A machine-learning approach for computation of fractional flow reserve from coronary computed tomography.

Fractional flow reserve (FFR) is a functional index quantifying the severity of coronary artery lesions and is clinically obtained using an invasive, catheter-based measurement. Recently, physics-based models have shown great promise in being able to noninvasively estimate FFR from patient-specific anatomical information, e.g., obtained from computed tomography scans of the heart and the coronary arteries. However, these models have high computational demand, limiting their clinical adoption. In this paper, we present a machine-learning-based model for predicting FFR as an alternative to physics-based approaches. The model is trained on a large database of synthetically generated coronary anatomies, where the target values are computed using the physics-based model. The trained model predicts FFR at each point along the centerline of the coronary tree, and its performance was assessed by comparing the predictions against physics-based computations and against invasively measured FFR for 87 patients and 125 lesions in total. Correlation between machine-learning and physics-based predictions was excellent (0.9994, P < 0.001), and no systematic bias was found in Bland-Altman analysis: mean difference was -0.00081 ± 0.0039. Invasive FFR ≤ 0.80 was found in 38 lesions out of 125 and was predicted by the machine-learning algorithm with a sensitivity of 81.6%, a specificity of 83.9%, and an accuracy of 83.2%. The correlation was 0.729 (P < 0.001). Compared with the physics-based computation, average execution time was reduced by more than 80 times, leading to near real-time assessment of FFR. Average execution time went down from 196.3 ± 78.5 s for the CFD model to ∼2.4 ± 0.44 s for the machine-learning model on a workstation with 3.4-GHz Intel i7 8-core processor.

[1]  C D Murray,et al.  The Physiological Principle of Minimum Work: I. The Vascular System and the Cost of Blood Volume. , 1926, Proceedings of the National Academy of Sciences of the United States of America.

[2]  D. Levin Invasive Evaluation (Coronary Arteriography) of the Coronary Artery Disease Patient: Clinical, Economic and Social Issues , 1982, Circulation.

[3]  Levin Dc Invasive evaluation (coronary arteriography) of the coronary artery disease patient: clinical, economic and social issues , 1982 .

[4]  R. Wilson,et al.  Effects of adenosine on human coronary arterial circulation. , 1990, Circulation.

[5]  R Fumero,et al.  The coronary bed and its role in the cardiovascular system: a review and an introductory single-branch model. , 1992, Journal of biomedical engineering.

[6]  T. Wonnacott,et al.  Relation between diameter and flow in major branches of the arch of the aorta. , 1992, Journal of biomechanics.

[7]  P. H. van der Voort,et al.  Measurement of fractional flow reserve to assess the functional severity of coronary-artery stenoses. , 1996, The New England journal of medicine.

[8]  S Aharinejad,et al.  Morphometry of human coronary arterial trees , 1998, The Anatomical record.

[9]  G S Kassab,et al.  On the design of the coronary arterial tree: a generalization of Murray's law , 1999 .

[10]  M. Kern,et al.  Coronary physiology revisited : practical insights from the cardiac catheterization laboratory. , 2000, Circulation.

[11]  Maria Siebes,et al.  Hyperemic Stenosis Resistance Index for Evaluation of Functional Coronary Lesion Severity , 2002, Circulation.

[12]  T. Ryan,et al.  The Coronary Angiogram and Its Seminal Contributions to Cardiovascular Medicine Over Five Decades , 2002, Circulation.

[13]  D. Bessems,et al.  On the propagation of pressure and flow waves through the patient specific arterial system , 2003 .

[14]  M. Olufsen,et al.  Numerical Simulation and Experimental Validation of Blood Flow in Arteries with Structured-Tree Outflow Conditions , 2000, Annals of Biomedical Engineering.

[15]  G. Kassab Scaling laws of vascular trees: of form and function. , 2006, American journal of physiology. Heart and circulatory physiology.

[16]  Y. Fung,et al.  The pattern of coronary arteriolar bifurcations and the uniform shear hypothesis , 2006, Annals of Biomedical Engineering.

[17]  Radford M. Neal Pattern Recognition and Machine Learning , 2007, Technometrics.

[18]  Fractional flow reserve after previous myocardial infarction. , 2007, European heart journal.

[19]  Yoshua. Bengio,et al.  Learning Deep Architectures for AI , 2007, Found. Trends Mach. Learn..

[20]  Patrick W Serruys,et al.  Comprehensive assessment of coronary artery stenoses: computed tomography coronary angiography versus conventional coronary angiography and correlation with fractional flow reserve in patients with stable angina. , 2008, Journal of the American College of Cardiology.

[21]  M. D. Di Carli,et al.  Methods and Limitations of Assessing New Noninvasive Tests: Part II: Outcomes-Based Validation and Reliability Assessment of Noninvasive Testing , 2008, Circulation.

[22]  Ghassan S Kassab,et al.  Scaling of myocardial mass to flow and morphometry of coronary arteries. , 2008, Journal of applied physiology.

[23]  Dorin Comaniciu,et al.  Four-Chamber Heart Modeling and Automatic Segmentation for 3-D Cardiac CT Volumes Using Marginal Space Learning and Steerable Features , 2008, IEEE Transactions on Medical Imaging.

[24]  Yan Chen,et al.  Accurate, fast, and robust vessel contour segmentation of CTA using an adaptive self-learning edge model , 2009, Medical Imaging.

[25]  U. Siebert,et al.  Fractional flow reserve versus angiography for guiding percutaneous coronary intervention. , 2009, The New England journal of medicine.

[26]  Volkmar Falk,et al.  Guidelines on Myocardial Revascularization the Task Force on Myocardial Revascularization of the European Society of Cardiology (esc) and the European Association for Cardio-thoracic Surgery (eacts) Developed with the Special Contribution of the European Association for Percutaneous Cardiovascular I , 2022 .

[27]  Peter J. Hunter,et al.  Hierarchical Cluster-based Partial Least Squares Regression (HC-PLSR) is an efficient tool for metamodelling of nonlinear dynamic models , 2011, BMC Systems Biology.

[28]  A. Lansky,et al.  Novel QCA methodologies and angiographic scores , 2011, The International Journal of Cardiovascular Imaging.

[29]  A. Dunning,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. , 2011, Journal of the American College of Cardiology.

[30]  Woo-Young Chung,et al.  Optimal intravascular ultrasound criteria and their accuracy for defining the functional significance of intermediate coronary stenoses of different locations. , 2011, JACC. Cardiovascular interventions.

[31]  A. Alwan Global status report on noncommunicable diseases 2010. , 2011 .

[32]  Michael C. McDaniel,et al.  Coronary Artery Wall Shear Stress Is Associated With Progression and Transformation of Atherosclerotic Plaque and Arterial Remodeling in Patients With Coronary Artery Disease , 2011, Circulation.

[33]  Correlation between fractional flow reserve and intravascular ultrasound lumen area in intermediate coronary artery stenosis. , 2011, EuroIntervention : journal of EuroPCR in collaboration with the Working Group on Interventional Cardiology of the European Society of Cardiology.

[34]  Michael J Pencina,et al.  Diagnostic accuracy of fractional flow reserve from anatomic CT angiography. , 2012, JAMA.

[35]  P. Damman,et al.  Diagnostic Accuracy of Combined Intracoronary Pressure and Flow Velocity Information During Baseline Conditions: Adenosine-Free Assessment of Functional Coronary Lesion Severity , 2012, Circulation. Cardiovascular interventions.

[36]  Dorin Comaniciu,et al.  A framework for personalization of coronary flow computations during rest and hyperemia , 2012, 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[37]  Nikola Jagic,et al.  Fractional flow reserve-guided PCI versus medical therapy in stable coronary disease. , 2012, The New England journal of medicine.

[38]  C. Macaya,et al.  Morphometric assessment of coronary stenosis relevance with optical coherence tomography: a comparison with fractional flow reserve and intravascular ultrasound. , 2012, Journal of the American College of Cardiology.

[39]  Sankey V. Williams,et al.  2012 ACCF/AHA/ACP/AATS/PCNA/SCAI/STS Guideline for the diagnosis and management of patients with stable ischemic heart disease: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines, and the American College of Physicians, American Ass , 2012, Journal of the American College of Cardiology.

[40]  Helmut Baumgartner,et al.  ESC / EACTS Guidelines on myocardial revascularization , 2014 .

[41]  A. Hughes,et al.  Development and validation of a new adenosine-independent index of stenosis severity from coronary wave-intensity analysis: results of the ADVISE (ADenosine Vasodilator Independent Stenosis Evaluation) study. , 2012, Journal of the American College of Cardiology.

[42]  Dorin Comaniciu,et al.  Non-Invasive Hemodynamic Assessment of Aortic Coarctation: Validation with In Vivo Measurements , 2013, Annals of Biomedical Engineering.

[43]  Puneet Sharma,et al.  A patient-specific reduced-order model for coronary circulation , 2012, 2012 9th IEEE International Symposium on Biomedical Imaging (ISBI).

[44]  Charles A. Taylor,et al.  Computational fluid dynamics applied to cardiac computed tomography for noninvasive quantification of fractional flow reserve: scientific basis. , 2013, Journal of the American College of Cardiology.

[45]  Dorin Comaniciu,et al.  Data-Driven Reduction of a Cardiac Myofilament Model , 2013, FIMH.

[46]  A. Hughes,et al.  Hybrid iFR-FFR decision-making strategy: implications for enhancing universal adoption of physiology-guided coronary revascularisation. , 2013, EuroIntervention.

[47]  Patricia V Lawford,et al.  Virtual fractional flow reserve from coronary angiography: modeling the significance of coronary lesions: results from the VIRTU-1 (VIRTUal Fractional Flow Reserve From Coronary Angiography) study. , 2013, JACC. Cardiovascular interventions.

[48]  William Wijns,et al.  Fractional flow reserve calculation from 3-dimensional quantitative coronary angiography and TIMI frame count: a fast computer model to quantify the functional significance of moderately obstructed coronary arteries. , 2014, JACC. Cardiovascular interventions.

[49]  Stefan Baumann,et al.  Comparison of diagnostic value of a novel noninvasive coronary computed tomography angiography method versus standard coronary angiography for assessing fractional flow reserve. , 2014, The American journal of cardiology.

[50]  Dorin Comaniciu,et al.  Marginal Space Learning for Medical Image Analysis , 2014, Springer New York.

[51]  U. Schoepf,et al.  Coronary CT angiography-derived fractional flow reserve correlated with invasive fractional flow reserve measurements – initial experience with a novel physician-driven algorithm , 2015, European Radiology.

[52]  Hiroshi Ito,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). , 2014, Journal of the American College of Cardiology.

[53]  William Wijns,et al.  Evolving concepts of angiogram: fractional flow reserve discordances in 4000 coronary stenoses. , 2014, European heart journal.

[54]  Michail I. Papafaklis,et al.  Fast virtual functional assessment of intermediate coronary lesions using routine angiographic data and blood flow simulation in humans: comparison with pressure wire - fractional flow reserve. , 2014, EuroIntervention : journal of EuroPCR in collaboration with the Working Group on Interventional Cardiology of the European Society of Cardiology.

[55]  A. Kono,et al.  Fractional flow reserve computed from noninvasive CT angiography data: diagnostic performance of an on-site clinician-operated computational fluid dynamics algorithm. , 2015, Radiology.

[56]  Dorin Comaniciu,et al.  A parameter estimation framework for patient-specific hemodynamic computations , 2015, J. Comput. Phys..

[57]  Khodor Haidar Hassan,et al.  Morphometric Study of the Right Coronary Artery , 2015 .

[58]  U. Schoepf,et al.  Diagnostic value of quantitative stenosis predictors with coronary CT angiography compared to invasive fractional flow reserve. , 2015, European journal of radiology.

[59]  J. Reiber,et al.  Image-based assessment of fractional flow reserve. , 2015, EuroIntervention : journal of EuroPCR in collaboration with the Working Group on Interventional Cardiology of the European Society of Cardiology.

[60]  D. Comaniciu,et al.  Noninvasive hemodynamic assessment, treatment outcome prediction and follow-up of aortic coarctation from MR imaging. , 2015, Medical physics.

[61]  Ya-jie Liu,et al.  Diagnostic performance of noninvasive fractional flow reserve derived from coronary computed tomography angiography in coronary artery disease: A systematic review and meta-analysis. , 2015, International journal of cardiology.

[62]  Dorin Comaniciu,et al.  Comparison of Fractional Flow Reserve Based on Computational Fluid Dynamics Modeling Using Coronary Angiographic Vessel Morphology Versus Invasively Measured Fractional Flow Reserve. , 2016, The American journal of cardiology.

[63]  Witold Rużyłło,et al.  Workstation-Based Calculation of CTA-Based FFR for Intermediate Stenosis. , 2016, JACC. Cardiovascular imaging.

[64]  A. Persson,et al.  Software-based on-site estimation of fractional flow reserve using standard coronary CT angiography data , 2016, Acta radiologica.

[65]  A. Kono,et al.  Coronary CT angiography derived fractional flow reserve: Methodology and evaluation of a point of care algorithm. , 2016, Journal of cardiovascular computed tomography.