Response to Letter Regarding Article, “Prediction of Progression of Coronary Artery Disease and Clinical Outcomes Using Vascular Profiling of Endothelial Shear Stress and Arterial Plaque Characteristics: The PREDICTION Study”

Machine Learning Coronary Artery Disease Prediction Based on Imaging and Non-Imaging Data

The prediction of obstructive atherosclerotic disease has significant clinical meaning for the decision making. In this study, a machine learning predictive model based on gradient boosting classifier is presented, aiming to identify the patients of high CAD risk and those of low CAD risk. The machine learning methodology includes five steps: the preprocessing of the input data, the class imbalance handling applying the Easy Ensemble algorithm, the recursive feature elimination technique implementation, the implementation of gradient boosting classifier, and finally the model evaluation, while the fine tuning of the presented model was implemented through a randomized search optimization of the model’s hyper-parameters over an internal 3-fold cross-validation. In total, 187 participants with suspicion of CAD previously underwent CTCA during EVINCI and ARTreat clinical studies and were prospectively included to undergo follow-up CTCA. The predictive model was trained using imaging data (geometrical and blood flow based) and non-imaging data. The overall predictive accuracy of the model was 0.81, using both imaging and non-imaging data. The innovative aspect of the proposed study is the combination of imaging-based data with the typical CAD risk factors to provide an integrated CAD risk-predictive model.

Coronary Computer Tomography Angiography in 2021-Acquisition Protocols, Tips and Tricks and Heading beyond the Possible

Recent technological advances, together with an increasing body of evidence from randomized trials, have placed coronary computer tomography angiography (CCTA) in the center of the diagnostic workup of patients with coronary artery disease. The method was proven reliable in the diagnosis of relevant coronary artery stenosis. Furthermore, it can identify different stages of the atherosclerotic process, including early atherosclerotic changes of the coronary vessel wall, a quality not met by other non-invasive tests. In addition, newer computational software can measure the hemodynamic relevance (fractional flow reserve) of a certain stenosis. In addition, if required, information related to cardiac and valvular function can be provided with specific protocols. Importantly, recent trials have highlighted the prognostic relevance of CCTA in patients with coronary artery disease, which helped establishing CCTA as the first-line method for the diagnostic work-up of such patients in current guidelines. All this can be gathered in one relatively fast examination with minimal discomfort for the patient and, with newer machines, with very low radiation exposure. Herein, we provide an overview of the current technical aspects, indications, pitfalls, and new horizons with CCTA, providing examples from our own clinical practice.

The Holistic Coronary Physiology Display: Calculation of the Flow Separation Index in Vessel-Specific Individual Flow Range during Fractional Flow Reserve Measurement Using 3D Coronary Reconstruction

In order to make optimal decisions on the treatment of atherosclerotic coronary heart disease (CHD), appropriate evaluation is necessary, including both the anatomical and physiological assessment of the coronary arteries. According to current guidelines, a fractional flow reserve (FFR)-based clinical decision is recommended, but coronary flow reserve (CFR) measurements and microvascular evaluation should also be considered in special cases for a detailed exploration of the coronary disease state. We aimed to generate an extended physiological evaluation during routine FFR measurement and define a new pathological flow-related prognostic factor. Fluid dynamic equations were applied to calculate CFR on the basis of the three-dimensional (3D) reconstruction of the invasively acquired coronary angiogram and the measured intracoronary pressure data. A new, potentially robust prognostic parameter of a coronary lesion called the "flow separation index" (FS_i), which is thought to detect the pathological flow amount through a stenosis was introduced in a vessel-specific flow range. Correlations between FS_i and the clinically established physiological indices (CFR and FFR) were determined. The FS_i was calculated in 19 vessels of 16 patients, including data from the pre- and post-stent revascularization treatment of 3 patients. There was no significant correlation between the FS_i and the CFR (r = -0.23, p = 0.34); however, there was significant negative correlation between the FS_i and the FFR (r = -0.66, p = 0.002). An even stronger correlation was found between the FS_i and the ratio of the resting pressure ratio and the FFR (r = 0.92, p < 0.0001). The diagnostic power of the FS_i for predicting the FFR value of <0.80, as a gold standard prognostic factor, was tested by receiver operating characteristic analysis. FS_i > 0.022 proved to be the cutoff value of the prediction of a pathologically low FFR with a 0.856 area under the curve (95% confidence interval: 0.620 to 0.972). The present flow-pressure-velocity display provides a comprehensive summary of patient-specific pathophysiology in CHD. The consequences of epicardial stenoses can be evaluated together with their complex relations to microvascular conditions. Based on these values, clinical decision-making concerning both pharmacological therapy and percutaneous or surgical revascularization may be more precisely guided.