European Radiology

, Volume 28, Issue 6, pp 2604–2611 | Cite as

Spectral multi-energy CT texture analysis with machine learning for tissue classification: an investigation using classification of benign parotid tumours as a testing paradigm

  • Eiman Al Ajmi
  • Behzad Forghani
  • Caroline Reinhold
  • Maryam Bayat
  • Reza ForghaniEmail author
Head and Neck



There is a rich amount of quantitative information in spectral datasets generated from dual-energy CT (DECT). In this study, we compare the performance of texture analysis performed on multi-energy datasets to that of virtual monochromatic images (VMIs) at 65 keV only, using classification of the two most common benign parotid neoplasms as a testing paradigm.


Forty-two patients with pathologically proven Warthin tumour (n = 25) or pleomorphic adenoma (n = 17) were evaluated. Texture analysis was performed on VMIs ranging from 40 to 140 keV in 5-keV increments (multi-energy analysis) or 65-keV VMIs only, which is typically considered equivalent to single-energy CT. Random forest (RF) models were constructed for outcome prediction using separate randomly selected training and testing sets or the entire patient set.


Using multi-energy texture analysis, tumour classification in the independent testing set had accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of 92%, 86%, 100%, 100%, and 83%, compared to 75%, 57%, 100%, 100%, and 63%, respectively, for single-energy analysis.


Multi-energy texture analysis demonstrates superior performance compared to single-energy texture analysis of VMIs at 65 keV for classification of benign parotid tumours.

Key Points

• We present and validate a paradigm for texture analysis of DECT scans.

• Multi-energy dataset texture analysis is superior to single-energy dataset texture analysis.

• DECT texture analysis has high accura\cy for diagnosis of benign parotid tumours.

• DECT texture analysis with machine learning can enhance non-invasive diagnostic tumour evaluation.


Multidetector computed tomography Radiography dual-energy scanned projection Computer-assisted diagnosis Artificial intelligence Head and neck neoplasms 



Dual-energy CT


Negative predictive value


Positive predictive value


Region of interest


Virtual monochromatic image



The authors state that this work has not received any funding.

Compliance with ethical standards:


The scientific guarantor of this publication is R. Forghani.

Conflict of interest

The authors of this manuscript declare relationships with the following companies:

R.F. has acted as a consultant for GE Healthcare (DECT/GSI Neuro 510K image review/study for FDA clearance).

R.F. has been featured as a speaker at lunch-and-learn sessions, Applications of Dual-Energy CT in Neuroradiology and Head and Neck Imaging, sponsored by GE Healthcare at the 27th and 28th Annual Meetings of the Eastern Neuroradiological Society (2015, 2016; no personal financial compensation or travel support).

R.F. is a shareholder and previously acted as a consultant for Real Time Medical Inc. (teleradiology company), but the activities of this company have no relation to the topic of this study.

Statistics and biometry

B.F. has significant statistical and informatics expertise and performed the mathematical and statistical analyses.

Ethical approval

Ethics approval was obtained from the institutional review board of the Jewish General Hospital (CIUSSS West-Central Montreal).

Informed consent

Written informed consent was waived by the institutional review board.


• retrospective

• experimental

• performed at one institution.

Supplementary material

330_2017_5214_MOESM1_ESM.docx (27 kb)
ESM 1 (DOCX 27 kb)


  1. 1.
    Zhang H, Graham CM, Elci O et al (2013) Locally advanced squamous cell carcinoma of the head and neck: CT texture and histogram analysis allow independent prediction of overall survival in patients treated with induction chemotherapy. Radiology 269:801–809CrossRefPubMedGoogle Scholar
  2. 2.
    Aerts HJ, Velazquez ER, Leijenaar RT et al (2014) Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics approach. Nat Commun 5:4006PubMedPubMedCentralGoogle Scholar
  3. 3.
    Parmar C, Grossmann P, Bussink J, Lambin P, Aerts HJ (2015) Machine Learning methods for Quantitative Radiomic Biomarkers. Sci Rep 5:13087CrossRefPubMedPubMedCentralGoogle Scholar
  4. 4.
    Dang M, Lysack JT, Wu T et al (2015) MRI texture analysis predicts p53 status in head and neck squamous cell carcinoma. AJNR Am J Neuroradiol 36:166–170CrossRefPubMedGoogle Scholar
  5. 5.
    Buch K, Fujita A, Li B, Kawashima Y, Qureshi MM, Sakai O (2015) Using Texture Analysis to Determine Human Papillomavirus Status of Oropharyngeal Squamous Cell Carcinomas on CT. AJNR Am J Neuroradiol 36:1343–1348CrossRefPubMedGoogle Scholar
  6. 6.
    Leijenaar RT, Carvalho S, Hoebers FJ et al (2015) External validation of a prognostic CT-based radiomic signature in oropharyngeal squamous cell carcinoma. Acta Oncol 54:1423–1429CrossRefPubMedGoogle Scholar
  7. 7.
    Liu J, Mao Y, Li Z et al (2016) Use of texture analysis based on contrast-enhanced MRI to predict treatment response to chemoradiotherapy in nasopharyngeal carcinoma. J Magn Reson Imaging 44:445–455CrossRefPubMedGoogle Scholar
  8. 8.
    Vallieres M, Freeman CR, Skamene SR, El Naqa I (2015) A radiomics model from joint FDG-PET and MRI texture features for the prediction of lung metastases in soft-tissue sarcomas of the extremities. Phys Med Biol 60:5471–5496CrossRefPubMedGoogle Scholar
  9. 9.
    Foncubierta-Rodriguez A, Jimenez del Toro OA, Platon A, Poletti PA, Muller H, Depeursinge A (2013) Benefits of texture analysis of dual energy CT for Computer-Aided pulmonary embolism detection. Conf Proc IEEE Eng Med Biol Soc 2013:3973–3976Google Scholar
  10. 10.
    Jorge Oldan, Miao He, Teresa Wu, Alvin C. Silva, Jing Li, J. Ross Mitchell, William M. Pavlicek, Michael C. Roarke, Amy K. Hara, (2014) Pilot Study: Evaluation of Dual-Energy Computed Tomography Measurement Strategies for Positron Emission Tomography Correlation in Pancreatic Adenocarcinoma. Journal of Digital Imaging 27 (6):824–832CrossRefPubMedPubMedCentralGoogle Scholar
  11. 11.
    Depeursinge A, Foncubierta-Rodriguez A, Vargas A et al (2013) Rotation-covariant texture analysis of 4D dual-energy CT as an indicator of local pulmonary perfusionBiomedical Imaging (ISBI), 2013 I.E. 10th International Symposium on, San Francisco, CA, USA, pp 145–148Google Scholar
  12. 12.
    McCollough CH, Leng S, Yu L, Fletcher JG (2015) Dual- and Multi-Energy CT: Principles, Technical Approaches, and Clinical Applications. Radiology 276:637–653CrossRefPubMedPubMedCentralGoogle Scholar
  13. 13.
    Forghani R (2015) Advanced dual-energy CT for head and neck cancer imaging. Expert Rev Anticancer Ther.
  14. 14.
    Srinivasan A, Parker RA, Manjunathan A, Ibrahim M, Shah GV, Mukherji SK (2013) Differentiation of benign and malignant neck pathologies: preliminary experience using spectral computed tomography. J Comput Assist Tomogr 37:666–672CrossRefPubMedGoogle Scholar
  15. 15.
    Lam S, Gupta R, Levental M, Yu E, Curtin HD, Forghani R (2015) Optimal Virtual Monochromatic Images for Evaluation of Normal Tissues and Head and Neck Cancer Using Dual-Energy CT. AJNR Am J Neuroradiol 36:1518–1524CrossRefPubMedGoogle Scholar
  16. 16.
    Wichmann JL, Noske EM, Kraft J et al (2014) Virtual monoenergetic dual-energy computed tomography: optimization of kiloelectron volt settings in head and neck cancer. Invest Radiol 49:735–741CrossRefPubMedGoogle Scholar
  17. 17.
    Forghani R, Levental M, Gupta R, Lam S, Dadfar N, Curtin HD (2015) Different spectral Hounsfield unit curve and high-energy virtual monochromatic image characteristics of squamous cell carcinoma compared with nonossified thyroid cartilage. AJNR Am J Neuroradiol 36:1194–1200CrossRefPubMedGoogle Scholar
  18. 18.
    Matsumoto K, Jinzaki M, Tanami Y, Ueno A, Yamada M, Kuribayashi S (2011) Virtual monochromatic spectral imaging with fast kilovoltage switching: improved image quality as compared with that obtained with conventional 120-kVp CT. Radiology 259:257–262CrossRefPubMedGoogle Scholar
  19. 19.
    Forghani R, Kelly H, Yu E et al (2017) Low-Energy Virtual Monochromatic Dual-Energy Computed Tomography Images for the Evaluation of Head and Neck Squamous Cell Carcinoma: A Study of Tumor Visibility Compared With Single-Energy Computed Tomography and User Acceptance. J Comput Assist Tomogr.
  20. 20.
    Lam S, Gupta R, Kelly H, Curtin HD, Forghani R (2015) Multiparametric Evaluation of Head and Neck Squamous Cell Carcinoma Using a Single-Source Dual-Energy CT with Fast kVp Switching: State of the Art. Cancers (Basel) 7:2201–2216CrossRefGoogle Scholar
  21. 21.
    Ueno Y, Forghani B, Forghani R et al (2017) Endometrial Carcinoma: MR Imaging-based Texture Model for Preoperative Risk Stratification-A Preliminary Analysis. Radiology.
  22. 22.
    Albrecht MH, Scholtz JE, Kraft J et al (2015) Assessment of an Advanced Monoenergetic Reconstruction Technique in Dual-Energy Computed Tomography of Head and Neck Cancer. Eur Radiol 25:2493–2501CrossRefPubMedGoogle Scholar
  23. 23.
    Parikh J, Selmi M, Charles-Edwards G et al (2014) Changes in primary breast cancer heterogeneity may augment midtreatment MR imaging assessment of response to neoadjuvant chemotherapy. Radiology 272:100–112CrossRefPubMedGoogle Scholar
  24. 24.
    De Cecco CN, Ganeshan B, Ciolina M et al (2015) Texture analysis as imaging biomarker of tumoral response to neoadjuvant chemoradiotherapy in rectal cancer patients studied with 3-T magnetic resonance. Invest Radiol 50:239–245CrossRefPubMedGoogle Scholar
  25. 25.
    Breiman L (2001) Random Forests. Machine Learning 45:5–32CrossRefGoogle Scholar
  26. 26.
    Caruana R, Karampatziakis N, Yessenalina A (2008) An Empirical Evaluation of Supervised Learning in High Dimensions. ICML '08. The 25th International Conference on Machine Learning. ACM, Helsinki, Finland, pp 96-103Google Scholar
  27. 27.
    Raman SP, Schroeder JL, Huang P et al (2015) Preliminary data using computed tomography texture analysis for the classification of hypervascular liver lesions: generation of a predictive model on the basis of quantitative spatial frequency measurements--a work in progress. J Comput Assist Tomogr 39:383–395PubMedGoogle Scholar
  28. 28.
    Raman SP, Chen Y, Schroeder JL, Huang P, Fishman EK (2014) CT texture analysis of renal masses: pilot study using random forest classification for prediction of pathology. Acad Radiol 21:1587–1596CrossRefPubMedPubMedCentralGoogle Scholar
  29. 29.
    Hastie T, Tibshirani R, Friedman J (2009) The Elements of Statistical Learning Data Mining, Inference, and Prediction, Second Edition edn. Springer Science+Business Media, LLCGoogle Scholar
  30. 30.
    Liaw A, Wiener M (2002) Classification and Regression by random Forest. R News 2:18–22Google Scholar
  31. 31.
    Mileto A, Marin D (2017) Dual-Energy Computed Tomography in Genitourinary Imaging. Radiol Clin North Am 55:373–391CrossRefPubMedGoogle Scholar
  32. 32.
    Mallinson PI, Coupal TM, McLaughlin PD, Nicolaou S, Munk PL, Ouellette HA (2016) Dual-Energy CT for the Musculoskeletal System. Radiology 281:690–707CrossRefPubMedGoogle Scholar
  33. 33.
    Machida H, Tanaka I, Fukui R et al (2016) Dual-Energy Spectral CT: Various Clinical Vascular Applications. Radiographics 36:1215–1232CrossRefPubMedGoogle Scholar
  34. 34.
    Forghani R, Mukherji SK (2017) Advanced dual-energy CT applications for the evaluation of the soft tissues of the neck. Clin Radiol.
  35. 35.
    Forghani R, Roskies M, Liu X et al (2016) Dual-Energy CT Characteristics of Parathyroid Adenomas on 25-and 55-Second 4D-CT Acquisitions: Preliminary Experience. J Comput Assist Tomogr 40:806–814CrossRefPubMedGoogle Scholar

Copyright information

© European Society of Radiology 2017

Authors and Affiliations

  1. 1.Department of RadiologyJewish General HospitalMontrealCanada
  2. 2.Department of Diagnostic RadiologyMcGill UniversityMontrealCanada
  3. 3.Department of Radiology, Royal Victoria HospitalMcGill University Health CentreMontrealCanada
  4. 4.Segal Cancer Centre and Lady Davis Institute for Medical ResearchJewish General HospitalMontrealCanada
  5. 5.Department of Otolaryngology-Head and Neck SurgeryJewish General HospitalMontrealCanada

Personalised recommendations