CARDIAC PROCEDURE

CROSS-REFERENCE OF RELATED APPLICATION

[0001] This application claims priority to U.S. Provisional Application No.

61/441,947, filed February 11, 2011, the entire contents of which are hereby incorporated by reference.

[0002] This invention was made with Government support of Grant Nos. HL094610,

HLl 03428, HL099275 and HLl 03090, awarded by NIH, and Grant No. CBET-0933029, awarded by NSF. The U.S. Government has certain rights in this invention.

BACKGROUND

1. Field of Invention

[0003] The field of the currently claimed embodiments of this invention relates to systems and methods for planning patient-specific cardiac procedures.

2. Discussion of Related Art

[0004] Ventricular tachyarrhythmia (VT) frequently occurs in the setting of myocardial infarction (MI). Catheter-based ablation is a promising procedure which has become first-line therapy for many types of cardiac arrhythmias (E. Delacretaz, W. G.

Stevenson, Catheter ablation of ventricular tachycardia in patients with coronary heart disease: part I: Mapping. Pacing Clin. Electrophysiol. 24, 1261-1277 (2001); K. Soejima, M. Suzuki, W. H. Maisel, C. B. Brunckhorst, E. Delacretaz, L. Blier, S. Tung, H. Khan, W. G. Stevenson, Catheter ablation in patients with multiple and unstable ventricular tachycardias after myocardial infarction: short ablation lines guided by reentry circuit isthmuses and sinus rhythm mapping. Circulation. 104, 664-669 (2001)). However, catheter ablation has achieved low levels of success in eliminating MI- related VT; only 58% initial success rate and 71% eventual success rate following repeated procedures, with complications rate as high as 8% of the treated population (D. J. Callans, E. Zado, B. H. Sarter, D. Schwartzman, C. D. Gottlieb, F. E. Marchlinski, Efficacy of radiofrequency catheter ablation for ventricular tachycardia in healed myocardial infarction. Am. J. Cardiol. 82, 429- 432 (1998)).

[0005] The low efficacy of catheter ablation for infarct-related VT stems from the fact that current voltage and pace mapping techniques to identify the targets of ablation are associated with numerous limitations, including ambiguities in correlating maps with anatomy (J. Dong, D. Dalai, D. Scherr, A. Cheema, S. Nazarian, K. Bilchick, I. Almasry, A. Cheng, C. A. Henrikson, D. Spragg, J. E. Marine, R. D. Berger, H. Calkins, Impact of heart rhythm status on registration accuracy of the left atrium for catheter ablation of atrial fibrillation. J. Cardiovasc.

Electrophysiol. 18, 1269-1276 (2007)), and insufficient resolution in identifying ablation targets, resulting from the point-by-point sampling nature of current mapping techniques (J. Brugada, A. Berruezo, A. Cuesta, J. Osca, E. Chueca, X. Fosch, L. Wayar, L. Mont, Nonsurgical transthoracic epicardial radiofrequency ablation: an alternative in incessant ventricular tachycardia. J. Am. Coll. Cardiol. 41, 2036-2043 (2003); E. Sosa, M. Scanavacca, A. d'Avila, F. Oliveira, J. A. Ramires, Nonsurgical transthoracic epicardial catheter ablation to treat recurrent ventricular tachycardia occurring late after myocardial infarction. J. Am. Coll. Cardiol. 35, 1442-1449 (2000); H. Zhong, J. M. Lacomis, D. Schwartzman, On the accuracy of CartoMerge for guiding posterior left atrial ablation in man Heart Rhythm. 4, 595-602 (2007)). Furthermore, the complex 3D pathways along which the cardiac impulse propagates around/through the zone of infarct during VT, are difficult to reconstruct on the basis of electrical interrogation of ventricular surfaces only (J. M. de Bakker, F. J. van Capelle, M. J. Janse, A. A. Wilde, R. Coronel, A. E. Becker, K. P.

Dingemans, N. M. van Hemel, R. N. Hauer, Reentry as a cause of ventricular tachycardia in patients with chronic ischemic heart disease: electrophysiologic and anatomic correlation. Circulation. 77, 589-606 (1988); N. Peters, A. Wit, Myocardial architecture and ventricular arrhythmogenesis.

Circulation. 97, 1746-1754 (1998)). These limitations prolong procedure duration, greatly increasing the risk of chamber perforation, thromboemboli, and radiation overexposure, and limit the success of the therapy.

[0006] New approaches that deliver swift and accurate identification of optimal infarct-related VT ablation targets will dramatically improve the efficacy of the therapy and increase its tolerance while reducing post-procedure complications. This will result in a dramatic medical and economic impact on both the lives of patients and the society at large.

[0007] In addition, heart failure is a major cause of morbidity and mortality, contributing significantly to global health expenditure. Heart failure patients often exhibit contractile dyssynchrony, which diminishes cardiac systolic function. Cardiac

^synchronization therapy (CRT), a treatment modality that employs bi-ventricular (bi-V) pacing to re-coordinate the contraction of the heart, is a valuable therapeutic option for such patients. CRT has been shown to improve heart failure symptoms and reduce hospitalization, yet approximately 30% of patients fail to respond to the therapy. The poor predictive ability of current approaches to identify potential responders to CRT reflects the incomplete understanding of the complex pathophysiologic and electromechanical factors that need to be considered to achieve optimal resynchronization in each dyssychronous heart.

[0008] Therefore, there remains a need for improved systems and methods for planning patient-specific cardiac procedures.

SUMMARY

[0009] A method of planning a patient-specific cardiac procedure according to an embodiment of the current invention includes receiving three-dimensional imaging data of a patient's heart, simulating at least one of electrophysiological or electromechanical activity of at least a portion of the patient's heart using the three-dimensional imaging data, and planning the patient-specific cardiac procedure based on the simulating. The cardiac procedure is for providing a preselected alteration of at least one of electrophysiological or electromechanical behavior of the patient's heart.

[0010] A computer-readable medium according to an embodiment of the current invention includes computer-executable code for planning a patient-specific cardiac procedure, the computer-executable code includes instructions that, when executed by the computer, causes the computer to receive three-dimensional imaging data of a patient's heart, simulate at least one of electrophysiological or electromechanical activity of at least a portion of the patient's heart using the three-dimensional imaging data, and provide simulation results for planning the patient-specific cardiac procedure. The cardiac procedure is for providing a preselected alteration of at least one of electrophysiological or electromechanical behavior of the patient's heart.

[0011] A system for planning a patient-specific cardiac procedure according to an embodiment of the current invention includes a data processor configured with computer- executable code, the computer-executable code comprising instructions that, when executed by the data processor, causes the data processor to receive magnetic resonance three- dimensional imaging data of a patient's heart, simulate at least one of electrophysiological or electromechanical activity of at least a portion of the patient's heart using the three- dimensional imaging data, and provide simulation results for planning the patient-specific cardiac procedure. The cardiac procedure is for providing a preselected alteration of at least one of electrophysiological or electromechanical behavior of the patient's heart.

BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Further objectives and advantages will become apparent from a consideration of the description, drawings, and examples.

[0013] Figures 1A-1G show the model creation pipeline according to an embodiment of the current invention. A. Ex-vivo MRIs of an infarcted canine heart with corresponding segmentation. B. 3D model with epicardium rendered semi-transparent. C. Streamline representation of fibers obtained from DTMRI. D. In-vivo MRI of infarcted pig heart with corresponding segmentation. E. Model with epicardium rendered semi-transparent. F. Streamlines of approximated fibers. G. Action potentials of healthy myocytes and GZ cells.

[0014] Figures 2A-2C show VT induction in the canine heart. A. Transmembrane potential maps during PES (GZ outlined in white). B. Activation map, VT morphology 1, demonstrating figure-of-eight reentry on the epicardium and RV endocardium. Reentry circuits are organized around two I-type filaments (pink lines) located within GZ with endpoints on the epicardium and RV endocardium. C. VT morphology 2, manifested as figure-of-eight reentry on epicardium and breakthroughs on endocardium (white dots). Reentry was organized around two I-type filaments with endpoints on the epicardium and scar.

[0015] Figures 3A-3C show morphological erosion of GZ. A. Model with GZ volume 64% of the original. B. and C. Activation maps of the two VT morphologies induced after PES; both manifest as breakthrough on the epicardium and are organized around a U- type filament with endpoints on the scar.

[0016] Figures 4A and 4B show targeting filaments for ablation. A. Ablating the site of U-type filament formation shown in Fig. 3B-3C results in VT non-inducibility. B. Ablating only one of I-type filaments shown in Fig. 2C results in ablation failure. The resulting reentry is organized around a different I-type filament with endpoints on epicardium and scar.

[0017] Figures 5A-5F show sensitivity of filament position to GZ electrophysiological properties. A. Model with 20% scar in GZ (white speckles ). B. Time needed to fully activate GZ as a function of scar density in GZ. C. and D. Activation maps, filaments, and pseudo-ECGs for GZs composed of 20% and 60% scar. E. and F. Same for GZ composed of 20% and 60% normal myocardium.

[0018] Figures 6A-6F show retrospective validation of successful ablation in swine hearts according to an embodiment of the current invention. A. Pre- and post-ablation in- vivo MRI; reconstruction of LV endocardium from CT (rendered semi-transparently) with the locations of catheter tip (white dots). B. Model reconstructed from pre-ablation MRI (epicardium and GZ rendered semi-transparent). C. Simulated VT activation map, with reentry organized around a filament with endpoints on the RV side of septum and scar. D. Incorporating experimental ablation in the model also results in VT non-inducibility. E. Targeted ablation of the filament renders VT non-inducible. F. Another example of heart with successful VT ablation. Corresponding VT simulation and predicted optimal ablation target are also shown. [0019] Figures 7A-7F show retrospective validation of failed ablation in swine hearts. A. Pre- and post-ablation in-vivo MRI; reconstruction of LV endocardium from CT. B. Model reconstruction (epi- and endocardium rendered semi-transparent). C. Simulated VT activation map, with reentry figure-of-eight reentry on anterior RV. D. Incorporating experimental ablation in the model also results in ablation failure. E. Simulation-predicted optimal ablation target. F. Another example of heart with failed ablation with corresponding simulated Mi-related VT and predicted optimal ablation target.

[0020] Figure 8 is a schematic illustration of an overall approach to image-based modeling of DHF electromechanics according to an embodiment of the current invention. The geometrical images of hearts in the electrical and mechanical components are of the remodeled DHF canine heart. The light blue arrows indicate where the remodeling aspects of slowed conduction, deranged calcium, increased stiffness were incorporated into the model.

[0021] Figure 9 A shows transmural short-axis electrical maps during LBBB for normal (top) and DHF (bottom) hearts. Figure 9B shows transmural short-axis EMD maps during LBBB for normal (top) and DHF (bottom) hearts. Figure 9C shows correlation between electrical activation and mechanical activation time during LBBB for normal (top) and DHF (bottom) hearts.

[0022] Figure 10 A shows a computational mesh for the mechanical component of the MRI-based electromechanical model of post-MI canine heart according to an embodiment of the current invention. The infarct zone (scar with peri-infarct) is demarcated in blue. Figure 10B shows transmural distribution of fiber strain during sinus rhythm at two instants of time. Reference state is the unloaded state.

[0023] Figure 11 A provides a map of the percentage increase in dP/dt _max as a function of the LV pacing site. Red dots denote LV pacing sites. Figure 11B provides correlation of longitudinal distance between LV pacing site and region with longest EMD, and percentage increase in dP/dt _max . [0024] Figure 12A is a bar graph of stroke work (left), total ventricular energy consumption (middle) and myocardial efficiency (right) during LBBB and following CRT. Values are normalized to LBBB values. Figure 12B shows distribution of ATP consumption during LBBB and following CRT.

DETAILED DESCRIPTION

[0025] Some embodiments of the current invention are discussed in detail below. In describing embodiments, specific terminology is employed for the sake of clarity. However, the invention is not intended to be limited to the specific terminology so selected. A person skilled in the relevant art will recognize that other equivalent components can be employed and other methods developed without departing from the broad concepts of the current invention. All references cited anywhere in this specification, including the Background and Detailed Description sections, are incorporated by reference as if each had been individually incorporated.

[0026] According to some embodiments of the current invention, we introduce a novel multiscale electrophysiological modeling methodology, which we term "virtual electrophysiology (EP) lab," which we apply to the prediction of the optimal targets of catheter ablation of infarct-related monomorphic ventricular tachycardia (VT) in individual hearts. Determining what constitutes an optimal target of VT ablation is based on a novel mechanistic understanding of the organization of VT in myocardial infarction (MI) obtained in the present study. Predicting where these optimal ablation targets are located in an individual heart with its specific infarct morphology is based on MRI-based multiscale computational modeling of electrophysiology in that heart according to some embodiments of the current invention. We take advantage of advanced image-processing and computational-anatomy tools (H. Ashikaga, T. Sasano, J. Dong, M. M. Zviman, R. Evers, B. Hopenfeld, V. Castro, R. H. Helm, T. Dickfeld, S. Nazarian, J. K. Donahue, R. D. Berger, H. Calkins, M. R. Abraham, E. Marban, A. C. Lardo, E. R. McVeigh, H. R. Halperin, Magnetic resonance-based anatomical analysis of scar-related ventricular tachycardia: implications for catheter ablation Circ. Res. 101, 939-947 (2007); M. F. Beg, P. A. Helm, E. McVeigh, M. I. Miller, R. L. Winslow, Computational cardiac anatomy using MRI Magn. Reson. Med. 52, 1167-1174 (2004)), a high- throughput pipeline for MRI-based individualized heart model generation (F. Vadakkumpadan, H. Arevalo, A. J. Prassl, J. Chen, F. Kickinger, P. Kohl, G. Plank, N. Trayanova, Image-based models of cardiac structure in health and disease Wiley Interdisciplinary Reviews: Systems Biology and Medicine. 2, 489-506 (2010)), and sophisticated numerical simulation and analysis approaches (E. Vigmond, M. Hughes, G. Plank, L. J. Leon, Computational tools for modeling electrical activity in cardiac tissue. J. Electrocardiol. 36, 69-74 (2003)) to evaluate the VT circuits associated with the individual infarct morphology and to predict the optimal targets of VT ablation in the given heart. This approach paves the way for a major paradigm shift in the clinical procedure of VT ablation, where identification of the optimal ablation targets in each individual heart would be carried out non-invasively by the present simulation methodology prior to the clinical procedure. Delivery of catheter ablation will then be minimally-invasive, swift and precise, eradicating all infarct-related VTs.

[0027] We first present a new mechanistic understanding of VT maintenance in infarcted hearts and how this new understanding allows for the accurate prediction, by means of computational modeling of arrhythmia in the individual heart, of the optimal targets of VT ablation. We then provide examples of the success of this "virtual EP lab" approach in accurately identifying the optimal ablation targets in a retrospective animal study.

Computational Modeling of Electrophysiology in Individual Ex-vivo and In-vivo Infarcted Hearts

[0028] Ex-vivo heart models. To understand the mechanisms maintaining VT in MI and how that knowledge can be used to determine the optimal targets of infarct-related VT ablation, we used a biophysically-detailed model of an individual canine heart with MI reconstructed from high-resolution ex-vivo MRI and diffusion tensor (DT)-MRI scans. Figures 1 A- 1C present the generation of the geometrical and structural aspects of the canine heart model. Infarcted tissue in the ventricles is discriminated from the rest of the myocardium, with the infarct further segmented out (Figure 1A) into infarct scar and remodeled myocardium, the latter often referred to as border, peri-infarct, or gray zone (A. Schmidt, C. F. Azevedo, A. Cheng, S. N. Gupta, D. A. Bluemke, T. K. Foo, G. Gerstenblith, R. G. Weiss, E. Marban, G. F. Tomaselli, J. A. Lima, K. C. Wu, Infarct tissue heterogeneity by magnetic resonance imaging identifies enhanced cardiac arrhythmia susceptibility in patients with left ventricular dysfunction. Circulation. 115, 2006-2014 (2007)) based on the appearance of remodeled tissue in clinical MR images. Below we use the term gray zone (GZ). The resulting infarct segmentation shows strands of GZ tissue interdigitated with the electrically inert scar tissue, forming numerous channels within the scar. Separation of the atria from the ventricles completes the geometric reconstruction of the model (Figure IB). Fiber orientation is based on DT-MRI data (Figure 1C). A similar approach for ex-vivo MRI- based heart reconstruction has been used in our recent studies (J. D. Moreno, Z. I. Zhu, P. C. Yang, J. R. Bankston, M. T. Jeng, C. Kang, L. Wang, J. D. Bayer, D. J. Christini, N. A. Trayanova, C. M. Ripplinger, R. S. Kass, C. E. Clancy, A computational model to predict the effects of class I anti-arrhythmic drugs on ventricular rhythms Sci. Transl. Med. 3, 98ra83 (2011); K. S. McDowell, H. J. Arevalo, M. M. Maleckar, N. A. Trayanova, Susceptibility to arrhythmia in the infarcted heart depends on myofibroblast density Biophys. J. 101, 1307-1315 (2011); V. Gurev, T. Lee, J. Constantino, H. Arevalo, N. A. Trayanova, Models of cardiac electromechanics based on individual hearts imaging data: image-based electromechanical models of the heart Biomech. Model. Mechanobiol. 10, 295-306 (2011)).

[0029] The canine heart is characterized with an extensive GZ (P. Ursell, P. Gardner,

A. Albala, J. F. Jr., A. Wit, Structural and electrophysiological changes in the epicardial border zone of myocardial infarcts during infarct healing. Circ. Res. 56, 436-452 (1985)), while infarcted swine (the animal model used to demonstrate the capabilities of our approach, see below) and human hearts have been shown to be arrhythmogenic with smaller GZs. To create cardiac geometrical models with the same infarct scar but with different (smaller) GZ volumes, the GZ was "morphologically eroded," decreasing GZ volume while preserving object topology.

[0030] In-vivo heart models. To demonstrate that our simulation methodology can successfully predict the optimal ablation targets, we conducted a retrospective modeling study of infarct-related VT ablation in swine hearts. Models were generated from in-vivo MRI scans pre-ablation (Figure ID). To the best of our knowledge, this is the first development and application of heart models from in-vivo MRI. While the general electrophysiological model generation pipeline was similar to that for ex-vivo hearts, there were also significant differences. The first was in the segmentation process (Figure ID), where the ventricles were segmented by fitting cubic splines around manually identified landmark points demarcating the epicardial and endocardial surfaces; the full in- vivo geometrical model is shown in Figure IE. The second difference consisted in the fact that fiber orientation could not be acquired in this case. Therefore, fiber orientation was assigned in the in- vivo reconstructed hearts (Figure IF) using a novel geometry-driven approach (J. D. Bayer, R. Blake, G. Plank, Trayanova N, Novel rule based algorithm for assigning myocardial fiber orientation to computation heart models. Ann Biomed Eng. , (in submission) (2012)).

[0031] Assigning electrophysiological properties. The ex-vivo and in- vivo electrophysiological ventricular models were completed by assigning different electrophysiological properties to normal and GZ tissue; biophysically-detailed models of the action potentials (see Methods, below) in these regions are shown in Figure 1G. Scar and ablation lesions were assumed electrically insulating.

The 3D Organizing Centers of Infarct-related Monomorphic VT are Contained within the Infarct GZ

[0032] Using the canine ex-vivo model (GZ volume=5.0cm ), we simulated programmed electrical stimulation (PES) delivered from 27 different endocardial sites. The PES protocol consisted of pacing at a cycle length of 300ms for 6 beats followed by one or two premature extrastimuli delivered at shorter intervals until VT was induced, similar to protocols used in experimental studies (H. Ashikaga, T. Sasano, J. Dong, M. M. Zviman, R. Evers, B. Hopenfeld, V. Castro, R. H. Helm, T. Dickfeld, S. Nazarian, J. K. Donahue, R. D. Berger, H. Calkins, M. R. Abraham, E. Marban, A. C. Lardo, E. R. McVeigh, H. R. Halperin, Magnetic resonance-based anatomical analysis of scar-related ventricular tachycardia: implications for catheter ablation Circ. Res. 101, 939-947 (2007); T. Sasano, A. D. McDonald, K. Kikuchi, J. K. Donahue, Molecular ablation of ventricular tachycardia after myocardial infarction. 12, 1256-1258 (2006)). Monomorphic VT was induced in the model following PES from 8 out of the 27 pacing sites. VT persisted for the entire 2s of simulated time interval. Pseudo-ECGs were calculated in all cases, as described in Methods.

[0033] For all VTs induced, reentry initiation took place within the GZ. Figure 2A presents transmembrane potential maps depicting the events leading to reentry initiation for PES from the LV apex. The GZ exhibited slowed conduction and longer recovery time compared to the surrounding healthy tissue (Figure 2A, 2.1s). This resulted in conduction block (Figure 2A, 2.2s), wavebreak, and reentry formation (Figure 2A, 2.5s). For all PES sites resulting in VT induction, the reentrant circuit was manifested as a figure-of-eight pattern on the epicardium.

[0034] VT morphologies induced from the eight pacing sites were not all unique.

Comparison of pseudo-ECGs demonstrated two distinct VT morphologies. The first VT morphology resulted from PES at two sites, both on RV, and had an average cycle length of 190ms. The reentrant circuit was a figure-of-eight pattern on the epicardium and RV endocardium (Figure 2B). To gain further insight into the spatiotemporal organization of the VT circuit, the organizing centers of reentry (the filaments) were calculated (see Methods). For this VT morphology, the reentry revolved around two I-type filaments with endpoints at the epicardium and RV endocardium (Figure 2B, pink lines). The filaments were fully contained within the GZ and the endpoints remained in the same locations for the duration of the VT.

[0035] The second VT morphology resulted from PES at six LV endocardial sites.

The average cycle length, 222±17ms, was longer than that of the first VT morphology. The figure-of-eight reentry on the epicardium had chirality opposite to that of the first VT morphology, and was manifested as breakthroughs on LV and RV endocardial surfaces (Figure 2C). This was due to the reentrant activity being organized around two I-type filaments with endpoints at the epicardium and the infarct scar. Since the filaments did not extend to the endocardium, no rotational activity was observed there. Both filaments were stably located within the GZ throughout the VT duration.

[0036] Decreasing GZ volume by morphological erosion to values found in arrhythmogenic swine (K. H. Schuleri, M. Centola, R. T. George, L. C. Amado, K. S. Evers, K. Kitagawa, A. L. Vavere, R. Evers, J. M. Hare, C. Cox, E. R. McVeigh, J. A. C. Lima, A. C. Lardo, Characterization of Peri-Infarct Zone Heterogeneity by Contrast-Enhanced Multidetector Computed Tomography J. Am. Coll. Cardiol 53, 1699 <last_page> 1707 (2009)) or human hearts (A. Schmidt, C. F. Azevedo, A. Cheng, S. N. Gupta, D. A. Bluemke, T. K. Foo, G. Gerstenblith, R. G. Weiss, E. Marban, G. F. Tomaselli, J. A. Lima, K. C. Wu, Infarct tissue heterogeneity by magnetic resonance imaging identifies enhanced cardiac arrhythmia susceptibility in patients with left ventricular dysfunction. Circulation. 115, 2006-2014 (2007)) with MI resulted in the GZ becoming intramural and no longer extending to the epicardium as in the canine ventricles. In the model with GZ at 64% of the original volume (Figure 3 A, GZ=3.23cm ³ ), PES from the same 27 endocardial sites induced 9 VTs (average cycle length 227±23ms) with two distinct pseudo-ECG morphologies (Figures 3B-3C). For both VT morphologies, the VT manifested itself as a breakthrough on both endo- and epicardium (Figures 3B-3C), with a figure-of- eight intramural pattern. The reentrant activity was organized around a single U-type filament attached with both ends to the scar and fully contained within the GZ.

[0037] Further reduction of GZ to 37% of the original volume (1.88cm ³ ) resulted in

VT induction by PES from 7 sites with an average VT cycle length of 196±7ms; all VTs had the same morphology. VT was similarly organized around a U-type filament located in its entirety within GZ, which remained stable for the duration of the simulation. Reentry was again intramural with breakthroughs on both epi- and endocardial surfaces.

Decreasing GZ below critical size results in VT non-inducibility

[0038] Further morphological erosion of GZ resulting in the critical GZ volume of

12.6% of the original (0.76cm ³ ) resulted in inability to induce VT from any pacing site. In this case, the GZ volume was too small to support filament formation. No VT could be induced for any GZ volume below this critical value. These results indicate that there is a minimum GZ volume necessary to support filament formation in this heart.

[0039] The critical GZ volume obtained in our simulations is comparable to that reported in experiments. Using in-vivo MRI with late gadolinium enhancement (LGE) of pig hearts with MI, Estner et al (H. L. Estner, M. M. Zviman, D. Herzka, F. Miller, V. Castro, S. Nazarian, H. Ashikaga, Y. Dori, R. D. Berger, H. Calkins, A. C. Lardo, H. R. Halperin, The Critical Isthmus Sites of Ischemic Ventricular Tachycardia are in Zones of Tissue Heterogeneity, Visualized by Magnetic Resonance Imaging Heart Rhythm. (2011)) found that hearts with non-inducible VT had GZ volumes 13±5% of total infarct volume. These findings match ours: in the model where GZ volume was reduced to 12.6% of total infarct volume (the critical value of 0.76cm, ³ as described above), VT was not inducible. Our simulations demonstrate that large GZ volumes were able to support a larger number of stable filaments, resulting in multiple VT morphologies arising from the same infarct geometry (Figures2-3). Intermediate GZ volumes were able to support typically a single filament giving rise to the same VT morphology regardless of PES site, while GZ volumes below the critical value resulted in VT non-inducibility due to insufficient amount of electrically remodelled tissue to support reentrant activity.

Targeting of filaments in the GZ for optimal VT ablation

[0040] The results presented above suggest that targeting GZ with catheter ablation to decrease its size and bring it below the critical volume for sustaining reentrant activity would result in successful termination of VT. This approach has recently been validated in a retrospective study that showed that successful ablation sites, as determined during standard electrophysiological study, co-localized with GZ (H. L. Estner, M. M. Zviman, D. Herzka, F. Miller, V. Castro, S. Nazarian, H. Ashikaga, Y. Dori, R. D. Berger, H. Calkins, A. C. Lardo, H. R. Halperin, The Critical Isthmus Sites of Ischemic Ventricular Tachycardia are in Zones of Tissue Heterogeneity, Visualized by Magnetic Resonance Imaging Heart Rhythm. (2011)). Results from recent clinical ablations studies have demonstrated a significant benefit from encircling the infarct scar with ablation lesions (J. M. Frapier, J. J. Hubaut, J. L. Pasquie, P. A. Chaptal, Large encircling cryoablation without mapping for ventricular tachycardia after anterior myocardial infarction: Long-term outcome J Thorac. Cardiovasc. Surg. 116, 578 <last_page> 583 (1998); R. G. De Maria, M. Mukaddirov, P. Rouviere, E. Barbotte, B. Celton, B. Albat, J. Frapier, Long-Term Outcomes After Cryoablation for Ventricular Tachycardia During Surgical Treatment of Anterior Ventricular Aneurysms Pacing and Clinical Electrophysiology. 28, SI 68-171 (2005)). However, such an approach results in increased damage to functioning myocardium that could lead to depressed ventricular function (K. Soejima, M. Suzuki, W. H. Maisel, C. B. Brunckhorst, E. Delacretaz, L. Blier, S. Tung, H. Khan, W. G. Stevenson, Catheter ablation in patients with multiple and unstable ventricular tachycardias after myocardial infarction: short ablation lines guided by reentry circuit isthmuses and sinus rhythm mapping. Circulation. 104, 664-669 (2001); H. H. Khan, W. H. Maisel, C. Ho, M. Suzuki, K. Soejima, S. Solomon, W. G. Stevenson, Effect of radiofrequency catheter ablation of ventricular tachycardia on left ventricular function in patients with prior myocardial infarction J. Interv. Card. Electrophysiol. 7, 243-247 (2002)); current clinical guidelines encourage targeted approaches that minimize the ablation lesion (E. M. Aliot, W. G. Stevenson, J. M. Almendral-Garrote, F. Bogun, C. H. Calkins, E. Delacretaz, P. D. Bella, G. Hindricks, P. Jais, M. E. Josephson, J. Kautzner, G. N. Kay, K. -. Kuck, B. B. Lerman, F. Marchlinski, V. Reddy, M. -. Schalij, R. Schilling, K. Soejima, D. Wilber, EHRA/HRS Expert Consensus on Catheter Ablation of Ventricular Arrhythmias: Developed in a partnership with the European Heart Rhythm Association (EHRA), a Registered Branch of the European Society of Cardiology (ESC), and the Heart Rhythm Society (HRS); in collaboration with the American College of Cardiology (ACC) and the American Heart Association (AHA) Europace. 11, 771 <last_page> 817 (2009; 2009)).

[0041] Based on the new mechanistic insight regarding VT maintenance in the zone of infarct, and specifically, the fact that the scroll-wave filaments were anchored in specific regions of the GZ while being fully contained within the GZ, as described above, we hypothesized that ablating GZ region(s) containing the scroll wave filament(s) would terminate all VTs. To test this hypothesis, ablation lesions that encompass the scroll-wave filaments were implemented in all models with different GZ volumes; the lesions were assumed electrically inactive.

[0042] Figure 4A presents ablation of the tissue in which the U-shaped scroll-wave filament sustaining each of the VTs shown in Figures 3B-3C was formed. Model ablation was successful and VT could no longer be induced by PES from any of the 27 sites (Figure 4A). Similar was the outcome of ablation in the other models (different GZ volume) where VT was sustained by a single U-shaped filament (not shown here). Importantly, while the ablation in Figure 4 A decreased GZ volume (to 2.84cm ³ ), it did not bring it below the critical level for which VT can no longer be initiated (0.76cm ³ ). Thus, targeting the scroll-wave filaments sustaining VT, all of which are contained within GZ, is the most effective way of terminating infarct-related VT. Ablating the tissue in which the two I-shaped scroll wave filaments sustaining the VTs in Figure 2 resided also resulted in VT non-inducibility. The larger GZ (and thus longer filaments) required more extensive ablation lesions to encompass the filaments.

[0043] Figure 4B shows ablation of only one of the two filaments sustaining reentry in Figure 2, resulting in failure to terminate all VTs. Following ablation, the modified morphology and size of the GZ resulted in VT being sustained by a different I-type filament, with ends attached to the scar and epicardium. U 2012/024759

[0044] The simulations described above demonstrate that accurate identification of the optimal ablation targets in each individual heart could be carried out by determining, by means of individual MRI-based heart modeling, the locations of all scroll-wave filaments that sustain VT in the particular MI heart. Ablating the tissue in which the filaments resided successfully terminates all VTs. The simulation results also demonstrate that even when the first ablation attempt could be unsuccessful (because of, for instance, the filaments being difficult to fully access by an endo- or epicardial catheter approach in the EP lab), the new ablation targets (new filaments) associated with the modified VT substrate (modified GZ morphology and size) can be calculated again from the model, providing a dynamic update of the optimal ablation targets.

Sensitivity of VT filaments to GZ model parameters

[0045] The simulation results described above clearly demonstrate the paramount role that the GZ plays in establishing the locations, number and type of the scroll-wave filament(s) that sustain monomorphic VT in the MI heart. Since the filaments are the optimal ablation targets, accurate identification of their spatial positioning in each individual heart is key to the clinical translation of this simulation-guidance-of-VT-ablation approach. In the models presented here, GZ is represented as a homogenous region characterized with average electrophysiological remodeling (homogenous changes in ionic currents and conductivities, the latter resulting from Cx43 downregulation and lateralization). However, histological examinations of infarcted tissue have shown that voxels identified as GZ from MR scans correspond to microscopically heterogeneous mixtures of viable myocardium and infarct scar (H. Arheden, M. Saeed, C. B. Higgins, D. W. Gao, P. C. Ursell, J. Bremerich, R. Wyttenbach, M. W. Dae, M. F. Wendland, Reperfused rat myocardium subjected to various durations of ischemia: estimation of the distribution volume of contrast material with echo-planar MR imaging Radiology. 215, 520-528 (2000)). Representing GZ as a heterogeneous region would significantly complicate the model. Most importantly, the degree of GZ heterogeneity is difficult to assess from MRI scans, which would render the clinical utility of the simulation guidance approach to VT ablation difficult to ascertain. Therefore, we performed simulations to determine the sensitivity of the spatial position of the VT filaments to the degree of GZ structural heterogeneity. [0046] Similarly, we represent the ionic current remodeling in the GZ as a set of homogeneous ion current conductance changes, with specific data derived from the extensive literature on the canine epicardial border zone properties, as described in Methods. However, ionic current downregulation could be different in different animal species and the human, and data may not be readily available; furthermore, there could be significant variability in GZ ionic current remodeling between individual hearts. Demonstrating that physiological variations in GZ ionic current remodeling do not affect the accurate prediction of scroll- wave filaments by our modeling approach makes the clinical translation of the approach feasible because it eliminates the need to obtain information about the GZ electrophysological properties in each individual heart. Therefore, we also performed simulations to determine the sensitivity of the spatial position of the VT filaments to the degree of ionic current remodeling in the GZ.

[0047] In both sets of simulations, we deemed the spatial position of the VT filaments not sensitive to a particular set of GZ model parameters when this position remained within the approximate volume of a single clinical ablation lesion. As shown by Lardo et al (A. C. Lardo, E. R. McVeigh, P. Jumrussirikul, R. D. Berger, H. Calkins, J. Lima, H. R. Halperin, Visualization and temporal/spatial characterization of cardiac radiofrequency ablation lesions using magnetic resonance imaging Circulation. 102, 698-705 (2000)), the size of a typical single ablation lesion is 9.4±0.05mm by 6.7±.05mm by 3.4±2.1mm.

[0048] In the first set of simulations, micro-regions of scar were randomly distributed in the models throughout the GZ volume at varying densities (10% to 90% of GZ volume in 10% steps, Figure 5 A shows the 20% case), although histological studies have demonstrated that scar infiltration in the GZ is only up to 40% of GZ volume (H. Arheden, M. Saeed, C. B. Higgins, D. W. Gao, P. C. Ursell, J. Bremerich, R. Wyttenbach, M. W. Dae, M. F. Wendland, Reperfused rat myocardium subjected to various durations of ischemia: estimation of the distribution volume of contrast material with echo-planar MR imaging Radiology. 215, 520-528 (2000)). Following PES, the locations of the resulting filaments were compared to those in the corresponding homogeneous GZ model. Incorporation of scar in the GZ resulted in conduction slowing within GZ. The total time it took to fully activate the GZ increased with increased degree of scar density (Figure 5B). For the heterogeneous cases with GZ U 2012/024759 composed of up to 40% scar, all induced VT morphologies were identical to that in control (Fig.2B). Figure 5C shows the activation maps and filament locations for the model that incorporated 20% scar in the GZ. VT cycle length was 2% longer than in control, with VT again manifested as a figure-of-eight reentry on the epicardium and breakthrough on the endocardium. Most importantly, the filaments remained in the same spatial position, with accuracy fully within one clinical lesion.

[0049] As GZ scar density increased to more than 70%, wavefronts did not fully propagate through GZ, rendering it functionally identical to the scar; VT was also not inducible. In Figure 5D (60% scar), VT cycle length was 26% longer than in control. The VT was manifested as 6 reentries on the epicardium, with multiple filaments densely packed within the GZ. Despite the more complex VT spatiotemporal dynamics, filaments remained within the same general area as in control.

[0050] In the second set of simulations, we similarly incorporated random micro- regions in the GZ at increasing density, this time composed of normal myocardium. The simulations revealed that models with unchanged GZ conductivities but GZ composition incorporating up to 80% normal tissue exhibited the same VT morphology as in control; VT cycle lengths also did not differ significantly from the control (188.1±0.76ms). Increasing the amount of GZ normal tissue to 90% and 100% rendered VTs non-inducible. Figures 5E- 5F show the activation maps and filament locations for the VTs induced in models with 20% and 60% normal tissue in GZ. In both cases, there were slight changes in the activation pattern within the GZ as compared to control, but the reentrant patterns remained the same. Again, the filaments remained in the same spatial position, with accuracy within one clinical lesion.

[0051] These simulations demonstrate that scroll- wave filament locations are not particularly sensitive to the composition of the GZ and are determined predominately by GZ morphology and size. The results also strengthen significantly the possibility of clinical translation of the proposed "virtual EP lab" approach for identifying the optimal VT ablation targets since they demonstrate that the approach needs only the acquisition of the clinical MR and a model with "average" electrophysiological properties. Predicting the optimal ablation targets: a retrospective animal study

[0052] To demonstrate that our modeling approach can successfully be used to predict the optimal ablation targets, we conducted a retrospective animal study. Five pigs underwent intracardiac electrophysiological (EP) study to ablate post-MI VT; in-vivo MRIs with LGE pre- and post-ablation were also acquired (see Methods). Of the five hearts, ablation succeeded and VT was non-inducible during a follow up EP study 1 week post- ablation; in the other 3, ablation failed. All swine hearts were reconstructed from the in-vivo MRI scans, ventricular electrophysiological models created, and PES simulated to determine arrhythmogenicity, VT morphology, and filament locations.

[0053] Figure 6 presents experimental and simulation results for the two hearts where post-MI VT was chronically ablated. Figure 6A shows pre- and post-ablation MRIs of one pig heart, as well as the LV endocardium reconstructed from CT with the locations of the catheter tip delivering ablation (white dots) registered on the endocardial surface. In this heart, lesions were created throughout the septum, successfully terminating VT. Comparing pre- and post-ablation MRJs demonstrated that the ablated region co-localized with the GZ in the septum (Figure 6A). The computational model accurately reconstructed the septal infarct with islands of viable GZ (Figure 6B). Following PES, the induced VT organized around an I-type filament located on the septum with endpoints on the RV endocardium and scar (Figure 6C). Implementing in the model the experimental lesions also successfully terminated VT (Figure 6D), demonstrating excellent correspondence between model and experiment. However, the experimental ablation lesions were fairly extensive (Figure 6D). The simulations demonstrated that a smaller ablation lesion at the location of the filament would have successfully terminated VT (Figure 6E).

[0054] Figure 6F shows another example of excellent correspondence between experiment and simulation prediction. In this animal, ablation delivered at a relatively small area on the septum chronically rendered VT non-inducible. The simulations revealed that the presence of GZ in this area resulted in the formation of two I-type filaments; thus this region was the optimal ablation target. [0055] For the three animals where VT was not chronically ablated, the post-ablation

MRI revealed that ablation was not delivered at the GZ. Figure 7A shows that in one animal the heart was ablated along the posterior portion of the LV wall near the RV insertion point. MRI with LGE post-ablation revealed that the lesions were distal from the infarct, the latter located primarily at the septum and anterior portion of the LV (Figure 7A). Using the pre- ablation reconstruction shown in Figure 7B and simulating PES revealed that reentry was organized around two I-type filaments in the GZ located at the anterior portion of the RV (Figure 7C). Simulations that incorporated the experimental ablation lesions correctly predicted that VT would remain inducible (Figure 7D), with morphology similar to pre- ablation. Ablating the site of filament formation at the GZ in the RV resulted in ablation success in-silico (Figure 7E).

[0056] Figure 7F shows another example where experimental ablation was delivered at sites distal from the GZ. In this case, the heart was ablated at the anterior LV wall near the RV insertion point and the LV free wall near the base. The simulations correctly predicted VT termination failure when the experimental lesions were incorporated in the model. Modeling was able to determine that GZ in RV harbored the filaments and was the optimal ablation target. Similar was the case of failed ablation in the third pig (results not shown).

DISCUSSION

[0057] Ventricular ablation is currently offered to MI patients with recurrent infarct- related VT. A catheter is inserted through veins for an endocardial approach to map the electrical activity of the heart following an arrhythmia induction protocol, so that the targets of ablation can be identified (E. M. Aliot, W. G. Stevenson, J. M. Almendral-Garrote, F. Bogun, C. H. Calkins, E. Delacretaz, P. D. Bella, G. Hindricks, P. Jais, M. E. Josephson, J. Kautzner, G. N. Kay, K. -. Kuck, B. B. Lerman, F. Marchlinski, V. Reddy, M. -. Schalij, R. Schilling, K. Soejima, D. Wilber, EHRA/HRS Expert Consensus on Catheter Ablation of Ventricular Arrhythmias: Developed in a partnership with the European Heart Rhythm Association (EHRA), a Registered Branch of the European Society of Cardiology (ESC), and the Heart Rhythm Society (HRS); in collaboration with the American College of Cardiology (ACC) and the American Heart Association (AHA) Europace. 11, 771 <last_page> 817 (2009; 2009)). Mapping is a meticulous process, lasting for several hours, in which information regarding the electrical activity on the ventricular surface is collected from the tip of a roving catheter on a point-by-point basis (J. Brugada, A. Berruezo, A. Cuesta, J. Osca, E. Chueca, X. Fosch, L. Wayar, L. Mont, Nonsurgical transthoracic epicardial radiofrequency ablation: an alternative in incessant ventricular tachycardia. J. Am. Coll. Cardiol. 41, 2036-2043 (2003); E. Sosa, M. Scanavacca, A. d'Avila, F. Oliveira, J. A. Ramires, Nonsurgical transthoracic epicardial catheter ablation to treat recurrent ventricular tachycardia occurring late after myocardial infarction. J. Am. Coll. Cardiol. 35, 1442-1449 (2000)). The generated maps are thus typically of low resolution, and the propagation pathways, as identified from the maps, are only surface manifestations of the 3D reentrant circuits during infarct-related VT. Furthermore, mapping is best performed during sustained VT. However, patients often exhibit a hemodynamic intolerance to the sustained VT induced in the EP lab, which does not allow the time needed for detailed mapping. Therefore, infarct-related VT ablation remains a relatively high-risk procedure with a success rate far from desired (D. J. Callans, E. Zado, B. H. Sarter, D. Schwartzman, C. D. Gottlieb, F. E. Marchlinski, Efficacy of radiofrequency catheter ablation for ventricular tachycardia in healed myocardial infarction. Am. J. Cardiol. 82, 429-432 (1998)).

[0058] The above examples demonstrate some concepts of a non-invasive approach to identify the optimal infarct-related VT ablation targets according to an embodiment of the current invention. This embodiment includes using, prior to the clinical procedure, an MRI- based subject-specific multiscale electrophysiological model of the heart to analyze post-MI VT circuits and to determine the optimal ablation targets. The approach allows for a full 3D visualization and analysis of these circuits. As the results of the present study demonstrate, the optimal ablation targets are the scroll-wave filaments that sustain VTs, which were found to be stably located in the GZ. Once the optimal ablation targets are determined and visualized by the present approach, ablation delivery could be swift and precise, eradicating, with a minimal number of lesions, all infarct-related VTs. This will dramatically improve the efficacy of ablation, increase the tolerance for the procedure, and reduce post-procedure complications and long-term deleterious effects resulting from the lengthy invasive mapping and the numerous unnecessary ablation lesions. Translating the approach presented here into the clinic will constitute a dramatic shift in the paradigm of infarct-related VT ablation procedure. [0059] Importantly, the approach presented here allows for careful ablation procedure planning. It is possible that the locations of the scroll-wave filaments are not accessible, in their entirety, by any clinical endo- or epicardial approach. The subject-specific heart model could then be used to play out scenarios of tiered ablation approaches, where the filament locations are re-calculated following an initial model ablation that does not eradicate all filaments, much like in Figure 4B. Such targeted "filament modification" approach (electrophysiologically equivalent to a targeted GZ substrate modification) in situations of restricted access to the sites will still terminate all VTs with a minimum number of ablation lesions.

[0060] The predictive capabilities of the approach were tested here in a retrospective animal study. We demonstrated that when the experimental lesions resulting in VT termination were implemented in the model, the latter also predicted VT termination; similarly, failed ablation in the experiment was also a failure to ablate VT in the model. Importantly, comparisons between experimental ablation lesions that eradicated VT and the optimal ablation lesions predicted by the model revealed that ablation could have succeeded with a much smaller number of ablations and an overall lesion size that is dramatically smaller than the one delivered in the experiment (Figure 6).

[0061] A novel mechanistic insight from the results presented is that the scroll- wave filaments sustaining infarct-related monomorphic VT (i.e. the optimal ablation targets) are fully contained within the GZ; the simulation results also provided information regarding how GZ size determines filament type. GZ was thus the arrhythmogenic substrate that promoted wavebreak and reentry formation. Our findings are consistent with observations of epicardial reentrant activity anchored to regions of slow conduction within GZ, made during electrical mapping of post-MI VT (H. Ashikaga, T. Sasano, J. Dong, M. M. Zviman, R. Evers, B. Hopenfeld, V. Castro, R. H. Helm, T. Dickfeld, S. Nazarian, J. K. Donahue, R. D. Berger, H. Calkins, M. R. Abraham, E. Marban, A. C. Lardo, E. R. McVeigh, H. R. Halperin, Magnetic resonance-based anatomical analysis of scar-related ventricular tachycardia: implications for catheter ablation Circ. Res. 101, 939-947 (2007)). Similarly, recent clinical studies using contrast-enhanced cardiac MRI have shown that GZ extent correlates with arrhythmia susceptibility in patients with ischemic cardiomyopathy (A. Schmidt, C. F. Azevedo, A. Cheng, S. N. Gupta, D. A. Bluemke, T. K. Foo, G. Gerstenblith, R. G. Weiss, E. Marban, G. F. Tomaselli, J. A. Lima, K. C. Wu, Infarct tissue heterogeneity by magnetic resonance imaging identifies enhanced cardiac arrhythmia susceptibility in patients with left ventricular dysfunction. Circulation. 115, 2006-2014 (2007); A. T. Yan, A. J. Shayne, K. A. Brown, S. N. Gupta, C. W. Chan, T. M. Luu, M. F. Di Carli, H. G. Reynolds, W. G. Stevenson, R. Y. Kwong, Characterization of the peri-infarct zone by contrast- enhanced cardiac magnetic resonance imaging is a powerful predictor of post-myocardial infarction mortality. Circulation. 114, 32-39 (2006); S. D. Roes, C. J. Borleffs, R. J. van der Geest, J. J. Westenberg, N. A. Marsan, T. A. Kaandorp, J. H. Reiber, K. Zeppenfeld, H. J. Lamb, A. de Roos, M. J. Schalij, J. J. Bax, Infarct tissue heterogeneity assessed with contrast-enhanced MRI predicts spontaneous ventricular arrhythmia in patients with ischemic cardiomyopathy and implantable cardioverter-defibrillator Circ. Cardiovasc. Imaging. 2, 183-190 (2009)). The filament locations in GZ were found to not be sensitive to the structural composition of the GZ. Our simulations show that the presence of up to 40% of scar in GZ does not affect filament spatial position; even with 70% scar in GZ, the induced VTs still had filaments located in approximately the same general region as in the model without scar tissue in GZ (Figure 5C-5D). The filaments spatial position was also not much sensitive to the degree of ionic current remodeling in GZ. For a given PES site, GZ morphology and size were found to be the main determinants of filament number, location, and type. Our simulations demonstrated that approximating the GZ as a homogeneously remodeled tissue with slowed conduction is sufficient to predict the locations of post-MI VT filaments.

[0062] The "virtual EP lab" presented here incorporates advanced image-processing for in- vivo MRJ-based subject-specific heart model generation as well as sophisticated numerical simulation and analysis approaches. Should the methodology be successfully implemented in the clinic, it will constitute a major leap forward in the integration of computational modeling, traditionally a basic-science discipline, in the diagnosis and treatment of cardiac disease.

[0063] In conclusion, we presented examples of a novel approach for accurate identification of infarct-related VT ablation targets according to an embodiment of the current invention. In this example, we focused on monomorphic VT, and demonstrated the predictive capabilities of the approach in a retrospective animal study. METHODS

Datasets and Model Creation

[0064] Ex-vivo canine heart: Details regarding the image acquisition and reconstruction of the infarcted canine heart from ex-vivo MRI and DTMRI images were fully described elsewhere (F. Vadakkumpadan, H. Arevalo, A. J. Prassl, J. Chen, F. Kickinger, P. Kohl, G. Plank, N. Trayanova, Image-based models of cardiac structure in health and disease Wiley Interdisciplinary Reviews: Systems Biology and Medicine. 2, 489-506 (2010)).

[0065] In-vivo swine hearts: The infarcted swine hearts imaging and EP study were previously described (H. L. Estner, M. M. Zviman, D. Herzka, F. Miller, V. Castro, S. Nazarian, H. Ashikaga, Y. Dori, R. D. Berger, H. Calkins, A. C. Lardo, H. R. Halperin, The Critical Isthmus Sites of Ischemic Ventricular Tachycardia are in Zones of Tissue Heterogeneity, Visualized by Magnetic Resonance Imaging Heart Rhythm. (2011) (H. L. Estner, M. M. Zviman, D. Herzka, F. Miller, V. Castro, S. Nazarian, H. Ashikaga, Y. Dori, R. D. Berger, H. Calkins, A. C. Lardo, H. R. Halperin, The Critical Isthmus Sites of Ischemic Ventricular Tachycardia are in Zones of Tissue Heterogeneity, Visualized by Magnetic Resonance Imaging Heart Rhythm. (2011)). For this example, we used a subset of the datasets (5 pigs) to prospectively validate our simulation methodology. In these pigs, MI was induced via occlusion of the mid-left anterior descending coronary artery. Four weeks after MI induction, the animals underwent in-vivo contrast-enhanced MRI at a resolution of 976χ976χ4000μηι ³ . One day post-MRI, the animals underwent a full EP study to ablate VT. 7-9 days post-ablation, the animals underwent a follow up EP study to determine if ablation resulted in VT non-inducibility. A post-ablation in-vivo MRI was also performed.

[0066] Figures ID- IF illustrate the reconstruction of hearts from pre-ablation in-vivo

MRI. First, ventricles were segmented f om the rest of the torso by fitting closed splines through a set of landmark points placed manually along the epi- and endocardial boundaries (Figure ID). These splines were then tricubically interpolated to obtain a segmentation of the ventricles with a resolution of 400x400x400um ³ . To segment the infarct, the MR images were tricubically interpolated to the same resolution as the ventricular segmentation. Gray- level thresholding was used to segment healthy myocardium, infarct scar, and GZ (A. Schmidt, C. F. Azevedo, A. Cheng, S. N. Gupta, D. A. Bluemke, T. K. Foo, G. Gerstenblith, R. G. Weiss, E. Marban, G. F. Tomaselli, J. A. Lima, K. C. Wu, Infarct tissue heterogeneity by magnetic resonance imaging identifies enhanced cardiac arrhythmia susceptibility in patients with left ventricular dysfunction. Circulation. 115, 2006-2014 (2007)). We used software tools developed by our group to create finite element meshes of the hearts that incorporated adaptive element sizing that preserves fine details of the geometry including infarct surfaces.

[0067] Electrophysiological Parameters: In healthy ventricular myocardium, passive electrical properties were defined using normal conductivity values (L. Clerc, Directional differences of impulse spread in trabecular muscle from mammalian heart. J. Physiol. (Lond. ). 255, 335-346 (1976)) and ionic kinetics were described by the Luo-Rudy II model of ventricular action potential (C. Luo, Y. Rudy, A dynamic model of the cardiac ventricular action potential. II. Afterdepolarizations, triggered activity, and potentiation. Circ Res. 74, 1097-1113 (1994)) (Figure 1G). The GZ incorporated experimentally determined changes that resulted in decreased transverse conductivity (J. Yao, W. Hussain, P. Patel, N. Peters, P. Boyden, A. Wit, Remodeling of gap junctional channel function in epicardial border zone of healing canine infarcts. Circ. Res. 92, 437-443 (2003)) and action potential with decreased excitability and increased duration (J. Pu, P. Boyden, Alterations of NaS' -S currents in myocytes from epicardial border zone of the infarcted heart. A possible ionic mechanism for reduced excitability and postrepolarization refractoriness. Circ. Res. 81, 110-119 (1997); W. Dun, S. Baba, T. Yagi, P. A. Boyden, Dynamic remodeling of K+ and Ca2+ currents in cells that survived in the epicardial border zone of canine healed infarcted heart. Am. J. Physiol. Heart Circ. Physiol. 287, HI 046-54 (2004); M. Jiang, C. Cabo, J. Yao, P. Boyden, G. Tseng, Delayed rectifier K currents have reduced amplitudes and altered kinetics in myocytes from infarcted canine ventricle. Cardiovasc. Res. 48, 34-43 (2000)).

Simulation Protocol and Analysis

[0068] Mathematical description of cardiac tissue was based on monodomain equations. The software CARP was used to solve this system of equations on a parallel computing system (E. Vigmond, M. Hughes, G. Plank, L. J. Leon, Computational tools for modeling electrical activity in cardiac tissue. J. Electrocardiol. 36, 69-74 (2003)). [0069] To classify the induced VT morphologies, pseudo-ECGs were calculated by taking the difference of extracellular potentials between two points separated by 18cm in an isotropic bath surrounding the hearts. The extracellular potentials were approximated using an integral equation by Gima et al (K. Gima, Y. Rudy, Ionic Current Basis of Electrocardiographic Waveforms: A Model Study. Circ Res. 90, 889-896 (2002)).

[0070] Scroll-wave filaments were determined by converting transmembrane potential maps into phase angle maps, and then determining the nodes where the integral of the phase angles of surrounding nodes was ±2π (C. Larson, L. Dragnev, N. Trayanova, Analysis of electrically-induced reentrant circuits in a sheet of myocardium. Ann Biomed Eng. 31, 768-80 (2003)). These nodes correspond to locations of phase singularities, which are the filament building blocks.

[0071] Some other aspects of the current invention are directed computational models of ventricular electromechanics in providing a new level of understanding of the relationship between electrical and mechanical activation in the heart, and how this understanding can be utilized to provide improved cardiac ^synchronization therapy (CRT) strategies.

Dyssynchronous Heart Failure

[0072] Heart failure is a major cardiovascular disease affecting 5 million people in the US alone, and is associated with high morbidity and mortality rates (Lloyd- Jones et al., 2009). (The references cited in the following examples are listed below for convenience.) The syndrome is characterized with impaired pump function due to the deleterious remodeling of the ventricles, from the organ down to the molecular level, which significantly alters the electrical and mechanical behavior of the heart. High-resolution magnetic resonance imaging (MRI) and diffusion tensor (DT) MRI scans (Helm et al., 2006) have shown that in dyssynchronous heart failure (DHF) there is a substantial remodeling of ventricular geometry and structure. At the organ level, the ventricles become dilated and wall thickness is reduced. At the tissue level, laminar sheet angle is altered, and the transmural gradient in fiber orientation is increased. Because chamber geometry and sheet structure are major determinants of LV mechanics (Cheng et al., 2008; LeGrice et al., 1995), the mechanical deformation of the failing heart is markedly different. Furthermore, altered heart geometry as well as fiber and sheet orientations directly affect 3D electrical propagation (Hooks et al., 2007) in the failing heart.

[0073] Heart failure is also characterized with remodeling of the electrophysiological and mechanical properties at the cellular and subcellular levels. Studies (Akar et al., 2007; Akar et al., 2004) have shown that the gap junctional protein connexin43 (Cx43) is redistributed from the intercalated disk to the lateral myocyte borders and that the amount of hypophosphorylated Cx43 is increased, leading to reduced conduction velocity in heart failure. There is a considerable downregulation of the membrane potassium channels carrying the Ito and IK1 currents (Kaab et al., 1996) and of the intracellular Ca2+ ATPase (SERCA) pump (O'Rourke et al., 1999), and upregulation of the Na-Ca exchanger (NCX) (O'Rourke et al., 1999). Remodeled ionic currents and Ca2+ handling result in altered Ca2+ transients, which, in turn, impair active tension development by the myofilaments in the cell. Finally, differential expression of collagen isoforms (Marijianowski et al., 1995) and altered ratio of titin (Wu et al., 2002) (an intrasarcomeric protein that modulates myofilament passive tension) isoforms results in increased myocardial stiffness.

[0074] Because of the combined effects of chamber, contractile, and electrophysiological remodeling, the ability of the LV to efficiently pump blood is severely compromised in heart failure patients. Furthermore, a subset of these patients exhibits abnormal electrical conduction that delays activation of one portion of the ventricle relative to another (intraventricular conduction delay due to left bundle branch block, LBBB). This results in contractile dyssynchrony (dyssynchronous heart failure, DHF), which further diminishes cardiac systolic function and energetic efficiency.

Cardiac Resynchronization Therapy

[0075] CRT is an established therapy for DHF patients. CRT typically employs bi-V pacing, with an endocardial right ventricular (RV) pacing lead and an epicardial LV pacing lead, to re-coordinate contraction (Bleeker et al., 2006a). CRT has been shown to acutely and chronically improve systolic function (Nelson et al., 2000) of the heart and to reverse the detrimental remodeling (Sutton et al., 2006) associated with heart failure. Clinical trials of CRT have consistently demonstrated improvement in heart failure symptoms, exercise tolerance, quality of life, and a reduction in recurrent hospitalizations (Auricchio et al., 2003).

[0076] Although CRT reduces morbidity and mortality (Cleland et al., 2005), approximately 30% of patients fail to respond to the therapy (Kass, 2005). This reflects the poor predictive capability of current approaches to identify potential responders to CRT. The QRS duration (QRS >150 ms), widely used in clinical trials as a basic component of the inclusion criteria for CRT, does not provide an indication of the degree of mechanical dyssynchrony (Fauchier et al., 2003). Indeed, patients with long QRS duration may not exhibit mechanical dyssynchrony and those with short QRS complexes may present with significant dyssynchrony in contraction (Auricchio et al., 1999; Fauchier et al., 2002; Pitzalis et al., 2002). Measurements of mechanical dyssynchrony by Doppler echocardiography (Bax et al., 2004; Yu et al., 2002) reveal only local dyssynchrony, while the complex deformations in DHF are global. In recent clinical trials, Doppler echocardiography was characterized by lack of repeatability and low predictive value (Beshai et al, 2007; Chung et al., 2008; Miyazaki et al.). The poor predictive capability of the above measures indicates an incomplete understanding of the relation between the electrical and mechanical events in DHF.

[0077] The presence of myocardial infarction (MI) is an additional reason for lack of response to CRT. Placement of a pacing electrode at or near the infarct scar may result in ineffective pacing and thus in failure of resynchronization. Since infarction modulates electromechanical interactions, it also alters the mechanism of CRT. Bleeker et al. (Bleeker et al., 2006b) documented that patients with transmural posterolateral scar have a much lower response rate to CRT than those without scar, 14% vs. 81%. Increased scar volume has been found to result in unfavorable response to CRT (Adelstein and Saba, 2007). Infarct location and scar transmurality are considered important (Choi et al., 2001; White et al., 2006) yet unknown factors that affect the relationship between electrical activation and contraction and contribute to diminished CRT efficacy.

[0078] Finally, the location of LV pacing has been shown to play an important role in

CRT efficacy (Butter et al., 2000; Helm et al., 2007; St John Sutton et al., 2003). Currently, LV pacing lead is implanted in a tributary of the coronary sinus, as in epicardial bi-V pacing (Butter et al., 2001). However, for a small class of patients unsuitable for transvenous bi-V, a transseptal approach has been developed that allows endocardial bi-V pacing (Leclercq et al., 1999). Recent studies have brought to light the potential proarrhythmic effect of epicardial bi-V pacing (Fish et al., 2005), resulting from the reversal of the direction of electrical propagation in the LV. Furthermore, new findings indicate that endocardial bi-V pacing might be associated with improved resynchronization in canine models (Howard et al., 2011; van Deursen et al., 2009) and humans (Spragg et al., 2010). Thus determining the optimal location of LV pacing lead placement remains a problem.

[0079] Multi-scale computational modeling of electromechanics in the normal and failing heart is provided to address these problems according to some embodiments of the current invention. Recent advancements in cardiac computational modeling, numerical algorithms and image processing techniques have enabled the development of detailed tomographically-reconstructed heart models that integrate functions from the molecular level to the electromechanical interactions in the intact organ. According to this embodiment of the current invention, we employ such models to provide approaches to optimizing CRT therapy. To achieve this goal, this embodiment focuses on exploiting knowledge regarding the electromechanical delay in the heart as well as myocardial efficiency.

Electromechanical delay in the heart and how it can be used to optimize CRT

Significance of electromechanical delay

[0080] The time period between the local electrical depolarization and the onset of local myofiber shortening (mechanical activation) in the intact ventricles can last as much as tens of milliseconds. This electromechanical delay (EMD) is a function of the myocyte's intrinsic latent period between membrane depolarization and myofilament activation in the excitation-contraction process (Cordeiro et al., 2004), but is also dependent on the local myofiber mechanical loading conditions in the intact heart. Acute CRT therapy affects only the component of EMD that is due to the loading conditions, but has no influence on the cell- intrinsic E-C coupling latency (Russell et al., 2011). Thus, by understanding EMD and its distribution that is due to the loading conditions, one could suggest potential avenues for CRT optimization. Alternatively, since most echocardiography-based dyssynchrony measurements are affected by the timing of myofiber shortening onset, ascertaining the mechanisms underlying the EMD distribution may improve or lead to the development of novel indices of electromechanical dyssynchrony to identify potential CRT responders.

Electromechanical delay in the normal heart

[0081] The first computational study to assess the 3D distribution of EMD was by

Usyk and McCulloch (Usyk and McCulloch, 2003). In this study, the authors employed an electromechanical model of the normal canine ventricles to determine the 3D EMD distribution during sinus rhythm and following LV pacing. With this early model, which in fact was the first whole-heart electromechanical model developed, the authors demonstrated that EMD may be both positive and negative, indicating that myofiber shortening may precede electrical activation in the whole heart. A more recent study by Gurev et al. (Gurev et al., 2010) have expounded on this work by providing thorough analysis of the 3D EMD distribution in the normal rabbit heart and its dependence on the loading conditions (i.e. on the electrical activation sequence). Simulations of electromechanical activity during sinus rhythm and LV epicardial pacing were conducted and compared to determine the effect of electrical activation pattern on the 3D distribution of EMD. The simulation results revealed that the 3D distribution of EMD was heterogeneous and depended on the electrical activation sequence. The distributions were markedly different for sinus rhythm and epicardial pacing. During sinus rhythm, the distribution was longer at the epicardium compared to the endocardium and longer at the base compared to the apex. Following epicardial pacing, the distribution was markedly different: the posterior wall exhibited longer EMD compared to the anterior wall. Mechanistic analysis of the electromechanical behavior revealed that the late-depolarized regions were characterized with significant myofiber pre-stretch caused by the contraction of the early-depolarized regions. This pre-stretch, in turn, delayed myofiber shortening onset, and resulted in longer EMD there.

Assessment of EMD in DHF

[0082] The pumping inefficiency of the DHF heart arises from deleterious remodeling of cardiac electromechanical properties, from the sub-cellular to the organ level, and is thus expected to change the 3D EMD distribution. Determining the 3D EMD distribution in the setting of DHF and exploiting the mechanistic insight into the relation between electrical activation and mechanical contraction could offer clues to improvement in CRT delivery. In this section, we present our new image-based electromechanical model of the failing canine ventricles, and employ it to determine how the 3D distribution of EMD is altered in the setting of DHF according to an embodiment of the current invention.

[0083] A schematic of the electromechanical model of the failing heart is shown in

Figure 8. Briefly, the electromechanical model is composed of two main components, an electrical and a mechanical component. The electrical component, which contains a biophysically-detailed representation of myocyte membrane kinetics, simulates the propagation of the action potential, while the mechanical component, which incorporates a biophysical model of myofilament dynamics, describes the active contraction and resultant deformation of the ventricles. Details regarding the basics of the model pertaining to the normal ventricles can be found in (Gurev et al., 2011). The model is generic, and could be used with any geometry, image-based or idealized. In the studies presented in this section, the geometry of the electromechanical model was generated from MR images of DHF canine ventricles (Figure 8), and the fiber and sheet architecture were constructed from DTMR images of the same DHF canine ventricles, using a recently developed methodology (Gurev et al., 2011). Using MR and DTMR images for the reconstruction of the DHF ventricular geometry and fiber and sheet geometry allows for the inclusion of the structural remodeling of the DHF ventricles.

[0084] To take into account the remodeling of the passive electromechanical tissue properties associated with DHF, the following changes were incorporated into the model (Figure 8). First, electrical conductivities were reduced by 20% from the normal values to represent the slowed conduction in DHF (Akar et al., 2004). To account for the increased stiffness of the failing myocardial tissue (Wu et al., 2002), the passive scaling constant in the strain-energy function was increased fivefold. Finally, to incorporate deranged calcium handling associated with the DHF heart (O'Rourke et al., 1999), the peak amplitude and relaxation rate of the calcium transient function, which served as an input model of the myofilament dynamics (Rice et al., 2008), was reduced to 70% and increased by 30% of the 24759 normal values, respectively. Since CRT patients exhibit a left bundle branch block (LBBB) type of electrical activation, LBBB was simulated in both models by stimulating the endocardial surface at discrete locations as if the electrical activity was emanating from the activation of the corresponding branch of the Purkinje network.

[0085] Figure 9A presents the transmural electrical activation map in LBBB for the

DHF heart in a short axis view; the same map for the normal canine heart but with LBBB is also shown. This comparison allows the examination of the differences in electrical activation that arise from heart failure remodeling. In both hearts the depolarization wave travels from the right ventricular wall, through the septum and to the left ventricular lateral wall; the mechanical activation follows the same general direction (data not shown). The electromechanical activation patterns are consistent with previous experimental results (Becker et al., 1958; Leclercq et al., 2002). To assess the contribution of the detrimental remodeling associated with DHF to altering the 3D EMD distribution, the resultant transmural EMD maps for the normal and DHF hearts are compared in Figure 9B. The maps reveal that the 3D EMD distribution is heterogeneous in both the normal and failing hearts: the late-activated lateral wall is associated with an extended EMD as compared to the septum. However, in the DHF heart, EMD is longer that in the normal heart. The differences in EMD are particularly pronounced at the lateral wall (green in normal vs red in DHF, Figure 9B).

[0086] To further examine the relationship between electrical and mechanical activation in DHF, the electrical and mechanical activation times at 24 different locations in the left ventricular mid-wall in the normal and DHF canine ventricles are presented in Figure 9C. A linear regression analysis was performed. The slopes of the regression lines, which were obtained through least squares fitting, are greater than 1 in the normal and DHF hearts, indicating that the time interval between depolarization and onset of myofiber shortening is extended at the late-activated regions. However, this interval increases at a greater rate in the DHF heart (1.25 vs 1.63). These data demonstrate that the detrimental changes to the electromechanical properties in the DHF heart results in a prolonged EMD, particularly at the LV anterior wall during LBBB activation sequence. Initial modeling efforts towards the assessment of contractile dyssynchrony and EMD in the infarcted heart

[0087] As mentioned above, the presence of MI is also expected to alter the EMD distribution in the ventricles. This section presents the initial modeling effort towards constructing an electromechanical model of the infarcted ventricles, which could be used to then assess the distribution of EMD in MI, and suggest possible improvements in CRT therapy in DHF patients with ischemic cardiomyopathy.

[0088] The model was reconstructed from MRI and DTMRI scans of canine ventricles with a 4-week old infarct. Briefly, the zone of infarct was segmented from the healthy myocardium using the fractional anisotropy values calculated from the DTMRI data, and the infarct zone was then divided into the akinetic scar and the partially viable peri- infarct using level set thresholding. Further details regarding the infarct segmentation can be found in Vadakkumpadan et al. (Vadakkumpadan et al., 2010). The mechanics finite elements mesh is shown in Figure 10A; the infarct zone (scar with peri-infarct), is demarcated in blue.

[0089] Because the scar is primarily composed of necrotic collagen, it was modeled as an insulator in the electrical component of the model. In the mechanics component of the model, the passive stiffness constant was increased 1500% and no active tension was generated in the scar. In the peri-infarct, the transverse electrical conductivity was reduced by 90% to reflect the disorganization of Cx43. To account for the mechanical changes in the peri-infarct, the active tension was reduced to 10% and the passive stiffness constant was increased fifteen fold (Tyberg et al, 1970; Walker et al., 2005).

[0090] An example simulation of paced propagation using the MRI-based electromechanical model of the infarcted canine ventricles is shown in Figure 10B. Owing to the tethering of the adjacent myocardium to the akinetic scar, myocyte shortening was impaired at the LV anterior wall, resulting in regional dyssynchrony. These data suggest that the EMD distribution in the MI heart will be markedly altered by the presence of an infarct; this distribution will depend on the specific scar location and transmurality. Electromechanical models with realistic topography of MI can be used to construct 3D maps of EMD for various scar locations and degrees of transmurality to reveal the mechanisms by which infarction alter the electromechanical activity of the heart.

Using the EMD distribution to guide CRT optimization

[0091] Suboptimal placement of the LV lead constitutes a major reason underlying the high non-response rate to CRT. To date, there is no clear consensus as to where to place the LV pacing lead to achieve optimal CRT response. Previous studies (Ansalone et al., 2002; Howard et al., 2011; Suffoletto et al., 2006) have indicated that the site of latest electrical or the latest mechanical activation was associated with a greater hemodynamic benefit to CRT; however, recent data (Derval et al., 2010; Fung et al., 2009; Spragg et al., 2010) suggests there is a lack of concordance between the site of latest electrical or mechanical activation and CRT response. In a pilot study (abstract) by Constantino et al. (Constantino et al., 2010), we proposed an alternative strategy to determine the LV pacing location in an effort to optimize the response to CRT: targeting the regions with the longest EMD. This section presents simulation results towards optimization of CRT employing this strategy. Using the image-based model of canine DHF electromechanics, as described above, CRT was delivered by pacing at the RV apex, with the LV pacing electrode placed at 18 different epicardial sites along the LV free wall. For each LV pacing site, response to CRT was assessed by calculating the percent change in maximal rise in LV pressure (dP/dt _max ) as compared to that in the DHF heart.

[0092] Using the transmural EMD maps that were constructed as described above, the region with the longest EMD was determined to be the endocardial surface of the lateral wall between the base and the mid-ventricles during LBBB. Figure 11A presents CRT response as a function of LV pacing location. Maximal hemodynamic benefit occurred when the LV pacing site was located near the base and mid- ventricle, which was within the region of longest EMD. The relationship between LV pacing location and longest EMD region is quantified in Figure 1 IB. For each pacing site, CRT response and the longitudinal distance between the pacing site and the center of the region with the longest EMD were plotted. Increase in dP/dt _max strongly correlated with the longitudinal distance between LV pacing site and the center of the region with longest EMD (r=-.86, p<0.05). [0093] These computational results demonstrate that targeting the region with the longest EMD results in greatest hemodynamic response to CRT. Thus determining the 3D EMD distribution in DHF could be used to guide the optimal placement of the LV pacing electrode for CRT. It remains to be determined whether the same approach will be applicable in the infarcted heart, where discord between electrical and mechanical activity is exacerbated by the presence of the infarct (Ashikaga et al., 2005).

Using energy consumption as a guide to CRT optimization

[0094] An alternative approach to identify the optimal LV pacing location for CRT may be to target the LV location that results in maximum increase in myocardial efficiency while simultaneously minimizing the heterogeneity in energy consumption. This suggestion is based on the fact that myocardial efficiency, the ratio of mechanical work performed by the ventricles to myocardial energy consumption, is markedly reduced in DHF patients (Suga, 1990). Furthermore, it has been shown that CRT improves myocardial efficiency in DHF patients (Lindner et al., 2006). Since current experimental techniques are limited by the inability to record local mechanical activity and energy consumption in the ventricles with high spatiotemporal resolution, subject-specific electromechanical models of the DHF ventricles that incorporate a biophysically-detailed representation of cardiac myofilament dynamics can be employed to analyze the effect of CRT on local energy consumption and total myocardial efficiency in the setting of DHF and potentially identify the optimal LV pacing location. Below we present such results.

[0095] Simulations of LBBB and CRT were performed using the MRI-based electromechanical model of DHF canine heart, as above. The use of the Rice et al. (Rice et al., 2008) representation of myofilament dynamics allowed for the calculation of local energy consumption. ATP consumption distribution was calculated by integrating over time the ATP consumption rate, a function of the ATP-consuming cross-bridge detachment rate and the single overlap fraction of thick filaments. The ATP consumption of the entire ventricle was then determined by spatially integrating the local ATP consumption throughout the entire ventricular volume, as done in our previous publication (Lim et al., 2012). Finally, mechanical work was calculated by integrating the area within the pressure-volume loop curve generated by the model, and myocardial efficiency was calculted as the ratio of mechanical work to total ATP consumption in the ventricles.

[0096] Consistent with experimental findings (Lindner et al., 2006; Lindner et al.,

2005; Ukkonen et al., 2003), CRT resulted in improvement in myocardial efficiency (Figure 12A). This stems from the fact that CRT increased the mechanical work performed by the ventricles without increasing the total ventricular energy consumption. Although total energy consumption was unaltered, CRT homogenized energy consumption throughout the ventricles by increasing energy consumption at the septum and reducing it at the lateral wall (Figure 12A); this finding is consistent with experimental data (Lindner et al., 2006; Lindner et al., 2005; Ukkonen et al., 2003). These results demonstrate that computational models of DHF electromechanics can accurately simulate the effects of CRT on myocardial efficiency and local energy consumption and can be used to determine the LV pacing location that minimizes the heterogeneity in energy consumption and maximizes myocardial efficiency.

Conclusions regarding CRT examples

[0097] A comprehensive characterization of the spatiotemporal electromechanical interactions in the DHF heart, without and with MI, is fundamental to the effort towards improving CRT efficacy. This example demonstrates that a biophysically-based model of ventricular electromechanics that incorporates representations from the scale of the protein to the intact organ is a powerful methodology to provide insight into the electromechanical interactions in the heart. This example highlights how the basic science insight into the electromechanical activity of the DHF heart gained from computational modeling can be exploited to guide improvements in CRT delivery according to an embodiment of the current invention. The simulation results presented here indicate that optimal CRT strategy in the DHF heart can be achieved by pacing at the LV location characterized with longest EMD. The same approach can be used to determine whether CRT can be also optimized by targeting the region with the longest EMD in the infarcted heart. In addition, computational modeling could also aid in the identification of the LV pacing location that results in maximal myocardial efficiency and most beneficial regional energy consumption. With new advancements in computational modeling and increased ubiquity of computers in the clinic, it will not be long before electromechanical models of DHF patients' hearts that are enriched with patient-specific data will serve as a bedside tool for diagnosis and treatment planning.

[0098] Adelstein, E. C. and Saba, S. (2007) Scar Burden by Myocardial Perfusion

Imaging Predicts Echocardiographic Response to Cardiac Resynchronization Therapy in Ischemic Cardiomyopathy. Am Heart J 153, 105-12.

[0099] Akar, F. G., Nass, R. D., Hahn, S., Cingolani, E., Shah, M., Hesketh, G. G.,

DiSilvestre, D., Tunin, R. S., Kass, D. A. and Tomaselli, G. F. (2007) Dynamic Changes in Conduction Velocity and Gap Junction Properties During Development of Pacing-Induced Heart Failure. Am J Physiol Heart Circ Physiol 293, H1223-30.

[00100] Akar, F. G., Spragg, D. D., Tunin, R. S., Kass, D. A. and Tomaselli, G. F. (2004) Mechanisms Underlying Conduction Slowing and Arrhythmogenesis in Nonischemic Dilated Cardiomyopathy. Circ Res 95, 717-25.

[00101] Ansalone, G., Giannantoni, P., Ricci, R., Trambaiolo, P., Fedele, F. and Santini, M. (2002) Doppler myocardial imaging to evaluate the effectiveness of pacing sites in patients receiving biventricular pacing. J Am Coll Cardiol 39, 489-99.

[00102] Ashikaga, H., Mickelsen, S. R, Ennis, D. B., Rodriguez, I., Kellman, P., Wen, H. and McVeigh, E. R. (2005) Electromechanical analysis of infarct border zone in chronic myocardial infarction. Am J Physiol Heart Circ Physiol 289, HI 099- 105.

[00103] Auricchio, A., Stellbrink, C, Block, M., Sack, S., Vogt, J., Bakker, P., Klein, H., Kramer, A., Ding, J., Salo, R., Tockman, B., Pochet, T. and Spinelli, J. (1999) Effect of Pacing Chamber and Atrioventricular Delay on Acute Systolic Function of Paced Patients with Congestive Heart Failure. The Pacing Therapies for Congestive Heart Failure Study Group. The Guidant Congestive Heart Failure Research Group. Circulation 99, 2993-3001.

[00104] Auricchio, A., Stellbrink, C, Butter, C, Sack, S., Vogt, J., Misier, A. R, Bocker, D., Block, M., Kirkels, J. H., Kramer, A. and Huvelle, E. (2003) Clinical Efficacy of Cardiac Resynchronization Therapy Using Left Ventricular Pacing in Heart Failure Patients Stratified by Severity of Ventricular Conduction Delay. J Am Coll Cardiol 42, 2109-16. [00105] Bax, J. J., Bleeker, G. B., Marwick, T. H., Molhoek, S. G., Boersma, E., Steendijk, P., van der Wall, E. E. and Schalij, M. J. (2004) Left Ventricular Dyssynchrony Predicts Response and Prognosis after Cardiac Resynchronization Therapy. J Am Coll Cardiol 44, 1834-40.

[00106] Becker, R. A., Scher, A. M. and Erickson, R. V. (1958) Ventricular excitation in experimental left bundle branch block. Am Heart J 55, 547-56.

[00107] Beshai, J. F., Grimm, R. A., Nagueh, S. F., Baker, J. FL, Beau, S. L., Greenberg, S. M., Pires, L. A. and Tchou, P. J. (2007) Cardiac-resynchronization therapy in heart failure with narrow QRS complexes. N EngJMed 357, 2461-2471.

[00108] Bleeker, G. B., Bax, J. J., Steendijk, P., Schalij, M. J. and van der Wall, E. E. (2006a) Left Ventricular Dyssynchrony in Patients with Heart Failure: Pathophysiology, Diagnosis and Treatment. Nat Clin Pract Cardiovasc Med 3, 213-9.

[00109] Bleeker, G. B., Kaandorp, T. A., Lamb, H. J., Boersma, E., Steendijk, P., de Roos, A., van der Wall, E. E., Schalij, M. J. and Bax, J. J. (2006b) Effect of Posterolateral Scar Tissue on Clinical and Echocardiographic Improvement after Cardiac Resynchronization Therapy. Circulation 113, 969-76.

[00110] Butter, C, Auricchio, A., Stellbrink, C, Fleck, E., Ding, J., Yu, Y., Huvelle, E. and Spinelli, J. (2001) Effect of Resynchronization Therapy Stimulation Site on the Systolic Function of Heart Failure Patients. Circulation 104, 3026-9.

[00111] Butter, C, Auricchio, A., Stellbrink, C, Schlegl, M., Fleck, E., Horsch, W., Huvelle, E., Ding, J. and Kramer, A. (2000) Should Stimulation Site Be Tailored in the Individual Heart Failure Patient? Am J Cardiol 86, 144K-151K.

[00112] Cheng, A., Nguyen, T. C, Malinowski, M., Daughters, G. T., Miller, D. C. and Ingels, N. B. (2008) Heterogeneity of left ventricular wall thickening mechanisms. Circulation 118, 713-721. [00113] Choi, K. M., Kim, R. J., Gubernikoff, G., Vargas, J. D., Parker, M. and Judd, R. M. (2001) Transmural Extent of Acute Myocardial Infarction Predicts Long-Term Improvement in Contractile Function. Circulation 104, 1101-7.

[00114] Chung, E. S., Leon, A. R., Tavazzi, L., Sun, J. P., Nihoyannopoulos, P., Merlino, J., Abraham, W. T., Ghio, S., Leclercq, C. and Bax, J. J. (2008) Results of the Predictors of Response to CRT (PROSPECT) trial. Circulation 117, 2608-2616.

[00115] Cleland, J. G., Daubert, J. C, Erdmann, E., Freemantle, N., Gras, D., Kappenberger, L. and Tavazzi, L. (2005) The Effect of Cardiac Resynchronization on Morbidity and Mortality in Heart Failure. N Engl J Med 352, 1539-49.

[00116] Constantino, J., Gurev, V. and Trayanova, N. (2010) Optimal cardiac resynchronization therapy is achieved by pacing from the LV region with the longest electromechanical delay. Heart Rhythm 7, SI 64- 165.

[00117] Cordeiro, J. M., Greene, L., Heilmann, C, Antzelevitch, D. and Antzelevitch, C. (2004) Transmural heterogeneity of calcium activity and mechanical function in the canine left ventricle. Am J Physiol Heart Circ Physiol 286, H1471-9.

[00118] Derval, N., Steendijk, P., Gula, L. J., Deplagne, A., Laborderie, J., Sacher, F., Knecht, S., Wright, M., Nault, I., Ploux, S., Ritter, P., Bordachar, P., Lafitte, S., Reant, P., Klein, G. J., Narayan, S. M., Garrigue, S., Hocini, M., Haissaguerre, M., Clementy, J. and Jais, P. (2010) Optimizing hemodynamics in heart failure patients by systematic screening of left ventricular pacing sites: the lateral left ventricular wall and the coronary sinus are rarely the best sites. J Am Coll Cardiol 55, 566-75.

[00119] Fauchier, L., Marie, O., Casset-Senon, D., Babuty, D., Cosnay, P. and Fauchier, J. P. (2002) Interventricular and Intraventricular Dyssynchrony in Idiopathic Dilated Cardiomyopathy: A Prognostic Study with Fourier Phase Analysis of Radionuclide Angioscintigraphy. J Am Coll Cardiol 40, 2022-30.

[00120] Fauchier, L., Marie, O., Casset-Senon, D., Babuty, D., Cosnay, P. and Fauchier, J. P. (2003) Reliability of QRS Duration and Morphology on Surface Electrocardiogram to Identify Ventricular Dyssynchrony in Patients with Idiopathic Dilated Cardiomyopathy. Am J Cardiol 92, 341-4.

[00121] Fish, J. M., Brugada, J. and Antzelevitch, C. (2005) Potential Proarrhythmic Effects of Biventricular Pacing. J Am Coll Cardiol 46, 2340-7.

[00122] Fung, J. W., Lam, Y. Y., Zhang, Q., Yip, G. W., Chan, W. W., Chan, G. C, Chan, J. Y. and Yu, C. M. (2009) Effect of left ventricular lead concordance to the delayed contraction segment on echocardiographic and clinical outcomes after cardiac resynchronization therapy. J Cardiovasc Electrophysiol 20, 530-5.

[00123] Gurev, V., Constantino, J., Rice, J. J. and Trayanova, N. A. (2010) Distribution of electromechanical delay in the heart: insights from a three-dimensional electromechanical model. Biophys J99, 745-54.

[00124] Gurev, V., Lee, T., Constantino, J., Arevalo, H. and Trayanova, N. A. (2011) Models of cardiac electromechanics based on individual hearts imaging data: image-based electromechanical models of the heart. Biomech Model Mechanobiol 10, 295-306.

[00125] Helm, P. A., Younes, L., Beg, M. F., Ennis, D. B., Leclercq, C, Faris, O. P., McVeigh, E., Kass, D., Miller, M. I. and Winslow, R. L. (2006) Evidence of Structural Remodeling in the Dyssynchronous Failing Heart. Circ Res 98, 125-32.

[00126] Helm, R. H., Byrne, M., Helm, P. A., Daya, S. K., Osman, N. F., Tunin, R., Halperin, H. R., Berger, R. D., Kass, D. A. and Lardo, A. C. (2007) Three-Dimensional Mapping of Optimal Left Ventricular Pacing Site for Cardiac Resynchronization. Circulation 115, 953-61.

[00127] Hooks, D. A., Trew, M. L., Caldwell, B. J., Sands, G. B., LeGrice, I. J. and Smaill, B. H. (2007) Laminar Arrangement of Ventricular Myocytes Influences Electrical Behavior of the Heart. Circ Res 101, el 03- 12.

[00128] Howard, E. J., Covell, J. W., Mulligan, L. J., McCulloch, A. D., Omens, J. H. and Kerckhoffs, R. C. (2011) Improvement in pump function with endocardial biventricular pacing increases with activation time at the left ventricular pacing site in failing canine hearts. Am J Physiol Heart Circ Physiol 301, H1447-55.

[00129] Kaab, S., Nuss, H. B., Chiamvimonvat, N., O'Rourke, B., Pak, P. H., Kass, D. A., Marban, E. and Tomaselli, G. F. (1996) Ionic Mechanism of Action Potential Prolongation in Ventricular Myocytes from Dogs with Pacing-Induced Heart Failure. Circ Res IS, 262-73.

[00130] Kass, D. A. (2005) Cardiac Resynchronization Therapy. J Cardiovasc Electrophysiol 16 Suppl 1, S35-41.

[00131] Leclercq, C, Faris, O., Tunin, R., Johnson, J., Kato, R., Evans, F., Spinelli, J., Halperin, H., McVeigh, E. and Kass, D. A. (2002) Systolic improvement and mechanical resynchronization does not require electrical synchrony in the dilated failing heart with left bundle-branch block. Circulation 106, 1760-3.

[00132] Leclercq, F., Hager, F. X., Macia, J. C, Mariottini, C. J., Pasquie, J. L. and Grolleau, R. (1999) Left Ventricular Lead Insertion Using a Modified Transseptal Catheterization Technique: A Totally Endocardial Approach for Permanent Biventricular Pacing in End-Stage Heart Failure. Pacing Clin Electrophysiol 22, 1570-5.

[00133] LeGrice, I. J., Takayama, Y. and Covell, J. W. (1995) Transverse Shear Along Myocardial Cleavage Planes Provides a Mechanism for Normal Systolic Wall Thickening. Circ Res 77, 182-93.

[00134] Lindner, O., Sorensen, J., Vogt, J., Fricke, E., Bailer, D., Horstkotte, D. and Burchert, W. (2006) Cardiac efficiency and oxygen consumption measured with 1 lC-acetate PET after long-term cardiac resynchronization therapy. JNucl Med 47, 378-83.

[00135] Lindner, O., Vogt, J., Kammeier, A., Wielepp, P., Holzinger, J., Bailer, D., Lamp, B., Hansky, B., Korfer, R., Horstkotte, D. and Burchert, W. (2005) Effect of cardiac resynchronization therapy on global and regional oxygen consumption and myocardial blood flow in patients with non-ischaemic and ischaemic cardiomyopathy. Eur Heart J 26, 70-6. 4759

[00136] Lloyd- Jones, D., Adams, R., Carnethon, M., De Simone, G., Ferguson, T. B., Flegal, ., Ford, E., Furie, K., Go, A., Greenlund, K., Haase, N., Hailpern, S., Ho, M., Howard, V., Kissela, B., Kittner, S., Lackland, D., Lisabeth, L., Marelli, A., McDermott, M., Meigs, J., Mozaffarian, D., Nichol, G., O'Donnell, C, Roger, V., Rosamond, W., Sacco, R., Sorlie, P., Stafford, R., Steinberger, J., Thorn, T., Wasserthiel-Smoller, S., Wong, N., Wylie- Rosett, J. and Hong, Y. (2009) Heart Disease and Stroke Statistics-2009 Update: A Report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation 119, 480-6.

[00137] Marijianowski, M. M., Teeling, P., Mann, J. and Becker, A. E. (1995) Dilated Cardiomyopathy Is Associated with an Increase in the Type I/Type III Collagen Ratio: A Quantitative Assessment. J Am Coll Cardiol 25, 1263-72.

[00138] Miyazaki, C, Redfield, M. M., Powell, B. D., Lin, G. M., Herges, R. M., Hodge, D. O., Olson, L. J., Hayes, D. L., Espinosa, R. E., Rea, R. F., Bruce, C. J., Nelson, S. M., Miller, F. A. and Oh, J. K. (2010) Dyssynchrony indices to predict response to cardiac resynchronization therapy: a comprehensive prospective single-center study. Circ Heart Fail 3, 565-73.

[00139] Nelson, G. S., Berger, R. D., Fetics, B. J., Talbot, M., Spinelli, J. C, Hare, J. M. and Kass, D. A. (2000) Left Ventricular or Biventricular Pacing Improves Cardiac Function at Diminished Energy Cost in Patients with Dilated Cardiomyopathy and Left Bundle-Branch Block. Circulation 102, 3053-9.

[00140] O'Rourke, B., Kass, D. A., Tomaselli, G. F., Kaab, S., Tunin, R. and Marban, E. (1999) Mechanisms of Altered Excitation-Contraction Coupling in Canine Tachycardia- Induced Heart Failure, I: Experimental Studies. Circ Res 84, 562-70.

[00141] Pitzalis, M. V., Iacoviello, M., Romito, R., Massari, F., Rizzon, B., Luzzi, G., Guida, P., Andriani, A., Mastropasqua, F. and Rizzon, P. (2002) Cardiac Resynchronization Therapy Tailored by Echocardiographic Evaluation of Ventricular Asynchrony. J Am Coll Cardiol 40, 1615-22. 12 024759

[00142] Rice, J. J., Wang, F., Bers, D. M. and de Tombe, P. P. (2008) Approximate model of cooperative activation and crossbridge cycling in cardiac muscle using ordinary differential equations. Biophys J 95, 2368-90.

[00143] Russell, K., Smiseth, O. A., Gjesdal, O., Qvigstad, E., Norseng, P. A., Sjaastad, I., Opdahl, A., Skulstad, H., Edvardsen, T. and Remme, E. W. (2011) Mechanism of prolonged electromechanical delay in late activated myocardium during left bundle branch block. Am J Physiol Heart Circ Physiol 301, H2334-43.

[00144] Spragg, D. D., Dong, J., Fetics, B. J., Helm, R., Marine, J. E., Cheng, A., Henrikson, C. A., Kass, D. A. and Berger, R. D. (2010) Optimal left ventricular endocardial pacing sites for cardiac resynchronization therapy in patients with ischemic cardiomyopathy. J Am Coll Cardiol 56, 774-81.

[00145] St John Sutton, M. G., Plappert, T., Abraham, W. T., Smith, A. L., DeLurgio, D. B., Leon, A. R., Loh, E., Kocovic, D. Z., Fisher, W. G., Ellestad, M., Messenger, J., Kruger, K., Hilpisch, K. E. and Hill, M. R. (2003) Effect of Cardiac Resynchronization Therapy on Left Ventricular Size and Function in Chronic Heart Failure. Circulation 107, 1985-90.

[00146] Suffoletto, M. S., Dohi, K., Cannesson, M., Saba, S. and Gorcsan, J., 3rd. (2006) Novel speckle-tracking radial strain from routine black-and-white echocardiographic images to quantify dyssynchrony and predict response to cardiac resynchronization therapy. Circulation 113, 960-8.

[00147] Suga, H. (1990) Ventricular energetics. Physiol Rev 70, 247-77.

[00148] Sutton, M. G., Plappert, T., Hilpisch, K. E., Abraham, W. T., Hayes, D. L. and Chinchoy, E. (2006) Sustained Reverse Left Ventricular Structural Remodeling with Cardiac Resynchronization at One Year Is a Function of Etiology: Quantitative Doppler Echocardiographic Evidence from the Multicenter Insync Randomized Clinical Evaluation (MIRACLE). Circulation 113, 266-72. [00149] Tyberg, J. V., Yeatman, L. A., Parmley, W. W., Urschel, C. W. and Sonnenblick, E. H. (1970) Effects of hypoxia on mechanics of cardiac contraction. Am J Physiol 218, 1780-8.

[00150] Ukkonen, H., Beanlands, R. S., Burwash, L G., de Kemp, R. A., Nahmias, C, Fallen, E., Hill, M. R. and Tang, A. S. (2003) Effect of cardiac ^synchronization on myocardial efficiency and regional oxidative metabolism. Circulation 107, 28-31.

[00151] Usyk, T. P. and McCulloch, A. D. (2003) Relationship between Regional Shortening and Asynchronous Electrical Activation in a Three-Dimensional Model of Ventricular Electromechanics. JCardiovasc Electrophysiol 14, SI 96-202.

[00152] Vadakkumpadan, F., Arevalo, H., Prassl, A. J., Chen, J., Kickinger, F., Kohl, P., Plank, G. and Trayanova, N. (2010) Image-based models of cardiac structure in health and disease. Wiley Interdiscip Rev Syst Biol Med 2, 489-506.

[00153] van Deursen, C, van Geldorp, I. E., Rademakers, L. M., van Hunnik, A., Kuiper, M., Klersy, C, Auricchio, A. and Prinzen, F. W. (2009) Left ventricular endocardial pacing improves resynchronization therapy in canine left bundle-branch hearts. Circ Arrhythm Electrophysiol 2, 580-7.

[00154] Walker, J. C, Ratcliffe, M. B., Zhang, P., Wallace, A. W., Fata, B., Hsu, E. W., Saloner, D. and Guccione, J. M. (2005) MRI-based finite-element analysis of left ventricular aneurysm. Am J Physiol Heart Circ Physiol 289, H692-700.

[00155] White, J. A., Yee, R., Yuan, X., Krahn, A., Skanes, A., Parker, M., Klein, G. and Drangova, M. (2006) Delayed Enhancement Magnetic Resonance Imaging Predicts Response to Cardiac Resynchronization Therapy in Patients with Intraventricular Dyssynchrony. J Am Coll Cardiol 48, 1953-60.

[00156] Wu, Y., Bell, S. P., Trombitas, K., Witt, C. C, Labeit, S., LeWinter, M. M. and Granzier, H. (2002) Changes in Titin Isoform Expression in Pacing-Induced Cardiac Failure Give Rise to Increased Passive Muscle Stiffness. Circulation 106, 1384-9. [00157] Yu, C. M., Chau, E., Sanderson, J. E., Fan, K., Tang, M. O., Fung, W. H., Lin, H., Kong, S. L., Lam, Y. M., Hill, M. R. and Lau, C. P. (2002) Tissue Doppler Echocardiographic Evidence of Reverse Remodeling and Improved Synchronicity by Simultaneously Delaying Regional Contraction after Biventricular Pacing Therapy in Heart Failure. Circulation 105, 438-45.

[00158] The above provides some examples according to particular embodiments of the current invention. The broad concepts of the current invention are not limited to only these particular examples. More generally, a method of planning a patient-specific cardiac procedure according to an embodiment of the current invention includes receiving three- dimensional imaging data of a patient's heart, simulating at least one of electrophysiological or electromechanical activity of at least a portion of the patient's heart using the three- dimensional imaging data, and planning the patient-specific cardiac procedure based on the simulating. The cardiac procedure is for providing a preselected alteration of at least one of electrophysiological or electromechanical behavior of the patient's heart.

[00159] The three-dimensional imaging data can be MRI data as described in the examples above. However, the broad concepts of the current invention are not limited to that particular example. The three-dimensional imaging data can be can be at least one of magnetic resonance imaging (MRI), computed tomography (CT), positron emission tomography (PET), ultrasound, or nuclear tracer three-dimensional imaging data, for example. The method of planning a patient-specific cardiac procedure can further include receiving additional patient-specific data in addition to the three-dimensional imaging data. For example, some embodiments can include receiving at least one of biopsy data, electrocardiogram data, recorded data from an implantable device (pace maker, defibrillator, etc.), or invasive electrical mapping data (e.g., endoscopic). The simulating can then use the additional patient-specific data for the simulation.

[00160] The simulating at least one of electrophysiological or electromechanical activity of at least the portion of the patient's heart can include constructing a geometric model of the portion of the patient's heart. The geometric model can include normal tissue regions and remodeled tissue regions that are determined for the patient using the three- dimensional imaging data. The term "remodeled tissue" can include infarct scar, infarct border (gray) zone, fibrosis, or other disease-related structural, electrophysiological or contractile changes in the heart. The simulating can further include estimating tissue fiber orientations in the geometric model of the portion of the patient's heart. The estimation of fiber orientations can be done in a variety of ways. For example, the fiber orientations can be calculated using a Laplace-Dirichlet method to define the local transmural and apicobasal axes at each spatial location in the ventricles , (J. D. Bayer, R. Blake, G. Plank, Trayanova N, Novel rule based algorithm for assigning myocardial fiber orientation to computation heart models. Ann Biomed Eng. , (in review) (2012), the entire contents of which are incorporated herein by reference). Another approach could utilize pre-compiled data (i.e., atlas data), which can be mapped into the specific size and shape of the patient's heart (Image-Based Estimation of Ventricular Fiber Orientations for Personalized Modeling of Cardiac Electrophysiology, Vadakkumpadan F, Arevalo H, Ceritoglu C, Miller M, Trayanova N., IEEE Trans Med Imaging. 2012 Jan 18. [Epub ahead of print], the entire contents of which are incorporated herein by reference).

[00161] A method of planning a patient-specific cardiac procedure according to an embodiment of the current invention can be directed to planning an ablation procedure to alleviate a ventricular arrhythmia. In this embodiment, the geometric model of the at least said portion of the patient's heart includes at least a geometric model of right and left ventricles of the patient's heart. The remodeled tissue regions in this case can be segmented into a plurality of different regions based on the three-dimensional imaging data. The plurality of different regions can include scar tissue regions, normal tissues regions, and transition zones, for example, between normal and scar tissue regions. The transition zones include infarct border zone tissue regions (we also refer to these zones as GZ, gray zones). The simulating in this case can be simulating electrophysiological activity of at least the right and left ventricles of the patient's heart. The ventricular arrhythmia can be ventricular tachycardia or ventricular fibrillation, for example. In an embodiment of the current invention, the planning of the patient-specific cardiac procedure includes identifying organizing centers of ventricular tachycardia from the simulation of electrophysiological activity. In some embodiments, the planning the patient-specific cardiac procedure can include identifying a critical pathway for ventricular tachycardia or three-dimensional scroll- wave filaments corresponding to the organizing centers as well as other slow conducting pathways through and around areas of scar tissue that are a part of a ventricular tachycardia circuit and further planning ablation to coincide with at least a portion of the three- dimensional scroll-wave filaments or other critical pathways for the ventricular tachycardia. (See above for some specific examples.)

[00162] In a method of planning a patient-specific cardiac procedure according to other embodiments of the current invention, the geometric model of the portion of the patient's heart includes a geometric model of at least right and left atria of the patient's heart. In this embodiment, the remodeled tissue regions are fibrotic tissue regions. The simulating at least one of electrophysiological or electromechanical activity is simulating electrophysiological activity of at least the right and left atria of the patient's heart. This embodiment is for planning an ablation procedure to alleviate atrial fibrillation.

[00163] In another embodiment, the simulating at least one of electrophysiological or electromechanical activity is simulating electromechanical activity of at least a portion of the patient's heart. The planning can include determining whether the patient is a suitable candidate for cardiac resynchronization therapy. If the patient is a suitable candidate for cardiac resynchronization therapy, further embodiments can include planning a location in which to attach at least one pacing lead for cardiac resynchronization therapy. Further embodiments can include, planning the location in which to attach at least one pacing lead for cardiac resynchronization therapy based on regions of longest electromechanical delay or regions of latest electrical or mechanical activation as determined from the simulation. Further embodiments can include, planning the location in which to attach at least one pacing lead for cardiac resynchronization therapy based on local or global energy consumption or myocardial efficiency, as determined from the simulation. Myocardial efficiency is the ratio of mechanical work performed by the ventricles to myocardial energy consumption.

[00164] The embodiments discussed in this specification are intended to explain concepts of the invention. However, the invention is not intended to be limited to the specific terminology selected and the particular examples described. The above-described embodiments of the invention may be modified or varied, without departing from the invention, as appreciated by those skilled in the art in light of the above teachings. It is therefore to be understood that, within the scope of the claims and their equivalents, the invention may be practiced otherwise than as specifically described.