Database labeled with subcellular pathologies developed by multi-scale heart simulator, UT-Heart

12 lead ECG DATA : ECG1,ECG2,ECG3,ECG4,ECG5CLICK

Echocardiographic parameters list : UCG SummaryCLICK

We have developed a multi-scale multi-physics heart simulator 'UT-Heart', in which more than 20,000,000 cell models of cardiac electrophysiology are implemented while reproducing the microscopic tissue structure in a realistic 3D heart model [1-3](for more information please visit our website at http://ut-heart.com/). In addition, the human ventricular myocyte model of electrophysiology by O’Hara et al[4, 5] was coupled with the spatially-detailed cardiac sarcomere model of our own[6] to simulate the contractile behavior of the heart. The details of the model can be seen in the literature listed below. Using this heart model, we created an ECG dataset by repeating the beating heart simulations while varying the activities of functional molecules, tissue structure, morphology, and the loading condition of the heart under normal and failing conditions as reported in the literatures. Concomitant changes in pumping function were evaluated and summarized in the format of echocardiographic parameters. File name of each 12 lead ECG data was formatted to indicate how each simulation parameter was modified (please see the following example).

Example of filename：ECG.Kr1Ks1to1K11Na1Ca0NCX1SR1MK0G0LV2EX0CELL0.out

Kr* : Conductances of potassium current (IKr) was set at (0) normal or (1) 70% of normal.
Ks* : Conductances of potassium current (IKs) was set at (0) normal or (1) 60% of normal.
to* : Conductances of potassium current (Ito) was set at (0) normal or (1) 60% of normal.
K1* : Conductances of potassium current (IK1) was set at (0) normal or (1) 70% of normal.
Na*: Conductance of sodium current (INa/INaL) were set at (0) normal or (1) 80%/200% of normal.
Ca*: Conductance of L-type calcium current (ICa,L) was set at (0) normal or (1) 80% of normal.
NCX*: The activity of sodium-calcium exchanger current (NCX) was set at normal (0) or 200% of normal.
SR*: The activity of Calcium Uptake via SERCA Pump (Jup) was set at normal (0) or 80% of normal.
MK*: The fraction of active CaMK binding sites at equilibrium and Ca2+ leakage from the NSR were set at normal (0) or 150% of normal.
G*: Conduction velocity in fiber/cross-fiber directions was set at either 100%/100% (0) or 120%/80% (1) of normal.
LV*: The morphology of LV was quantified by the sphericity index (SI=long axis length/diameter) and set at either 0.53, 0.68 or 0.48 for normal (0), dilated (1) and hypertrophic (2) models respectively The thickness of LV was 10mm for the normal model and 13mm for the hypertrophic model. Only for the hypertrophic model (LV2), conduction velocity (fiber/cross fiber) was set at 130%/100%.
EX*: Ventricular activation pattern was modified as normal (0), left axis deviation (1), right axis deviation (2), left anterior fascicular block (3) and an incomplete right bundle branch block (4).
CELL*: We arranged the three-type of ventricular cell models, i.e., endocardial cell, M-cell, and epicardial cell. From endocardial side to epicardial side, cells were distributed in the following manner (in volume ratio). 10%/75%/15% (0), 100%/0%/0% (1).

Each file contains digitized ECG data sampled at 500 Hz, with 12 lead of data in column wise. Numeric data can be converted to mV by multiplying 0.00488.

References
[1] Sugiura S, Okada J, Washio T, Hisada T. UT-Heart: A Finite Element Model Designed for the Multiscale and Multiphysics Integration of our Knowledge on the Human Heart. In: Cortassa S, Aon MA, editors. Computational Systems Biology in Medicine and Biotechnology: Methods and Protocols. New York, NY: Springer US; 2022. p. 221-45.
[2] Okada J, Yoshinaga T, Kurokawa J, Washio T, Furukawa T, Sawada K, Sugiura S, Hisada T. Screening system for drug-induced arrhythmogenic risk combining a patch clamp and heart simulator. Science Advances, Vol.1, e1400142. 2015.
[3] Okada J, Yoshinaga T, Kurokawa J, Washio T, Furukawa T, Sawada K, Sugiura S, Hisada T. Arrhythmic hazard map for a 3D whole-ventricle model under multiple ion channel block. British Journal of Pharmacology, Vol.175, pp.3435-52. 2018.
[4] O'Hara T, Virag L, Varro A, Rudy Y. Simulation of the undiseased human cardiac ventricular action potential: model formulation and experimental validation. PLoS Computational Biology, Vol.7, e1002061. 2011.
[5] Dutta S, Strauss D, Colatsky T, Li Z, editors. Optimization of an in silico cardiac cell model for proarrhythmia risk assessment. 2016 Computing in Cardiology Conference (CinC); 2016 11-14 Sept. 2016.
[6] Washio T, Okada J, Sugiura S, Hisada T. Approximation for Cooperative Interactions of a Spatially-Detailed Cardiac Sarcomere Model. Cellular and Molecular Bioengineering, Vol.5, pp.113-26. 2012.

This work was supported by the Ministry of Education, Culture, Sports, Science and Technology (MEXT) under the “Program for Promoting Research on the Supercomputer Fugaku” （hp200121, hp210180, hp220178).