 16(10): 906–914PubMedCrossRefGoogle Scholar
  21. 21.
    Chang C, Lin CJ (2001) LIBSVM: a library for support vector machines.
  22. 22.
    Hsu C, Chang CC, Lin CJ (2003) A practical guide to support vector classification.
  23. 23.
    Ng SH et al (2004) Nodal metastases of nasopharyngeal carcinoma: patterns of disease on MRI and FDG PET. Eur J Nucl Med Mol Imaging 31(8): 1073–1080PubMedCrossRefGoogle Scholar
  24. 24.
    Ng SH et al (2009) Staging of untreated nasopharyngeal carcinoma with PET/CT: comparison with conventional imaging work-up. Eur J Nucl Med Mol Imaging 36(1): 12–22PubMedCrossRefGoogle Scholar
  25. 25.
    King AD et al (2008) The impact of 18F-FDG PET/CT on assessment of nasopharyngeal carcinoma at diagnosis. Br J Radiol 81(964): 291–298PubMedCrossRefGoogle Scholar
  26. 26.
    Aung W et al (2005) In-vivo PET imaging of inducible D2R reporter transgene expression using [11C]FLB 457 as reporter probe in living rats. Nucl Med Commun 26(3): 259–268PubMedCrossRefGoogle Scholar
  27. 27.
    Hara T et al (2008) Automated scoring system of standard uptake value for torso FDG-PET images. Proc SPIE 6915: 691534–6915344CrossRefGoogle Scholar
  28. 28.
    Griffeth LK (2005) Use of PET/CT scanning in cancer patients: technical and practical considerations. Proc (Bayl Univ Med Cent) 18(4): 321–330Google Scholar
  29. 29.
    Shih FY, Cheng SX (2005) Automatic seeded region growing for color image segmentation. Image Vis Comput 23(10): 877–886CrossRefGoogle Scholar
  30. 30.
    Loosli G, Canu S (2007) Comments on the “core vector machines: fast SVM training on very large data sets”. J Mach Learn Res 8: 291–301Google Scholar
  31. 31.
    Tsang IW, Kwok JT, Cheung PM (2005) Core vector machines: fast SVM training on very large data sets. J Mach Learn Res 6: 363–392Google Scholar

Copyright information

© The Author(s) 2011

Authors and Affiliations

  1. 1.Department of Diagnostic RadiologyThe University of Hong KongPok Fu LamHong Kong

Personalised recommendations