计算模型指导临床房颤消融的进展现状

Current progress of computational modeling for guiding clinical atrial fibrillation ablation 计算模型指导临床房颤消融的进展现状 1 first-author https://orcid.org/0000-0003-3867-6519 2 2 2 2 1 corresp Ling XIA, xialing@zju.edu.cn Ling XIA, xialing@zju.edu.cn xialing@zju.edu.cn https://orcid.org/0000-0002-1937-9693 College of Biomedical Engineering & Instrument Science, Zhejiang University, Hangzhou310027, China College of Biomedical Engineering & Instrument Science, Zhejiang University, Hangzhou310027, China School of Biomedical Engineering, Dalian University of Technology, Dalian116024, China School of Biomedical Engineering, Dalian University of Technology, Dalian116024, China Atrial fibrillation (AF) is one of the most common arrhythmias, associated with high morbidity, mortality, and healthcare costs, and it places a significant burden on both individuals and society. Anti-arrhythmic drugs are the most commonly used strategy for treating AF. However, drug therapy faces challenges because of its limited efficacy and potential side effects. Catheter ablation is widely used as an alternative treatment for AF. Nevertheless, because the mechanism of AF is not fully understood, the recurrence rate after ablation remains high. In addition, the outcomes of ablation can vary significantly between medical institutions and patients, especially for persistent AF. Therefore, the issue of which ablation strategy is optimal is still far from settled. Computational modeling has the advantages of repeatable operation, low cost, freedom from risk, and complete control, and is a useful tool for not only predicting the results of different ablation strategies on the same model but also finding optimal personalized ablation targets for clinical reference and even guidance. This review summarizes three-dimensional computational modeling simulations of catheter ablation for AF, from the early-stage attempts such as Maze III or circumferential pulmonary vein isolation to the latest advances based on personalized substrate-guided ablation. Finally, we summarize current developments and challenges and provide our perspectives and suggestions for future directions. 房颤（atrial fibrillation，AF）是最常见的心律失常之一，临床危害极大。虽然抗心律失常药物是治疗房颤最常用的策略，但是药物治疗因其有限的疗效和潜在的副作用而面临挑战。导管消融术作为一种替代方案被广泛用于治疗房颤患者。然而，由于房颤的机制还不完全清楚，消融术后的复发率仍然很高，尤其是持续性房颤。因此，何种消融策略是最优的一直是临床亟待解决的问题。计算模型具有可重复操作、低成本、低风险和灵活可控等优点，不仅可以在同一模型上预测不同消融策略的结果，还可以找到最佳的个性化消融靶点供临床参考。本综述总结了三维计算模型仿真房颤导管消融策略的进展，从早期的迷宫术或环肺静脉隔离术到基于特异性底物的个性化消融方式。最后，我们总结了目前的发展和面临的挑战，并对未来的发展方向提出了看法和建议。 Atrial fibrillation Catheter ablation Computational modeling Atrial fibrosis 房颤 导管消融 计算模型 心房纤维化 1 Introduction 1 Atrial fibrillation (AF) is one of the most common cardiac arrhythmias worldwide ( Rahman et al., 2014 Rahman et al., 2014 ; Patel et al., 2018 Patel et al., 2018 ), characterized by very chaotic and irregular propagation throughout the atria, leading to cardiac acceleration and the loss of mechanical contraction ( Heijman et al., 2016 Heijman et al., 2016 ; Nattel et al., 2020 Nattel et al., 2020 ). Approximately 1%–2% of the world's population suffers from AF, a percentage which is expected to double by 2050 ( Nishida and Nattel, 2014 Nishida and Nattel, 2014 ). In China, the prevalence of AF is 0.77%–1.03%, and there were 5.26 million patients over the age of 18 years in 2013; the prevalence also is increasing year by year ( Li et al., 2013 Li et al., 2013 ). Even more worrying, AF increases the risks of other serious diseases, mainly stroke and heart failure, which are associated with high morbidity and mortality ( Benjamin et al., 2017 Benjamin et al., 2017 ; Ahmed et al., 2019 Ahmed et al., 2019 ), high treatment costs ( Ha et al., 2017 Ha et al., 2017 ), and reduced quality of life ( Sławuta et al., 2020 Sławuta et al., 2020 ). Thus, the management of AF is critical and urgent for individuals and society. In general, treatment of AF is often attempted with anti-arrhythmic agents, electrical cardioversion, and maze/catheter ablation strategies. Because of limited efficacy and significant side effects in some cases ( Miller et al., 2019 Miller et al., 2019 ), treating AF with anti-arrhythmic agents is not the optimal choice. Ablation therapy (catheter ablation) is an alternative option, especially for AF patients who do not tolerate anti-arrhythmic agents ( Bhatti et al., 2019 Bhatti et al., 2019 ), and in terms of therapeutic effect, it seems to be more effective than anti-arrhythmic agents as well ( Hakalahti et al., 2015 Hakalahti et al., 2015 ). The principle of catheter ablation therapy is creating lesions with a catheter using radiofrequency energy or cryoenergy. The lesion forms a barrier to prevent the conduction of electrical activity, halting fibrillation ( Verma et al., 2015 Verma et al., 2015 ). Aggressive ablation pattern such as Cox's Maze III procedure ( Cox et al., 1991 Cox et al., 1991 ) has been found to have a high success rate in treating AF. This approach involves a complicated set of incisions made in a maze-like pattern on the left and right atria to block the abnormal impulses that cause AF. However, ablation requires open-heart surgery, which is time-consuming and may cause complications ( Weimar et al., 2012 Weimar et al., 2012 ). Pulmonary vein isolation (PVI) ( Haïssaguerre et al., 1998 Haïssaguerre et al., 1998 ) represents the cornerstone of current catheter ablation approaches for the treatment of AF. This approach achieves the goal of electrically isolating AF triggers in pulmonary veins (PVs) through circumferential ablation lesions in the PVs, thus successfully terminating AF, especially in patients with paroxysmal AF. However, the success rate of this strategy in patients with persistent AF (PeAF) remains limited as the triggers for PeAF may originate outside of PVs ( Dewire and Calkins, 2013 Dewire and Calkins, 2013 ; Conti et al., 2018 Conti et al., 2018 ; Pallisgaard et al., 2018 Pallisgaard et al., 2018 ). Although several different extra-PVI ablation approaches have been developed and attempted to increase the success rate of clinical procedures, whether patients with AF can benefit from these additional ablation strategies remains uncertain ( Latchamsetty and Morady, 2018 Latchamsetty and Morady, 2018 ). There is no one-size-fits-all ablation approach. Hence, the most efficient and personalized ablation strategies with the fewest ablation lesions need to be pursued. Virtual ablation with computational modeling has some unique advantages in comparison with clinical catheter ablation. For example, (1) different ablation strategies can be tested in the same biophysical model; (2) the effectiveness of any ablation strategies can be tested; (3) modeling parameters are flexibly controlled; (4) computational modeling costs less; and (5) lastly and most importantly, there is no risk to human life. Based on these advantages, computational modeling can provide additional information about the mechanisms of AF initiation and maintenance, and help to determine a personalized optimal ablation strategy. Over the past several decades, research groups worldwide have made many attempts with significant progress. This review provides a detailed description of virtual ablation for AF, from the initial attempt to the latest development. 2 Development of the computational modeling for atrial fibrillation ablation 1 2.1 Early development 2 Early-stage three-dimensional (3D) computational modeling for AF ablation is briefly summarized in Table 1 Table 1 . Dang et al. (2005) Dang et al. (2005) created the first life-size 3D biophysical model for evaluating ablation patterns and comparing ablation outcomes in silico with clinical data. They integrated a magnetic resonance imaging (MRI)-derived monolayer bi-atrial model with the homogeneous modified Luo-Rudy ionic model ( Luo and Rudy, 1991 Luo and Rudy, 1991 ). The monodomain reaction-diffusion function was solved by the finite volume approach to simulate wave propagation in cardiac tissue ( Virag et al., 2002 Virag et al., 2002 ). The gold standard Cox's Maze III and 12 other less invasive ablation patterns were simulated in this biophysical AF model. The results confirmed an excellent success rate for Maze III and illustrated that both right and left atrial ablation lesions should be included in less invasive ablation patterns. The work in this paper indicated that computational modeling could be useful for the development of new ablation patterns. 3D: three-dimensional; MRI: magnetic resonance imaging; FVM: finite volume method; FEM: finite element method; FDM: finite difference method; PVI: pulmonary vein isolation. Reference Image source Geometry Atrial myocyte model Fiber orientation Numerical solution Ablation pattern Dang et al., 2005 Dang et al., 2005 MRI Monolayer, bi-atrial Modified Luo-Rudy model No FVM Maze, modified maze Ruchat et al., 2007a, 2007b, 2007c MRI Monolayer, bi-atrial Modified Luo-Rudy model No FVM Maze, modified maze Rotter et al., 2007 Rotter et al., 2007 MRI Monolayer, dilated, bi-atrial Modified Luo-Rudy model No FVM PVI, linear ablation Reumann et al., 2008 Reumann et al., 2008 Visible female Multilayer, bi-atrial Cellular automaton Visible female data FEM PVI, linear ablation maze Gong et al., 2015 Gong et al., 2015 Heart specimen Bilayer, bi-atrial Modified Courtemanche model Image-based FDM Maze, modified maze Ruchat et al. (2007a Ruchat et al. (2007a , 2007b 2007b , 2007c 2007c ) used one MRI-derived atrial model to test the effectiveness of different AF ablation line patterns and defined more efficient ablation patterns for the modified maze. The results suggested that Maze III was the most efficient ablation pattern and that the mini-maze ablation pattern was an effective strategy for converting fibrillation to sinus rhythm. The authors also found that the more ablation lines there were, the more effectively AF was terminated, and a combination of ablation lines in the left and right atria enhanced the success rate. Notably, potential complications after AF ablation often led to atrial flutter ( Steinbeck et al., 2018 Steinbeck et al., 2018 ). Rotter et al. (2007) Rotter et al. (2007) used one MRI-derived atrial model to simulate five different ablation strategies and investigated their effects in AF termination and their propensity to form atrial flutter substrates. Furthermore, they expanded the same atrial geometry by 1.26 times to mimic the AF-introduced characteristic of dilated atria ( Nattel and Harada, 2014 Nattel and Harada, 2014 ). They demonstrated that PVI in combination with extra linear lesions had a higher success rate than PVI alone. There were a higher atrial flutter incidence and a decreased incidence of AF termination in the dilated atrial model compared with the nondilated model with four of the five ablation patterns. Another computational model for AF ablation was published by Reumann et al. (2008) Reumann et al. (2008) . They constructed a realistic multilayered bi-atrial anatomical model, including atrial thickness and detailed anatomical structures, based on the visible female dataset ( Seemann et al., 2006 Seemann et al., 2006 ) and combined it with an advanced cellular automaton model ( Reumann et al., 2007 Reumann et al., 2007 ) and isotropic conduction. AF was induced at 35 different locations on the left atrial surface via an s1/s2 stimulus induce-protocol ( Haïssaguerre et al., 1998 Haïssaguerre et al., 1998 ). Ten different ablation patterns were simulated to test the optimal patterns to terminate AF. The results of the simulation found that the ablation pattern with a single circumferential lesion surrounding all PVs displayed a lower success rate in all simulation tests ( Reumann et al., 2008 Reumann et al., 2008 ). By contrast, the success rate reached its maximum with the combination of a circumferential ablation pattern and at least two additional linear ablation lesions. However, it may be worthwhile to note that one circumferential lesion line around all PVs, rather than lines around each individual PV, is used in current clinical practice at most institutions to avoid complications and thus achieve a higher success rate of PVI ( Woods and Olgin, 2014 Woods and Olgin, 2014 ). In 2015, Xia's group used one detailed 3D AF model to study the efficacies of different ablation patterns ( Gong et al., 2015 Gong et al., 2015 ). The cardiac anatomical model was reconstructed from a healthy adult heart with a detailed conduction system and realistic fiber orientation ( Deng et al., 2012 Deng et al., 2012 ). The focus pacing protocol was used to initiate AF ( Gong et al., 2007 Gong et al., 2007 ). Eight different ablation patterns based on the standard Maze III were simulated and evaluated in terms of the time required for AF termination. The authors discovered that the ablation lines on the right atrium were connected to the superior and inferior vena cava, and one modified Maze III pattern with fewer ablation lines was as effective as the standard Maze III pattern. In general, because of the development of more realistic atrial models, these early studies mentioned above used a computational simulation model of the human atria to evaluate different ablation patterns and compared the results with both the standard Maze III procedure (as a reference) and clinical outcomes. These computational models demonstrated that modeling could effectively address specific clinical questions regarding the efficacies of AF ablation strategies. It should be noted that in the early work, most ablation strategies were non-personalized, and were basically developed by making small changes from Maze III or circum-PVI. At this point, more accurate heterogeneous electrical properties and atrial structural remodeling representation, especially fibrosis, have not been integrated with computational modeling. At the same time, most of the clinical questions that need to be addressed involved patient-specific ablation targets for patient treatment. Hence, non-personalized models and ablation strategies do not meet the essential requirements for clinical practice. 2.2 Connection between atrial fibrillation and fibrosis 2 In recent years, patient-specific AF models have been developed by more realistic representations of the atrial substrate, mainly because electroanatomic remodeling of the atria has been shown to be conducive to the progressive nature of the generation and maintenance of AF ( Nattel and Harada, 2014 Nattel and Harada, 2014 ). Moreover, structural remodeling, especially atrial fibrosis, plays a crucial role in this process and has been clinically demonstrated to be associated with AF ( Seitz et al., 2011 Seitz et al., 2011 ; Morgan et al., 2016 Morgan et al., 2016 ). Specifically, the interaction between fibroblasts and myocytes changes the conduction velocity (CV) and action potential duration (APD), both of which slow localized regional conduction, increase conduction heterogeneity, and provide an AF substrate, thus promoting the formation of reentry ( Nguyen et al., 2014 Nguyen et al., 2014 ; Cochet et al., 2018 Cochet et al., 2018 ). Hence, information on the degree and extent of fibrosis in patients with AF is extremely useful in obtaining the personalized atrial arrhythmogenic substrate to guide clinical catheter ablation. Voltage mapping during catheter ablation for AF is commonly used in clinical practice for the assessment of atrial fibrosis ( Rolf et al., 2014 Rolf et al., 2014 ). However, it has a number of drawbacks, such as weak correlation with the fibrotic substrate, lack of standardized and reproducible protocols, and the invasive nature. In contrast, advanced late gadolinium-enhancement-MRI (LGE-MRI) is a novel tool, which has been shown recently to be less invasive and more reproducible for assessing fibrotic substrate. The approach is to localize and quantify the substrate by providing 3D visualization of the remodeled atria ( Pontecorboli et al., 2017 Pontecorboli et al., 2017 ; Fochler et al., 2019 Fochler et al., 2019 ; Mărgulescu et al., 2019 Mărgulescu et al., 2019 ; Sim et al., 2019 Sim et al., 2019 ). These developments in image processing and electrical signal processing techniques contribute to mapping the patient-specific arrhythmogenic substrate of AF and identifying optimal ablation targets ( Nademanee et al., 2004 Nademanee et al., 2004 ; Sanders et al., 2005 Sanders et al., 2005 ; Takahashi et al., 2008 Takahashi et al., 2008 ; Ganesan et al., 2013 Ganesan et al., 2013 ; Seitz et al., 2017 Seitz et al., 2017 ), permitting developments of custom-tailored ablation strategies. A workflow for a patient-specific model to predict the optimal AF ablation strategy is shown ( Fig. 1 Fig. 1 ). Table 2 Table 2 provides an overview of the published patient-specific models of AF and their properties. The details of the progress in this area are discussed in the following sections. AF: atrial fibrillation; n : number of patients; NS: numerical solution; CT: computed tomography; MRI: magnetic resonance imaging; CRN: Courtemanche-Ramirez-Nattel; FK: Fenton-Karma; DTMRI: diffusion tensor magnetic resonance imaging; FEM: finite element method; FDM: finite difference method; PVI: pulmonary vein isolation; CFAE: complex fractionated atrial electrogram; PS: phase singularity; ShEn: Shannon entropy; DF: dominant frequency; IAC: interatrial connection; RD: reentrant driver; AS: activation sequence; LAAI: left atrial appendage isolation. Reference n Image source Geometry Cell model Fibrosis distribution Fiber orientation NS Ablation pattern Guide for clinical practice Hwang et al., 2014 Hwang et al., 2014 20 CT Monolayer, left atrial Modified, CRN model No No FEM PVI, linear ablation, CFAE No Hwang et al., 2016 Hwang et al., 2016 1 CT Monolayer, left atrial Modified, CRN model No No FEM PS-based, ShEn-base,DF-based, CFAE No Lim et al., 2017 Lim et al., 2017 10 CT Monolayer, left atrial Modified CRN model No No FEM DF-based, PS-based No Shim et al., 2017 Shim et al., 2017 108 CT Monolayer, left atrial Modified CRN model No No FEM PVI, linear ablation, CFAE Yes Kim et al., 2019 Kim et al., 2019 118 CT Monolayer, left atrial Modified CRN model No No FDM PVI, linear ablation, CFAE Yes Lim et al., 2020b Lim et al., 2020b 27 CT Monolayer, left atrial Modified CRN model Clinical voltagemap-based Atlas-based FDM PS-based, DF-based Yes Lim et al., 2020a Lim et al., 2020a 10 CT Monolayer, bi-atrial Modified CRN model Clinical voltagemap-based Atlas-based FDM PVI, IAC-based No McDowell et al., 2015 McDowell et al., 2015 4 MRI Bilayer, left atrial Krummen model+fibroblasts Thresold-based Atlas-based FEM PS-based No Zahid et al., 2016 Zahid et al., 2016 10 MRI Bilayer, bi-atrial Modified CRN model Thresold-based Atlas-based FEM Graph-based No Deng et al., 2017 Deng et al., 2017 12 MRI Bilayer, bi-atrial Modified CRN model Thresold-based Atlas-based FEM RD-based No Hakim et al., 2018 Hakim et al., 2018 12 MRI Bilayer, bi-atrial Modified CRN model Thresold-based Atlas-based FEM RD-based No Boyle et al., 2018b Boyle et al., 2018b 11 MRI Bilayer, bi-atrial Modified CRN model Thresold-based Atlas-based FEM RD-based No Boyle et al., 2018a Boyle et al., 2018a 12 MRI Bilayer, bi-atrial Modified CRN model Thresold-based Atlas-based FEM RD-based No Boyle et al., 2019 Boyle et al., 2019 10 MRI Bilayer, bi-atrial Modified CRN model Thresold-based Atlas-based FEM RD-based Yes Ali et al., 2019 Ali et al., 2019 12 MRI Bilayer, bi-atrial Modified CRN model Thresold-based Atlas-based FEM PVI, RD-based No Shade et al., 2020 Shade et al., 2020 32 MRI Bilayer, bi-atrial Modified CRN model Thresold-based Atlas-based FEM RD-based No Bayer et al., 2016 Bayer et al., 2016 1 CT Bilayer, bi-atrial Modified CRN model Possibility method Rule-based FEM PVI, linear ablation,AS-based, PS-based No Roney et al., 2018 Roney et al., 2018 12 MRI Bilayer, bi-atrial Modified CRN model Possibility method, thresold-based Atlas-based FEM PVI, IAC-based No Roney et al., 2020 Roney et al., 2020 50 MRI Bilayer, left atrial Modified CRN model Thresold-based DTMRI-based FEM PVI, linear ablation,PS-based No Alessandrini et al., 2018 Alessandrini et al., 2018 1 MRI Bilayer, left atrial Modified CRN model No Rule-based FEM Basket Cathter-based No Roy et al., 2020 Roy et al., 2020 6 MRI Bilayer, left atrial Modified FK model Thresold-based No FDM PVI, linear ablation,RD-based No Gharaviri et al., 2021 Gharaviri et al., 2021 1 MRI Bilayer, bi-atrial Modified CRN model Possibility method DTMRI-based FDM PVI, linear ablation, LAAI No 2.3 Optimal atrial fibrillation ablation strategies using personalized computational models 2 PaK's group used virtual in silico modeling to study the efficacy of different ablation methods and to guide catheter ablation in the last decade. Specifically, Hwang et al. (2014) Hwang et al. (2014) used the computational platform devolved by their group to perform patient-specific virtual AF ablation and validate the outcomes of in silico AF ablation with the empirically chosen clinical ablations. In their study, 20 left atrial models were created from 20 patients' clinical spiral computed tomography (CT) merged with 3D electroanatomical map data. Five different virtual ablation protocols (using PVI, PVI with three different additional ablation lines, and complex fraction atrial electrogram (CFAE)) were applied to the models to test their efficacies. They found that circumferential pulmonary vein isolation (CPVI) with additional ablation lines was the most effective ablation pattern to terminate AF in the left atrium. Moreover, the CFAE-guided approach did not yield better success rates, which was in agreement with the clinical findings ( Kim et al., 2017 Kim et al., 2017 ). Hwang et al. (2016) Hwang et al. (2016) tested the efficacies of four different rotor ablation methods, including phase singularity (PS), dominant frequency (DF), Shannon entropy, and CFAE-cycle length (CFAE-CL). The results showed that virtual ablation of the DF area terminated AF, but not ablation of the areas detected by the PS, Shannon entropy, or CFAE-CL method. Another study by Lim et al. (2017) Lim et al. (2017) demonstrated that the outcomes of the DF-based ablation strategy were affected by CV, and the DF ablation pattern could not terminate AF under persistent AF conditions. Shim et al. (2017) Shim et al. (2017) conducted a randomized prospective study to evaluate the feasibility of computational modeling-guided ablation. In their study, 108 patients with PeAF were randomly assigned to the in silico ( n =53) and empirical ablation groups ( n =55). In the in silico ablation group, five different ablation patterns (using PVI, three additional linear ablations, and CPVI+CFAE ablation) were tested in each CT-derived left atrial model, and the most rapid pattern for terminating AF was achieved in clinical practice. They concluded that in silico ablation was feasible under clinical time constraints. Meanwhile, comparing recurrence rates for patients after follow-up, the outcome of the computational simulation ablation was not inferior to that of real clinical ablation procedures. Kim et al. (2019) Kim et al. (2019) updated the outcome of the clinical feasibility study on computational modeling-guided AF ablation. During the mean 31.5 months follow-up, the clinical recurrence rate after computational modeling-guided ablation was significantly lower than that after empirical ablation (20.8% vs. 40.0%). Meanwhile, there was a better long-term rhythm outcome of PeAF ablation in modeling-guided ablation. However, we would like to note that CPVI was much more commonly chosen in the empirical group (30.9%) than in the modeling-guided ablation group (1.0%). In contrast, extensive ablation strategies were used more widely in the modeling-guided ablation group, which may partially explain the better long-term outcomes. Lim et al. (2020b) Lim et al. (2020b) developed an improved simulation platform reflecting anatomy, fibrosis, fiber orientation, and electrophysiology using personal map data obtained during AF ablation procedures. They tested the accuracy of the improved simulation platform retrospectively in 17 patients and also tested its feasibility prospectively in ten patients during clinical procedures. The authors found that their improved platform could be applied in clinical AF ablation procedures after further validation. Based on this improved platform, Lim et al. (2020a) Lim et al. (2020a) reporting their tests of the impact of ablating interatrial connections (IACs) after CPVI. They discovered that the extra interatrial ablation improved the success rate in terminating AF, which was especially apparent with a cavotricuspid isthmus ablation. Trayanova's group performed the most comprehensive computational modeling-guided AF ablation. McDowell et al. (2015) McDowell et al. (2015) were the first to combine patient-specific atrial structure and fibrosis distribution as quantified from clinical LGE-MRI in a proof-of-concept study that addressed the following challenges: (1) the association between fibrosis distribution and AF and (2) the detection of optimal ablation targets. The authors demonstrated that AF could not be induced when the ablation area encompassed a meander of persistent reentrant drivers (RDs) and that patient-specific fibrosis distribution was crucial in initiating and maintaining AF and detecting the optimal ablation targets. Based on the methodology described by McDowell et al. (2015) McDowell et al. (2015) , Zahid et al. (2016) Zahid et al. (2016) developed an automatic approach using the principles of graph theory to predict optimal left atrial flutter ablation targets. The novel algorithm was named the "minimal cut algorithm," and was designed to locate the smallest amount of tissue separating the left atrium in two disconnected components. Boyle et al. (2018a Boyle et al. (2018a , 2018b) 2018b) published two articles to validate the accuracies of RDs predicted from computational modeling and those observed by electrocardiographic imaging (ECGI) ( Haissaguerre et al., 2016 Haissaguerre et al., 2016 ) and focal impulse and rotor mapping (FIRM) ( Narayan et al., 2012 Narayan et al., 2012 ), respectively. They found that simulations revealed additional areas harboring RDs that can never be identified with FIRM (termed "latent RDs"). Long-term outcomes were better in patients in which there were fewer latent RD sites, suggesting that latent RD sites may be a potential explanation for failed FIRM-guided ablation. Furthermore, more exhaustive PeAF substrate ablation targets should be identified based on computational modeling compared with FIRM-guided ablation. ECGI-guided ablation of targets in regions harboring RDs in both ECGI and simulations may lead to much better results when compared with ablation of non-overlapping RDs from ECGI targets. Thus, a combination of modeling and ECGI may improve the success rate in terminating AF. As discussed above, fibrotic remodeling has a more vital link with the location of RDs. Hence, electrophysiological properties such as APD and CV in the patient-specific computational model may affect optimal ablation targets. Deng et al. (2017) Deng et al. (2017) conducted a sensitivity study to investigate the influence of 10% variation of atrial APD or CV on exact locations of RDs in 12 patient-specific computational AF models. They discovered that changes in electrophysiological properties led to variation in the likelihood that RDs would be anchored to a specific site, meaning that RD trajectories based on personalized ablation strategy alone may not produce a satisfactory effect. A further study based on these results was published by Hakim et al. (2018) Hakim et al. (2018) . They found that a total of 21 emergent RDs were observed in the 9/12 atrial models by re-applying the same AF-induced protocol to the post-ablation model under average electrophysiology conditions. A total of 71% of emergent RD locations were close to sites in pre-ablation simulation with ±10% variation in APD or CV conditions. Hence, repeating simulation and ablation of emergent RDs could alleviate the uncertainty caused by electrophysiology parameters. In 2019, Trayanova's group used an innovative approach termed OPTIMA (optimal target identification via modeling of arrhythmogenesis) to predict the optimal ablation target and steer patients' ablation procedures ( Boyle et al., 2019 Boyle et al., 2019 ). They constructed ten patient-specific PeAF bi-atrial models integrating realistic atlas-based fiber orientation and fibrosis distribution. Then, they performed a set of custom-tailored ablation lesions by connecting RDs to the nearest anatomic barriers. AF induction pacing protocol was repeated to observe new emergent RDs until each of the 40 pacing sites failed to induce arrhythmias. The OPTIMA personalized ablation targets were imported into the clinical electroanatomic navigation (CARTO) maps by a series of co-registered steps and used to steer the patients' clinical ablation procedures. During follow-up, none of the patients in this study had a recurrence of PeAF, and only one patient was required to repeat the ablation due to paroxysmal AF and atrial flutter. Ali et al. (2019) Ali et al. (2019) developed 24 models based on pre- and post-PVI LGE-MRI from 12 AF patients to investigate the association between apparently successful PVI and AF recurrence post-PVI. The results indicated that both preserved RDs missed during PVI and the emergence of new RDs post-ablation contributed to post-PVI AF recurrence. In the latest article on AF ablation published by Trayanova's group, Shade et al. (2020) Shade et al. (2020) developed a personalized approach to predict the recurrence rate post-PVI via machine learning (ML) and their previous MRI-derived AF computational model simulation. The ML classifier used deductive and inductive features extracted from LGE-MRI-based AF results and raw pre-procedure LGE-MRI images as the input features. Ten-fold nested cross-validation was used to train, validate, and test the classifier. Feature selection was unbiased via random forests. A quadratic discriminant analysis classifier was used to predict the probable AF recurrence rate post-PVI. Ultimately, their ML algorithm predicted AF recurrence post-PVI with an average validation sensitivity of 82%. Several European groups have also made significant improvements in the virtual simulation of AF ablation. Bayer et al. (2016) Bayer et al. (2016) tested new radiofrequency ablation strategies for AF termination. One realistic CT-derived bilayer bi-atrial model integrated with rule-based fiber orientation and statistically distributed fibrosis was constructed in this study. Three ablation strategies were used for the AF termination test: (1) PVI+addition lines; (2) detection of ablation areas near the RDs by phase mapping; and (3) lines streaming the sequence of electrical activation during sinus rhythm. The third ablation strategy was found to be the most effective ablation pattern to terminate AF, and it was not inferior to the personalized substrate-guided ablation patterns. Roney et al. (2018) Roney et al. (2018) used patient-specific simulation to predict the efficacy of ablation of IACs for the treatment of PeAF. In their study, 12 LGE-MRI-derived bi-atrial bilayer models were created. The models were integrated with electrophysiological heterogeneity, fiber orientation, and IACs. Two different approaches (a probabilistic approach and fibrotic remodeling, including conductivity and ionic conductance changes) were used to model atrial fibrosis. All three IACs and PVI can either be ablated or not. The outcome of the ablation showed that 75% ablation patterns in combination with IACs were useful in all 12 computational models and the extent of fibrosis in the right atria had an inverse correlation with the success rate of terminating AF when combining fibrotic remodeling. Roney et al. (2020) Roney et al. (2020) compared left atrial ablation techniques, in which they constructed 20 paroxysmal AF and 30 PeAF left atrial bilayer models integrated with diffusion tensor MRI atlas-based fiber orientation and fibrosis distribution. Six different ablation approaches were used on reconstruction models to identify the optimal ablation approach for every patient. Subsequently, they used three sets of metrics as the input variables of the ML random forest classifier to predict the binary ablation response. They discovered that the optimal ablation strategies varied between individuals. Both the properties of patient-specific fibrosis and driver locations played crucial roles in determining the optimal ablation approach. Meanwhile, the distribution of ablation lesions also should be considered. Alessandrini et al. (2018) Alessandrini et al. (2018) implemented a computational framework to benchmark basket catheter-guided ablation in silico. They created one magnetic resonance angiography-derived left atrium integrated with rule-based fiber orientation and a uniform wall thickness of 3 mm. The trajectory density of the induced reentries in their model was treated as a rotor-ablation target. The simulation results revealed that the phase maps derived from intracardiac electrograms could be a powerful tool to map atrial activation patterns. Yet, they may mislead physicians due to inaccurate localization of the rotor tip depending on electrode resolution and distance to the wall. Roy et al. (2020) Roy et al. (2020) applied five different ablation strategies on six left atrial models with patient-specific geometry and fibrosis distribution to explore the association between the location of RDs and fibrosis distribution and to identify potential target areas for ablation procedures. They found that ablating the target areas and connecting them to the nearest boundary with linear lesions was superior compared with strategies such as ablating the target areas alone or clinically accepted PVI. Recently, the left atrial appendage (LAA) has received increasing attention for its role in the recurrence of AF. Gharaviri et al. (2021) Gharaviri et al. (2021) developed a model with wall thickness heterogeneities and realistic fiber orientations, performed LAA isolation, and investigated its impact on AF recurrence in three different degrees of fibrosis. The authors suggested that LAA isolation was an effective way to reduce recurrence, especially in those with high levels of fibrosis. 3 Challenges and future directions 1 Over the past two decades, significant progress has been made in the computational modeling of AF. However, there remain some challenges to be tackled. There is no direct evidence to prove the feasibility of using the LGE-MRI methodology to obtain the distribution of the fibrosis in remodeling atria ( Chrispin et al., 2016 Chrispin et al., 2016 ; Haissaguerre et al., 2016 Haissaguerre et al., 2016 ; Schade et al., 2016 Schade et al., 2016 ; Sohns et al., 2017 Sohns et al., 2017 ). More studies are needed to validate and ascertain connected findings. Personalized high-density electroanatomical substrate mapping data can play a supporting role in identifying fibrosis. Another important aspect is that the spatial resolution of most clinical images is not good enough to detect the microscopic structural changes that conducted microanatomic reentry as a substrate for AF initiation, maintenance, and ablation ( Zhao et al., 2017 Zhao et al., 2017 ); therefore, higher-resolution spatial images will be required to improve the accuracy of computational models. Furthermore, the representation of patient-specific fiber orientation is complex and yet critical for simulated atrial arrhythmia ( Ho and Sánchez-Quintana, 2009 Ho and Sánchez-Quintana, 2009 ). The methods of measuring fiber orientation and the incorporation of fiber orientation into computational models should be pursued ( Pashakhanloo et al., 2016 Pashakhanloo et al., 2016 ; Roney et al., 2021 Roney et al., 2021 ). Future atrial models for AF ablation could incorporate ganglionated plexi, which play an essential role in the pathophysiology of AF ( Bayer et al., 2019 Bayer et al., 2019 ). Moreover, since the patient-specific electrophysiological data can only be acquired in an invasive and time-consuming way, most computational models of AF ablation use the "uniform atrial cells model" ( Courtemanche et al., 1998 Courtemanche et al., 1998 ) and non-specific representation of fibrotic remodeling in fibrotic regions ( Pedrotty et al., 2009 Pedrotty et al., 2009 ). However, changing the electrophysiological parameters has been proved to significantly influence the simulation results of AF ( Saha et al., 2018 Saha et al., 2018 ). Thus, novel approaches such as noninvasive body surface potential map recordings ( Giffard-Roisin et al., 2017 Giffard-Roisin et al., 2017 ) that can measure personalized electrophysiological information should be introduced for computational modeling of AF ablation. ML techniques, including recent deep-learning approaches, have the potential to give new insights into AF that are difficult to observe and analyze with existing methods; they could also enable further automation of the entire process of computational simulation to eliminate manual choosing bias ( Cantwell et al., 2019 Cantwell et al., 2019 ; Trayanova et al., 2021 Trayanova et al., 2021 ). In virtual ablation studies, ablation lesions are performed simultaneously, whereas in clinical ablation procedures, lesions are performed sequentially. This difference leads to deviations between the results which should be addressed in the future. In addition to the sequential characteristics of ablation, the automatic identification of ablation targets is also essential. Therefore, more accurate and time-saving algorithms, such as those based on directed networks ( Vandersickel et al., 2019 Vandersickel et al., 2019 ), need to be further designed and innovated to minimize human operations and the time required for ablation. Additionally, the ultimate purpose of computational simulations is to guide clinical AF ablation. Therefore, the results of computational simulations must be tested against the most stringent clinical and experimental criteria so that they can be safely applied to the actual clinical procedure. Finally, in order to guide rather than merely predict clinical ablation, computational simulations need to be carried out in real time alongside clinical ablation procedures. Computational modeling must be simple enough for cardiologists to manipulate the entire simulation process. The duration of simulation should also meet the essential clinical requirements, all of which demand tremendous advances in hardware or software to improve computing power and computational algorithms ( Kaboudian et al., 2019 Kaboudian et al., 2019 ). 4 Conclusions 1 In summary, although virtual ablation of AF by personalized computational modeling has been developed rapidly during the last decade, it has not become a routine clinical tool. The long simulation time does not meet the requirements of actual clinical practice, and the computational modeling approach is not simple enough for cardiologists to manipulate the entire simulation process. Furthermore, computational modeling is mainly used to predict short-term rather than long-term clinical outcomes, and long-term outcomes are of most concern to cardiologists. Nonetheless, there is great potential for personalized computational modeling to become an alternative approach to drive new developments in AF treatment. References Ahmed A , Ullah W , Hussain I , et al. , 2019 . Atrial fibrillation: a leading cause of heart failure-related hospitalizations; a dual epidemic . Am J Cardiovasc Dis , 9 ( 5 ): 109 - 115 . Atrial fibrillation: a leading cause of heart failure-related hospitalizations; a dual epidemic Alessandrini M , Valinoti M , Unger L , et al. , 2018 . A computational framework to benchmark basket catheter guided ablation in atrial fibrillation . Front Physiol , 9 : 1251 . https://doi.org/10.3389/fphys.2018.01251 https://doi.org/10.3389/fphys.2018.01251 A computational framework to benchmark basket catheter guided ablation in atrial fibrillation Ali RL , Hakim JB , Boyle PM , et al. , 2019 . Arrhythmogenic propensity of the fibrotic substrate after atrial fibrillation ablation: a longitudinal study using magnetic resonance imaging-based atrial models . Cardiovasc Res , 115 ( 12 ): 1757 - 1765 . https://doi.org/10.1093/cvr/cvz083 https://doi.org/10.1093/cvr/cvz083 Arrhythmogenic propensity of the fibrotic substrate after atrial fibrillation ablation: a longitudinal study using magnetic resonance imaging-based atrial models Bayer JD , Roney CH , Pashaei A , et al. , 2016 . Novel radiofrequency ablation strategies for terminating atrial fibrillation in the left atrium: a simulation study . Front Physiol , 7 : 108 . https://doi.org/10.3389/fphys.2016.00108 https://doi.org/10.3389/fphys.2016.00108 Novel radiofrequency ablation strategies for terminating atrial fibrillation in the left atrium: a simulation study Bayer JD , Boukens BJ , Krul SPJ , et al. , 2019 . Acetylcholine delays atrial activation to facilitate atrial fibrillation . Front Physiol , 10 : 1105 . https://doi.org/10.3389/fphys.2019.01105 https://doi.org/10.3389/fphys.2019.01105 Acetylcholine delays atrial activation to facilitate atrial fibrillation Benjamin EJ , Blaha MJ , Chiuve SE , et al. , 2017 . Heart Disease and Stroke Statistics-2017 Update: a report from the American Heart Association . Circulation , 135 ( 10 ): e146 - e603 . https://doi.org/10.1161/cir.0000000000000485 https://doi.org/10.1161/cir.0000000000000485 Heart Disease and Stroke Statistics-2017 Update: a report from the American Heart Association Bhatti A , Oakeshott P , Dhinoja M , et al. , 2019 . Ablation therapy in atrial fibrillation . BMJ , 367 : l6428 . https://doi.org/10.1136/bmj.l6428 https://doi.org/10.1136/bmj.l6428 Ablation therapy in atrial fibrillation Boyle PM , Hakim JB , Zahid S , et al. , 2018a . Comparing reentrant drivers predicted by image-based computational modeling and mapped by electrocardiographic imaging in persistent atrial fibrillation . Front Physiol , 9 : 414 . https://doi.org/10.3389/fphys.2018.00414 https://doi.org/10.3389/fphys.2018.00414 Comparing reentrant drivers predicted by image-based computational modeling and mapped by electrocardiographic imaging in persistent atrial fibrillation Boyle PM , Hakim JB , Zahid S , et al. , 2018b . The fibrotic substrate in persistent atrial fibrillation patients: comparison between predictions from computational modeling and measurements from focal impulse and rotor mapping . Front Physiol , 9 : 1151 . https://doi.org/10.3389/fphys.2018.01151 https://doi.org/10.3389/fphys.2018.01151 The fibrotic substrate in persistent atrial fibrillation patients: comparison between predictions from computational modeling and measurements from focal impulse and rotor mapping Boyle PM , Zghaib T , Zahid S , et al. , 2019 . Computationally guided personalized targeted ablation of persistent atrial fibrillation . Nat Biomed Eng , 3 ( 11 ): 870 - 879 . https://doi.org/10.1038/s41551-019-0437-9 https://doi.org/10.1038/s41551-019-0437-9 Computationally guided personalized targeted ablation of persistent atrial fibrillation Cantwell CD , Mohamied Y , Tzortzis KN , et al. , 2019 . Rethinking multiscale cardiac electrophysiology with machine learning and predictive modelling . Comput Biol Med , 104 : 339 - 351 . https://doi.org/10.1016/j.compbiomed.2018.10.015 https://doi.org/10.1016/j.compbiomed.2018.10.015 Rethinking multiscale cardiac electrophysiology with machine learning and predictive modelling Chrispin J , Gucuk Ipek E , Zahid S , et al. , 2016 . Lack of regional association between atrial late gadolinium enhancement on cardiac magnetic resonance and atrial fibrillation rotors . Heart Rhythm , 13 ( 3 ): 654 - 660 . https://doi.org/10.1016/j.hrthm.2015.11.011 https://doi.org/10.1016/j.hrthm.2015.11.011 Lack of regional association between atrial late gadolinium enhancement on cardiac magnetic resonance and atrial fibrillation rotors Cochet H , Dubois R , Yamashita S , et al. , 2018 . Relationship between fibrosis detected on late gadolinium-enhanced cardiac magnetic resonance and re-entrant activity assessed with electrocardiographic imaging in human persistent atrial fibrillation . JACC Clin Electrophysiol , 4 ( 1 ): 17 - 29 . https://doi.org/10.1016/j.jacep.2017.07.019 https://doi.org/10.1016/j.jacep.2017.07.019 Relationship between fibrosis detected on late gadolinium-enhanced cardiac magnetic resonance and re-entrant activity assessed with electrocardiographic imaging in human persistent atrial fibrillation Conti S , Weerasooriya R , Novak P , et al. , 2018 . Contact force sensing for ablation of persistent atrial fibrillation: a randomized, multicenter trial . Heart Rhythm , 15 ( 2 ): 201 - 208 . https://doi.org/10.1016/j.hrthm.2017.10.010 https://doi.org/10.1016/j.hrthm.2017.10.010 Contact force sensing for ablation of persistent atrial fibrillation: a randomized, multicenter trial Courtemanche M , Ramirez RJ , Nattel S , 1998 . Ionic mechanisms underlying human atrial action potential properties: insights from a mathematical model . Am J Physiol , 275 ( 1 ): H301 - H321 . https://doi.org/10.1152/ajpheart.1998.275.1.H301 https://doi.org/10.1152/ajpheart.1998.275.1.H301 Ionic mechanisms underlying human atrial action potential properties: insights from a mathematical model Cox JL , Schuessler RB , Jr D'Agostino HJ , et al. , 1991 . The surgical treatment of atrial fibrillation . III. Development of a definitive surgical procedure. J Thorac Cardiovasc Surg , 101 ( 4 ): 569 - 583 . The surgical treatment of atrial fibrillation Dang L , Virag N , Ihara Z , et al. , 2005 . Evaluation of ablation patterns using a biophysical model of atrial fibrillation . Ann Biomed Eng , 33 ( 4 ): 465 - 474 . https://doi.org/10.1007/s10439-005-2502-7 https://doi.org/10.1007/s10439-005-2502-7 Evaluation of ablation patterns using a biophysical model of atrial fibrillation Deng DD , Jiao PF , Ye XS , et al. , 2012 . An image-based model of the whole human heart with detailed anatomical structure and fiber orientation . Comput Math Methods Med , 2012 : 891070 . https://doi.org/10.1155/2012/891070 https://doi.org/10.1155/2012/891070 An image-based model of the whole human heart with detailed anatomical structure and fiber orientation Deng DD , Murphy MJ , Hakim JB , et al. , 2017 . Sensitivity of reentrant driver localization to electrophysiological parameter variability in image-based computational models of persistent atrial fibrillation sustained by a fibrotic substrate . Chaos , 27 ( 9 ): 093932 . https://doi.org/10.1063/1.5003340 https://doi.org/10.1063/1.5003340 Sensitivity of reentrant driver localization to electrophysiological parameter variability in image-based computational models of persistent atrial fibrillation sustained by a fibrotic substrate Dewire J , Calkins H , 2013 . Update on atrial fibrillation catheter ablation technologies and techniques . Nat Rev Cardiol , 10 ( 10 ): 599 - 612 . https://doi.org/10.1038/nrcardio.2013.121 https://doi.org/10.1038/nrcardio.2013.121 Update on atrial fibrillation catheter ablation technologies and techniques Fochler F , Yamaguchi T , Kheirkahan M , et al. , 2019 . Late gadolinium enhancement magnetic resonance imaging guided treatment of post-atrial fibrillation ablation recurrent arrhythmia . Circ Arrhythm Electrophysiol , 12 ( 8 ): e007174 . https://doi.org/10.1161/circep.119.007174 https://doi.org/10.1161/circep.119.007174 Late gadolinium enhancement magnetic resonance imaging guided treatment of post-atrial fibrillation ablation recurrent arrhythmia Ganesan AN , Kuklik P , Lau DH , et al. , 2013 . Bipolar electrogram Shannon entropy at sites of rotational activation: implications for ablation of atrial fibrillation . Circ Arrhythm Electrophysiol , 6 ( 1 ): 48 - 57 . https://doi.org/10.1161/circep.112.976654 https://doi.org/10.1161/circep.112.976654 Bipolar electrogram Shannon entropy at sites of rotational activation: implications for ablation of atrial fibrillation Gharaviri A , Pezzuto S , Potse M , et al. , 2021 . Left atrial appendage electrical isolation reduces atrial fibrillation recurrences: a simulation study . Circ Arrhythm Electrophysiol , 14 ( 1 ): e009230 . https://doi.org/10.1161/CIRCEP.120.009230 https://doi.org/10.1161/CIRCEP.120.009230 Left atrial appendage electrical isolation reduces atrial fibrillation recurrences: a simulation study Giffard-Roisin S , Jackson T , Fovargue L , et al. , 2017 . Noninvasive personalization of a cardiac electrophysiology model from body surface potential mapping . IEEE Trans Bio-Med Eng , 64 ( 9 ): 2206 - 2218 . https://doi.org/10.1109/TBME.2016.2629849 https://doi.org/10.1109/TBME.2016.2629849 Noninvasive personalization of a cardiac electrophysiology model from body surface potential mapping Gong YF , Xie FG , Stein KM , et al. , 2007 . Mechanism underlying initiation of paroxysmal atrial flutter/atrial fibrillation by ectopic foci: a simulation study . Circulation , 115 ( 16 ): 2094 - 2102 . https://doi.org/10.1161/circulationaha.106.656504 https://doi.org/10.1161/circulationaha.106.656504 Mechanism underlying initiation of paroxysmal atrial flutter/atrial fibrillation by ectopic foci: a simulation study Gong YL , Gao Y , Lu ZH , et al. , 2015 . Preliminary simulation study of atrial fibrillation treatment procedure based on a detailed human atrial model . J Clin Trial Cardiol , 2 ( 4 ): 1 - 9 . https://doi.org/10.15226/2374-6882/2/4/00130 https://doi.org/10.15226/2374-6882/2/4/00130 Preliminary simulation study of atrial fibrillation treatment procedure based on a detailed human atrial model Ha ACT , Wijeysundera HC , Birnie DH , et al. , 2017 . Real-world outcomes, complications, and cost of catheter-based ablation for atrial fibrillation: an update . Curr Opin Cardiol , 32 ( 1 ): 47 - 52 . https://doi.org/10.1097/hco.0000000000000348 https://doi.org/10.1097/hco.0000000000000348 Real-world outcomes, complications, and cost of catheter-based ablation for atrial fibrillation: an update Haïssaguerre M , Jaïs P , Shah DC , et al. , 1998 . Spontaneous initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins . N Engl J Med , 339 ( 10 ): 659 - 666 . https://doi.org/10.1056/NEJM199809033391003 https://doi.org/10.1056/NEJM199809033391003 Spontaneous initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins Haissaguerre M , Shah AJ , Cochet H , et al. , 2016 . Intermittent drivers anchoring to structural heterogeneities as a major pathophysiological mechanism of human persistent atrial fibrillation . J Physiol , 594 ( 9 ): 2387 - 2398 . https://doi.org/10.1113/jp270617 https://doi.org/10.1113/jp270617 Intermittent drivers anchoring to structural heterogeneities as a major pathophysiological mechanism of human persistent atrial fibrillation Hakalahti A , Biancari F , Nielsen JC , et al. , 2015 . Radiofrequency ablation vs . antiarrhythmic drug therapy as first line treatment of symptomatic atrial fibrillation : systematic review and meta-analysis. Europace , 17 ( 3 ): 370 - 378 . https://doi.org/10.1093/europace/euu376 https://doi.org/10.1093/europace/euu376 Radiofrequency ablation vs Hakim JB , Murphy MJ , Trayanova NA , et al. , 2018 . Arrhythmia dynamics in computational models of the atria following virtual ablation of re-entrant drivers . Europace , 20(S3):iii45-iii54. https://doi.org/10.1093/europace/euy234 https://doi.org/10.1093/europace/euy234 Arrhythmia dynamics in computational models of the atria following virtual ablation of re-entrant drivers Heijman J , Algalarrondo V , Voigt N , et al. , 2016 . The value of basic research insights into atrial fibrillation mechanisms as a guide to therapeutic innovation: a critical analysis . Cardiovasc Res , 109 ( 4 ): 467 - 479 . https://doi.org/10.1093/cvr/cvv275 https://doi.org/10.1093/cvr/cvv275 The value of basic research insights into atrial fibrillation mechanisms as a guide to therapeutic innovation: a critical analysis Ho SY , Sánchez-Quintana D , 2009 . The importance of atrial structure and fibers . Clin Anat , 22 ( 1 ): 52 - 63 . https://doi.org/10.1002/ca.20634 https://doi.org/10.1002/ca.20634 The importance of atrial structure and fibers Hwang M , Kwon SS , Wi J , et al. , 2014 . Virtual ablation for atrial fibrillation in personalized in - silico three-dimensional left atrial modeling: comparison with clinical catheter ablation . Prog Biophys Mol Biol , 116 ( 1 ): 40 - 47 . https://doi.org/10.1016/j.pbiomolbio.2014.09.006 https://doi.org/10.1016/j.pbiomolbio.2014.09.006 Virtual ablation for atrial fibrillation in personalized in-silico three-dimensional left atrial modeling: comparison with clinical catheter ablation Hwang M , Song JS , Lee YS , et al. , 2016 . Electrophysiological rotor ablation in in - silico modeling of atrial fibrillation: comparisons with dominant frequency, Shannon entropy, and phase singularity . PLoS ONE , 11 ( 2 ): e0149695 . https://doi.org/10.1371/journal.pone.0149695 https://doi.org/10.1371/journal.pone.0149695 Electrophysiological rotor ablation in in-silico modeling of atrial fibrillation: comparisons with dominant frequency, Shannon entropy, and phase singularity Kaboudian A , Cherry EM , Fenton FH , 2019 . Real-time interactive simulations of large-scale systems on personal computers and cell phones: toward patient-specific heart modeling and other applications . Sci Adv , 5 ( 3 ): eaav6019 . https://doi.org/10.1126/sciadv.aav6019 https://doi.org/10.1126/sciadv.aav6019 Real-time interactive simulations of large-scale systems on personal computers and cell phones: toward patient-specific heart modeling and other applications Kim IS , Lim B , Shim J , et al. , 2019 . Clinical usefulness of computational modeling-guided persistent atrial fibrillation ablation: updated outcome of multicenter randomized study . Front Physiol , 10 : 1512 . https://doi.org/10.3389/fphys.2019.01512 https://doi.org/10.3389/fphys.2019.01512 Clinical usefulness of computational modeling-guided persistent atrial fibrillation ablation: updated outcome of multicenter randomized study Kim TH , Uhm JS , Kim JY , et al. , 2017 . Does additional electrogram-guided ablation after linear ablation reduce recurrence after catheter ablation for longstanding persistent atrial fibrillation? A prospective randomized study. J Am Heart Assoc , 6 ( 2 ): e004811 . https://doi.org/10.1161/jaha.116.004811 https://doi.org/10.1161/jaha.116.004811 Does additional electrogram-guided ablation after linear ablation reduce recurrence after catheter ablation for longstanding persistent atrial fibrillation? Latchamsetty R , Morady F , 2018 . Atrial fibrillation ablation . Annu Rev Med , 69 : 53 - 63 . https://doi.org/10.1146/annurev-med-041316-090015 https://doi.org/10.1146/annurev-med-041316-090015 Atrial fibrillation ablation Li Y , Wu YF , Chen KP , et al. , 2013 . Prevalence of atrial fibrillation in China and its risk factors . Biomed Environ Sci , 26 ( 9 ): 709 - 716 . https://doi.org/10.3967/0895-3988.2013.09.001 https://doi.org/10.3967/0895-3988.2013.09.001 Prevalence of atrial fibrillation in China and its risk factors Lim B , Hwang M , Song JS , et al. , 2017 . Effectiveness of atrial fibrillation rotor ablation is dependent on conduction velocity: an in - silico 3-dimensional modeling study . PLoS ONE , 12 ( 12 ): e0190398 . https://doi.org/10.1371/journal.pone.0190398 https://doi.org/10.1371/journal.pone.0190398 Effectiveness of atrial fibrillation rotor ablation is dependent on conduction velocity: an in-silico 3-dimensional modeling study Lim B , Park JW , Hwang M , et al. , 2020a . Electrophysiological significance of the interatrial conduction including cavo-tricuspid isthmus during atrial fibrillation . J Physiol , 598 ( 17 ): 3597 - 3612 . https://doi.org/10.1113/jp279660 https://doi.org/10.1113/jp279660 Electrophysiological significance of the interatrial conduction including cavo-tricuspid isthmus during atrial fibrillation Lim B , Kim J , Hwang M , et al. , 2020b . In situ procedure for high-efficiency computational modeling of atrial fibrillation reflecting personal anatomy, fiber orientation, fibrosis, and electrophysiology . Sci Rep , 10 : 2417 . https://doi.org/10.1038/s41598-020-59372-x https://doi.org/10.1038/s41598-020-59372-x In situ procedure for high-efficiency computational modeling of atrial fibrillation reflecting personal anatomy, fiber orientation, fibrosis, and electrophysiology Luo CH , Rudy Y , 1991 . A model of the ventricular cardiac action potential . Depolarization , repolarization, and their interaction. Circ Res, 68 ( 6 ): 1501 - 1526 . https://doi.org/10.1161/01.res.68.6.1501 https://doi.org/10.1161/01.res.68.6.1501 A model of the ventricular cardiac action potential Mărgulescu AD , Nuñez-Garcia M , Alarcón F , et al. , 2019 . Reproducibility and accuracy of late gadolinium enhancement cardiac magnetic resonance measurements for the detection of left atrial fibrosis in patients undergoing atrial fibrillation ablation procedures . Europace , 21 ( 5 ): 724 - 731 . https://doi.org/10.1093/europace/euy314 https://doi.org/10.1093/europace/euy314 Reproducibility and accuracy of late gadolinium enhancement cardiac magnetic resonance measurements for the detection of left atrial fibrosis in patients undergoing atrial fibrillation ablation procedures McDowell KS , Zahid S , Vadakkumpadan F , et al. , 2015 . Virtual electrophysiological study of atrial fibrillation in fibrotic remodeling . PLoS ONE , 10 ( 2 ): e0117110 . https://doi.org/10.1371/journal.pone.0117110 https://doi.org/10.1371/journal.pone.0117110 Virtual electrophysiological study of atrial fibrillation in fibrotic remodeling Miller CAS , Maron MS , Estes NAM III , et al. , 2019 . Safety, side effects and relative efficacy of medications for rhythm control of atrial fibrillation in hypertrophic cardiomyopathy . Am J Cardiol , 123 ( 11 ): 1859 - 1862 . https://doi.org/10.1016/j.amjcard.2019.02.051 https://doi.org/10.1016/j.amjcard.2019.02.051 Safety, side effects and relative efficacy of medications for rhythm control of atrial fibrillation in hypertrophic cardiomyopathy Morgan R , Colman MA , Chubb H , et al. , 2016 . Slow conduction in the border zones of patchy fibrosis stabilizes the drivers for atrial fibrillation: insights from multi-scale human atrial modeling . Front Physiol , 7 : 474 . https://doi.org/10.3389/fphys.2016.00474 https://doi.org/10.3389/fphys.2016.00474 Slow conduction in the border zones of patchy fibrosis stabilizes the drivers for atrial fibrillation: insights from multi-scale human atrial modeling Nademanee K , McKenzie J , Kosar E , et al. , 2004 . A new approach for catheter ablation of atrial fibrillation: mapping of the electrophysiologic substrate . J Am Coll Cardiol , 43 ( 11 ): 2044 - 2053 . https://doi.org/10.1016/j.jacc.2003.12.054 https://doi.org/10.1016/j.jacc.2003.12.054 A new approach for catheter ablation of atrial fibrillation: mapping of the electrophysiologic substrate Narayan SM , Krummen DE , Shivkumar K , et al. , 2012 . Treatment of atrial fibrillation by the ablation of localized sources: CONFIRM (Conventional Ablation for Atrial Fibrillation With or Without Focal Impulse and Rotor Modulation) trial . J Am Coll Cardiol , 60 ( 7 ): 628 - 636 . https://doi.org/10.1016/j.jacc.2012.05.022 https://doi.org/10.1016/j.jacc.2012.05.022 Treatment of atrial fibrillation by the ablation of localized sources: CONFIRM (Conventional Ablation for Atrial Fibrillation With or Without Focal Impulse and Rotor Modulation) trial Nattel S , Harada M , 2014 . Atrial remodeling and atrial fibrillation: recent advances and translational perspectives . J Am Coll Cardiol , 63 ( 22 ): 2335 - 2345 . https://doi.org/10.1016/j.jacc.2014.02.555 https://doi.org/10.1016/j.jacc.2014.02.555 Atrial remodeling and atrial fibrillation: recent advances and translational perspectives Nattel S , Heijman J , Zhou LP , et al. , 2020 . Molecular basis of atrial fibrillation pathophysiology and therapy: a translational perspective . Circ Res , 127 ( 1 ): 51 - 72 . https://doi.org/10.1161/circresaha.120.316363 https://doi.org/10.1161/circresaha.120.316363 Molecular basis of atrial fibrillation pathophysiology and therapy: a translational perspective Nguyen TP , Qu ZL , Weiss JN , 2014 . Cardiac fibrosis and arrhythmogenesis: the road to repair is paved with perils . J Mol Cell Cardiol , 70 : 83 - 91 . https://doi.org/10.1016/j.yjmcc.2013.10.018 https://doi.org/10.1016/j.yjmcc.2013.10.018 Cardiac fibrosis and arrhythmogenesis: the road to repair is paved with perils Nishida K , Nattel S , 2014 . Atrial fibrillation compendium: historical context and detailed translational perspective on an important clinical problem . Circ Res , 114 ( 9 ): 1447 - 1452 . https://doi.org/10.1161/circresaha.114.303466 https://doi.org/10.1161/circresaha.114.303466 Atrial fibrillation compendium: historical context and detailed translational perspective on an important clinical problem Pallisgaard JL , Gislason GH , Hansen J , et al. , 2018 . Temporal trends in atrial fibrillation recurrence rates after ablation between 2005 and 2014: a nationwide Danish cohort study . Eur Heart J , 39 ( 6 ): 442 - 449 . https://doi.org/10.1093/eurheartj/ehx466 https://doi.org/10.1093/eurheartj/ehx466 Temporal trends in atrial fibrillation recurrence rates after ablation between 2005 and 2014: a nationwide Danish cohort study Pashakhanloo F , Herzka DA , Ashikaga H , et al. , 2016 . Myofiber architecture of the human atria as revealed by submillimeter diffusion tensor imaging . Circ Arrhythm Electrophysiol , 9 ( 4 ): e004133 . https://doi.org/10.1161/circep.116.004133 https://doi.org/10.1161/circep.116.004133 Myofiber architecture of the human atria as revealed by submillimeter diffusion tensor imaging Patel NJ , Atti V , Mitrani RD , et al. , 2018 . Global rising trends of atrial fibrillation: a major public health concern . Heart , 104 ( 24 ): 1989 - 1990 . https://doi.org/10.1136/heartjnl-2018-313350 https://doi.org/10.1136/heartjnl-2018-313350 Global rising trends of atrial fibrillation: a major public health concern Pedrotty DM , Klinger RY , Kirkton RD , et al. , 2009 . Cardiac fibroblast paracrine factors alter impulse conduction and ion channel expression of neonatal rat cardiomyocytes . Cardiovasc Res , 83 ( 4 ): 688 - 697 . https://doi.org/10.1093/cvr/cvp164 https://doi.org/10.1093/cvr/cvp164 Cardiac fibroblast paracrine factors alter impulse conduction and ion channel expression of neonatal rat cardiomyocytes Pontecorboli G , Figueras i Ventura RM , Carlosena A , et al. , 2017 . Use of delayed-enhancement magnetic resonance imaging for fibrosis detection in the atria: a review . Europace , 19 ( 2 ): 180 - 189 . https://doi.org/10.1093/europace/euw053 https://doi.org/10.1093/europace/euw053 Use of delayed-enhancement magnetic resonance imaging for fibrosis detection in the atria: a review Rahman F , Kwan GF , Benjamin EJ , 2014 . Global epidemiology of atrial fibrillation . Nat Rev Cardiol , 11 ( 11 ): 639 - 654 . https://doi.org/10.1038/nrcardio.2014.118 https://doi.org/10.1038/nrcardio.2014.118 Global epidemiology of atrial fibrillation Reumann M , Bohnert J , Osswald B , et al. , 2007 . Multiple wavelets, rotors, and snakes in atrial fibrillation—a computer simulation study . J Electrocardiol , 40 ( 4 ): 328 - 334 . https://doi.org/10.1016/j.jelectrocard.2006.12.016 https://doi.org/10.1016/j.jelectrocard.2006.12.016 Multiple wavelets, rotors, and snakes in atrial fibrillation—a computer simulation study Reumann M , Bohnert J , Seemann G , et al. , 2008 . Preventive ablation strategies in a biophysical model of atrial fibrillation based on realistic anatomical data . IEEE Trans Biomed Eng , 55 ( 2 ): 399 - 406 . https://doi.org/10.1109/tbme.2007.912672 https://doi.org/10.1109/tbme.2007.912672 Preventive ablation strategies in a biophysical model of atrial fibrillation based on realistic anatomical data Rolf S , Kircher S , Arya A , et al. , 2014 . Tailored atrial substrate modification based on low-voltage areas in catheter ablation of atrial fibrillation . Circ Arrhythm Electrophysiol , 7 ( 5 ): 825 - 833 . https://doi.org/10.1161/CIRCEP.113.001251 https://doi.org/10.1161/CIRCEP.113.001251 Tailored atrial substrate modification based on low-voltage areas in catheter ablation of atrial fibrillation Roney CH , Williams SE , Cochet H , et al. , 2018 . Patient-specific simulations predict efficacy of ablation of interatrial connections for treatment of persistent atrial fibrillation . Europace , 20(S3):iii55-iii68. https://doi.org/10.1093/europace/euy232 https://doi.org/10.1093/europace/euy232 Patient-specific simulations predict efficacy of ablation of interatrial connections for treatment of persistent atrial fibrillation Roney CH , Beach ML , Mehta AM , et al. , 2020 . In silico comparison of left atrial ablation techniques that target the anatomical, structural, and electrical substrates of atrial fibrillation . Front Physiol , 11 : 1145 . https://doi.org/10.3389/fphys.2020.572874 https://doi.org/10.3389/fphys.2020.572874 In silico comparison of left atrial ablation techniques that target the anatomical, structural, and electrical substrates of atrial fibrillation Roney CH , Bendikas R , Pashakhanloo F , et al. , 2021 . Constructing a human atrial fibre atlas . Ann Biomed Eng , 49 ( 1 ): 233 - 250 . https://doi.org/10.1007/s10439-020-02525-w https://doi.org/10.1007/s10439-020-02525-w Constructing a human atrial fibre atlas Rotter M , Dang L , Jacquemet V , et al. , 2007 . Impact of varying ablation patterns in a simulation model of persistent atrial fibrillation . Pace-Pacing Clin Electrophysiol , 30 ( 3 ): 314 - 321 . https://doi.org/10.1111/j.1540-8159.2007.00671.x https://doi.org/10.1111/j.1540-8159.2007.00671.x Impact of varying ablation patterns in a simulation model of persistent atrial fibrillation Roy A , Varela M , Chubb H , et al. , 2020 . Identifying locations of re-entrant drivers from patient-specific distribution of fibrosis in the left atrium . PLoS Comput Biol , 16 ( 9 ): e1008086 . https://doi.org/10.1371/journal.pcbi.1008086 https://doi.org/10.1371/journal.pcbi.1008086 Identifying locations of re-entrant drivers from patient-specific distribution of fibrosis in the left atrium Ruchat P , Virag N , Dang L , et al. , 2007a . A biophysical model of atrial fibrillation ablation: what can a surgeon learn from a computer model? Europace , 9(S6):vi71-vi76. https://doi.org/10.1093/europace/eum209 https://doi.org/10.1093/europace/eum209 A biophysical model of atrial fibrillation ablation: what can a surgeon learn from a computer model? Ruchat P , Dang L , Virag N , et al. , 2007b . A biophysical model of atrial fibrillation to define the appropriate ablation pattern in modified maze . Eur J Cardio-Thorac Surg , 31 ( 1 ): 65 - 69 . https://doi.org/10.1016/j.ejcts.2006.10.015 https://doi.org/10.1016/j.ejcts.2006.10.015 A biophysical model of atrial fibrillation to define the appropriate ablation pattern in modified maze Ruchat P , Dang L , Schlaepfer J , et al. , 2007c . Use of a biophysical model of atrial fibrillation in the interpretation of the outcome of surgical ablation procedures . Eur J Cardio-Thorac Surg , 32 ( 1 ): 90 - 95 . https://doi.org/10.1016/j.ejcts.2007.02.031 https://doi.org/10.1016/j.ejcts.2007.02.031 Use of a biophysical model of atrial fibrillation in the interpretation of the outcome of surgical ablation procedures Saha M , Roney CH , Bayer JD , et al. , 2018 . Wavelength and fibrosis affect phase singularity locations during atrial fibrillation . Front Physiol , 9 : 1207 . https://doi.org/10.3389/fphys.2018.01207 https://doi.org/10.3389/fphys.2018.01207 Wavelength and fibrosis affect phase singularity locations during atrial fibrillation Sanders P , Berenfeld O , Hocini M , et al. , 2005 . Spectral analysis identifies sites of high-frequency activity maintaining atrial fibrillation in humans . Circulation , 112 ( 6 ): 789 - 797 . https://doi.org/10.1161/circulationaha.104.517011 https://doi.org/10.1161/circulationaha.104.517011 Spectral analysis identifies sites of high-frequency activity maintaining atrial fibrillation in humans Schade A , Nentwich K , Costello-Boerrigter LC , et al. , 2016 . Spatial relationship of focal impulses, rotors and low voltage zones in patients with persistent atrial fibrillation . J Cardiovasc Electrophysiol , 27 ( 5 ): 507 - 514 . https://doi.org/10.1111/jce.12913 https://doi.org/10.1111/jce.12913 Spatial relationship of focal impulses, rotors and low voltage zones in patients with persistent atrial fibrillation Seemann G , Höper C , Sachse FB , et al. , 2006 . Heterogeneous three-dimensional anatomical and electrophysiological model of human atria . Philos Trans Roy Soc A-Math Phys Eng Sci , 364 ( 1843 ): 1465 - 1481 . https://doi.org/10.1098/rsta.2006.1781 https://doi.org/10.1098/rsta.2006.1781 Heterogeneous three-dimensional anatomical and electrophysiological model of human atria Seitz J , Horvilleur J , Lacotte J , et al. , 2011 . Correlation between AF substrate ablation difficulty and left atrial fibrosis quantified by delayed-enhancement cardiac magnetic resonance . Pacing Clin Electrophysiol , 34 ( 10 ): 1267 - 1277 . https://doi.org/10.1111/j.1540-8159.2011.03148.x https://doi.org/10.1111/j.1540-8159.2011.03148.x Correlation between AF substrate ablation difficulty and left atrial fibrosis quantified by delayed-enhancement cardiac magnetic resonance Seitz J , Bars C , Théodore G , et al. , 2017 . AF ablation guided by spatiotemporal electrogram dispersion without pulmonary vein isolation: a wholly patient-tailored approach . J Am Coll Cardiol , 69 ( 3 ): 303 - 321 . https://doi.org/10.1016/j.jacc.2016.10.065 https://doi.org/10.1016/j.jacc.2016.10.065 AF ablation guided by spatiotemporal electrogram dispersion without pulmonary vein isolation: a wholly patient-tailored approach Shade JK , Ali RL , Basile D , et al. , 2020 . Preprocedure application of machine learning and mechanistic simulations predicts likelihood of paroxysmal atrial fibrillation recurrence following pulmonary vein isolation . Circ Arrhythm Electrophysiol , 13 ( 7 ): e008213 . https://doi.org/10.1161/circep.119.008213 https://doi.org/10.1161/circep.119.008213 Preprocedure application of machine learning and mechanistic simulations predicts likelihood of paroxysmal atrial fibrillation recurrence following pulmonary vein isolation Shim J , Hwang M , Song JS , et al. , 2017 . Virtual in - silico modeling guided catheter ablation predicts effective linear ablation lesion set for longstanding persistent atrial fibrillation: multicenter prospective randomized study . Front Physiol , 8 : 792 . https://doi.org/10.3389/fphys.2017.00792 https://doi.org/10.3389/fphys.2017.00792 Virtual in-silico modeling guided catheter ablation predicts effective linear ablation lesion set for longstanding persistent atrial fibrillation: multicenter prospective randomized study Sim I , Razeghi O , Karim R , et al. , 2019 . Reproducibility of atrial fibrosis assessment using CMR imaging and an open source platform . JACC Cardiovasc Imaging , 12 ( 10 ): 2076 - 2077 . https://doi.org/10.1016/j.jcmg.2019.03.027 https://doi.org/10.1016/j.jcmg.2019.03.027 Reproducibility of atrial fibrosis assessment using CMR imaging and an open source platform Sławuta A , Jacek P , Mazur G , et al. , 2020 . Quality of life and frailty syndrome in patients with atrial fibrillation . Clin Interv Aging , 15 : 783 - 795 . https://doi.org/10.2147/cia.s248170 https://doi.org/10.2147/cia.s248170 Quality of life and frailty syndrome in patients with atrial fibrillation Sohns C , Lemes C , Metzner A , et al. , 2017 . First-in-man analysis of the relationship between electrical rotors from noninvasive panoramic mapping and atrial fibrosis from magnetic resonance imaging in patients with persistent atrial fibrillation . Circ Arrhythm Electrophysiol , 10 ( 8 ): e004419 . https://doi.org/10.1161/circep.116.004419 https://doi.org/10.1161/circep.116.004419 First-in-man analysis of the relationship between electrical rotors from noninvasive panoramic mapping and atrial fibrosis from magnetic resonance imaging in patients with persistent atrial fibrillation Steinbeck G , Sinner MF , Lutz M , et al. , 2018 . Incidence of complications related to catheter ablation of atrial fibrillation and atrial flutter: a nationwide in-hospital analysis of administrative data for Germany in 2014 . Eur Heart J , 39 ( 45 ): 4020 - 4029 . https://doi.org/10.1093/eurheartj/ehy452 https://doi.org/10.1093/eurheartj/ehy452 Incidence of complications related to catheter ablation of atrial fibrillation and atrial flutter: a nationwide in-hospital analysis of administrative data for Germany in 2014 Takahashi Y , O'Neill MD , Hocini M , et al. , 2008 . Characterization of electrograms associated with termination of chronic atrial fibrillation by catheter ablation . J Am Coll Cardiol , 51 ( 10 ): 1003 - 1010 . https://doi.org/10.1016/j.jacc.2007.10.056 https://doi.org/10.1016/j.jacc.2007.10.056 Characterization of electrograms associated with termination of chronic atrial fibrillation by catheter ablation Trayanova NA , Popescu DM , Shade JK , 2021 . Machine learning in arrhythmia and electrophysiology . Circ Res , 128 ( 4 ): 544 - 566 . https://doi.org/10.1161/CIRCRESAHA.120.317872 https://doi.org/10.1161/CIRCRESAHA.120.317872 Machine learning in arrhythmia and electrophysiology Vandersickel N , van Nieuwenhuyse E , van Cleemput N , et al. , 2019 . Directed networks as a novel way to describe and analyze cardiac excitation: directed graph mapping . Front Physiol , 10 : 1138 . https://doi.org/10.3389/fphys.2019.01138 https://doi.org/10.3389/fphys.2019.01138 Directed networks as a novel way to describe and analyze cardiac excitation: directed graph mapping Verma A , Jiang CY , Betts TR , et al. , 2015 . Approaches to catheter ablation for persistent atrial fibrillation . N Engl J Med , 372 ( 19 ): 1812 - 1822 . https://doi.org/10.1056/NEJMoa1408288 https://doi.org/10.1056/NEJMoa1408288 Approaches to catheter ablation for persistent atrial fibrillation Virag N , Jacquemet V , Henriquez CS , et al. , 2002 . Study of atrial arrhythmias in a computer model based on magnetic resonance images of human atria . Chaos , 12 ( 3 ): 754 - 763 . https://doi.org/10.1063/1.1483935 https://doi.org/10.1063/1.1483935 Study of atrial arrhythmias in a computer model based on magnetic resonance images of human atria Weimar T , Schena S , Bailey MS , et al. , 2012 . The Cox-Maze procedure for lone atrial fibrillation: a single-center experience over 2 decades . Circ Arrhythm Electrophysiol , 5 ( 1 ): 8 - 14 . https://doi.org/10.1161/circep.111.963819 https://doi.org/10.1161/circep.111.963819 The Cox-Maze procedure for lone atrial fibrillation: a single-center experience over 2 decades Woods CE , Olgin J , 2014 . Atrial fibrillation therapy now and in the future: drugs, biologicals, and ablation . Circ Res , 114 ( 9 ): 1532 - 1546 . https://doi.org/10.1161/CIRCRESAHA.114.302362 https://doi.org/10.1161/CIRCRESAHA.114.302362 Atrial fibrillation therapy now and in the future: drugs, biologicals, and ablation Zahid S , Whyte KN , Schwarz EL , et al. , 2016 . Feasibility of using patient-specific models and the "minimum cut" algorithm to predict optimal ablation targets for left atrial flutter . Heart Rhythm , 13 ( 8 ): 1687 - 1698 . https://doi.org/10.1016/j.hrthm.2016.04.009 https://doi.org/10.1016/j.hrthm.2016.04.009 Feasibility of using patient-specific models and the "minimum cut" algorithm to predict optimal ablation targets for left atrial flutter Zhao JC , Hansen BJ , Wang YF , et al. , 2017 . Three-dimensional integrated functional, structural, and computational mapping to define the structural "fingerprints" of heart-specific atrial fibrillation drivers in human heart ex vivo . J Am Heart Assoc , 6 ( 8 ): e005922 . https://doi.org/10.1161/jaha.117.005922 https://doi.org/10.1161/jaha.117.005922 Three-dimensional integrated functional, structural, and computational mapping to define the structural "fingerprints" of heart-specific atrial fibrillation drivers in human heart ex vivo Zhenghong WU and Yunlong LIU wrote the manuscript. Zhenghong WU created the figure and tables. Zhenghong WU, Lv TONG, and Diandian DONG edited the manuscript. Dongdong DENG and Ling XIA contributed the framework of the manuscript. All authors have read and approved the final manuscript. This work was supported by the National Natural Science Foundation of China (Nos. 81901841 and 61527811), the Key Research and Development Program of Zhejiang Province (No. 2020C03016), and the Dalian University of Technology (No. DUT18RC(3)068), China. 10.1631/jzus.B2000727 1673-1581（2021）10-0805-13 This work was supported by the National Natural Science Foundation of China (Nos. 81901841 and 61527811), the Key Research and Development Program of Zhejiang Province (No. 2020C03016), and the Dalian University of Technology (No. 2020-11-13 2021-03-26 2021-10-15 2022-05-09 2021-09-02 Copyright © Zhejiang University and Springer-Verlag GmbH Germany, part of Springer Nature 2021 2021 浙江大学出版社 Zhenghong WU, Yunlong LIU, Lv TONG, Diandian DONG, Dongdong DENG, and Ling XIA declare that they have no conflict of interest. This article does not contain any studies with human or animal subjects performed by any of the authors. Compliance with ethics guidelines 浙江大学学报（英文版）（B辑：生物医学和生物技术） 10 22