ABSTRACTThe accurate non‐invasive detection and estimation of central aortic pressure waveforms (CAPW) are crucial for reliable treatments of cardiovascular system diseases. But the accuracy and practicality of current estimation methods need to be improved. Our study combines a meta‐learning neural network and a physics‐driven method to accurately estimate CAPW based on personalized physiological indicators. We collected data from 260 patients who underwent catheterization surgery, using measured CAPW and personalized physiological indicators (e.g., weight, body mass index (BMI), radial mean arterial pressure (MAP), heart rate (HR), cardiac output (CO), radial systolic blood pressure (SBP), and radial diastolic blood pressure (DBP)) as input for neural network training. The output of the neural network are the Gaussian characteristic parameters of the single‐period decomposed CAPW. The neural network model was constructed using the model‐agnostic meta‐learning (MAML) algorithm framework. Applying the physical characteristics of CAPW to the loss function, served to increase the constraints on the output and improve the accuracy of CAPW estimation. To verify the accuracy of the model, we compared measured and estimated CAPW in 52 patients. The results are consistent with a normalized root mean square error (NRMSE) of 0.0206. The predictions had low biases, namely SBP: 4.97 ± 4.42 mmHg, DBP: 4.78 ± 5.98 mmHg, and MAP: 0.35 ± 3.36 mmHg. The results demonstrate the accuracy and practicability of the approach to estimate CAPW. It can provide personalized parameters to calculate myocardial ischemia indicators (e.g., instantaneous wave‐free ratio [iFR] and fractional flow reserve [FFR]) and may contribute to the early monitoring and prevention of cardiovascular diseases.

The purpose of the current study was to investigate the effects of left/right coronary artery flow distribution on calculation of fractional flow reserve derived from coronary computed tomography angiography (FFRct) in different dominant types. First, 195 patients were collected to count the distribution ratios of the three categories: right dominance (RD), balanced dominance (BD), and left dominance (LD). Ratios of diameters of the left/right coronary arteries (D-LCA:D-RCA) of the three types were calculated and used to represent the ratio of flow distribution (Q(LCA):Q(RCA)) in the dominant type method. The other method was known as the fixed ratio method (Q(LCA):Q(RCA) = 6:4). Second, a total of 73 patients with coronary artery disease (CAD) were enrolled for numerical calculation. A 0D/3D geometric multiscale model was used for the numerical simulation of FFR and the results of the fixed ratio method and the dominant type method were recorded as F-FFRct and D-FFRct. Lastly, invasive FFR(clinic-FFR)was used as a standard to evaluate the consistency and diagnostic performance of F-FFRct and D-FFRct. Corresponding flow distributions for the dominant type method were Q(LCA):Q(RCA) = 5:5 for RD, Q(LCA):Q(RCA) = 5.5:4.5 for BD, and Q(LCA):Q(RCA) = 6:4 for LD. D-FFRct showed a better correlation than F-FFRct (r = 0.85 vs. r = 0.81, both p < .001); the AUC (95%CI) were 0.974 (0.906-0.997, p < .0001) and 0.960 (0.886-0.992, p < .0001). Accuracy, specificity, sensitivity, positive predictive value (PPV) and negative predictive values (NPV) for D-FFRct and F-FFRct were 94.52%, 93.75%, 94.74%, 83.33%, 98.18% and 90.41%, 87.50%, 91.23%, 73.68%, 96.30%, respectively. Overall, the left/right coronary artery flow distribution was affected by the dominant type and the dominant type method was superior to the fixed ratio method in detecting coronary ischemic lesions.

BACKGROUND AND OBJECTIVE:The functional assessment of the severity of coronary stenosis from coronary computed tomography angiography (CCTA)-derived fractional flow reserve (FFR) has recently attracted interest. However, existing algorithms run at high computational cost. Therefore, this study proposes a fast calculation method of FFR for the diagnosis of ischemia-causing coronary stenosis. METHODS:We combined CCTA and machine learning to develop a simplified single-vessel coronary model for rapid calculation of FFR. First, a zero-dimensional model of single-vessel coronary was established based on CCTA, and microcirculation resistance was determined through the relationship between coronary pressure and flow. In addition, a coronary stenosis model based on machine learning was introduced to determine stenosis resistance. Computational FFR (cFFR) was then obtained by combining the zero-dimensional model and the stenosis model with inlet boundary conditions for resting (cFFRr) and hyperemic (cFFRh) aortic pressure, respectively. We retrospectively analyzed 75 patients who underwent clinically invasive FFR (iFFR), and verified the model accuracy by comparison of cFFR with iFFR. RESULTS:The average computing time of cFFR was less than 2 s. The correlations between cFFRr and cFFRh with iFFR were r = 0.89 (p < 0.001) and r = 0.90 (p < 0.001), respectively. Diagnostic accuracy, sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, negative likelihood ratio for cFFRr and cFFRh were 90.7%, 95.0%, 89.1%, 76.0%, 98.0%, 8.7, 0.1 and 92.0%, 95.0%, 90.9%, 79.2%, 98.0%, 10.5, 0.1, respectively. CONCLUSIONS:The proposed model enables rapid prediction of cFFR and exhibits high diagnostic performance in selected patient cohorts. The model thus provides an accurate and time-efficient computational tool to detect ischemia-causing stenosis and assist with clinical decision-making.

Background and objectives A quick calculation approach of steady-state fractional flow reserve (FFRss) based on computed tomography angiography (CTA) images is a reliable non-invasive way of calculate FFR, the assumptions used in the research should be study further to increase forecast accuracy. The effect of inlet and outlet boundary conditions on FFRss was investigated. Methods 15 patients who had been diagnosed with coronary artery disease were enrolled in this study. We investigated the sensitivity of calculating FFR to boundary circumstances, using invasive FFR as a benchmark. There are two types of inlet: (1) aortic pressure based on clinically measured. (2) mean pressure calculated based on physiological formula; we further studied the outlet changes of FFRss under different coronary vasodilation responses (24%, 48%, 72%). Results According to the calculate FFR results of all patients, FFRSST (based on the clinical experiment) and FFRSSM (based on the physiological formula) {r = 0.99, [95% confidence interval (CI):0.0.94 to 1.14] (p < 0.001)}. Although the pressure difference between the two pressure boundary conditions is 15 mmHg, the calculated FFR result does not change significantly. The microcirculation resistance of the outlet gradually rose as the vasodilation state changed, and the computed FFR increased. Conclusions A numerical analysis of the effects of proximal and distal boundary constraints of computational models on computed CT-FFR is presented. The findings revealed that distal boundary circumstances (hyperemic vasodilation response of coronary micro-vessels) have a significant impact on FFR, providing evidence to guide the development and application of a computational model for estimating FFR.

This study aimed to examine whether the ratio of vessel-specific coronary arterial lumen volume to the fraction of myocardial mass (VR/MR) affects myocardial ischemia. We proposed a calculation method for VR/MR, and compared the ratio of total epicardial coronary arterial lumen volume to left ventricular myocardial mass (V/M) with VR/MR in predicting myocardial ischemia. VR/MR and V/M were computed using data from 205 patients with 241 stenosis vessel who underwent coronary computed tomography angiography (CTA), quantitative coronary angiography, and fractional flow reserve. The vessel-specific coronary arterial lumen volume (VR) was obtained from CTA by segmenting the coronary arterial lumen volume, while the vessel-specific fraction of myocardial mass (MR) was obtained by allometric scaling. The VR/MR was then calculated. The cut-off values of V/M (23.55 mm3/g) and VR/MR (12.98 mm3/g) were used to define equal groups of ischemic and non-ischemic patients, respectively. Using these cut-off values, the accuracy, specificity, sensitivity, positive predictive value, and negative predictive value of V/M were 60%, 76%, 45%, 57%, and 66%, and of VR/MR were 87%, 92%, 77%, 89%, and 83%, respectively. Patients have different VR/MR values in different stenotic coronary arteries. Clinically, VR/MR is a quantitative indicator of the risk of myocardial ischemia.

Background Fractional flow reserve (FFR) is considered to be the criterion standard for the clinical diagnosis of functional myocardial ischemia. In this study, we explored the effect of the coronary arterial diameter derived from coronary computed tomography angiography on FFR. Method We retrospectively reviewed the clinical information of 131 patients with moderate coronary artery stenosis. To compare the mean diameter of stenotic vessels, patients were divided into ischemic and nonischemic groups. According to the clinical statistics of the diameter of the ischemic group and the nonischemic group, we established 8 ideal models of coronary artery diameter of 4 mm (40%, 50%, 60%, and 70% stenosis) and diameter of 3 mm (40%, 50%, 60%, and 70% stenosis). Two sets of numerical simulation experiments were carried out: experiment 1 evaluated the variation rate of CT-based computation of non-invasive fractional flow reserve (FFRCT) with vessel diameters of 4 mm and 3 mm under different stenosis rates, and experiment 2 explored the variation of FFRCT with vessel diameters of 4 mm and 3 mm under different cardiac outputs. We simulated changes in the flow of narrow blood vessels by changes in cardiac output. Results According to clinical statistics, the mean ± SD diameter of stenotic vessels in the ischemic and nonischemic groups was 3.67 ± 0.77 mm and 3.31 ± 0.64 mm (P < 0.05 for difference), respectively. In experiment 1, the FFRCT of coronary with a diameter of 4 mm was 0.86, 0.80, 0.66, and 0.35, and that with a diameter of 3 mm was 0.90, 0.84, 0.71, and 0.50, respectively. In experiment 2, the FFRCT of the coronary vessel diameter of 4 mm was 0.84, 0.80, 0.76, and 0.72, respectively. The FFRCT coronary vessels with a diameter of 3 mm were 0.87, 0.84, 0.80, and 0.76, respectively. Conclusions As the stenosis increases, compared with narrow blood vessel of small diameter, the narrow blood vessel with larger diameter is accompanied by faster flow rate changes and is more prone to ischemia.

OBJECTIVE:To explore the influence of the blood flow-diameter scaling laws of $ \mathrm{Q}\mathrm{\alpha }{\mathrm{D}}^{3} $, $ \mathrm{Q}\mathrm{\alpha }{\mathrm{D}}^{2.7} $ and $ \text{Q}\alpha \text{D}{}^{7}\!\!\diagup\!\!{}_{3}\; $ on the numerical simulation of fraction flow reserve based on CTA images and to find the optimal exponents.METHODS:1) 26 patients with coronary artery disease were screened according to the inclusion criteria; 2) Microcirculation resistance (Rm) was calculated under the 3, 2.7 and 7/3 power of the flow-diameter scaling law, which were recorded as 3Rm, 2.7Rm and 7/3Rm, respectively; 3) 3Rm, 2.7Rm and 7/3Rm were used as exit boundary conditions to simulate FFRCT, quoted as 3FFRCT, 2.7FFRCT and 7/3FFRCT, respectively; 4) The correlation and diagnostic performance between three kinds of FFRCT and FFR were analyzed.RESULTS:The p-values of comparing 3Rm, 2.7Rm and 7/3Rm with FFR were 0.004, 0.005 and 0.010, respectively; the r value between 7/3FFRCT and FFR (0.96) was better than that of 3FFRCT (0.95) and 2.7FFRCT (0.95); the 95% LoA between 7/3FFRCT and FFR (-0.08~0.11) was smaller than that of 3FFRCT (-0.10~0.12) and 2.7FFRCT (-0.09~0.11); the AUC and accuracy of 7/3FFRCT [0.962 (0.805-0.999), 96.15%] were the same as those of 2.7FFRCT [0.962 (0.805-0.999), 96.15%] and better than those of 3FFRCT [0.944 (0.777-0.996), 92.3%]. The prediction threshold of 7/3FFRCT (0.791) was closer to 0.8 than that of 3FFRCT (0.816) and 2.7FFRCT (0.787).CONCLUSION:The blood flow-diameter scaling law affects the FFRCT simulation by influencing the exit boundary condition Rm of the calculation. With $ Q\alpha D{}^{7}\!\!\diagup\!\!{}_{3}\; $, FFRCT had the highest diagnostic performance. The blood flow-diameter scaling law provides theoretical support for the blood flow distribution in the bifurcated vessel and improves the FFRCT model.

Background and Objectives: With the development of medical imaging and computational fluid dynamics (CFD), a fast calculation method of steady-state fractional flow reserve (FFRss) based on CTA images has been applied to predict myocardial ischemia. Whilst this is a reliable non-invasive method of calculating FFR, assumptions involved in the analysis still need to be investigated for further improvement in predictive accuracy. In this study, we analyzed the influence of inlet and outlet boundary conditions on FFR SS. Methods: A clinical trial was carried out in Peking University People's Hospital. We enrolled 15 patients with coronary heart disease. All patients underwent coronary CTA examination, aortic pressure measurement and FFR catheter surgery. In order to better reflect the relationship between different entrance boundary conditions and FFRss, we used the invasive measurement of FFR as the standard, and calculated FFRss for 15 patients with coronary heart disease with different degrees of stenosis. The boundary conditions are divided into two groups: (1) pressure calculated based on physiological formula; (2) clinically measured aortic pressure. Based on the boundary conditions calculated by the physiological formula, we further studied the changes of FFRss under different coronary vasodilation responses (12%-48% baseline). Results: The research results show that although the pressure difference between the two pressure boundary conditions is 15mmHg, the FFRss calculation result does not change significantly. With the change of the vasodilation state, the microcirculation resistance of the exit boundary condition gradually increased, and the calculated FFR value increased. Conclusions: We found that changes in the microcirculation resistance of coronary stenoses have a huge impact on FFRss. The changes in FFRss values caused by boundary conditions are due to the overestimation of the vasodilation response. Therefore, individualizing the hyperemia state of the stenotic vessel microcirculation resistance value is an effective method to improve the calculation of FFR.