Machine learning–based MRI texture analysis to predict occult lymph node metastasis in early-stage oral tongue squamous cell carcinoma

Abstract

Objectives

To develop and compare several machine learning models to predict occult cervical lymph node (LN) metastasis in early-stage oral tongue squamous cell cancer (OTSCC) from preoperative MRI texture features.

Materials and methods

We retrospectively enrolled 116 patients with early-stage OTSCC (cT1-2N0) who had been surgically treated by tumor excision and elective neck dissection (END). For each patient, we extracted 86 texture features from T2-weighted imaging (T2WI) and contrast-enhanced T1-weighted imaging (ceT1WI), respectively. Dimension reduction was performed in three consecutive steps: reproducibility analysis, collinearity analysis, and information gain algorithm. Models were created using six machine learning methods, including logistic regression (LR), random forest (RF), naïve Bayes (NB), support vector machine (SVM), AdaBoost, and neural network (NN). Their performance was assessed using tenfold cross-validation.

Results

Occult LN metastasis was pathologically detected in 42.2% (49/116) of the patients. No significant association was identified between node status and patients’ gender, age, or clinical T stage. Dimension reduction steps selected 6 texture features. The NB model gave the best overall performance, which correctly classified the nodal status in 74.1% (86/116) of the carcinomas, with an AUC of 0.802.

Conclusion

Machine learning–based MRI texture analysis offers a feasible tool for preoperative prediction of occult cervical node metastasis in early-stage OTSCC.

Key Points

• A machine learningbased MRI texture analysis approach was adopted to predict occult cervical node metastasis in early-stage OTSCC with no evidence of node involvement on conventional images.

• Six texture features from T2WI and ceT1WI of preoperative MRI were selected to construct the predictive model.

• After comparing six machine learning methods, naïve Bayes (NB) achieved the best performance by correctly identifying the node status in 74.1% of the patients, using tenfold cross-validation.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5

Abbreviations

AUC:

Area under the curve

ceT1WI:

Contrast-enhanced T1-weighted imaging

DOI:

Depth of invasion

END:

Elective neck dissection

GLCM:

Gray-level co-occurrence matrix

GLDM:

Gray-level dependence matrix

GLRLM:

Gray-level run-length matrix

GLSZM:

Gray-level size zone matrix

HNSCC:

Head and neck squamous cell carcinoma

ICCs:

Intraclass correlation coefficients

LN:

Lymph node

LR:

Logistic regression

MRI:

Magnetic resonance imaging

NB:

Naïve Bayes

NN:

Neural network

OTSCC:

Oral tongue squamous cell carcinoma

RF:

Random forest

ROC:

Receiver operator characteristic

ROI:

Region of interests

SVM:

Support vector machine

T2WI:

T2-Weighted imaging

TE:

Echo time

TR:

Repetition time

VOI:

Volume of interest

References

  1. 1.

    Chen W, Zheng R, Baade PD et al (2016) Cancer statistics in China, 2015. CA Cancer J Clin 66:115–132

    Article  Google Scholar 

  2. 2.

    Greenberg JS, El Naggar AK, Mo V, Roberts D, Myers JN (2003) Disparity in pathologic and clinical lymph node staging in oral tongue carcinoma. Implication for therapeutic decision making. Cancer 98:508–515

    Article  Google Scholar 

  3. 3.

    Oh LJ, Phan K, Kim SW et al (2020) Elective neck dissection versus observation for early-stage oral squamous cell carcinoma: systematic review and meta-analysis. Oral Oncol 105:104661

    CAS  Article  Google Scholar 

  4. 4.

    Kelner N, Rodrigues PC, Bufalino A et al (2015) Activin A immunoexpression as predictor of occult lymph node metastasis and overall survival in oral tongue squamous cell carcinoma. Head Neck 37:479–486

    Article  Google Scholar 

  5. 5.

    Yuen AP, Ho CM, Chow TL et al (2009) Prospective randomized study of selective neck dissection versus observation for N0 neck of early tongue carcinoma. Head Neck 31:765–772

    Article  Google Scholar 

  6. 6.

    Kelly HR, Curtin HD (2017) Chapter 2 Squamous cell carcinoma of the head and neck-imaging evaluation of regional lymph nodes and implications for management. Semin Ultrasound CT MR 38:466–478

    Article  Google Scholar 

  7. 7.

    Goel V, Parihar PS, Parihar A et al (2016) Accuracy of MRI in prediction of tumour thickness and nodal stage in oral tongue and gingivobuccal cancer with clinical correlation and staging. J Clin Diagn Res 10:Tc01–Tc05

    PubMed  PubMed Central  Google Scholar 

  8. 8.

    Liang L, Luo X, Lian Z et al (2017) Lymph node metastasis in head and neck squamous carcinoma: efficacy of intravoxel incoherent motion magnetic resonance imaging for the differential diagnosis. Eur J Radiol 90:159–165

    Article  Google Scholar 

  9. 9.

    Noij DP, Pouwels PJW, Ljumanovic R et al (2015) Predictive value of diffusion-weighted imaging without and with including contrast-enhanced magnetic resonance imaging in image analysis of head and neck squamous cell carcinoma. Eur J Radiol 84:108–116

    Article  Google Scholar 

  10. 10.

    Connolly M, Srinivasan A (2018) Diffusion-weighted imaging in head and neck cancer: technique, limitations, and applications. Magn Reson Imaging Clin N Am 26:121–133

  11. 11.

    Yamada I, Yoshino N, Hikishima K et al (2018) Oral carcinoma: clinical evaluation using diffusion kurtosis imaging and its correlation with histopathologic findings. Magn Reson Imaging 51:69–78

    Article  Google Scholar 

  12. 12.

    Romeo V, Cuocolo R, Ricciardi C et al (2020) Prediction of tumor grade and nodal status in oropharyngeal and oral cavity squamous-cell carcinoma using a radiomic approach. Anticancer Res 40:271–280

    Article  Google Scholar 

  13. 13.

    Park JH, Bae YJ, Choi BS et al (2019) Texture analysis of multi-shot echo-planar diffusion-weighted imaging in head and neck squamous cell carcinoma: the diagnostic value for nodal metastasis. J Clin Med 8:1767

    CAS  Article  Google Scholar 

  14. 14.

    Huang YQ, Liang CH, He L et al (2016) Development and validation of a radiomics nomogram for preoperative prediction of lymph node metastasis in colorectal cancer. J Clin Oncol 34:2157–2164

    Article  Google Scholar 

  15. 15.

    Wu S, Zheng J, Li Y et al (2018) Development and validation of an MRI-based radiomics signature for the preoperative prediction of lymph node metastasis in bladder cancer. EBioMedicine 34:76–84

    Article  Google Scholar 

  16. 16.

    Forghani R, Chatterjee A, Reinhold C et al (2019) Head and neck squamous cell carcinoma: prediction of cervical lymph node metastasis by dual-energy CT texture analysis with machine learning. Eur Radiol 29:6172–6181

    Article  Google Scholar 

  17. 17.

    Kocak B, Durmaz ES, Ates E, Kaya OK, Kilickesmez O (2019) Unenhanced CT texture analysis of clear cell renal cell carcinomas: a machine learning-based study for predicting histopathologic nuclear grade. AJR Am J Roentgenol W1–W8. https://doi.org/10.2214/AJR.18.20742

  18. 18.

    Shrout PE, Fleiss JL (1979) Intraclass correlations: uses in assessing rater reliability. Psychol Bull 86:420–428

    CAS  Article  Google Scholar 

  19. 19.

    Daghistani TA, Elshawi R, Sakr S et al (2019) Predictors of in-hospital length of stay among cardiac patients: a machine learning approach. Int J Cardiol 288:140–147

    Article  Google Scholar 

  20. 20.

    Lambin P, Leijenaar RTH, Deist TM et al (2017) Radiomics: the bridge between medical imaging and personalized medicine. Nat Rev Clin Oncol 14:749–762

    Article  Google Scholar 

  21. 21.

    Bayanati H, Thornhill RE, Souza CA et al (2015) Quantitative CT texture and shape analysis: can it differentiate benign and malignant mediastinal lymph nodes in patients with primary lung cancer? Eur Radiol 25:480–487

    Article  Google Scholar 

  22. 22.

    Andersen MB, Harders SW, Ganeshan B et al (2016) CT texture analysis can help differentiate between malignant and benign lymph nodes in the mediastinum in patients suspected for lung cancer. Acta Radiol 57:669–676

    Article  Google Scholar 

  23. 23.

    Liu S, Shi H, Ji C et al (2018) Preoperative CT texture analysis of gastric cancer: correlations with postoperative TNM staging. Clin Radiol 73:756.e751–756.e759

    Google Scholar 

  24. 24.

    Kuno H, Garg N (2019) CT texture analysis of cervical lymph nodes on contrast-enhanced [(18)F] FDG-PET/CT images to differentiate nodal metastases from reactive lymphadenopathy in HIV-positive patients with head and neck squamous cell carcinoma. AJNR Am J Neuroradiol 40:543–550

  25. 25.

    Kan Y, Dong D, Zhang Y et al (2019) Radiomic signature as a predictive factor for lymph node metastasis in early-stage cervical cancer. J Magn Reson Imaging 49:304–310

    Article  Google Scholar 

  26. 26.

    Wang T, Gao T, Yang J et al (2019) Preoperative prediction of pelvic lymph nodes metastasis in early-stage cervical cancer using radiomics nomogram developed based on T2-weighted MRI and diffusion-weighted imaging. Eur J Radiol 114:128–135

    Article  Google Scholar 

  27. 27.

    Yu YY, Zhang R, Dong RT et al (2019) Feasibility of an ADC-based radiomics model for predicting pelvic lymph node metastases in patients with stage IB-IIA cervical squamous cell carcinoma. Br J Radiol 92:20180986

    Article  Google Scholar 

  28. 28.

    Cui X, Wang N, Zhao Y et al (2019) Preoperative prediction of axillary lymph node metastasis in breast cancer using radiomics features of DCE-MRI. Sci Rep 9:2240

    Article  Google Scholar 

  29. 29.

    Han L, Zhu Y, Liu Z et al (2019) Radiomic nomogram for prediction of axillary lymph node metastasis in breast cancer. Eur Radiol 29:3820–3829

    Article  Google Scholar 

  30. 30.

    Liu C, Ding J, Spuhler K et al (2019) Preoperative prediction of sentinel lymph node metastasis in breast cancer by radiomic signatures from dynamic contrast-enhanced MRI. J Magn Reson Imaging 49:131–140

    Article  Google Scholar 

  31. 31.

    Friedman NGD, Goldszmidt M (1997) Bayesian network classifiers. Mach Learn 29:131–163

    Article  Google Scholar 

  32. 32.

    Rodolico V, Barresi E, Di Lorenzo R et al (2004) Lymph node metastasis in lower lip squamous cell carcinoma in relation to tumour size, histologic variables and p27Kip1 protein expression. Oral Oncol 40:92–98

    CAS  Article  Google Scholar 

  33. 33.

    Okura M, Iida S, Aikawa T et al (2008) Tumor thickness and paralingual distance of coronal MR imaging predicts cervical node metastases in oral tongue carcinoma. AJNR Am J Neuroradiol 29:45–50

    CAS  Article  Google Scholar 

  34. 34.

    Kwon M, Moon H, Nam SY et al (2016) Clinical significance of three-dimensional measurement of tumour thickness on magnetic resonance imaging in patients with oral tongue squamous cell carcinoma. Eur Radiol 26:858–865

    Article  Google Scholar 

  35. 35.

    Lam P, Au-Yeung KM, Cheng PW et al (2004) Correlating MRI and histologic tumor thickness in the assessment of oral tongue cancer. AJR Am J Roentgenol 182:803–808

    Article  Google Scholar 

Download references

Funding

This study has received funding by National Scientific Foundation of China (Grant number: 91859202, 81771901, to Xiaofeng Tao). Youth Medical Talents-Medical Imaging Practitioner Program (to Ying Yuan), Shanghai Municipal Health Commission (Grant number: 20194Y0104 to Jiliang Ren).

Author information

Affiliations

Authors

Corresponding author

Correspondence to Xiaofeng Tao.

Ethics declarations

Guarantor

The scientific guarantor of this publication is Xiaofeng Tao.

Conflict of interest

The authors of this manuscript declare no relationships with any companies whose products or services may be related to the subject matter of the article.

Statistics and biometry

One of the authors has significant statistical expertise.

Informed consent

Written informed consent was waived by the Institutional Review Board.

Ethical approval

Institutional Review Board approval was obtained.

Methodology

• retrospective

• case-control study

• performed at one institution

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Yuan, Y., Ren, J. & Tao, X. Machine learning–based MRI texture analysis to predict occult lymph node metastasis in early-stage oral tongue squamous cell carcinoma. Eur Radiol (2021). https://doi.org/10.1007/s00330-021-07731-1

Download citation

Keywords

  • Squamous cell carcinoma of head and neck
  • Lymphatic metastasis
  • Magnetic resonance imaging
  • Machine learning
  • Computer-assisted diagnosis