A High-Fidelity Cardiac Digital Twin for Assessing Rate-Dependent Responses: A Comparative Multi-scale Study of TT2 and Tomek Models in Stratified 3D Ventricles

Q: What is the clinical evidence from this study?

Population: Myocardial Infarction. Intervention: Tomek human ventricular ionic model vs. ten Tusscher-Panfilov (TT2) human ventricular ionic model. Primary outcome: Mean inter-layer action potential duration at 80% repolarization (APD80) in 3D ventricular model at BCL 4000 ms.

Q: What does this research mean for the field?

Tissue-level electrical coupling in three-dimensional ventricles substantially alters intrinsic cellular electrophysiological dynamics across pacing rates, producing divergent multi-scale fidelity between TT2 and Tomek ionic models. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

Zhenyin Fu; Zhaokai Kong; Lucheng Xu; Mingfeng Jiang; Ling Xia; Wenlong Xu; Ruiqing Dong; Dongdong Deng

doi:10.48550/arxiv.2611.14995

November 11, 2026Open Access

A High-Fidelity Cardiac Digital Twin for Assessing Rate-Dependent Responses: A Comparative Multi-scale Study of TT2 and Tomek Models in Stratified 3D Ventricles

View Full Paper

Ask AI

Bookmark

Share

Key result

Tomek ventricular model increases inter-layer APD80 by ~35% with rate slowing, outperforming TT2.

Why the study?

The predictive behavior of cardiac digital twins depends strongly on the selected ionic model and the representation of transmural electrophysiological heterogeneity.

Does the Tomek ionic model differ from the ten Tusscher-Panfilov (TT2) model in simulating rate-dependent action potential duration (APD80) and transmural repolarization gradients across single-cell and 3D patient-specific ventricular models?

Key Points

To evaluate how different cellular ionic formulations (TT2 versus Tomek) and tissue-level electrical coupling affect rate-dependent electrophysiological responses in a multiscale cardiac digital twin.
Reconstructed a 3D patient-specific ventricular mesh at approximately 0.4 mm resolution from 3.0 T CMR-LGE imaging data.
Applied Laplace-Dirichlet transmural coordinate mapping to stratify the ventricular wall into epicardial, mid-myocardial, and endocardial domains.
Simulated single-cell (0D) and 3D ventricles in openCARP across pacing basic cycle lengths from 600 ms to 4000 ms using simultaneous endocardial stimulation.
Tomek single cells maintained a more hyperpolarized resting membrane potential (-88.34 mV to -88.92 mV) than TT2 single cells (-84.90 mV to -86.32 mV).
TT2 single-cell models showed divergent action potential duration gradients across cell layers (∆APD of 70.88 ms at 500 ms cycle length and 86.18 ms at 600 ms cycle length), whereas Tomek displayed synchronous prolongation across all cell layers.
Tissue-level electrotonic coupling substantially reshaped the intrinsic repolarization dynamics observed at the isolated single-cell scale across both model frameworks.

Structured PICO

Does the Tomek ionic model differ from the ten Tusscher-Panfilov (TT2) model in simulating rate-dependent action potential duration (APD80) and transmural repolarization gradients across single-cell and 3D patient-specific ventricular models?

P

Population

Patient-specific 3D ventricular computational models reconstructed from cardiac MRI of myocardial infarction patients, evaluated across single-cell and tissue scales.

I

Intervention

Tomek human ventricular ionic model implemented with transmural phenotype assignments (epicardial, mid-myocardial, and endocardial layers) and multi-point Simultaneous Endocardial Stimulation (SES) across basic cycle lengths (BCLs) from 500 to 5000 ms (600 to 4000 ms in 3D simulations).

C

Comparator

ten Tusscher-Panfilov (TT2) human ventricular ionic model implemented under identical 0D and 3D stratified ventricular configurations and pacing protocols, alongside published experimental canine ventricular data for qualitative benchmarking.

O

Outcome

Rate-dependent action potential duration at 80% repolarization (APD80) and transmural APD gradient (ΔAPD = APDmax − APDmin) across ten transmural layers.surrogate

Main Result

Absolute Event Rate: 346.68% vs 345.82%

Tissue-level electrotonic coupling in 3D cardiac digital twins substantially reshapes single-cell rate-dependent action potential duration and transmural dispersion profiles, demonstrating critical fidelity differences between the TT2 and Tomek ionic models.

Limitations

Experimental benchmark data used canine hearts and APD60, whereas computational simulations modeled human ventricles and measured APD80, allowing only qualitative comparisons.
Transmural partition ratio was held fixed across all simulations rather than re-optimized for each ionic model.
The Purkinje network was not explicitly reconstructed, relying instead on a multi-point Simultaneous Endocardial Stimulation (SES) approximation.
Experimental canine-heart benchmark data used APD60 whereas the present simulations used APD80, preventing direct comparison of absolute APD values or calculation of pointwise errors.
At extended BCLs (2000-5000 ms), the Tomek model manifests a sustained APD prolongation that diverges from the stable plateau behavior typically observed in experimental preparations.
The His-Purkinje system was approximated using a multi-point Simultaneous Endocardial Stimulation (SES) strategy rather than explicitly reconstructing the Purkinje network.

Ask about this paper

Answers are grounded in this paper and cite their sources.

Implication

Tomek versus TT2 choice markedly alters simulated rate-dependent repolarization in 3D models; leaves open optimal formulation for clinical digital twins.

Abstract

Cardiac digital twins (CDTs) provide a computational platform for personalized cardiac modeling, but their predictive behavior depends strongly on the selected ionic model and the representation of transmural electrophysiological heterogeneity. In this study, we developed a multiscale CDT validation framework that integrates physiologically optimized hierarchical configurations. Using single-cell (0D) models and three-dimensional (3D) patient-specific ventricular models, we implemented the ten Tusscher-Panfilov model (TT2) and Tomek ionic model and compared their electrophysiological responses across a range of pacing frequencies, with basic cycle lengths (BCLs) from 600 to 4000 ms. The results show that tissue-level electrical coupling substantially reshapes the intrinsic dynamical properties observed at the single-cell level. By benchmarking model outputs against clinical data, this study quantifies fidelity differences between the two ionic models across multiple scales and provides a scientific basis for ionic model selection and CDT construction.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Fu et al. (2026) studied Myocardial Infarction. Tomek human ventricular ionic model vs. ten Tusscher-Panfilov (TT2) human ventricular ionic model was evaluated on Mean inter-layer action potential duration at 80% repolarization (APD80) in 3D ventricular model at BCL 4000 ms. In 3D ventricular models, the Tomek model increased mean inter-layer APD80 by 35.11% from BCL 600 to 4000 ms, compared to 7.79% with the TT2 model.

synapsesocial.com/papers/6af88c530e4d8f4b089c1bb5 https://doi.org/10.48550/arxiv.2611.14995

View Full Paper

Ask AI

Bookmark

Share

A High-Fidelity Cardiac Digital Twin for Assessing Rate-Dependent Responses: A Comparative Multi-scale Study of TT2 and Tomek Models in Stratified 3D Ventricles

Key Points

Structured PICO

Main Result

Limitations

Ask about this paper

Implication

Abstract

Discussion

Cite This Study

A High-Fidelity Cardiac Digital Twin for Assessing Rate-Dependent Responses: A Comparative Multi-scale Study of TT2 and Tomek Models in Stratified 3D Ventricles

Key Points

Structured PICO

Main Result

Limitations

Ask about this paper

Implication

Abstract

Discussion

Cite This Study

Keep reading

Keep reading