In Silico Classifiers for the Assessment of Drug Proarrhythmicity

In silico cardiac safety profile of drugs and their potential to induce clinical cardiotoxicity

During the cardiac safety pharmacology decision-making process, a precise evaluation of the proarrhythmic liabilities of new drug candidates is essential to prevent serious side effects like torsade de pointes (TdP), which could result in sudden death. Based in part on in silico reconstruction of the human ventricular cardiomyocyte action potential (AP), the Comprehensive in vitro Proarrhthymia Assay (CiPA) initiative aims to achieve this goal. Because it is based on a significant amount of human data, the O'Hara-Rudy dynamic model is the first algorithm approved by the CiPA initiative for AP modeling. This algorithm was used to build a new database (www.scaptest.com for Safe Cardiac Action Potential test) in order to fully describe the in silico cardiac safety profile of a very large set of 200 compounds based on their concentration required to induce effects on AP shape and time course (resting membrane potential, AP maximal amplitude, AP duration) as well as on various predictive safety parameters extrapolated from this AP (triangulation, transmural dispersion of repolarization, reverse use dependence, likelihood of inducing early afterdepolarization (EAD), occurrence of EAD, threshold leading to EAD). This description of the in silico cardiac safety profile helps warn against the use of certain compounds possibly inducing cardiac safety issues (such as TdP or inexcitability) at least at high concentrations.

Elevating performance and interpretability of in silico classifiers for drug proarrhythmia risk evaluations using multi-biomarker approach with ranking algorithm

BACKGROUND AND OBJECTIVE

Using electrophysiological simulations and machine learning to predict drug proarrhythmia risk has gained popularity due to its effectiveness. The leading in silico drug assessment system mainly uses a single biomarker (qNet) to predict proarrhythmia risk, offering good performance and straightforward interpretation. Other advanced classifiers incorporating additional physiological biomarkers provide better predictive capabilities but are less intuitive. Thus, a method that accommodates multiple biomarkers while maintaining interpretability is needed.

METHODS

We enhance the current best ordinal logistic regression (OLR) model by adding more physiological biomarkers to overcome its limitations. We also introduce a general torsade metric score (TMS) for multi-biomarker approaches to facilitate easier interpretation. Additionally, a novel ranking algorithm based on a simple multi-criteria decision analysis method is employed to evaluate various classifiers against standard proarrhythmia risk criteria efficiently.

RESULTS

Our proposed method demonstrates that using multiple well-known biomarkers yields better performance than using qNet alone. Some accepted multi-biomarker OLR models do not incorporate qNet yet outperform those that do. Moreover, some ill-performing biomarkers when utilized individually can show improved performance in combination with other biomarkers.

CONCLUSION

The proposed approach offers an effective way of utilizing multiple biomarkers, including well-known ones, providing practical alternatives for proarrhythmia risk assessment. The interpretability of the accepted models is straightforward, thanks to the TMS thresholds for multi-biomarker OLR models that allow direct evaluation of the classification prediction of individual drugs.

Evaluation of cardiac pro-arrhythmic risks using the artificial neural network with ToR-ORd in silico model output

Torsades de pointes (TdP) is a type of ventricular arrhythmia that can lead to sudden cardiac death. Drug-induced TdP has been an important concern for researchers and international regulatory boards. The Comprehensive in vitro Proarrhythmia Assay (CiPA) initiative was proposed that integrates in vitro testing and computational models of cardiac ion channels and human cardiomyocyte cells to evaluate the proarrhythmic risk of drugs. The TdP risk classification performance using only a single TdP metric may require some improvements because of information limitations and the instability of generalizing results. This study evaluates the performance of TdP metrics from the in silico simulations of the Tomek-O'Hara Rudy (ToR-ORd) ventricular cell model for classifying the TdP risk of drugs. We utilized these metrics as an input to an artificial neural network (ANN)-based classifier. The ANN model was optimized through hyperparameter tuning using the grid search (GS) method to find the optimal model. The study outcomes show an area under the curve (AUC) value of 0.979 for the high-risk category, 0.791 for the intermediate-risk category, and 0.937 for the low-risk category. Therefore, this study successfully demonstrates the capability of the ToR-ORd ventricular cell model in classifying the TdP risk into three risk categories, providing new insights into TdP risk prediction methods.

HPC Framework for Performing in Silico Trials Using a 3D Virtual Human Cardiac Population as Means to Assess Drug-Induced Arrhythmic Risk

Following the 3 R's principles of animal research-replacement, reduction, and refinement-a high-performance computational framework was produced to generate a platform to perform human cardiac in-silico clinical trials as means to assess the pro-arrhythmic risk after the administrations of one or combination of two potentially arrhythmic drugs. The drugs assessed in this study were hydroxychloroquine and azithromycin. The framework employs electrophysiology simulations on high-resolution three-dimensional, biventricular human heart anatomies including phenotypic variabilities, so as to determine if differential QT-prolongation responds to drugs as observed clinically. These simulations also reproduce sex-specific ionic channel characteristics. The derived changes in the pseudo-electrocardiograms, calcium concentrations, as well as activation patterns within 3D geometries were evaluated for signs of induced arrhythmia. The virtual subjects could be evaluated at two different cycle lengths: at a normal heart rate and at a heart rate associated with stress as means to analyze the proarrhythmic risks after the administrations of hydroxychloroquine and azithromycin. Additionally, a series of experiments performed on reanimated swine hearts utilizing Visible Heart® methodologies in a four-chamber working heart model were performed to verify the arrhythmic behaviors observed in the in silico trials.The obtained results indicated similar pro-arrhythmic risk assessments within the virtual population as compared to published clinical trials (21% clinical risk vs 21.8% in silico trial risk). Evidence of transmurally heterogeneous action potential prolongations after providing a large dose of hydroxychloroquine was found as the observed mechanisms for elicited arrhythmias, both in the in vitro and the in silico models. The proposed workflow for in silico clinical drug cardiotoxicity trials allows for reproducing the complex behavior of cardiac electrophysiology in a varied population, in a matter of a few days as compared to the months or years it requires for most in vivo human clinical trials. Importantly, our results provided evidence of the common phenotype variants that produce distinct drug-induced arrhythmogenic outcomes.

Assessing drug safety by identifying the axis of arrhythmia in cardiomyocyte electrophysiology

Many classes of drugs can induce fatal cardiac arrhythmias by disrupting the electrophysiology of cardiomyocytes. Safety guidelines thus require all new drugs to be assessed for pro-arrhythmic risk prior to conducting human trials. The standard safety protocols primarily focus on drug blockade of the delayed-rectifier potassium current (IKr). Yet the risk is better assessed using four key ion currents (IKr, ICaL, INaL, IKs). We simulated 100,000 phenotypically diverse cardiomyocytes to identify the underlying relationship between the blockade of those currents and the emergence of ectopic beats in the action potential. We call that relationship the axis of arrhythmia. It serves as a yardstick for quantifying the arrhythmogenic risk of any drug from its profile of multi-channel block alone. We tested it on 109 drugs and found that it predicted the clinical risk labels with an accuracy of 88.1-90.8%. Pharmacologists can use our method to assess the safety of novel drugs without resorting to animal testing or unwieldy computer simulations.

Combining pharmacokinetic and electrophysiological models for early prediction of drug-induced arrhythmogenicity

BACKGROUND AND OBJECTIVE

In silico methods are gaining attention for predicting drug-induced Torsade de Pointes (TdP) in different stages of drug development. However, many computational models tended not to account for inter-individual response variability due to demographic covariates, such as sex, or physiologic covariates, such as renal function, which may be crucial when predicting TdP. This study aims to compare the effects of drugs in male and female populations with normal and impaired renal function using in silico methods.

METHODS

Pharmacokinetic models considering sex and renal function as covariates were implemented from data published in pharmacokinetic studies. Drug effects were simulated using an electrophysiologically calibrated population of cellular models of 300 males and 300 females. The population of models was built by modifying the endocardial action potential model published by O'Hara et al. (2011) according to the experimentally measured gene expression levels of 12 ion channels.

RESULTS

Fifteen pharmacokinetic models for CiPA drugs were implemented and validated in this study. Eight pharmacokinetic models included the effect of renal function and four the effect of sex. The mean difference in action potential duration (APD) between male and female populations was 24.9 ms (p<0.05). Our simulations indicated that women with impaired renal function were particularly susceptible to drug-induced arrhythmias, whereas healthy men were less prone to TdP. Differences between patient groups were more pronounced for high TdP-risk drugs. The proposed in silico tool also revealed that individuals with impaired renal function, electrophysiologically simulated with hyperkalemia (extracellular potassium concentration [K+]o = 7 mM) exhibited less pronounced APD prolongation than individuals with normal potassium levels. The pharmacokinetic/electrophysiological framework was used to determine the maximum safe dose of dofetilide in different patient groups. As a proof of concept, 3D simulations were also run for dofetilide obtaining QT prolongation in accordance with previously reported clinical values.

CONCLUSIONS

This study presents a novel methodology that combines pharmacokinetic and electrophysiological models to incorporate the effects of sex and renal function into in silico drug simulations and highlights their impact on TdP-risk assessment. Furthermore, it may also help inform maximum dose regimens that ensure TdP-related safety in a specific sub-population of patients.

Machine learning approach to evaluate TdP risk of drugs using cardiac electrophysiological model including inter-individual variability

Introduction: Predicting ventricular arrhythmia Torsade de Pointes (TdP) caused by drug-induced cardiotoxicity is essential in drug development. Several studies used single biomarkers such as qNet and Repolarization Abnormality (RA) in a single cardiac cell model to evaluate TdP risk. However, a single biomarker may not encompass the full range of factors contributing to TdP risk, leading to divergent TdP risk prediction outcomes, mainly when evaluated using unseen data. We addressed this issue by utilizing multi-in silico features from a population of human ventricular cell models that could capture a representation of the underlying mechanisms contributing to TdP risk to provide a more reliable assessment of drug-induced cardiotoxicity. Method: We generated a virtual population of human ventricular cell models using a modified O'Hara-Rudy model, allowing inter-individual variation. IC 50 and Hill coefficients from 67 drugs were used as input to simulate drug effects on cardiac cells. Fourteen features (dVm dt repol , dVm dt max , Vm peak , Vm resting , APD tri , APD 90 , APD 50 , Ca peak , Ca diastole , Ca tri , CaD 90 , CaD 50 , qNet, qInward) could be generated from the simulation and used as input to several machine learning models, including k-nearest neighbor (KNN), Random Forest (RF), XGBoost, and Artificial Neural Networks (ANN). Optimization of the machine learning model was performed using a grid search to select the best parameter of the proposed model. We applied five-fold cross-validation while training the model with 42 drugs and evaluated the model's performance with test data from 25 drugs. Result: The proposed ANN model showed the highest performance in predicting the TdP risk of drugs by providing an accuracy of 0.923 (0.908-0.937), sensitivity of 0.926 (0.909-0.942), specificity of 0.921 (0.906-0.935), and AUC score of 0.964 (0.954-0.975). Discussion and conclusion: According to the performance results, combining the electrophysiological model including inter-individual variation and optimization of machine learning showed good generalization ability when evaluated using the unseen dataset and produced a reliable drug-induced TdP risk prediction system.

Uncertainty assessment of proarrhythmia predictions derived from multi-level in silico models

In silico methods can be used for an early assessment of arrhythmogenic properties of drug candidates. However, their use for decision-making is conditioned by the possibility to estimate the predictions' uncertainty. This work describes our efforts to develop uncertainty quantification methods for the predictions produced by multi-level proarrhythmia models. In silico models used in this field usually start with experimental or predicted IC50 values that describe drug-induced ion channel blockade. Using such inputs, an electrophysiological model computes how the ion channel inhibition, exerted by a drug in a certain concentration, translates to an altered shape and duration of the action potential in cardiac cells, which can be represented as arrhythmogenic risk biomarkers such as the APD90. Using this framework, we identify the main sources of aleatory and epistemic uncertainties and propose a method based on probabilistic simulations that replaces single-point estimates predicted using multiple input values, including the IC50s and the electrophysiological parameters, by distributions of values. Two selected variability types associated with these inputs are then propagated through the multi-level model to estimate their impact on the uncertainty levels in the output, expressed by means of intervals. The proposed approach yields single predictions of arrhythmogenic risk biomarkers together with value intervals, providing a more comprehensive and realistic description of drug effects on a human population. The methodology was tested by predicting arrhythmogenic biomarkers on a series of twelve well-characterised marketed drugs, belonging to different arrhythmogenic risk classes.

Insights from an electro-mechanical heart failure cell model: Role of SERCA enhancement on arrhythmogenesis and myocyte contraction

BACKGROUND AND OBJECTIVE

Structural and electrical remodeling in heart failure predisposes the heart to ventricular arrhythmias. Computer modeling approaches, used to complement experimental results, can provide a more mechanistic knowledge of the biophysical phenomena underlying cardiac pathologies. Indeed, previous in-silico studies have improved the understanding of the electrical correlates of heart failure involved in arrhythmogenesis; however, information on the crosstalk between electrical activity, intracellular Ca²⁺ and contraction is still incomplete. This study aims to investigate the electro-mechanical behavior of virtual failing human ventricular myocytes to help in the development of therapies, which should ideally target pump failure and arrhythmias at the same time.

METHODS

We implemented characteristic remodeling of heart failure with reduced ejection fraction by including reported changes in ionic conductances, sarcomere function and cell structure (e.g. T-tubules disarray). Model parametrization was based on published experimental data and the outcome of simulations was validated against experimentally observed patterns. We focused on two aspects of myocardial dysfunction central in heart failure: altered force-frequency relationship and susceptibility to arrhythmogenic early afterdepolarizations. Because biological variability is a major problem in the generalization of in-silico findings based on a unique set of model parameters, we generated and evaluated a population of models.

RESULTS

The population-based approach is crucial in robust identification of parameters at the core of abnormalities and in generalizing the outcome of their correction. As compared to non-failing ones, failing myocytes had prolonged repolarization, a higher incidence of early afterdepolarizations, reduced contraction and a shallower force-frequency relationship, all features peculiar of heart failure. Component analysis applied to the model population identified reduced SERCA function as a relevant contributor to most of these derangements, which were largely reverted or diminished by restoration of SERCA function alone.

CONCLUSIONS

These simulated results encourage the development of strategies comprising SERCA stimulation and highlight the need to evaluate both electrical and mechanical outcomes.

Application of machine learning to improve the efficiency of electrophysiological simulations used for the prediction of drug-induced ventricular arrhythmia

BACKGROUND AND OBJECTIVE

In silico prediction of drug-induced ventricular arrhythmia often requires computationally intensive simulations, making its application tedious and non-interactive. This inconvenience can be mitigated using matrices of precomputed simulation results, allowing instantaneous computation of biomarkers such as action potential duration at 90% of the repolarisation (APD₉₀). However, preparing such matrices can be computationally intensive for the method developers, limiting the range of simulated conditions. In this work, we aim to optimise the generation of these matrices so that they can be obtained with less effort and for a broader range of input values.

METHODS

Machine learning methods were applied, building models trained with only a small fraction of the originally simulated results. The predictive performances of the models were assessed by comparing their predicted values with the actual simulation results, using percentual mean absolute error and mean relative error, as well as the percentage of data with a relative error below 5%.

RESULTS

Our method obtained highly accurate estimations of the original values, leading to a nearly one hundred-fold decrease in computation time. This method also allows precomputing more complex matrices, describing the effect of more ion channels on the APD_90. The best results were obtained by applying Support Vector Machine models, which yielded errors below 1% in most cases. This approach was further validated by predicting the APD₉₀ of a set of 12 CiPA compounds and exporting the optimal settings for predicting APD₉₀ using a different set of ion channels, always with satisfactory results.

CONCLUSIONS

The proposed method effectively reduces the computational effort required to generate matrices of precomputed electrophysiological simulation values. The same approach can be applied in other fields where computationally costly simulations are applied repeatedly using slightly different input values.

Application of Convolutional Neural Networks Using Action Potential Shape for In-Silico Proarrhythmic Risk Assessment

This study proposes a convolutional neural network (CNN) model using action potential (AP) shapes as input for proarrhythmic risk assessment, considering the hypothesis that machine-learning features automatically extracted from AP shapes contain more meaningful information than do manually extracted indicators. We used 28 drugs listed in the comprehensive in vitro proarrhythmia assay (CiPA), consisting of eight high-risk, eleven intermediate-risk, and nine low-risk torsadogenic drugs. We performed drug simulations to generate AP shapes using experimental drug data, obtaining 2000 AP shapes per drug. The proposed CNN model was trained to classify the TdP risk into three levels, high-, intermediate-, and low-risk, based on in silico AP shapes generated using 12 drugs. We then evaluated the performance of the proposed model for 16 drugs. The classification accuracy of the proposed CNN model was excellent for high- and low-risk drugs, with AUCs of 0.914 and 0.951, respectively. The model performance for intermediate-risk drugs was good, at 0.814. Our proposed model can accurately assess the TdP risks of drugs from in silico AP shapes, reflecting the pharmacokinetics of ionic currents. We need to secure more drugs for future studies to improve the TdP-risk-assessment robustness.

Accelerating Digitalization in Healthcare with the InSilicoTrials Cloud-Based Platform: Four Use Cases

The use of in silico trials is expected to play an increasingly important role in the development and regulatory evaluation of new medical products. Among the advantages that in silico approaches offer, is that they permit testing of drug candidates and new medical devices using virtual patients or computational emulations of preclinical experiments, allowing to refine, reduce or even replace time-consuming and costly benchtop/in vitro/ex vivo experiments as well as the involvement of animals and humans in in vivo studies. To facilitate and widen the adoption of in silico trials, InSilicoTrials Technologies has developed a cloud-based platform, hosting healthcare simulation tools for different bench, preclinical and clinical evaluations, and for diverse disease areas. This paper discusses four use cases of in silico trials performed using the InSilicoTrials.com platform. The first application illustrates how in silico approaches can improve the early preclinical assessment of drug-induced cardiotoxicity risks. The second use case is a virtual reproduction of a bench test for the safety assessment of transcatheter heart valve substitutes. The third and fourth use cases are examples of virtual patients generation to evaluate treatment effects in multiple sclerosis and prostate cancer patients, respectively.

qInward variability-based in-silico proarrhythmic risk assessment of drugs using deep learning model

Many researchers have suggested evaluation methods and Torsades de Pointes (TdP) metrics to assess the proarrhythmic risk of a drug based on the in silico simulation, as part of the Comprehensive in-vitro Proarrhythmia Assay (CiPA) project. In the previous study, we validated the robustness of 12 in silico features using the ordinal logistic regression (OLR) model by comparing the classification performances of metrics according to the in-vitro experimental datasets used; however, the OLR model using 12 in silico features did not provide desirable results. This study proposed a convolutional neural network (CNN) model using the variability of promising in silico TdP metrics hypothesizing that the variability of in silico features based on beats has more information than the single value of in silico features. We performed the action potential (AP) simulation using a human ventricular myocyte model to calculate seven in silico features representing the electrophysiological cell states of drug effects over 1,000 beats: qNet, qInward, intracellular calcium duration at returning to 50% baseline (CaD50) and 90% baseline (CaD90), AP duration at 50% repolarization (APD50) and 90% repolarization (APD90), and dVm/dtMax_repol. The proposed CNN classifier was trained using 12 train drugs and tested using 16 test drugs among CiPA drugs. The torsadogenic risk of drugs was classified as high, intermediate, and low risks. We determined the CNN classifier by comparing the classification performance according to the variabilities of seven in silico biomarkers computed from the in silico drug simulation using the Chantest dataset. The proposed CNN classifier performed the best when using qInward variability to classify the TdP-risk drugs with 0.94 AUC for high risk and 0.93 AUC for low risk. In addition, the final CNN classifier was validated using the qInward variability obtained after merging three in-vitro datasets, but the model performance decreased to a moderate level of 0.75 and 0.78 AUC. These results suggest the need for the proposed CNN model to be trained and tested using various types of drugs.

Cross clinical-experimental-computational qualification of in silico drug trials on human cardiac purkinje cells for proarrhythmia risk prediction

The preclinical identification of drug-induced cardiotoxicity and its translation into human risk are still major challenges in pharmaceutical drug discovery. The ICH S7B Guideline and Q&A on Clinical and Nonclinical Evaluation of QT/QTc Interval Prolongation and Proarrhythmic Potential promotes human in silico drug trials as a novel tool for proarrhythmia risk assessment. To facilitate the use of in silico data in regulatory submissions, explanatory control compounds should be tested and documented to demonstrate consistency between predictions and the historic validation data. This study aims to quantify drug-induced electrophysiological effects on in silico cardiac human Purkinje cells, to compare them with existing in vitro rabbit data, and to assess their accuracy for clinical pro-arrhythmic risk predictions. The effects of 14 reference compounds were quantified in simulations with a population of in silico human cardiac Purkinje models. For each drug dose, five electrophysiological biomarkers were quantified at three pacing frequencies, and results compared with available in vitro experiments and clinical proarrhythmia reports. Three key results were obtained: 1) In silico, repolarization abnormalities in human Purkinje simulations predicted drug-induced arrhythmia for all risky compounds, showing higher predicted accuracy than rabbit experiments; 2) Drug-induced electrophysiological changes observed in human-based simulations showed a high degree of consistency with in vitro rabbit recordings at all pacing frequencies, and depolarization velocity and action potential duration were the most consistent biomarkers; 3) discrepancies observed for dofetilide, sotalol and terfenadine are mainly caused by species differences between humans and rabbit. Taken together, this study demonstrates higher accuracy of in silico methods compared to in vitro animal models for pro-arrhythmic risk prediction, as well as a high degree of consistency with in vitro experiments commonly used in safety pharmacology, supporting the potential for industrial and regulatory adoption of in silico trials for proarrhythmia prediction.

Considering population variability of electrophysiological models improves the in silico assessment of drug-induced torsadogenic risk

BACKGROUND AND OBJECTIVE

In silico tools are known to aid in drug cardiotoxicity assessment. However, computational models do not usually consider electrophysiological variability, which may be crucial when predicting rare adverse events such as drug-induced Torsade de Pointes (TdP). In addition, classification tools are usually binary and are not validated using an external data set. Here we analyze the role of incorporating electrophysiological variability in the prediction of drug-induced arrhythmogenic-risk, using a ternary classification and two external validation datasets.

METHODS

The effects of the 12 training CiPA drugs were simulated at three different concentrations using a single baseline model and an electrophysiologically calibrated population of models. 9 biomarkers related with action potential (AP), calcium dynamics and net charge were measured for each simulated concentration. These biomarkers were used to build ternary classifiers based on Support Vector Machines (SVM) methodology. Classifiers were validated using two external drug sets: the 16 validation CiPA drugs and 81 drugs from CredibleMeds database.

RESULTS

Population of models allowed to obtain different AP responses under the same pharmacological intervention and improve the prediction of drug-induced TdP with respect to the baseline model. The classification tools based on population of models achieve an accuracy higher than 0.8 and a mean classification error (MCE) lower than 0.3 for both validation drug sets and for the two electrophysiological action potential models studied (Tomek et al. 2020 and a modified version of O'Hara et al. 2011). In addition, simulations with population of models allowed the identification of individuals with lower conductances of I_Kr, I_Ks, and I_NaK and higher conductances of I_CaL, I_NaL, and I_NCX, which are more prone to develop TdP.

CONCLUSIONS

The methodology presented here provides new opportunities to assess drug-induced TdP-risk, taking into account electrophysiological variability and may be helpful to improve current cardiac safety screening methods.

Proarrhythmic risk assessment of drugs by dVm /dt shapes using the convolutional neural network

Comprehensive in vitro Proarrhythmia Assay (CiPA) projects for assessing proarrhythmic drugs suggested a logistic regression model using qNet as the Torsades de Pointes (TdP) risk assessment biomarker, obtained from in silico simulation. However, using a single in silico feature, such as qNet, cannot reflect whole characteristics related to TdP in the entire action potential (AP) shape. Thus, this study proposed a deep convolutional neural network (CNN) model using differential action potential shapes to classify three proarrhythmic risk levels: high, intermediate, and low, considering both characteristics related to TdP not only in the depolarization phase but also the repolarization phase of AP shape. We performed an in silico simulation and got AP shapes with drug effects using half-maximal inhibitory concentration and Hill coefficients of 28 drugs released by CiPA groups. Then, we trained the deep CNN model with the differential AP shapes of 12 drugs and tested it with those of 16 drugs. Our model had a better performance for classifying the proarrhythmic risk of drugs than the traditional logistic regression model using qNet. The classification accuracy was 98% for high-risk level drugs, 94% for intermediate-risk level drugs, and 89% for low-risk level drugs.

Assessment of Drug Proarrhythmicity Using Artificial Neural Networks With in silico Deterministic Model Outputs

As part of the Comprehensive in vitro Proarrhythmia Assay initiative, methodologies for predicting the occurrence of drug-induced torsade de pointes via computer simulations have been developed and verified recently. However, their predictive performance still requires improvement. Herein, we propose an artificial neural networks (ANN) model that uses nine multiple input features, considering the action potential morphology, calcium transient morphology, and charge features to further improve the performance of drug toxicity evaluation. The voltage clamp experimental data for 28 drugs were augmented to 2,000 data entries using an uncertainty quantification technique. By applying these data to the modified O'Hara Rudy in silico model, nine features (dVm/dt_max, AP_resting, APD90, APD50, Ca_resting, CaD90, CaD50, qNet, and qInward) were calculated. These nine features were used as inputs to an ANN model to classify drug toxicity into high-risk, intermediate-risk, and low-risk groups. The model was trained with data from 12 drugs and tested using the data of the remaining 16 drugs. The proposed ANN model demonstrated an AUC of 0.92 in the high-risk group, 0.83 in the intermediate-risk group, and 0.98 in the low-risk group. This was higher than the classification performance of the method proposed in previous studies.