Automated feature extraction from large cardiac electrophysiological data sets

Jurkiewicz, John; Kroboth, Stacie; Zlochiver, Sharon; Hinow, Peter

doi:10.1016/j.jelectrocard.2021.02.003

Cited by 3 publications

(4 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Noise spuriously reduces measures of predictability such as DET, and can hence compromise the differentiation of arrhythmia categories presented here. One compelling option to address this is to pre-process the membrane potential time series before attempting to calculate any RQA measures, and indeed automated noise removal that preserves action potentials has been recently demonstrated 59 . Options for making RQA robust to any residual noise after pre-processing are also available.…”

Section: Discussionmentioning

confidence: 99%

Recurrence quantification analysis for fine-scale characterisation of arrhythmic patterns in cardiac tissue

Halfar¹,

Lawson²,

Santos³

et al. 2023

Sci Rep

View full text Add to dashboard Cite

This paper uses recurrence quantification analysis (RQA) combined with entropy measures and organization indices to characterize arrhythmic patterns and dynamics in computer simulations of cardiac tissue. We performed different simulations of cardiac tissues of sizes comparable to the human heart atrium. In these simulations, we observed four classic arrhythmic patterns: a spiral wave anchored to a highly fibrotic region resulting in sustained re-entry, a meandering spiral wave, fibrillation, and a spiral wave anchored to a scar region that breaks up into wavelets away from the main rotor. A detailed analysis revealed that, within the same simulation, maps of RQA metrics could differentiate regions with regular AP propagation from ones with chaotic activity. In particular, the combination of two RQA metrics, the length of the longest diagonal string of recurrence points and the mean length of diagonal lines, was able to identify the location of rotor tips, which are the active elements that maintain spiral waves and fibrillation. By proposing low-dimensional models based on the mean value and spatial correlation of metrics calculated from membrane potential time series, we identify RQA-based metrics that successfully separate the four different types of cardiac arrhythmia into distinct regions of the feature space, and thus might be used for automatic classification, in particular distinguishing between fibrillation driven by self-sustaining chaos and that created by a persistent rotor and wavebreak. We also discuss the practical applicability of such an approach.

show abstract

Section: Discussionmentioning

confidence: 99%

Recurrence quantification analysis for fine-scale characterisation of arrhythmic patterns in cardiac tissue

Halfar¹,

Lawson²,

Santos³

et al. 2023

Sci Rep

View full text Add to dashboard Cite

show abstract

“…This results in a considerably higher class accuracy standard deviation than overall accuracy as the performances offset each other. Much of the current space in using ML to classify disease state from solely EP is assessing cardiac arrhythmias [ 14 , 15 , 40 ] which present with a number of simplifications compared to assessing neuronal pathologies. These include the ease of obtaining and accessibility of electrocardiograms (ECGs) compared to Electroencephalograms (EEGs) or MEA recordings of isogenic neurons as well as the significantly higher quality signal obtained from ECGs compared to the methods of recording neuronal signals.…”

Section: Discussionmentioning

confidence: 99%

“…It is of current interest to assess the ability of electrophysiology to have similar predictive power. Recent studies have shown that ML algorithms can be used to extract features from Keywords: Machine learning, Neural network, Support vector machine, Gaussian naïve bayes, Decision tree, Gradient boosting decision tree, Epilepsy, Dravet syndrome, SCN1A, In silico, BRIAN2, Diagnosis, Genomics, Electrophysiology, Microelectrode array electrophysiological data [14]. Furthermore, it has been shown that electrophysiological data can be utilized in isolation by machine learning algorithms to identify pathologic states or perturbations [15,16].…”

Section: Introductionmentioning

confidence: 99%

“…It is of current interest to assess the ability of electrophysiology to have similar predictive power. Recent studies have shown that ML algorithms can be used to extract features from electrophysiological data [ 14 ]. Furthermore, it has been shown that electrophysiological data can be utilized in isolation by machine learning algorithms to identify pathologic states or perturbations [ 15 , 16 ].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Utilizing machine learning algorithms to predict subject genetic mutation class from in silico models of neuronal networks

Kress

Chan

Garcia

et al. 2022

BMC Med Inform Decis Mak

View full text Add to dashboard Cite

Background Epilepsy is the fourth-most common neurological disorder, affecting an estimated 50 million patients globally. Nearly 40% of patients have uncontrolled seizures yet incur 80% of the cost. Anti-epileptic drugs commonly result in resistance and reversion to uncontrolled drug-resistant epilepsy and are often associated with significant adverse effects. This has led to a trial-and-error system in which physicians spend months to years attempting to identify the optimal therapeutic approach. Objective To investigate the potential clinical utility from the context of optimal therapeutic prediction of characterizing cellular electrophysiology. It is well-established that genomic data alone can sometimes be predictive of effective therapeutic approach. Thus, to assess the predictive power of electrophysiological data, machine learning strategies are implemented to predict a subject’s genetically defined class in an in silico model using brief electrophysiological recordings obtained from simulated neuronal networks. Methods A dynamic network of isogenic neurons is modeled in silico for 1-s for 228 dynamically modeled patients falling into one of three categories: healthy, general sodium channel gain of function, or inhibitory sodium channel loss of function. Data from previous studies investigating the electrophysiological and cellular properties of neurons in vitro are used to define the parameters governing said models. Ninety-two electrophysiological features defining the nature and consistency of network connectivity, activity, waveform shape, and complexity are extracted for each patient network and t-tests are used for feature selection for the following machine learning algorithms: Neural Network, Support Vector Machine, Gaussian Naïve Bayes Classifier, Decision Tree, and Gradient Boosting Decision Tree. Finally, their performance in accurately predicting which genetic category the subjects fall under is assessed. Results Several machine learning algorithms excel in using electrophysiological data from isogenic neurons to accurately predict genetic class with a Gaussian Naïve Bayes Classifier predicting healthy, gain of function, and overall, with the best accuracy, area under the curve, and F1. The Gradient Boosting Decision Tree performs the best for loss of function models indicated by the same metrics. Conclusions It is possible for machine learning algorithms to use electrophysiological data to predict clinically valuable metrics such as optimal therapeutic approach, especially when combining several models.

show abstract

A review on machine learning approaches in cardiac tissue engineering

Kalkunte,

Cisneros,

Castillo

et al. 2024

Front. Biomater. Sci.

View full text Add to dashboard Cite

Cardiac tissue engineering (CTE) holds promise in addressing the clinical challenges posed by cardiovascular disease, the leading global cause of mortality. Human induced pluripotent stem cells (hiPSCs) are pivotal for cardiac regeneration therapy, offering an immunocompatible, high density cell source. However, hiPSC-derived cardiomyocytes (hiPSC-CMs) exhibit vital functional deficiencies that are not yet well understood, hindering their clinical deployment. We argue that machine learning (ML) can overcome these challenges, by improving the phenotyping and functionality of these cells via robust mathematical models and predictions. This review paper explores the transformative role of ML in advancing CTE, presenting a primer on relevant ML algorithms. We focus on how ML has recently addressed six key address six key challenges in CTE: cell differentiation, morphology, calcium handling and cell-cell coupling, contraction, and tissue assembly. The paper surveys common ML models, from tree-based and probabilistic to neural networks and deep learning, illustrating their applications to better understand hiPSC-CM behavior. While acknowledging the challenges associated with integrating ML, such as limited biomedical datasets, computational costs of learning data, and model interpretability and reliability, we examine suggestions for improvement, emphasizing the necessity for more extensive and diverse datasets that incorporate temporal and imaging data, augmented by synthetic generative models. By integrating ML with mathematical models and existing expert knowledge, we foresee a fruitful collaboration that unites innovative data-driven models with biophysics-informed models, effectively closing the gaps within CTE.

show abstract

Automated feature extraction from large cardiac electrophysiological data sets

Cited by 3 publications

References 39 publications

Recurrence quantification analysis for fine-scale characterisation of arrhythmic patterns in cardiac tissue

Recurrence quantification analysis for fine-scale characterisation of arrhythmic patterns in cardiac tissue

Utilizing machine learning algorithms to predict subject genetic mutation class from in silico models of neuronal networks

A review on machine learning approaches in cardiac tissue engineering

Contact Info

Product

Resources

About