Usefulness of Machine Learning-Based Detection and Classification of Cardiac Arrhythmias With 12-Lead Electrocardiograms

The accuracy of Gemini, GPT-4, and GPT-4o in ECG analysis: A comparison with cardiologists and emergency medicine specialists

INTRODUCTION

GPT-4, GPT-4o and Gemini advanced, which are among the well-known large language models (LLMs), have the capability to recognize and interpret visual data. When the literature is examined, there are a very limited number of studies examining the ECG performance of GPT-4. However, there is no study in the literature examining the success of Gemini and GPT-4o in ECG evaluation. The aim of our study is to evaluate the performance of GPT-4, GPT-4o, and Gemini in ECG evaluation, assess their usability in the medical field, and compare their accuracy rates in ECG interpretation with those of cardiologists and emergency medicine specialists.

METHODS

The study was conducted from May 14, 2024, to June 3, 2024. The book "150 ECG Cases" served as a reference, containing two sections: daily routine ECGs and more challenging ECGs. For this study, two emergency medicine specialists selected 20 ECG cases from each section, totaling 40 cases. In the next stage, the questions were evaluated by emergency medicine specialists and cardiologists. In the subsequent phase, a diagnostic question was entered daily into GPT-4, GPT-4o, and Gemini Advanced on separate chat interfaces. In the final phase, the responses provided by cardiologists, emergency medicine specialists, GPT-4, GPT-4o, and Gemini Advanced were statistically evaluated across three categories: routine daily ECGs, more challenging ECGs, and the total number of ECGs.

RESULTS

Cardiologists outperformed GPT-4, GPT-4o, and Gemini Advanced in all three groups. Emergency medicine specialists performed better than GPT-4o in routine daily ECG questions and total ECG questions (p = 0.003 and p = 0.042, respectively). When comparing GPT-4o with Gemini Advanced and GPT-4, GPT-4o performed better in total ECG questions (p = 0.027 and p < 0.001, respectively). In routine daily ECG questions, GPT-4o also outperformed Gemini Advanced (p = 0.004). Weak agreement was observed in the responses given by GPT-4 (p < 0.001, Fleiss Kappa = 0.265) and Gemini Advanced (p < 0.001, Fleiss Kappa = 0.347), while moderate agreement was observed in the responses given by GPT-4o (p < 0.001, Fleiss Kappa = 0.514).

CONCLUSION

While GPT-4o shows promise, especially in more challenging ECG questions, and may have potential as an assistant for ECG evaluation, its performance in routine and overall assessments still lags behind human specialists. The limited accuracy and consistency of GPT-4 and Gemini suggest that their current use in clinical ECG interpretation is risky.

Diagnostic and Prognostic Electrocardiogram-Based Models for Rapid Clinical Applications

Leveraging artificial intelligence (AI) for the analysis of electrocardiograms (ECGs) has the potential to transform diagnosis and estimate the prognosis of not only cardiac but, increasingly, noncardiac conditions. In this review, we summarize clinical studies and AI-enhanced ECG-based clinical applications in the early detection, diagnosis, and estimating prognosis of cardiovascular diseases in the past 5 years (2019-2023). With advancements in deep learning and the rapid increased use of ECG technologies, a large number of clinical studies have been published. However, most of these studies are single-centre, retrospective, proof-of-concept studies that lack external validation. Prospective studies that progress from development toward deployment in clinical settings account for < 15% of the studies. Successful implementations of ECG-based AI applications that have received approval from the Food and Drug Administration have been developed through commercial collaborations, with approximately half of them being for mobile or wearable devices. The field is in its early stages, and overcoming several obstacles is essential, such as prospective validation in multicentre large data sets, addressing technical issues, bias, privacy, data security, model generalizability, and global scalability. This review concludes with a discussion of these challenges and potential solutions. By providing a holistic view of the state of AI in ECG analysis, this review aims to set a foundation for future research directions, emphasizing the need for comprehensive, clinically integrated, and globally deployable AI solutions in cardiovascular disease management.

Comparison of emergency medicine specialist, cardiologist, and chat-GPT in electrocardiography assessment

INTRODUCTION

ChatGPT, developed by OpenAI, represents the cutting-edge in its field with its latest model, GPT-4. Extensive research is currently being conducted in various domains, including cardiovascular diseases, using ChatGPT. Nevertheless, there is a lack of studies addressing the proficiency of GPT-4 in diagnosing conditions based on Electrocardiography (ECG) data. The goal of this study is to evaluate the diagnostic accuracy of GPT-4 when provided with ECG data, and to compare its performance with that of emergency medicine specialists and cardiologists.

METHODS

This study has received approval from the Clinical Research Ethics Committee of Hitit University Medical Faculty on August 21, 2023 (decision no: 2023-91). Drawing on cases from the "150 ECG Cases" book, a total of 40 ECG cases were crafted into multiple-choice questions (comprising 20 everyday and 20 more challenging ECG questions). The participant pool included 12 emergency medicine specialists and 12 cardiology specialists. GPT-4 was administered the questions in a total of 12 separate sessions. The responses from the cardiology physicians, emergency medicine physicians, and GPT-4 were evaluated separately for each of the three groups.

RESULTS

In the everyday ECG questions, GPT-4 demonstrated superior performance compared to both the emergency medicine specialists and the cardiology specialists (p < 0.001, p = 0.001). In the more challenging ECG questions, while Chat-GPT outperformed the emergency medicine specialists (p < 0.001), no significant statistical difference was found between Chat-GPT and the cardiology specialists (p = 0.190). Upon examining the accuracy of the total ECG questions, Chat-GPT was found to be more successful compared to both the Emergency Medicine Specialists and the cardiologists (p < 0.001, p = 0.001).

CONCLUSION

Our study has shown that GPT-4 is more successful than emergency medicine specialists in evaluating both everyday and more challenging ECG questions. It performed better compared to cardiologists on everyday questions, but its performance aligned closely with that of the cardiologists as the difficulty of the questions increased.

Interpreting Wide-Complex Tachycardia With the Use of Artificial Intelligence

BACKGROUND

Adopting artificial intelligence (AI) in medicine may improve speed and accuracy in patient diagnosis. We sought to develop an AI algorithm to interpret wide-complex tachycardia (WCT) electrocardiograms (ECGs) and compare its diagnostic accuracy with that of cardiologists.

METHODS

Using 3330 WCT ECGs (2906 supraventricular tachycardia [SVT] and 424 ventricular tachycardia [VT]), we created a training/validation (3131) and a test set (199 ECGs). A convolutional neural network structure using a modification of differentiable architecture search was developed to differentiate between SVT and VT.

RESULTS

The mean accuracy of electrophysiology (EP) cardiologists was 92.5% with sensitivity 91.7%, specificity 93.4%, positive predictive value 93.7%, and negative predictive value 91.7%. Non-EP cardiologists had an accuracy of 73.2 ± 14.4% with sensitivity, specificity, and positive and negative predictive values of 59.8 ± 18.2%, 93.8 ± 3.7%, 93.6 ± 2.3%, and 73.2 ± 14.4%, respectively. AI had superior sensitivity and accuracy (91.9% and 93.0%, respectively) than non-EP cardiologists and similar performance compared with EP cardiologists. Mean time to interpret each ECG varied from 10.1 to 13.8 seconds for EP cardiologists and from 3.1 to 16.6 seconds for non-EP cardiologists. AI required a mean of 0.0092 ± 0.0035 seconds for each ECG interpretation.

CONCLUSIONS

AI appears to diagnose WCT with accuracy superior to non-EP cardiologists and similar to EP cardiologists. Using AI to assist with ECG interpretations may improve patient care.

An Efficient and Private ECG Classification System Using Split and Semi-Supervised Learning

Electrocardiography (ECG) is a standard diagnostic tool for evaluating the overall heart's electrical activity and is vital for detecting many cardiovascular diseases. Classifying ECG recordings using deep neural networks has been investigated in literature and has shown very good performance. However, this performance assumes that the training data is centralized, which is often not the case in real-life scenarios, where data resides in multiple places and only a small portion of it is labeled. Therefore, in this work, we propose an ECG classification system that focuses on preserving data privacy and enhancing overall system efficiency. We analyzed the complexity of previously proposed deep learning-based models and showed that the temporal convolutional network-based models (TCN) were the most efficient. Then, we built on the TCN models a modified split-learning (SL) system that achieves the same classification performance as the basic SL but reduces the communication overhead between the server and the client by 71.7% as well as reducing the computations at the client by 46.5% compared to the original SL system based on the TCN network. Finally, we implement semi-supervised learning in our system to enhance its classification performance by 9.1%-15.7%, when the training data consists only of 10% labeled data. We have tested our proposed system on a test IoT setup and it achieved satisfactory classification accuracy while being private and energy efficient for green-AI applications.

A Deep Learning Approach for Atrial Fibrillation Classification Using Multi-Feature Time Series Data from ECG and PPG

Atrial fibrillation is a prevalent cardiac arrhythmia that poses significant health risks to patients. The use of non-invasive methods for AF detection, such as Electrocardiogram and Photoplethysmogram, has gained attention due to their accessibility and ease of use. However, there are challenges associated with ECG-based AF detection, and the significance of PPG signals in this context has been increasingly recognized. The limitations of ECG and the untapped potential of PPG are taken into account as this work attempts to classify AF and non-AF using PPG time series data and deep learning. In this work, we emploted a hybrid deep neural network comprising of 1D CNN and BiLSTM for the task of AF classification. We addressed the under-researched area of applying deep learning methods to transmissive PPG signals by proposing a novel approach. Our approach involved integrating ECG and PPG signals as multi-featured time series data and training deep learning models for AF classification. Our hybrid 1D CNN and BiLSTM model achieved an accuracy of 95% on test data in identifying atrial fibrillation, showcasing its strong performance and reliable predictive capabilities. Furthermore, we evaluated the performance of our model using additional metrics. The precision of our classification model was measured at 0.88, indicating its ability to accurately identify true positive cases of AF. The recall, or sensitivity, was measured at 0.85, illustrating the model's capacity to detect a high proportion of actual AF cases. Additionally, the F1 score, which combines both precision and recall, was calculated at 0.84, highlighting the overall effectiveness of our model in classifying AF and non-AF cases.

Current and Future Use of Artificial Intelligence in Electrocardiography

Artificial intelligence (AI) is increasingly used in electrocardiography (ECG) to assist in diagnosis, stratification, and management. AI algorithms can help clinicians in the following areas: (1) interpretation and detection of arrhythmias, ST-segment changes, QT prolongation, and other ECG abnormalities; (2) risk prediction integrated with or without clinical variables (to predict arrhythmias, sudden cardiac death, stroke, and other cardiovascular events); (3) monitoring ECG signals from cardiac implantable electronic devices and wearable devices in real time and alerting clinicians or patients when significant changes occur according to timing, duration, and situation; (4) signal processing, improving ECG quality and accuracy by removing noise/artifacts/interference, and extracting features not visible to the human eye (heart rate variability, beat-to-beat intervals, wavelet transforms, sample-level resolution, etc.); (5) therapy guidance, assisting in patient selection, optimizing treatments, improving symptom-to-treatment times, and cost effectiveness (earlier activation of code infarction in patients with ST-segment elevation, predicting the response to antiarrhythmic drugs or cardiac implantable devices therapies, reducing the risk of cardiac toxicity, etc.); (6) facilitating the integration of ECG data with other modalities (imaging, genomics, proteomics, biomarkers, etc.). In the future, AI is expected to play an increasingly important role in ECG diagnosis and management, as more data become available and more sophisticated algorithms are developed.

Implementation of an All-Day Artificial Intelligence-Based Triage System to Accelerate Door-to-Balloon Times

OBJECTIVE

To implement an all-day artificial intelligence (AI)-based system to facilitate chest pain triage in the emergency department.

METHODS

The AI-based triage system encompasses an AI model combining a convolutional neural network and long short-term memory to detect ST-elevation myocardial infarction (STEMI) on electrocardiography (ECG) and a clinical risk score (ASAP) to prioritize patients for ECG examination. The AI model was developed on 2907 twelve-lead ECGs: 882 STEMI and 2025 non-STEMI ECGs.

RESULTS

Between November 1, 2019, and October 31, 2020, we enrolled 154 consecutive patients with STEMI: 68 during the AI-based triage period and 86 during the conventional triage period. The mean ± SD door-to-balloon (D2B) time was significantly shortened from 64.5±35.3 minutes to 53.2±12.7 minutes (P=.007), with 98.5% vs 87.2% (P=.009) of D2B times being less than 90 minutes in the AI group vs the conventional group. Among patients with an ASAP score of 3 or higher, the median door-to-ECG time decreased from 30 minutes (interquartile range [IQR], 7-59 minutes) to 6 minutes (IQR, 4-30 minutes) (P<.001). The overall performances of the AI model in identifying STEMI from 21,035 ECGs assessed by accuracy, precision, recall, area under the receiver operating characteristic curve, F1 score, and specificity were 0.997, 0.802, 0.977, 0.999, 0.881, and 0.998, respectively.

CONCLUSION

Implementation of an all-day AI-based triage system significantly reduced the D2B time, with a corresponding increase in the percentage of D2B times less than 90 minutes in the emergency department. This system may help minimize preventable delays in D2B times for patients with STEMI undergoing primary percutaneous coronary intervention.

Artificial intelligence-assisted remote detection of ST-elevation myocardial infarction using a mini-12-lead electrocardiogram device in prehospital ambulance care

Objective

To implement an all-day online artificial intelligence (AI)-assisted detection of ST-elevation myocardial infarction (STEMI) by prehospital 12-lead electrocardiograms (ECGs) to facilitate patient triage for timely reperfusion therapy.

Methods

The proposed AI model combines a convolutional neural network and long short-term memory (CNN-LSTM) to predict STEMI on prehospital 12-lead ECGs obtained from mini-12-lead ECG devices equipped in ambulance vehicles in Central Taiwan. Emergency medical technicians (EMTs) from the 14 AI-implemented fire stations performed the on-site 12-lead ECG examinations using the mini portable device. The 12-lead ECG signals were transmitted to the AI center of China Medical University Hospital to classify the recordings as "STEMI" or "Not STEMI". In 11 non-AI fire stations, the ECG data were transmitted to a secure network and read by available on-line emergency physicians. The response time was defined as the time interval between the ECG transmission and ECG interpretation feedback.

Results

Between July 17, 2021, and March 26, 2022, the AI model classified 362 prehospital 12-lead ECGs obtained from 275 consecutive patients who had called the 119 dispatch centers of fire stations in Central Taiwan for symptoms of chest pain or shortness of breath. The AI's response time to the EMTs in ambulance vehicles was 37.2 ± 11.3 s, which was shorter than the online physicians' response time from 11 other fire stations with no AI implementation (113.2 ± 369.4 s, P < 0.001) after analyzing another set of 335 prehospital 12-lead ECGs. The evaluation metrics including accuracy, precision, specificity, recall, area under the receiver operating characteristic curve, and F1 score to assess the overall AI performance in the remote detection of STEMI were 0.992, 0.889, 0.994, 0.941, 0.997, and 0.914, respectively. During the study period, the AI model promptly identified 10 STEMI patients who underwent primary percutaneous coronary intervention (PPCI) with a median contact-to-door time of 18.5 (IQR: 16-20.8) minutes.

Conclusion

Implementation of an all-day real-time AI-assisted remote detection of STEMI on prehospital 12-lead ECGs in the field is feasible with a high diagnostic accuracy rate. This approach may help minimize preventable delays in contact-to-treatment times for STEMI patients who require PPCI.

A systematic review and Meta-data analysis on the applications of Deep Learning in Electrocardiogram

The success of deep learning over the traditional machine learning techniques in handling artificial intelligence application tasks such as image processing, computer vision, object detection, speech recognition, medical imaging and so on, has made deep learning the buzz word that dominates Artificial Intelligence applications. From the last decade, the applications of deep learning in physiological signals such as electrocardiogram (ECG) have attracted a good number of research. However, previous surveys have not been able to provide a systematic comprehensive review including biometric ECG based systems of the applications of deep learning in ECG with respect to domain of applications. To address this gap, we conducted a systematic literature review on the applications of deep learning in ECG including biometric ECG based systems. The study analyzed systematically, 150 primary studies with evidence of the application of deep learning in ECG. The study shows that the applications of deep learning in ECG have been applied in different domains. We presented a new taxonomy of the domains of application of the deep learning in ECG. The paper also presented discussions on biometric ECG based systems and meta-data analysis of the studies based on the domain, area, task, deep learning models, dataset sources and preprocessing methods. Challenges and potential research opportunities were highlighted to enable novel research. We believe that this study will be useful to both new researchers and expert researchers who are seeking to add knowledge to the already existing body of knowledge in ECG signal processing using deep learning algorithm.

Supplementary information

The online version contains supplementary material available at 10.1007/s12652-022-03868-z.

State-of-the-art Deep Learning Methods on Electrocardiogram Data: A Systematic Review (Preprint)

Background

Electrocardiogram (ECG) is one of the most common noninvasive diagnostic tools that can provide useful information regarding a patient’s health status. Deep learning (DL) is an area of intense exploration that leads the way in most attempts to create powerful diagnostic models based on physiological signals.

Objective

This study aimed to provide a systematic review of DL methods applied to ECG data for various clinical applications.

Methods

The PubMed search engine was systematically searched by combining “deep learning” and keywords such as “ecg,” “ekg,” “electrocardiogram,” “electrocardiography,” and “electrocardiology.” Irrelevant articles were excluded from the study after screening titles and abstracts, and the remaining articles were further reviewed. The reasons for article exclusion were manuscripts written in any language other than English, absence of ECG data or DL methods involved in the study, and absence of a quantitative evaluation of the proposed approaches.

Results

We identified 230 relevant articles published between January 2020 and December 2021 and grouped them into 6 distinct medical applications, namely, blood pressure estimation, cardiovascular disease diagnosis, ECG analysis, biometric recognition, sleep analysis, and other clinical analyses. We provide a complete account of the state-of-the-art DL strategies per the field of application, as well as major ECG data sources. We also present open research problems, such as the lack of attempts to address the issue of blood pressure variability in training data sets, and point out potential gaps in the design and implementation of DL models.

Conclusions

We expect that this review will provide insights into state-of-the-art DL methods applied to ECG data and point to future directions for research on DL to create robust models that can assist medical experts in clinical decision-making.

Optimal ECG-lead selection increases generalizability of deep learning on ECG abnormality classification

Deep learning (DL) has achieved promising performance in detecting common abnormalities from the 12-lead electrocardiogram (ECG). However, diagnostic redundancy exists in the 12-lead ECG, which could impose a systematic overfitting on DL, causing poor generalization. We, therefore, hypothesized that finding an optimal lead subset of the 12-lead ECG to eliminate the redundancy would help improve the generalizability of DL-based models. In this study, we developed and evaluated a DL-based model that has a feature extraction stage, an ECG-lead subset selection stage and a decision-making stage to automatically interpret multiple common ECG abnormality types. The data analysed in this study consisted of 6877 12-lead ECG recordings from CPSC 2018 (labelled as normal rhythm or eight types of ECG abnormalities, split into training (approx. 80%), validation (approx. 10%) and test (approx. 10%) sets) and 3998 12-lead ECG recordings from PhysioNet/CinC 2020 (labelled as normal rhythm or four types of ECG abnormalities, used as external text set). The ECG-lead subset selection module was introduced within the proposed model to efficiently constrain model complexity. It detected an optimal 4-lead ECG subset consisting of leads II, aVR, V1 and V4. The proposed model using the optimal 4-lead subset significantly outperformed the model using the complete 12-lead ECG on the validation set and on the external test dataset. The results demonstrated that our proposed model successfully identified an optimal subset of 12-lead ECG; the resulting 4-lead ECG subset improves the generalizability of the DL model in ECG abnormality interpretation. This study provides an outlook on what channels are necessary to keep and which ones may be ignored when considering an automated detection system for cardiac ECG abnormalities. This article is part of the theme issue 'Advanced computation in cardiovascular physiology: new challenges and opportunities'.

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.

Research Progress of ECG Monitoring Equipment and Algorithms Based on Polymer Materials

Heart diseases such as myocardial ischemia (MI) are the main causes of human death. The prediction of MI and arrhythmia is an effective method for the early detection, diagnosis, and treatment of heart disease. For the rapid detection of arrhythmia and myocardial ischemia, the electrocardiogram (ECG) is widely used in clinical diagnosis, and its detection equipment and algorithm are constantly optimized. This paper introduces the current progress of portable ECG monitoring equipment, including the use of polymer material sensors and the use of deep learning algorithms. First, it introduces the latest portable ECG monitoring equipment and the polymer material sensor it uses and then focuses on reviewing the progress of detection algorithms. We mainly introduce the basic structure of existing deep learning methods and enumerate the internationally recognized ECG datasets. This paper outlines the deep learning algorithms used for ECG diagnosis, compares the prediction results of different classifiers, and summarizes two existing problems of ECG detection technology: imbalance of categories and high computational overhead. Finally, we put forward the development direction of using generative adversarial networks (GAN) to improve the quality of the ECG database and lightweight ECG diagnosis algorithm to adapt to portable ECG monitoring equipment.

[An arrhythmia classification method based on deep learning parallel network model]

OBJECTIVE

We propose a parallel neural network classification method to improve the performance of classification of 4 types of arrhythmias: normal beat, supraventricular ectopic beat, ventricular ectopic beat and fused beat.

METHODS

Preprocessing was performed including denoising of ECG signal, segmentation of small-scale heartbeat and large-scale heartbeat and data enhancement. Based on deep learning theory, densely connected convolutional network was applied to improve the limitation of waveform feature extraction, and bidirectional long short-term memory network and efficient channel attention network were combined to enhance the function of time series features and important features of the waveform. The parallel network structure was adopted, and the waveform features of small- scale heartbeat and large-scale heartbeat were input to improve the accuracy of arrhythmia classification at the same time. Softmax was used to carry out the 4 classification tasks of arrhythmia by the parallel network model.

RESULTS

The proposed method was verified using MIT-BIH Arrhythmia Database and 3 groups of experiments. The experiments for comparing the classification performance of multiple parallel network models and that of each classification model under different heartbeat input methods showed that the proposed classification model had an overall accuracy, average sensitivity and average specificity of 99.36%, 96.08% and 99.41%, respectively. Convergence performance analysis of the parallel network classification model showed that the training time of the classification model was 41 s.

CONCLUSION

The parallel multi-network classification method can improve the average sensitivity, specificity and training time while maintaining a high overall accuracy, and may thus provide a new technical solution for clinical diagnosis of arrhythmia.

Deep Learning Classification of Unipolar Electrograms in Human Atrial Fibrillation: Application in Focal Source Mapping

Focal sources are potential targets for atrial fibrillation (AF) catheter ablation, but they can be time-consuming and challenging to identify when unipolar electrograms (EGM) are numerous and complex. Our aim was to apply deep learning (DL) to raw unipolar EGMs in order to automate putative focal sources detection. We included 78 patients from the Focal Source and Trigger (FaST) randomized controlled trial that evaluated the efficacy of adjunctive FaST ablation compared to pulmonary vein isolation alone in reducing AF recurrence. FaST sites were identified based on manual classification of sustained periodic unipolar QS EGMs over 5-s. All periodic unipolar EGMs were divided into training (n = 10,004) and testing cohorts (n = 3,180). DL was developed using residual convolutional neural network to discriminate between FaST and non-FaST. A gradient-based method was applied to interpret the DL model. DL classified FaST with a receiver operator characteristic area under curve of 0.904 ± 0.010 (cross-validation) and 0.923 ± 0.003 (testing). At a prespecified sensitivity of 90%, the specificity and accuracy were 81.9 and 82.5%, respectively, in detecting FaST. DL had similar performance (sensitivity 78%, specificity 89%) to that of FaST re-classification by cardiologists (sensitivity 78%, specificity 79%). The gradient-based interpretation demonstrated accurate tracking of unipolar QS complexes by select DL convolutional layers. In conclusion, our novel DL model trained on raw unipolar EGMs allowed automated and accurate classification of FaST sites. Performance was similar to FaST re-classification by cardiologists. Future application of DL to classify FaST may improve the efficiency of real-time focal source detection for targeted AF ablation therapy.

The Role of Artificial Intelligence and Machine Learning in Clinical Cardiac Electrophysiology

In recent years, artificial intelligence (AI) has found numerous applications in cardiology due in part to large digitized datasets and the evolution of high performance computing. In the discipline of cardiac electrophysiology (EP), a number of clinical, imaging, and electrical waveform data are considered in the diagnosis, prognostication and management of arrhythmias, which lend themselves well to automation through AI. But equally relevant, AI offers a unique opportunity to discover novel EP concepts and improve clinical care through its inherent, hierarchical tenets of self-learning. This review will focus on the application of AI in clinical EP and summarize state-of-the art, large, clinical studies in the following key domains: (1) ECG-based arrhythmia and disease classification, (2) atrial fibrillation source detection, (3) substrate and risk assessment for atrial fibrillation and ventricular tachyarrhythmias, and (4) predicting outcomes after cardiac resynchronization therapy. Many are small, single-center, proof-of-concept investigations, but they still demonstrate groundbreaking performance of deep learning, a subdomain of AI, which surpasses traditional statistical analysis. Larger studies, for instance classifying arrhythmias from ECG recordings, have further provided external validation of their high accuracy. Ultimately, the performance of AI is dependent on the quality of the input data and the rigor of algorithm development. The field is still nascent and several barriers will need to be overcome, including prospective validation in large, well-labelled datasets and more seamless information technology-based data collection/integration, before AI can be adopted into broader clinical EP practice. This review will conclude with a discussion of these challenges and future work.

Premature beats detection based on a novel convolutional neural network

Objective.Automatic detection of premature beats on long electrocardiogram (ECG) recordings is of great significance for clinical diagnosis. In this paper, we propose a novel deep learning model, the ECGDet, to detect premature beats, including premature ventricular contractions (PVCs) and supraventricular premature beats (SPBs) on single-lead long-term ECGs.Approach.The ECGDet is proposed based on a convolutional neural network and squeeze-and-excitation network. It outputs the probabilities that the ECG samples belong to a premature contraction. Non-max suppression was used to select the most appropriate locations for the premature beats. The ECGDet was trained and tested on the MIT-BIH arrhythmia database (MITDB) using a five-fold cross-validation approach. A novel loss calculation method was introduced in the model training process. Then it was tuned and further tested on the China Physiological Signal Challenge (2020) database (CPSCDB).Main results.The results showed that the average F1 value of PVC detection was 92.6%, while that of SPB detection was 72.2% on MITDB. The ECGDet bagged the 2nd place for PVC detection and ranked 7th place of SPB detection in the China Physiological Signal Challenge (2020).Significance.The proposed ECGDet can automatically detect premature heartbeats without manually extracting the features. This technique can be used for long-term ECG signal analysis and has potential for clinical applications.

Uncertainty-Aware Deep Learning-Based Cardiac Arrhythmias Classification Model of Electrocardiogram Signals

Deep Learning-based methods have emerged to be one of the most effective and practical solutions in a wide range of medical problems, including the diagnosis of cardiac arrhythmias. A critical step to a precocious diagnosis in many heart dysfunctions diseases starts with the accurate detection and classification of cardiac arrhythmias, which can be achieved via electrocardiograms (ECGs). Motivated by the desire to enhance conventional clinical methods in diagnosing cardiac arrhythmias, we introduce an uncertainty-aware deep learning-based predictive model design for accurate large-scale classification of cardiac arrhythmias successfully trained and evaluated using three benchmark medical datasets. In addition, considering that the quantification of uncertainty estimates is vital for clinical decision-making, our method incorporates a probabilistic approach to capture the model’s uncertainty using a Bayesian-based approximation method without introducing additional parameters or significant changes to the network’s architecture. Although many arrhythmias classification solutions with various ECG feature engineering techniques have been reported in the literature, the introduced AI-based probabilistic-enabled method in this paper outperforms the results of existing methods in outstanding multiclass classification results that manifest F1 scores of 98.62% and 96.73% with (MIT-BIH) dataset of 20 annotations, and 99.23% and 96.94% with (INCART) dataset of eight annotations, and 97.25% and 96.73% with (BIDMC) dataset of six annotations, for the deep ensemble and probabilistic mode, respectively. We demonstrate our method’s high-performing and statistical reliability results in numerical experiments on the language modeling using the gating mechanism of Recurrent Neural Networks.

Usefulness of multi-labelling artificial intelligence in detecting rhythm disorders and acute ST-elevation myocardial infarction on 12-lead electrocardiogram

Aims

To develop an artificial intelligence-based approach with multi-labelling capability to identify both ST-elevation myocardial infarction (STEMI) and 12 heart rhythms based on 12-lead electrocardiograms (ECGs).

Methods and results

We trained, validated, and tested a long short-term memory (LSTM) model for the multi-label diagnosis of 13 ECG patterns (STEMI + 12 rhythm classes) using 60 537 clinical ECGs from 35 981 patients recorded between 15 January 2009 and 31 December 2018. In addition to the internal test above, we conducted a real-world external test, comparing the LSTM model with board-certified physicians of different specialties using a separate dataset of 308 ECGs covering all 13 ECG diagnoses. In the internal test, the area under the curves (AUCs) of the LSTM model in classifying the 13 ECG patterns ranged between 0.939 and 0.999. For the external test, the LSTM model for multi-labelling of the 13 ECG patterns evaluated by AUC was 0.987 ± 0.021, which was superior to those of cardiologists (0.898 ± 0.113, P < 0.001), emergency physicians (0.820 ± 0.134, P < 0.001), internists (0.765 ± 0.155, P < 0.001), and a commercial algorithm (0.845 ± 0.121, P < 0.001). Of note, the LSTM model achieved an accuracy of 0.987, AUC of 0.997, and precision, recall, and F ₁ score of 0.952, 0.870, and 0.909, respectively, in detecting STEMI.

Conclusions

We demonstrated the usefulness of an LSTM model in the multi-labelling detection of both rhythm classes and STEMI in competitive testing against board-certified physicians. This AI tool exceeding the cardiologist-level performance in detecting STEMI and rhythm classes on 12-lead ECG may be useful in prioritizing chest pain triage and expediting clinical decision-making in healthcare.

The Ethics of Artificial Intelligence in Pathology and Laboratory Medicine: Principles and Practice

Growing numbers of artificial intelligence applications are being developed and applied to pathology and laboratory medicine. These technologies introduce risks and benefits that must be assessed and managed through the lens of ethics. This article describes how long-standing principles of medical and scientific ethics can be applied to artificial intelligence using examples from pathology and laboratory medicine.

Accurate deep neural network model to detect cardiac arrhythmia on more than 10,000 individual subject ECG records

BACKGROUND AND OBJECTIVE

Cardiac arrhythmia, which is an abnormal heart rhythm, is a common clinical problem in cardiology. Detection of arrhythmia on an extended duration electrocardiogram (ECG) is done based on initial algorithmic software screening, with final visual validation by cardiologists. It is a time consuming and subjective process. Therefore, fully automated computer-assisted detection systems with a high degree of accuracy have an essential role in this task. In this study, we proposed an effective deep neural network (DNN) model to detect different rhythm classes from a new ECG database.

METHODS

Our DNN model was designed for high performance on all ECG leads. The proposed model, which included both representation learning and sequence learning tasks, showed promising results on all 12-lead inputs. Convolutional layers and sub-sampling layers were used in the representation learning phase. The sequence learning part involved a long short-term memory (LSTM) unit after representation of learning layers.

RESULTS

We performed two different class scenarios, including reduced rhythms (seven rhythm types) and merged rhythms (four rhythm types) according to the records from the database. Our trained DNN model achieved 92.24% and 96.13% accuracies for the reduced and merged rhythm classes, respectively.

CONCLUSION

Recently, deep learning algorithms have been found to be useful because of their high performance. The main challenge is the scarcity of appropriate training and testing resources because model performance is dependent on the quality and quantity of case samples. In this study, we used a new public arrhythmia database comprising more than 10,000 records. We constructed an efficient DNN model for automated detection of arrhythmia using these records.