A patient-specific modelling method of blood circulatory system for the numerical simulation of enhanced external counterpulsation. (9th October 2020)

• Record Type: Journal Article
• Authors: Li, Bao; Wang, Hui; Li, Gaoyang; Liu, Jian; Zhang, Zhe; Gu, Kaiyun; Yang, Haisheng; Qiao, Aike; Du, Jianhang; Liu, Youjun
• Is Part Of: Journal of biomechanics. Volume 111(2020)
• Publication Date: 2020-10-09
• Subjects: Patient-specific modelling -- Blood circulatory system -- Enhanced external counterpulsation -- Optimization algorithm -- Lumped parameter model; Animal mechanics -- Periodicals; Biomechanics -- Periodicals; Biomechanics -- Periodicals; Mécanique animale -- Périodiques; Biomécanique -- Périodiques; Electronic journals; 571.4305
• DOI: 10.1016/j.jbiomech.2020.110002
• Languages: English
• ISSNs: 0021-9290

Abstract

Abstract: Lumped parameter model (LPM) is a common numerical model for hemodynamic simulation of human's blood circulatory system. The numerical simulation of enhanced external counterpulsation (EECP) is a typical biomechanical simulation process based on the LPM of blood circulatory system. In order to simulate patient-specific hemodynamic effects of EECP and develop best treatment strategy for each individual, this study developed an optimization algorithm to individualize LPM elements. Physiological data from 30 volunteers including approximate aortic pressure, cardiac output, ankle pressure and carotid artery flow were clinically collected as optimization objectives. A closed-loop LPM was established for the simulation of blood circulatory system. Aiming at clinical data, a sensitivity analysis for each element was conducted to identify the significant ones. We improved the traditional simulated annealing algorithm to iteratively optimize the sensitive elements. To verify the accuracy of the patient-specific model, 30 samples of simulated data were compared with clinical measurements. In addition, an EECP experiment was conducted on a volunteer to verify the applicability of the optimized model for the simulation of EECP. For these 30 samples, the optimization results show a slight difference between clinical data and simulated data. The average relative root mean square error is lower than 5%. For the subject of EECP experiment, the relative error of hemodynamic responses during EECP is lower than 10%. This slight error demonstrated a good state of optimization. The optimized modeling algorithm can effectively individualize the LPM for blood circulatory system, which is significant to the numerical simulation of patient-specific hemodynamics.