Skip to content
naulon
Get started

No retraction, correction or concern found when last checked on Oct 24, 2026.

Comparison of machine learning methods for classifying mediastinal lymph node metastasis of non-small cell lung cancer from 18F-FDG PET/CT images

Hongkai Wang, Zongwei Zhou, Yingci Li and 5 more

EJNMMI Research · 2017

doi:10.1186/s13550-017-0260-9

Abstract

Background. This study aimed to compare one state-of-the-art deep learning method and four classical machine learning methods for classifying mediastinal lymph node metastasis of non-small cell lung cancer (NSCLC) from 18F-FDG PET/CT images. Another objective was to compare the discriminative power of the recently popular PET/CT texture features with the widely used diagnostic features such as tumor size, CT value, SUV, image contrast, and intensity standard deviation. The four classical machine learning methods included random forests, support vector machines, adaptive boosting, and artificial neural network. The deep learning method was the convolutional neural networks (CNN). The five methods were evaluated using 1397 lymph nodes collected from PET/CT images of 168 patients, with corresponding pathology analysis results as gold standard. The comparison was conducted using 10 times 10-fold cross-validation based on the criterion of sensitivity, specificity, accuracy (ACC), and area under the ROC curve (AUC). For each classical method, different input features were compared to select the optimal feature set. Based on the optimal feature set, the classical methods were compared with CNN, as well as with human doctors from our institute.

Results. For the classical methods, the diagnostic features resulted in 81~85% ACC and 0.87~0.92 AUC, which were significantly higher than the results of texture features. CNN’s sensitivity, specificity, ACC, and AUC were 84, 88, 86, and 0.91, respectively. There was no significant difference between the results of CNN and the best classical method. The sensitivity, specificity, and ACC of human doctors were 73, 90, and 82, respectively. All the five machine learning methods had higher sensitivities but lower specificities than human doctors.

Conclusions. The present study shows that the performance of CNN is not significantly different from the best classical methods and human doctors for classifying mediastinal lymph node metastasis of NSCLC from PET/CT images. Because CNN does not need tumor segmentation or feature calculation, it is more convenient and more objective than the classical methods. However, CNN does not make use of the import diagnostic features, which have been proved more discriminative than the texture features for classifying small-sized lymph nodes. Therefore, incorporating the diagnostic features into CNN is a promising direction for future research.

Full text

Its licence does not allow us to show the full text. Read it at PubMed Central:

Read at PubMed Central

References 41

  1. 1.

    McField D, Bauer T. A review of noninvasive staging of the mediastinum for non-small cell lung carcinoma. Surg Oncol Clin N Am. 2011;20:681-90.

    Not in our library10.1016/j.soc.2011.07.002

  2. 2.

    Silvestri GA, Gould MK, Margolis ML, Tanoue LT, McCrory D, Toloza E, et al. Noninvasive staging of non-small cell lung cancer: ACCP evidenced-based clinical practice guidelines (2nd edition). Chest. 2007;132:178S-201S.

    Not in our library10.1378/chest.07-1360

  3. 3.

    Broderick SR, Meyers BF. PET staging of mediastinal lymph nodes in thoracic oncology. Thorac Surg Clin. 2012;22:161-6.

    Not in our library10.1016/j.thorsurg.2011.12.004

  4. 4.

    Kim SK, Allen-Auerbach M, Goldin J, Fueger BJ, Dahlbom M, Brown M, et al. Accuracy of PET/CT in characterization of solitary pulmonary lesions. J Nucl Med. 2007;48:214-20.

    Not in our library

  5. 5.

    Li X, Zhang H, Xing L, Ma H, Xie P, Zhang L, et al. Mediastinal lymph nodes staging by 18F-FDG PET/CT for early stage non-small cell lung cancer: a multicenter study. Radiother Oncol. 2012;102:246-50.

    Not in our library10.1016/j.radonc.2011.10.016

  6. 6.

    Schmidt-Hansen M, Baldwin DR, Hasler E, Zamora J, Abraira V, Roque IFM. PET-CT for assessing mediastinal lymph node involvement in patients with suspected resectable non-small cell lung cancer. Cochrane Database Syst Rev. 2014;11:CD009519.

    Not in our library10.1002/14651858.cd009519.pub2

  7. 7.

    De Leyn P, Vansteenkiste J, Cuypers P, Deneffe G, Van Raemdonck D, Coosemans W, et al. Role of cervical mediastinoscopy in staging of non-small cell lung cancer without enlarged mediastinal lymph nodes on CT scan. Eur J Cardiothorac Surg. 1997;12:706-12.

    Not in our library10.1016/s1010-7940(97)00253-4

  8. 8.

    Zheng Y, Sun X, Wang J, Zhang L, Di X, Xu Y. FDG-PET/CT imaging for tumor staging and definition of tumor volumes in radiation treatment planning in non-small cell lung cancer. Oncol Lett. 2014;7:1015-20.

    Not in our library10.3892/ol.2014.1874

  9. 9.

    Edet-Sanson A, Dubray B, Doyeux K, Back A, Hapdey S, Modzelewski R, et al. Serial assessment of FDG-PET FDG uptake and functional volume during radiotherapy (RT) in patients with non-small cell lung cancer (NSCLC). Radiother Oncol. 2012;102:251-7.

    Not in our library10.1016/j.radonc.2011.07.023

  10. 10.

    Wang J, Welch K, Wang L, Kong FM. Negative predictive value of positron emission tomography and computed tomography for stage T1-2N0 non-small-cell lung cancer: a meta-analysis. Clin Lung Cancer. 2012;13:81-9.

    Not in our library10.1016/j.cllc.2011.08.002

  11. 11.

    Liao CY, Chen JH, Liang JA, Yeh JJ, Kao CH. Meta-analysis study of lymph node staging by 18F-FDG PET/CT scan in non-small cell lung cancer: comparison of TB and non-TB endemic regions. Eur J Radiol. 2012;81:3518-23.

    Not in our library10.1016/j.ejrad.2012.02.007

  12. 12.

    Ambrosini V, Fanti S, Chengazi VU, Rubello D. Diagnostic accuracy of FDG PET/CT in mediastinal lymph nodes from lung cancer. Eur J Radiol. 2014;83:1301-2.

    Not in our library10.1016/j.ejrad.2014.04.035

  13. 13.

    Silvestri GA, Gonzalez AV, Jantz MA, Margolis ML, Gould MK, Tanoue LT, et al. Methods for staging non-small cell lung cancer: diagnosis and management of lung cancer, 3rd ed: American College of Chest Physicians evidence-based clinical practice guidelines. Chest. 2013;143:e211S-50S.

    Not in our library10.1378/chest.12-2355

  14. 14.

    Gonçalves VM, Delamaro ME. A systematic review on the evaluation and characteristics of computer-aided diagnosis systems. Revista Brasileira De Engenharia Biomedica. 2014;30:355-83.

    Not in our library10.1590/1517-3151.0517

  15. 15.

    Chicklore S, Goh V, Siddique M, Roy A, Marsden P, Cook GJR. Quantifying tumour heterogeneity in 18F-FDG PET/CT imaging by texture analysis. Eur J Nucl Med Mol Imaging. 2013;40:133-40.

    Not in our library10.1007/s00259-012-2247-0

  16. 16.

    Yu H, Caldwell C, Mah K, Mozeg D. Coregistered FDG PET/CT-based textural characterization of head and neck cancer for radiation treatment planning. IEEE Trans Med Imag. 2009;28:374–83.

    Not in our library10.1109/tmi.2008.2004425

  17. 17.

    Rui X, Kido S, Suga K, Hirano Y, Tachibana R, Muramatsu K, et al. Texture analysis on 18F-FDG PET/CT images to differentiate malignant and benign bone and soft-tissue lesions. Jpn Circ J. 2014;58:95-9.

    Not in our library10.1007/s12149-014-0895-9

  18. 18.

    Gao X, Chu C, Li Y, Lu P, Wang W, Liu W, et al. The method and efficacy of support vector machine classifiers based on texture features and multi-resolution histogram from (18)F-FDG PET-CT images for the evaluation of mediastinal lymph nodes in patients with lung cancer. Eur J Radiol. 2015;84:312-7.

    Not in our library10.1016/j.ejrad.2014.11.006

  19. 19.

    Vaidya M, Creach KM, Frye J, Dehdashti F, Bradley JD, Naqa IE. Combined PET/CT image characteristics for radiotherapy tumor response in lung cancer. Radiother Oncol J European Soc Therapeutic Radiol Oncol. 2012;102:239-45.

    Not in our library10.1016/j.radonc.2011.10.014

  20. 20.

    Lian C, Ruan S, Denœux T, Jardin F, Vera P. Selecting radiomic features from FDG-PET images for cancer treatment outcome prediction. Med Image Anal. 2016;32:257-68.

    Not in our library10.1016/j.media.2016.05.007

  21. 21.

    Brooks FJ, Grigsby PW. Low-order non-spatial effects dominate second-order spatial effects in the texture quantifier analysis of 18F-FDG-PET images. Plos One. 2015;10:1-17.

    Not in our library10.1371/journal.pone.0116574

  22. 22.

    Hatt M, Tixier F, Pierce L, Kinahan PE, Le Rest CC, Visvikis D. Characterization of PET/CT images using texture analysis: the past, the present… any future? European Journal of Nuclear Medicine and Molecular Imaging, 2016. 44(1): p. 1-15.

    Not in our library10.1007/s00259-016-3427-0

  23. 23.

    Shen D, Wu G, Suk H-I. Deep Learning in Medical Image Analysis[J]. Annual Review of Biomedical Engineering, 2017, 19(1). doi:10.1146/annurev-bioeng-071516-044442.

    Not in our library10.1146/annurev-bioeng-071516-044442

  24. 24.

    Roth HR, Lu L, Liu J, Yao J, Seff A, Cherry KM, et al. Improving computer-aided detection using convolutional neural networks and random view aggregation. IEEE Trans Med Imag. 2016;35:1170-81.

    Not in our library10.1109/tmi.2015.2482920

  25. 25.

    Roth HR, Lu L, Seff A, Cherry KM, Hoffman J, Wang S, et al. A new 2.5 D representation for lymph node detection using random sets ofdeep convolutional neural network observations[C]//International Conference on Medical Image Computing and Computer-Assisted Intervention, Boston, MA, USA. Springer International Publishing, 2014: 520-7.

    Not in our library10.1007/978-3-319-10404-1_65

  26. 26.

    Tajbakhsh N, Gotway MB, Liang J. Computer-aided pulmonary embolism detection using a novel vesselaligned multi-planar image representation and convolutional neural networks[C]//International Conference on Medical Image Computing and Computer-Assisted Intervention, Munich, Germany. Springer International Publishing, 2015: 62-69.

    Not in our library

  27. 27.

    Tajbakhsh N, Gurudu SR, Liang J. Automated polyp detection in colonoscopy videos using shape and context information. IEEE Trans Med Imag. 2016;35:630-44.

    Not in our library10.1109/tmi.2015.2487997

  28. 28.

    Xu J, Xiang L, Liu Q, Gilmore H, Wu J, Tang J, et al. Stacked sparse autoencoder (SSAE) for nuclei detection on breast cancer histopathology images. IEEE Trans Med Imag. 2016;35:119-30.

    Not in our library10.1109/tmi.2015.2458702

  29. 29.

    Xu Y, Zhu JY, Chang EI, Lai M, Tu Z. Weakly supervised histopathology cancer image segmentation and classification. Med Image Anal. 2014;18:591-604.

    Not in our library10.1016/j.media.2014.01.010

  30. 30.
    Ultrastructural Characterization of the Lower Motor System in a Mouse Model of Krabbe Disease · 2016
    No notice
  31. 31.

    Vallieres E, Shepherd FA, Crowley J, Van Houtte P, Postmus PE, Carney D, et al. The IASLC Lung Cancer Staging Project: proposals regarding the relevance of TNM in the pathologic staging of small cell lung cancer in the forthcoming (seventh) edition of the TNM classification for lung cancer. J Thorac Oncol. 2009;4:1049-59.

    Not in our library10.1097/jto.0b013e3181b27799

  32. 32.

    Francis T. Machine Learning: An Algorithmic Perspective. 2nd ed. 6000 Broken Sound Parkway,Suite 300: Chapman and Hall/CRC, Taylor & Francis Group; 2014. p.39-280.

    Not in our library

  33. 33.
    Deep learning · 2015
    No notice
  34. 34.

    Krizhevsky A, Sutskever I, Hinton GE. ImageNet classification with deep convolutional neural networks. Adv Neural Inf Proces Syst. 2012;25:2012.

    Not in our library

  35. 35.

    Qian N. On the momentum term in gradient descent learning algorithms. Neural Netw. 1999;12:145-51.

    Not in our library10.1016/s0893-6080(98)00116-6

  36. 36.
    Machine Learning methods for Quantitative Radiomic Biomarkers · 2015
    No notice
  37. 37.

    Fernandezdelgado M, Cernadas E, Barro S, Amorim D. Do we need hundreds of classifiers to solve real world classification problems. J Mach Learn Res. 2014;15:3133-81.

    Not in our library

  38. 38.

    Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition [J]. arXiv preprint arXiv:1409.1556v6, 2015.

    Not in our library

  39. 39.

    Szegedy C, Liu W, Jia Y, Sermanet P, Reed S, Anguelov D, et al. Going deeper with convolutions[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Boston, MA, USA. 2015: 1-9.

    Not in our library

  40. 40.

    He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, Nevada, USA. 2016: 770-8.

    Not in our library

All 41 references

Cited by papers we hold 1

Reads

Read by people on this page

None yet.

Checked or read by AI through naulon

None yet.

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.

Are you an author?

Is this your paper?

Claim it to add your note and be paid when agents cite it. Free.