Classifying Drugs by their Arrhythmogenic Risk Using Machine Learning

MultiCBlo: Enhancing predictions of compound-induced inhibition of cardiac ion channels with advanced multimodal learning

Predicting compound-induced inhibition of cardiac ion channels is crucial and challenging, significantly impacting cardiac drug efficacy and safety assessments. Despite the development of various computational methods for compound-induced inhibition prediction in cardiac ion channels, their performance remains limited. Most methods struggle to fuse multi-source data, relying solely on specific dataset training, leading to poor accuracy and generalization. We introduce MultiCBlo, a model that fuses multimodal information through a progressive learning approach, designed to predict compound-induced inhibition of cardiac ion channels with high accuracy. MultiCBlo employs progressive multimodal information fusion technology to integrate the compound's SMILES sequence, graph structure, and fingerprint, enhancing its representation. This is the first application of progressive multimodal learning for predicting compound-induced inhibition of cardiac ion channels, to our knowledge. The objective of this study was to predict the compound-induced inhibition of three major cardiac ion channels: hERG, Cav1.2, and Nav1.5. The results indicate that MultiCBlo significantly outperforms current models in predicting compound-induced inhibition of cardiac ion channels. We hope that MultiCBlo will facilitate cardiac drug development and reduce compound toxicity risks. Code and data are accessible at: https://github.com/taowang11/MultiCBlo. The online prediction platform is freely accessible at: https://huggingface.co/spaces/wtttt/PCICB.

Enhancing Mechanical Metamodels with a Generative Model-Based Augmented Training Dataset

Modeling biological soft tissue is complex in part due to material heterogeneity. Microstructural patterns, which play a major role in defining the mechanical behavior of these tissues, are both challenging to characterize, and difficult to simulate. Recently, machine learning (ML)-based methods to predict the mechanical behavior of heterogeneous materials have made it possible to more thoroughly explore the massive input parameter space associated with heterogeneous blocks of material. Specifically, we can train ML models to closely approximate computationally expensive heterogeneous material simulations where the ML model is trained on datasets of simulations with relevant spatial heterogeneity. However, when it comes to applying these techniques to tissue, there is a major limitation: the number of useful examples available to characterize the input domain under study is often limited. In this work, we investigate the efficacy of both ML-based generative models and procedural methods as tools for augmenting limited input pattern datasets. We find that a Style-based Generative Adversarial Network with an adaptive discriminator augmentation mechanism is able to successfully leverage just 1,000 example patterns to create authentic generated patterns. And, we find that diverse generated patterns with adequate resemblance to real patterns can be used as inputs to finite element simulations to meaningfully augment the training dataset. To enable this methodological contribution, we have created an open access Finite Element Analysis simulation dataset based on Cahn-Hilliard patterns. We anticipate that future researchers will be able to leverage this dataset and build on the work presented here.

Synergy between machine learning and natural products cheminformatics: Application to the lead discovery of anthraquinone derivatives

Cheminformatics utilizing machine learning (ML) techniques have opened up a new horizon in drug discovery. This is owing to vast chemical space expansion with rocketing numbers of expected hits and lead compounds that match druggable macromolecular targets, in particular from natural compounds. Due to the natural products' (NP) structural complexity, uniqueness, and diversity, they could occupy a bigger space in pharmaceuticals, allowing the industry to pursue more selective leads in the nanomolar range of binding affinity. ML is an essential part of each step of the drug design pipeline, such as target prediction, compound library preparation, and lead optimization. Notably, molecular mechanic and dynamic simulations, induced docking, and free energy perturbations are essential in predicting best binding poses, binding free energy values, and molecular mechanics force fields. Those applications have leveraged from artificial intelligence (AI), which decreases the computational costs required for such costly simulations. This review aimed to describe chemical space and compound libraries related to NPs. High-throughput screening utilized for fractionating NPs and high-throughput virtual screening and their strategies, and significance, are reviewed. Particular emphasis was given to AI approaches, ML tools, algorithms, and techniques, especially in drug discovery of macrocyclic compounds and approaches in computer-aided and ML-based drug discovery. Anthraquinone derivatives were discussed as a source of new lead compounds that can be developed using ML tools for diverse medicinal uses such as cancer, infectious diseases, and metabolic disorders. Furthermore, the power of principal component analysis in understanding relevant protein conformations, and molecular modeling of protein-ligand interaction were also presented. Apart from being a concise reference for cheminformatics, this review is a useful text to understand the application of ML-based algorithms to molecular dynamics simulation and in silico absorption, distribution, metabolism, excretion, and toxicity prediction.

Fast Characterization of Inducible Regions of Atrial Fibrillation Models With Multi-Fidelity Gaussian Process Classification

Computational models of atrial fibrillation have successfully been used to predict optimal ablation sites. A critical step to assess the effect of an ablation pattern is to pace the model from different, potentially random, locations to determine whether arrhythmias can be induced in the atria. In this work, we propose to use multi-fidelity Gaussian process classification on Riemannian manifolds to efficiently determine the regions in the atria where arrhythmias are inducible. We build a probabilistic classifier that operates directly on the atrial surface. We take advantage of lower resolution models to explore the atrial surface and combine seamlessly with high-resolution models to identify regions of inducibility. We test our methodology in 9 different cases, with different levels of fibrosis and ablation treatments, totalling 1,800 high resolution and 900 low resolution simulations of atrial fibrillation. When trained with 40 samples, our multi-fidelity classifier that combines low and high resolution models, shows a balanced accuracy that is, on average, 5.7% higher than a nearest neighbor classifier. We hope that this new technique will allow faster and more precise clinical applications of computational models for atrial fibrillation. All data and code accompanying this manuscript will be made publicly available at: https://github.com/fsahli/AtrialMFclass.

[Digital twins in surgery: achievements and limitations]

OBJECTIVE

To review the possible options for digital twin technology in surgery, as well as to build a patent and publication landscape for identifying technological and academic leaders of the frontier.

MATERIAL AND METHODS

Scientometric and patent analysis was performed.

RESULTS

Possible options for digital twin technology in surgical practice were reviewed. Development of scientific and technological trend «digital twins in surgery» in the world was assessed (2002 - idea, concept, definition; 2007-2014 - large-scale studies in academic sector; 2014 - active participation of regulators in translation of pilot digital models of patient organs into practical healthcare; 2014-2017 - large-scale studies in business sector; 2018-2021 - development of a market of medical services based on digital twin technologies in surgery). According to scientometric and patent analysis of digital twins in surgery, there is no a single Russian-language article on to this issue in journals indexed in WOS at the end of 2021. Our country ranks the 23^rd in the world regarding its share in the total number of patent applications for inventions.

CONCLUSION

Over a 20-year period, large-scale scientific projects have been carried out in the world to develop digital twin algorithms for surgery. Regulators were involved in the process of broadcasting their results into practical health care. Network interaction of all authors and beneficiaries of technological frontier occurred (research centers, hospitals, companies, manufacturers of medical equipment and information services). Technological ecosystems developed (startups, gazelles, investment seed capital). Technological leaders and key players in new market niches have been identified. Development of this field is insufficient in the Russian Federation. There are no qualified customers and companies in the real sector of economy that could become the beneficiaries of the frontier.

Combined In-silico and Machine Learning Approaches Toward Predicting Arrhythmic Risk in Post-infarction Patients

Background: Remodeling due to myocardial infarction (MI) significantly increases patient arrhythmic risk. Simulations using patient-specific models have shown promise in predicting personalized risk for arrhythmia. However, these are computationally- and time- intensive, hindering translation to clinical practice. Classical machine learning (ML) algorithms (such as K-nearest neighbors, Gaussian support vector machines, and decision trees) as well as neural network techniques, shown to increase prediction accuracy, can be used to predict occurrence of arrhythmia as predicted by simulations based solely on infarct and ventricular geometry. We present an initial combined image-based patient-specific in silico and machine learning methodology to assess risk for dangerous arrhythmia in post-infarct patients. Furthermore, we aim to demonstrate that simulation-supported data augmentation improves prediction models, combining patient data, computational simulation, and advanced statistical modeling, improving overall accuracy for arrhythmia risk assessment.

Methods: MRI-based computational models were constructed from 30 patients 5 days post-MI (the “baseline” population). In order to assess the utility biophysical model-supported data augmentation for improving arrhythmia prediction, we augmented the virtual baseline patient population. Each patient ventricular and ischemic geometry in the baseline population was used to create a subfamily of geometric models, resulting in an expanded set of patient models (the “augmented” population). Arrhythmia induction was attempted via programmed stimulation at 17 sites for each virtual patient corresponding to AHA LV segments and simulation outcome, “arrhythmia,” or “no-arrhythmia,” were used as ground truth for subsequent statistical prediction (machine learning, ML) models. For each patient geometric model, we measured and used choice data features: the myocardial volume and ischemic volume, as well as the segment-specific myocardial volume and ischemia percentage, as input to ML algorithms. For classical ML techniques (ML), we trained k-nearest neighbors, support vector machine, logistic regression, xgboost, and decision tree models to predict the simulation outcome from these geometric features alone. To explore neural network ML techniques, we trained both a three - and a four-hidden layer multilayer perceptron feed forward neural networks (NN), again predicting simulation outcomes from these geometric features alone. ML and NN models were trained on 70% of randomly selected segments and the remaining 30% was used for validation for both baseline and augmented populations.

Results: Stimulation in the baseline population (30 patient models) resulted in reentry in 21.8% of sites tested; in the augmented population (129 total patient models) reentry occurred in 13.0% of sites tested. ML and NN models ranged in mean accuracy from 0.83 to 0.86 for the baseline population, improving to 0.88 to 0.89 in all cases.

Conclusion: Machine learning techniques, combined with patient-specific, image-based computational simulations, can provide key clinical insights with high accuracy rapidly and efficiently. In the case of sparse or missing patient data, simulation-supported data augmentation can be employed to further improve predictive results for patient benefit. This work paves the way for using data-driven simulations for prediction of dangerous arrhythmia in MI patients.

Prediction of arrhythmia susceptibility through mathematical modeling and machine learning

At present, the QT interval on the electrocardiographic (ECG) waveform is the most common metric for assessing an individual's susceptibility to ventricular arrhythmias, with a long QT, or, at the cellular level, a long action potential duration (APD) considered high risk. However, the limitations of this simple approach have long been recognized. Here, we sought to improve prediction of arrhythmia susceptibility by combining mechanistic mathematical modeling with machine learning (ML). Simulations with a model of the ventricular myocyte were performed to develop a large heterogenous population of cardiomyocytes (n = 10,586), and we tested each variant's ability to withstand three arrhythmogenic triggers: 1) block of the rapid delayed rectifier potassium current (I_Kr Block), 2) augmentation of the L-type calcium current (I_CaL Increase), and 3) injection of inward current (Current Injection). Eight ML algorithms were trained to predict, based on simulated AP features in preperturbed cells, whether each cell would develop arrhythmic dynamics in response to each trigger. We found that APD can accurately predict how cells respond to the simple Current Injection trigger but cannot effectively predict the response to I_Kr Block or I_CaL Increase. ML predictive performance could be improved by incorporating additional AP features and simulations of additional experimental protocols. Importantly, we discovered that the most relevant features and experimental protocols were trigger specific, which shed light on the mechanisms that promoted arrhythmia formation in response to the triggers. Overall, our quantitative approach provides a means to understand and predict differences between individuals in arrhythmia susceptibility.

Multiscale modeling meets machine learning: What can we learn?

Machine learning is increasingly recognized as a promising technology in the biological, biomedical, and behavioral sciences. There can be no argument that this technique is incredibly successful in image recognition with immediate applications in diagnostics including electrophysiology, radiology, or pathology, where we have access to massive amounts of annotated data. However, machine learning often performs poorly in prognosis, especially when dealing with sparse data. This is a field where classical physics-based simulation seems to remain irreplaceable. In this review, we identify areas in the biomedical sciences where machine learning and multiscale modeling can mutually benefit from one another: Machine learning can integrate physics-based knowledge in the form of governing equations, boundary conditions, or constraints to manage ill-posted problems and robustly handle sparse and noisy data; multiscale modeling can integrate machine learning to create surrogate models, identify system dynamics and parameters, analyze sensitivities, and quantify uncertainty to bridge the scales and understand the emergence of function. With a view towards applications in the life sciences, we discuss the state of the art of combining machine learning and multiscale modeling, identify applications and opportunities, raise open questions, and address potential challenges and limitations. We anticipate that it will stimulate discussion within the community of computational mechanics and reach out to other disciplines including mathematics, statistics, computer science, artificial intelligence, biomedicine, systems biology, and precision medicine to join forces towards creating robust and efficient models for biological systems.

Establishing the Biofidelity of a Multiphysics Finite Element Model of the Human Heart

PURPOSE

Accelerating development of new therapeutic cardiac devices remains a clinical and technical priority. High-performance computing and the emergence of functional and complex in silico models of human anatomy can be an engine to accelerate the commercialization of innovative, safe, and effective devices.

METHODS

An existing three-dimensional, nonlinear model of a human heart with flow boundary conditions was evaluated. Its muscular tissues were exercised using electrophysiological boundary conditions, creating a dynamic, electro-mechanical simulation of the kinetics of the human heart. Anatomic metrics were selected to characterize the functional biofidelity of the model based on their significance to the design of cardiac devices. The model output was queried through the cardiac cycle and compared to in vivo literature values.

RESULTS

For the kinematics of mitral and aortic valves and curvature of coronary vessels, the model's performance was at or above the 95th percentile range of the in vivo data from large patient cohorts. One exception was the kinematics of the tricuspid valve. The model's mechanical use environment would subject devices to generally conservative use conditions.

CONCLUSIONS

This conservative simulated use environment for heart-based medical devices, and its judicious application in the evaluation of medical devices is justified, but careful interpretation of the results is encouraged.

Precision medicine in human heart modeling : Perspectives, challenges, and opportunities

Precision medicine is a new frontier in healthcare that uses scientific methods to customize medical treatment to the individual genes, anatomy, physiology, and lifestyle of each person. In cardiovascular health, precision medicine has emerged as a promising paradigm to enable cost-effective solutions that improve quality of life and reduce mortality rates. However, the exact role in precision medicine for human heart modeling has not yet been fully explored. Here, we discuss the challenges and opportunities for personalized human heart simulations, from diagnosis to device design, treatment planning, and prognosis. With a view toward personalization, we map out the history of anatomic, physical, and constitutive human heart models throughout the past three decades. We illustrate recent human heart modeling in electrophysiology, cardiac mechanics, and fluid dynamics and highlight clinically relevant applications of these models for drug development, pacing lead failure, heart failure, ventricular assist devices, edge-to-edge repair, and annuloplasty. With a view toward translational medicine, we provide a clinical perspective on virtual imaging trials and a regulatory perspective on medical device innovation. We show that precision medicine in human heart modeling does not necessarily require a fully personalized, high-resolution whole heart model with an entire personalized medical history. Instead, we advocate for creating personalized models out of population-based libraries with geometric, biological, physical, and clinical information by morphing between clinical data and medical histories from cohorts of patients using machine learning. We anticipate that this perspective will shape the path toward introducing human heart simulations into precision medicine with the ultimate goals to facilitate clinical decision making, guide treatment planning, and accelerate device design.

Artificial intelligence in cardiovascular medicine

PURPOSE OF REVIEW

Artificial intelligence is a broad set of sophisticated computer-based statistical tools that have become widely available. Cardiovascular medicine with its large data repositories, need for operational efficiency and growing focus on precision care is set to be transformed by artificial intelligence. Applications range from new pathophysiologic discoveries to decision support for individual patient care to optimization of system-wide logistical processes.

RECENT FINDINGS

Machine learning is the dominant form of artificial intelligence wherein complex statistical algorithms 'learn' by deducing patterns in datasets. Supervised machine learning uses classified large data to train an algorithm to accurately predict the outcome, whereas in unsupervised machine learning, the algorithm uncovers mathematical relationships within unclassified data. Artificial multilayered neural networks or deep learning is one of the most successful tools. Artificial intelligence has demonstrated superior efficacy in disease phenomapping, early warning systems, risk prediction, automated processing and interpretation of imaging, and increasing operational efficiency.

SUMMARY

Artificial intelligence demonstrates the ability to learn through assimilation of large datasets to unravel complex relationships, discover prior unfound pathophysiological states and develop predictive models. Artificial intelligence needs widespread exploration and adoption for large-scale implementation in cardiovascular practice.

Identification of Drug-Induced Multichannel Block and Proarrhythmic Risk in Humans Using Continuous T Vector Velocity Effect Profiles Derived From Surface Electrocardiograms

We present continuous T vector velocity (TVV) effect profiles as a new method for identifying drug effects on cardiac ventricular repolarization. TVV measures the temporal change in the myocardial action potential distribution during repolarization. The T vector dynamics were measured as the time required to reach p percent of the total T vector trajectory length, denoted as Tr(p), with p in {1, …, 100%}. The Tr(p) values were individually corrected for heart rate at each trajectory length percentage p. Drug effects were measured by evaluating the placebo corrected changes from baseline of Tr(p)c jointly for all p using functional mixed effects models. The p-dependent model parameters were implemented as cubic splines, providing continuous drug effect profiles along the entire ventricular repolarization process. The effect profile distributions were approximated by bootstrap simulations. We applied this TVV-based analysis approach to ECGs available from three published studies that were conducted in the CiPA context. These studies assessed the effect of 10 drugs and drug combinations with different ion channel blocking properties on myocardial repolarization in a total of 104 healthy volunteers. TVV analysis revealed that blockade of outward potassium currents alone presents an effect profile signature of continuous accumulation of delay throughout the entire repolarization interval. In contrast, block of inward sodium or calcium currents involves acceleration, which accumulates during early repolarization. The balance of blocking inward versus outward currents was reflected in the percentage p_zero of the T vector trajectory length where accelerated repolarization transitioned to delayed repolarization. Binary classification using a threshold p_zero = 43% separated predominant hERG channel blocking drugs with potentially higher proarrhythmic risk (moxifloxacin, dofetilide, quinidine, chloroquine) from multichannel blocking drugs with low proarrhythmic risk (ranolazine, verapamil, lopinavir/ritonavir) with sensitivity 0.99 and specificity 0.97. The TVV-based effect profile provides a detailed view of drug effects throughout the entire ventricular repolarization interval. It enables the evaluation of drug-induced blocks of multiple cardiac repolarization currents from clinical ECGs. The proposed p_zero parameter enhances identification of the proarrhythmic risk of a drug beyond QT prolongation, and therefore constitutes an important tool for cardiac arrhythmia risk assessment.