Computing pulsatile blood flow of coronary artery under incomplete boundary conditions
WenJun Pu,Yan Chen,Shuai Zhao,Tiantong Yu,Heqiang Lin,Haokao Gao,Songyun Xie,Xi Zhang,Bohui Zhang,Chengxiang Li,Kun Lian,Xinzhou Xie
DOI: https://doi.org/10.1016/j.medengphy.2024.104193
Abstract:Background: Accurate measurement of pulsatile blood flow in the coronary arteries enables coronary wave intensity analysis, which can serve as an indicator for assessing coronary artery physiology and myocardial viability. Computational fluid dynamics (CFD) methods integrating coronary angiography images and fractional flow reserve (FFR) offer a novel approach for computing mean coronary blood flow. However, previous methods neglect the inertial effect of blood flow, which may have great impact on pulsatile blood flow calculation. To improve the accuracy of pulsatile blood flow calculation, a novel CFD based method considering the inertia term is proposed. Methods: A flow resistance model based on Pressure-Flow vs.Time curves is proposed to model the resistance of the epicardial artery. The parameters of the flow resistance model can be fitted from the simulated pulsating flow rates and pressure drops of a specific mode. Then, pulsating blood flow can be calculated by combining the incomplete pressure boundary conditions under pulsating conditions which are easily obtained in clinic. Through simulation experiments, the effectiveness of the proposed method is validated in idealized and reconstructed 3D model of coronary artery. The impacts of key parameters for generating the simulated pulsating flow rates and pressure drops on the accuracy of pulsatile blood flow calculation are also investigated. Results: For the idealized model, the previously proposed Pressure-Flow model has a significant leading effect on the computed blood flow waveform in the moderate model, and this leading effect disappears with the increase of the degree of stenosis. The improved model proposed in this paper has no leading effect, the root mean square error (RMSE) of the proposed model is low (the left coronary mode:≤0.0160, the right coronary mode:≤0.0065) for all simulated models, and the RMSE decreases with an increase of stenosis. The RMSE is consistently small (≤0.0217) as the key parameters of the proposed method vary in a large range. It is verified in the reconstructed model that the proposed model significantly reduces the RMSE of patients with moderate stenosis (the Pressure-Flow model:≤0.0683, the Pressure-Flow vs.Time model:≤0.0297), and the obtained blood flow waveform has a higher coincidence with the simulated reference waveform. Conclusions: This paper confirms that ignoring the effect of inertia term can significantly affect the accuracy of calculating pulsatile blood flow in moderate stenosis lesions, and the new method proposed in this paper can significantly improves the accuracy of calculating pulsatile blood flow in moderate stenosis lesions. The proposed method provides a convenient clinical method for obtaining pressure-synchronized blood flow, which is expected to facilitate the application of waveform analysis in the diagnosis of coronary artery disease.