Coronary CT Angiography–derived Fractional Flow Reserve: Machine Learning Algorithm versus Computational Fluid Dynamics Modeling

Published Online:

The results demonstrate that coronary CT angiography–derived fractional flow reserve using machine learning performs equally to computational fluid dynamics modeling in detecting lesion-specific ischemia; both algorithms outperform coronary CT angiography alone and quantitative coronary angiography.


To compare two technical approaches for determination of coronary computed tomography (CT) angiography–derived fractional flow reserve (FFR)—FFR derived from coronary CT angiography based on computational fluid dynamics (hereafter, FFRCFD) and FFR derived from coronary CT angiography based on machine learning algorithm (hereafter, FFRML)—against coronary CT angiography and quantitative coronary angiography (QCA).

Materials and Methods

A total of 85 patients (mean age, 62 years ± 11 [standard deviation]; 62% men) who had undergone coronary CT angiography followed by invasive FFR were included in this single-center retrospective study. FFR values were derived on-site from coronary CT angiography data sets by using both FFRCFD and FFRML. The performance of both techniques for detecting lesion-specific ischemia was compared against visual stenosis grading at coronary CT angiography, QCA, and invasive FFR as the reference standard.


On a per-lesion and per-patient level, FFRML showed a sensitivity of 79% and 90% and a specificity of 94% and 95%, respectively, for detecting lesion-specific ischemia. Meanwhile, FFRCFD resulted in a sensitivity of 79% and 89% and a specificity of 93% and 93%, respectively, on a per-lesion and per-patient basis (P = .86 and P = .92). On a per-lesion level, the area under the receiver operating characteristics curve (AUC) of 0.89 for FFRML and 0.89 for FFRCFD showed significantly higher discriminatory power for detecting lesion-specific ischemia compared with that of coronary CT angiography (AUC, 0.61) and QCA (AUC, 0.69) (all P < .0001). Also, on a per-patient level, FFRML (AUC, 0.91) and FFRCFD (AUC, 0.91) performed significantly better than did coronary CT angiography (AUC, 0.65) and QCA (AUC, 0.68) (all P < .0001). Processing time for FFRML was significantly shorter compared with that of FFRCFD (40.5 minutes ± 6.3 vs 43.4 minutes ± 7.1; P = .042).


The FFRML algorithm performs equally in detecting lesion-specific ischemia when compared with the FFRCFD approach. Both methods outperform accuracy of coronary CT angiography and QCA in the detection of flow-limiting stenosis.

© RSNA, 2018


  • 1. Koo BK, Erglis A, Doh JH, et al. Diagnosis of ischemia-causing coronary stenoses by noninvasive fractional flow reserve computed from coronary computed tomographic angiograms. Results from the prospective multicenter DISCOVER-FLOW (Diagnosis of Ischemia-Causing Stenoses Obtained Via Noninvasive Fractional Flow Reserve) study. J Am Coll Cardiol 2011;58(19):1989–1997. Crossref, MedlineGoogle Scholar
  • 2. Nakazato R, Park HB, Berman DS, et al. Noninvasive fractional flow reserve derived from computed tomography angiography for coronary lesions of intermediate stenosis severity: results from the DeFACTO study. Circ Cardiovasc Imaging 2013;6(6):881–889. Crossref, MedlineGoogle Scholar
  • 3. Nørgaard BL, Leipsic J, Gaur S, et al. Diagnostic performance of noninvasive fractional flow reserve derived from coronary computed tomography angiography in suspected coronary artery disease: the NXT trial (Analysis of Coronary Blood Flow Using CT Angiography: Next Steps). J Am Coll Cardiol 2014;63(12):1145–1155. Crossref, MedlineGoogle Scholar
  • 4. Kim HJ, Vignon-Clementel IE, Coogan JS, Figueroa CA, Jansen KE, Taylor CA. Patient-specific modeling of blood flow and pressure in human coronary arteries. Ann Biomed Eng 2010;38(10):3195–3209. Crossref, MedlineGoogle Scholar
  • 5. Taylor CA, Figueroa CA. Patient-specific modeling of cardiovascular mechanics. Annu Rev Biomed Eng 2009;11(1):109–134. Crossref, MedlineGoogle Scholar
  • 6. Taylor CA, Fonte TA, Min JK. Computational fluid dynamics applied to cardiac computed tomography for noninvasive quantification of fractional flow reserve: scientific basis. J Am Coll Cardiol 2013;61(22):2233–2241. Crossref, MedlineGoogle Scholar
  • 7. Itu L, Sharma P, Kamen A, Suciu C, Comaniciu D. Graphics processing unit accelerated one-dimensional blood flow computation in the human arterial tree. Int J Numer Methods Biomed Eng 2013;29(12):1428–1455. Crossref, MedlineGoogle Scholar
  • 8. Sharma P, Itu L, Zheng X, et al. A framework for personalization of coronary flow computations during rest and hyperemia. Conf Proc IEEE Eng Med Biol Soc 2012;2012:6665–6668. MedlineGoogle Scholar
  • 9. Renker M, Schoepf UJ, Wang R, et al. Comparison of diagnostic value of a novel noninvasive coronary computed tomography angiography method versus standard coronary angiography for assessing fractional flow reserve. Am J Cardiol 2014;114(9):1303–1308. Crossref, MedlineGoogle Scholar
  • 10. Coenen A, Lubbers MM, Kurata A, et al. Fractional flow reserve computed from noninvasive CT angiography data: diagnostic performance of an on-site clinician-operated computational fluid dynamics algorithm. Radiology 2015;274(3):674–683. LinkGoogle Scholar
  • 11. Kruk M, Wardziak Ł, Demkow M, et al. Workstation-based calculation of CTA-based FFR for intermediate stenosis. JACC Cardiovasc Imaging 2016;9(6):690–699. Crossref, MedlineGoogle Scholar
  • 12. Baumann S, Wang R, Schoepf UJ, et al. Coronary CT angiography-derived fractional flow reserve correlated with invasive fractional flow reserve measurements—initial experience with a novel physician-driven algorithm. Eur Radiol 2015;25(4):1201–1207. Crossref, MedlineGoogle Scholar
  • 13. Yang DH, Kim YH, Roh JH, et al. Diagnostic performance of on-site CT-derived fractional flow reserve versus CT perfusion. Eur Heart J Cardiovasc Imaging 2017;18(4):432–440. Crossref, MedlineGoogle Scholar
  • 14. Itu L, Rapaka S, Passerini T, et al. A machine-learning approach for computation of fractional flow reserve from coronary computed tomography. J Appl Physiol (1985) 2016;121(1):42–52. Crossref, MedlineGoogle Scholar
  • 15. Benton SM Jr, Tesche C, De Cecco CN, Duguay TM, Schoepf UJ, Bayer RR 2nd. Noninvasive derivation of fractional flow reserve from coronary computed tomographic angiography: a review. J Thorac Imaging 2017 Aug 16. [Epub ahead of print] Google Scholar
  • 16. Tesche C, De Cecco CN, Albrecht MH, et al. Coronary CT angiography-derived fractional flow reserve. Radiology 2017;285(1):17–33. LinkGoogle Scholar
  • 17. Task Force Members, Montalescot G, Sechtem U, et al. 2013 ESC guidelines on the management of stable coronary artery disease: the Task Force on the management of stable coronary artery disease of the European Society of Cardiology. Eur Heart J 2013;34(38):2949–3003. Crossref, MedlineGoogle Scholar
  • 18. Leipsic J, Abbara S, Achenbach S, et al. SCCT guidelines for the interpretation and reporting of coronary CT angiography: a report of the Society of Cardiovascular Computed Tomography Guidelines Committee. J Cardiovasc Comput Tomogr 2014;8(5):342–358. Crossref, MedlineGoogle Scholar
  • 19. Mantero S, Pietrabissa R, Fumero R. The coronary bed and its role in the cardiovascular system: a review and an introductory single-branch model. J Biomed Eng 1992;14(2):109–116. Crossref, MedlineGoogle Scholar
  • 20. Itu L, Sharma P, Kamen A, Suciu C, Comaniciu D. A novel coupling algorithm for computing blood flow in viscoelastic arterial models. Conf Proc IEEE Eng Med Biol Soc 2013;2013:727–730. MedlineGoogle Scholar
  • 21. Scanlon PJ, Faxon DP, Audet AM, et al. ACC/AHA guidelines for coronary angiography: executive summary and recommendations: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on Coronary Angiography) developed in collaboration with the Society for Cardiac Angiography and Interventions. Circulation 1999;99(17):2345–2357. Crossref, MedlineGoogle Scholar
  • 22. Tonino PA, De Bruyne B, Pijls NH, et al. Fractional flow reserve versus angiography for guiding percutaneous coronary intervention. N Engl J Med 2009;360(3):213–224. Crossref, MedlineGoogle Scholar
  • 23. DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 1988;44(3):837–845. Crossref, MedlineGoogle Scholar
  • 24. Min JK, Leipsic J, Pencina MJ, et al. Diagnostic accuracy of fractional flow reserve from anatomic CT angiography. JAMA 2012;308(12):1237–1245. Crossref, MedlineGoogle Scholar
  • 25. Ghekiere O, Dewilde W, Bellekens M, et al. Diagnostic performance of quantitative coronary computed tomography angiography and quantitative coronary angiography to predict hemodynamic significance of intermediate-grade stenoses. Int J Cardiovasc Imaging 2015;31(8):1651–1661. Crossref, MedlineGoogle Scholar
  • 26. Voros S, Rinehart S, Vazquez-Figueroa JG, et al. Prospective, head-to-head comparison of quantitative coronary angiography, quantitative computed tomography angiography, and intravascular ultrasound for the prediction of hemodynamic significance in intermediate and severe lesions, using fractional flow reserve as reference standard (from the ATLANTA I and II Study). Am J Cardiol 2014;113(1):23–29. Crossref, MedlineGoogle Scholar

Article History

Received June 6, 2017; revision requested August 23; final revision received January 9, 2018; accepted January 12.
Published online: Apr 10 2018
Published in print: July 2018