Detection of acute coronary occlusion with a novel mobile electrocardiogram device: a pilot study.

Aims: Many portable electrocardiogram (ECG) devices have been developed to monitor patients at home, but the majority of these devices are single lead and only intended for rhythm disorders. We developed the miniECG, a smartphone-sized portable device with four dry electrodes capable of recording a high-quality multi-lead ECG by placing the device on the chest. The aim of our study was to investigate the ability of the miniECG to detect occlusive myocardial infarction (OMI) in patients with chest pain. Methods and results: Patients presenting with acute chest pain at the emergency department of the University Medical Center Utrecht or Meander Medical Center, between May 2021 and February 2022, were included in the study. The clinical 12-lead ECG and the miniECG before coronary intervention were recorded. The recordings were evaluated by cardiologists and compared the outcome of the coronary angiography, if performed. A total of 369 patients were measured with the miniECG, 46 of whom had OMI. The miniECG detected OMI with a sensitivity and specificity of 65 and 92%, compared with 83 and 90% for the 12-lead ECG. Sensitivity of the miniECG was similar for different culprit vessels. Conclusion: The miniECG can record a multi-lead ECG and rule-in ST-segment deviation in patients with occluded or near-occluded coronary arteries from different culprit vessels without many false alarms. Further research is required to add automated analysis to the recordings and to show feasibility to use the miniECG by patients at home.

ECG-only explainable deep learning algorithm predicts the risk for malignant ventricular arrhythmia in phospholamban cardiomyopathy.

BACKGROUND: Phospholamban (PLN) p.(Arg14del) variant carriers are at risk for development of malignant ventricular arrhythmia (MVA). Accurate risk stratification allows timely implantation of intracardiac defibrillators and is currently performed with a multimodality prediction model. OBJECTIVE: This study aimed to investigate whether an explainable deep learning-based approach allows risk prediction with only electrocardiogram (ECG) data. METHODS: A total of 679 PLN p.(Arg14del) carriers without MVA at baseline were identified. A deep learning-based variational auto-encoder, trained on 1.1 million ECGs, was used to convert the 12-lead baseline ECG into its FactorECG, a compressed version of the ECG that summarizes it into 32 explainable factors. Prediction models were developed by Cox regression. RESULTS: The deep learning-based ECG-only approach was able to predict MVA with a C statistic of 0.79 (95% CI, 0.76-0.83), comparable to the current prediction model (C statistic, 0.83 [95% CI, 0.79-0.88]; P = .054) and outperforming a model based on conventional ECG parameters (low-voltage ECG and negative T waves; C statistic, 0.65 [95% CI, 0.58-0.73]; P < .001). Clinical simulations showed that a 2-step approach, with ECG-only screening followed by a full workup, resulted in 60% less additional diagnostics while outperforming the multimodal prediction model in all patients. A visualization tool was created to provide interactive visualizations (https://pln.ecgx.ai). CONCLUSION: Our deep learning-based algorithm based on ECG data only accurately predicts the occurrence of MVA in PLN p.(Arg14del) carriers, enabling more efficient stratification of patients who need additional diagnostic testing and follow-up.

ANOCA patients with and without coronary vasomotor dysfunction present with limited electrocardiographic remodeling.

Background: Coronary vasomotor dysfunction (CVDys) comprises coronary vasospasm (CVS) and/or coronary microvascular dysfunction (CMD) and is highly prevalent in patients with angina and non-obstructive coronary artery disease (ANOCA). Invasive coronary function testing (CFT) to diagnose CVDys is becoming more common, enabling pathophysiologic research of CVDys. This study aims to explore the electrophysiological characteristics of ANOCA patients with CVDys. Methods: We collected pre-procedural 12-lead electrocardiograms of ANOCA patients with CVS (n = 35), CMD (n = 24), CVS/CMD (n = 26) and patients without CVDys (CFT-, n = 23) who participated in the NL-CFT registry and underwent CFT. Heart axis and conduction times were compared between patients with CVS, CMD or CVS/CMD and patients without CVDys. Results: Heart axis, heart rate, PQ interval and QRS duration were comparable between the groups. A small prolongation of the QT-interval corrected with Bazett (QTcB) and Fridericia (QTcF) was observed in patients with CVDys compared to patients without CVDys (CVS vs CFT-: QTcB = 422 ± 18 vs 414 ± 18 ms (p = 0.14), QTcF = 410 ± 14 vs 406 ± 12 ms (p = 0.21); CMD vs CFT-: QTcB = 426 ± 17 vs 414 ± 18 ms (p = 0.03), QTcF = 413 ± 11 vs 406 ± 12 ms (p = 0.04); CVS/CMD vs CFT-: QTcB = 424 ± 17 vs 414 ± 18 ms (p = 0.05), QTcF = 414 ± 14 vs 406 ± 12 ms (p = 0.04)). Conclusions: Pre-procedural 12-lead electrocardiograms were comparable between patients with and without CVDys undergoing CFT except for a slightly longer QTc interval in patients with CVDys compared to patients without CVDys, suggesting limited cardiac remodeling in patients with CVDys.

Automatic triage of twelve-lead electrocardiograms using deep convolutional neural networks: a first implementation study.

Aims: Expert knowledge to correctly interpret electrocardiograms (ECGs) is not always readily available. An artificial intelligence (AI)-based triage algorithm (DELTAnet), able to support physicians in ECG prioritization, could help reduce current logistic burden of overreading ECGs and improve time to treatment for acute and life-threatening disorders. However, the effect of clinical implementation of such AI algorithms is rarely investigated. Methods and results: Adult patients at non-cardiology departments who underwent ECG testing as a part of routine clinical care were included in this prospective cohort study. DELTAnet was used to classify 12-lead ECGs into one of the following triage classes: normal, abnormal not acute, subacute, and acute. Performance was compared with triage classes based on the final clinical diagnosis. Moreover, the associations between predicted classes and clinical outcomes were investigated. A total of 1061 patients and ECGs were included. Performance was good with a mean concordance statistic of 0.96 (95% confidence interval 0.95-0.97) when comparing DELTAnet with the clinical triage classes. Moreover, zero ECGs that required a change in policy or referral to the cardiologist were missed and there was a limited number of cases predicted as acute that did not require follow-up (2.6%). Conclusion: This study is the first to prospectively investigate the impact of clinical implementation of an ECG-based AI triage algorithm. It shows that DELTAnet is efficacious and safe to be used in clinical practice for triage of 12-lead ECGs in non-cardiology hospital departments.

Deep neural network-based clustering of deformation curves reveals novel disease features in PLN pathogenic variant carriers.

Echocardiographic deformation curves provide detailed information on myocardial function. Deep neural networks (DNNs) may enable automated detection of disease features in deformation curves, and improve the clinical assessment of these curves. We aimed to investigate whether an explainable DNN-based pipeline can be used to detect and visualize disease features in echocardiographic deformation curves of phospholamban (PLN) p.Arg14del variant carriers. A DNN was trained to discriminate PLN variant carriers (n = 278) from control subjects (n = 621) using raw deformation curves obtained by 2D-speckle tracking in the longitudinal axis. A visualization technique was used to identify the parts of these curves that were used by the DNN for classification. The PLN variant carriers were clustered according to the output of the visualization technique. The DNN showed excellent discriminatory performance (C-statistic 0.93 [95% CI 0.87-0.97]). We identified four clusters with PLN-associated disease features in the deformation curves. Two clusters showed previously described features: apical post-systolic shortening and reduced systolic strain. The two other clusters revealed novel features, both reflecting delayed relaxation. Additionally, a fifth cluster was identified containing variant carriers without disease features in the deformation curves, who were classified as controls by the DNN. This latter cluster had a very benign disease course regarding development of ventricular arrhythmias. Applying an explainable DNN-based pipeline to myocardial deformation curves enables automated detection and visualization of disease features. In PLN variant carriers, we discovered novel disease features which may improve individual risk stratification. Applying this approach to other diseases will further expand our knowledge on disease-specific deformation patterns. Overview of the deep neural network-based pipeline for feature detection in myocardial deformation curves. Firstly, phospholamban (PLN) p.Arg14del variant carriers and controls were selected and a deep neural network (DNN) was trained to detect the PLN variant carriers. Subsequently, a clustering-based approach was performed on the attention maps of the DNN, which revealed 4 distinct phenotypes of PLN variant carriers with different prognoses. Moreover, a cluster without features and a benign prognosis was detected.

Electrocardiogram Devices for Home Use: Technological and Clinical Scoping Review.

BACKGROUND: Electrocardiograms (ECGs) are used by physicians to record, monitor, and diagnose the electrical activity of the heart. Recent technological advances have allowed ECG devices to move out of the clinic and into the home environment. There is a great variety of mobile ECG devices with the capabilities to be used in home environments. OBJECTIVE: This scoping review aimed to provide a comprehensive overview of the current landscape of mobile ECG devices, including the technology used, intended clinical use, and available clinical evidence. METHODS: We conducted a scoping review to identify studies concerning mobile ECG devices in the electronic database PubMed. Secondarily, an internet search was performed to identify other ECG devices available in the market. We summarized the devices' technical information and usability characteristics based on manufacturer data such as datasheets and user manuals. For each device, we searched for clinical evidence on the capabilities to record heart disorders by performing individual searches in PubMed and ClinicalTrials.gov, as well as the Food and Drug Administration (FDA) 510(k) Premarket Notification and De Novo databases. RESULTS: From the PubMed database and internet search, we identified 58 ECG devices with available manufacturer information. Technical characteristics such as shape, number of electrodes, and signal processing influence the capabilities of the devices to record cardiac disorders. Of the 58 devices, only 26 (45%) had clinical evidence available regarding their ability to detect heart disorders such as rhythm disorders, more specifically atrial fibrillation. CONCLUSIONS: ECG devices available in the market are mainly intended to be used for the detection of arrhythmias. No devices are intended to be used for the detection of other cardiac disorders. Technical and design characteristics influence the intended use of the devices and use environments. For mobile ECG devices to be intended to detect other cardiac disorders, challenges regarding signal processing and sensor characteristics should be solved to increase their detection capabilities. Devices recently released include the use of other sensors on ECG devices to increase their detection capabilities.

Machine-learning-derived heart and brain age are independently associated with cognition.

BACKGROUND AND PURPOSE: A heart age biomarker has been developed using deep neural networks applied to electrocardiograms. Whether this biomarker is associated with cognitive function was investigated. METHODS: Using 12-lead electrocardiograms, heart age was estimated for a population-based sample (N = 7779, age 40-85 years, 45.3% men). Associations between heart delta age (HDA) and cognitive test scores were studied adjusted for cardiovascular risk factors. In addition, the relationship between HDA, brain delta age (BDA) and cognitive test scores was investigated in mediation analysis. RESULTS: Significant associations between HDA and the Word test, Digit Symbol Coding Test and tapping test scores were found. HDA was correlated with BDA (Pearson's r = 0.12, p = 0.0001). Moreover, 13% (95% confidence interval 3-36) of the HDA effect on the tapping test score was mediated through BDA. DISCUSSION: Heart delta age, representing the cumulative effects of life-long exposures, was associated with brain age. HDA was associated with cognitive function that was minimally explained through BDA.

Feasibility of Linear Irreversible Electroporation Ablation in the Coronary Sinus.

INTRODUCTION: Previous studies demonstrated that the coronary sinus (CS) is an important target for ablation in persistent atrial fibrillation. However, radiofrequency ablation in the CS is associated with coronary vessel damage and tamponade. Animal data suggest irreversible electroporation (IRE) ablation can be a safe ablation modality in vicinity of coronary arteries. We investigated the feasibility of IRE in the CS in a porcine model. METHODS: Ablation and pacing was performed in the CS in six pigs (weight 60-75 kg) using a modified 9-French steerable linear hexapolar Tip-Versatile Ablation Catheter. Pacing maneuvers were performed from distal to proximal segments of the CS to assess atrial capture thresholds before and after IRE application. IRE ablations were performed with 100 J IRE pulses. After 3-week survival animals were euthanized and histological sections from the CS were analyzed. RESULTS: A total of 27 IRE applications in six animals were performed. Mean peak voltage was 1509 ± 36 V, with a mean peak current of 22.9 ± 1.0 A. No complications occurred during procedure and 3-week survival. At 30 min post ablation 100% isolation was achieved in all animals. At 3 weeks follow-up pacing thresholds were significant higher as compared to baseline. Histological analysis showed transmural ablation lesions in muscular sleeves surrounding the CS. CONCLUSION: IRE ablation of the musculature along the CS using a multi-electrode catheter is feasible in a porcine model.

Electrocardiogram-based deep learning improves outcome prediction following cardiac resynchronization therapy.

AIMS: This study aims to identify and visualize electrocardiogram (ECG) features using an explainable deep learning-based algorithm to predict cardiac resynchronization therapy (CRT) outcome. Its performance is compared with current guideline ECG criteria and QRSAREA. METHODS AND RESULTS: A deep learning algorithm, trained on 1.1 million ECGs from 251 473 patients, was used to compress the median beat ECG, thereby summarizing most ECG features into only 21 explainable factors (FactorECG). Pre-implantation ECGs of 1306 CRT patients from three academic centres were converted into their respective FactorECG. FactorECG predicted the combined clinical endpoint of death, left ventricular assist device, or heart transplantation [c-statistic 0.69, 95% confidence interval (CI) 0.66-0.72], significantly outperforming QRSAREA and guideline ECG criteria [c-statistic 0.61 (95% CI 0.58-0.64) and 0.57 (95% CI 0.54-0.60), P < 0.001 for both]. The addition of 13 clinical variables was of limited added value for the FactorECG model when compared with QRSAREA (Δ c-statistic 0.03 vs. 0.10). FactorECG identified inferolateral T-wave inversion, smaller right precordial S- and T-wave amplitude, ventricular rate, and increased PR interval and P-wave duration to be important predictors for poor outcome. An online visualization tool was created to provide interactive visualizations (https://crt.ecgx.ai). CONCLUSION: Requiring only a standard 12-lead ECG, FactorECG held superior discriminative ability for the prediction of clinical outcome when compared with guideline criteria and QRSAREA, without requiring additional clinical variables. End-to-end automated visualization of ECG features allows for an explainable algorithm, which may facilitate rapid uptake of this personalized decision-making tool in CRT.

Detection of Ischemic ST-Segment Changes Using a Novel Handheld ECG Device in a Porcine Model.

Background: Portable, smartphone-sized electrocardiography (ECG) has the potential to reduce time to treatment for patients suffering acute cardiac ischemia, thereby lowering the morbidity and mortality. In the UMC Utrecht, a portable, smartphone-sized, multi-lead precordial ECG recording device (miniECG 1.0, UMC Utrecht) was developed. Objectives: The purpose of this study was to investigate the ability of the miniECG to capture ischemic ECG changes in a porcine coronary occlusion model. Methods: In 8 animals, antero-septal myocardial infarction was induced by 75-minute occlusion of the left anterior descending artery, after the first or second diagonal. MiniECG and 12-lead ECG recordings were acquired simultaneously before, during and after coronary artery occlusion and ST-segment deviation was evaluated. Results: During the complete occlusion and reperfusion period, miniECG showed large ST-segment deviation in comparison to 12-lead ECG. MiniECG ST-segment deviation was observed within 1 minute for most animals. The miniECG was positive for ischemia (ie, ST-segment deviation ≥1 mm) for 99.7% (Q1-Q3: 99.6%-99.9%) of the occlusion time, while the 12-lead was only positive for 79.8% (Q1-Q3: 81.1%-98.7%) of the time (P = 0.018). ST-segment deviation reached maxima of 10.5 mm [95% CI: 6.5-14.5 mm] vs 5.0 mm [95% CI: 2.0-8.0 mm] for the miniECG vs 12-lead ECG, respectively. Conclusions: MiniECG ST-segment deviation was observed early and was of large magnitude during 75 minutes of porcine transmural antero-septal infarction. The miniECG was positive for ischemia for the complete occlusion period. These findings demonstrate the potential of the miniECG in the detection of cardiac ischemia. Although clinical research is required, data suggests that the miniECG is a promising tool for the detection of cardiac ischemia.

Multielectrode Contact Measurement Can Improve Long-Term Outcome of Pulmonary Vein Isolation Using Circular Single-Pulse Electroporation Ablation.

BACKGROUND: Irreversible electroporation (IRE) ablation is generally performed with multielectrode catheters. Electrode-tissue contact is an important predictor for the success of pulmonary vein (PV) isolation; however, contact force is difficult to measure with multielectrode ablation catheters. In a preclinical study, we assessed the feasibility of a multielectrode impedance system (MEIS) as a predictor of long-term success of PV isolation. In addition, we present the first-in-human clinical experience with MEIS. METHODS: In 10 pigs, one PV was ablated based on impedance (MEIS group), and the other PV was solely based on local electrogram information (EP group). IRE ablations were performed at 200 J. After 3 months, recurrence of conduction was assessed. Subsequently, in 30 patients undergoing PV isolation with IRE, MEIS was evaluated and MEIS contact values were compared to local electrograms. RESULTS: In the porcine study, 43 IRE applications were delivered in 19 PVs. Acutely, no reconnections were observed in either group. After 3 months, 0 versus 3 (P=0.21) PVs showed conduction recurrence in the MEIS and EP groups, respectively. Results from the clinical study showed a significant linear relation was found between mean MEIS value and bipolar dV/dt (r2=0.49, P<0.001), with a slope of 20.6 mV/s per Ohm. CONCLUSIONS: Data from the animal study suggest that MEIS values predict effective IRE applications. For the long-term success of electrical PV isolation with circular IRE applications, no significant difference in efficacy was found between ablation based on the measurement of electrode interface impedance and ablation using the classical EP approach for determining electrode-tissue contact. Experiences of the first clinical use of MEIS were promising and serve as an important basis for future research.

Life-threatening ventricular arrhythmia prediction in patients with dilated cardiomyopathy using explainable electrocardiogram-based deep neural networks.

AIMS: While electrocardiogram (ECG) characteristics have been associated with life-threatening ventricular arrhythmias (LTVA) in dilated cardiomyopathy (DCM), they typically rely on human-derived parameters. Deep neural networks (DNNs) can discover complex ECG patterns, but the interpretation is hampered by their 'black-box' characteristics. We aimed to detect DCM patients at risk of LTVA using an inherently explainable DNN. METHODS AND RESULTS: In this two-phase study, we first developed a variational autoencoder DNN on more than 1 million 12-lead median beat ECGs, compressing the ECG into 21 different factors (F): FactorECG. Next, we used two cohorts with a combined total of 695 DCM patients and entered these factors in a Cox regression for the composite LTVA outcome, which was defined as sudden cardiac arrest, spontaneous sustained ventricular tachycardia, or implantable cardioverter-defibrillator treated ventricular arrhythmia. Most patients were male (n = 442, 64%) with a median age of 54 years [interquartile range (IQR) 44-62], and median left ventricular ejection fraction of 30% (IQR 23-39). A total of 115 patients (16.5%) reached the study outcome. Factors F8 (prolonged PR-interval and P-wave duration, P < 0.005), F15 (reduced P-wave height, P = 0.04), F25 (increased right bundle branch delay, P = 0.02), F27 (P-wave axis P < 0.005), and F32 (reduced QRS-T voltages P = 0.03) were significantly associated with LTVA. CONCLUSION: Inherently explainable DNNs can detect patients at risk of LTVA which is mainly driven by P-wave abnormalities.

Safety and feasibility study of non-invasive robot-assisted high-intensity focused ultrasound therapy for the treatment of atherosclerotic plaques in the femoral artery: protocol for a pilot study.

INTRODUCTION: Peripheral arterial disease (PAD) is an atherosclerotic disease leading to stenosis and/or occlusion of the arterial circulation of the lower extremities. The currently available revascularisation methods have an acceptable initial success rate, but the long-term patency is limited, while surgical revascularisation is associated with a relatively high perioperative risk. This urges the need for development of less invasive and more effective treatment modalities. This protocol article describes a study investigating a new non-invasive technique that uses robot assisted high-intensity focused ultrasound (HIFU) to treat atherosclerosis in the femoral artery. METHODS AND ANALYSIS: A pilot study is currently performed in 15 symptomatic patients with PAD with a significant stenosis in the common femoral and/or proximal superficial femoral artery. All patients will be treated with the dual-mode ultrasound array system to deliver imaging-guided HIFU to the atherosclerotic plaque. Safety and feasibility are the primary objectives assessed by the technical feasibility of this therapy and the 30-day major complication rate as primary endpoints. Secondary endpoints are angiographic and clinical success and quality of life. ETHICS AND DISSEMINATION: Ethical approval for this study was obtained in 2019 from the Medical Ethics Committee of the University Medical Center Utrecht, the Netherlands. Data will be presented at national and international conferences and published in a peer-reviewed journal. TRIAL REGISTRATION NUMBER: NL7564.

Artificial Intelligence to Improve Risk Prediction with Nuclear Cardiac Studies.

PURPOSE OF REVIEW: As machine learning-based artificial intelligence (AI) continues to revolutionize the way in which we analyze data, the field of nuclear cardiology provides fertile ground for the implementation of these complex analytics. This review summarizes and discusses the principles regarding nuclear cardiology techniques and AI, and the current evidence regarding its performance and contribution to the improvement of risk prediction in cardiovascular disease. There is a growing body of evidence on the experimentation with and implementation of machine learning-based AI on nuclear cardiology studies both concerning SPECT and PET technology for the improvement of risk-of-disease (classification of disease) and risk-of-events (prediction of adverse events) estimations. These publications still report objective divergence in methods either utilizing statistical machine learning approaches or deep learning with varying architectures, dataset sizes, and performance. Recent efforts have been placed into bringing standardization and quality to the experimentation and application of machine learning-based AI in cardiovascular imaging to generate standards in data harmonization and analysis through AI. Machine learning-based AI offers the possibility to improve risk evaluation in cardiovascular disease through its implementation on cardiac nuclear studies. AI in improving risk evaluation in nuclear cardiology. * Based on the 2019 ESC guidelines.

Characteristics and time course of acute and chronic myocardial lesion formation after electroporation ablation in the porcine model.

INTRODUCTION: Electroporation ablation creates deep and wide myocardial lesions. No data are available on time course and characteristics of acute lesion formation. METHODS: For the acute phase of myocardial lesion development, seven pigs were investigated. Single 200 J applications were delivered at four different epicardial right ventricular sites using a linear suction device, yielding a total of 28 lesions. Timing of applications was designed to yield lesions at seven time points: 0, 10, 20, 30, 40, 50, and 60 min, with four lesions per time point. After killing, lesion characteristics were histologically investigated. For the chronic phase of myocardial lesion development, tissue samples were used from previously conducted studies where tissue was obtained at 3 weeks and 3 months after electroporation ablation. RESULTS: Acute myocardial lesions induce a necrosis pattern with contraction band necrosis and interstitial edema, immediately present after electroporation ablation. No further histological changes such as hemorrhage or influx of inflammatory cells occurred in the first hour. After 3 weeks, the lesions consisted of sharply demarcated loose connective tissue that further developed to more fibrotic scar tissue after 3 months without additional changes. Within the scar tissue, arteries and nerves were unaffected. CONCLUSION: Electroporation ablation immediately induces contraction band necrosis and edema without additional tissue changes in the first hour. After 3 weeks, a sharply demarked scar has been developed that remains stable during follow-up of 3 months. This is highly relevant for clinical application of electroporation ablation in terms of the electrophysiological endpoint and waiting period after ablation.

Improving explainability of deep neural network-based electrocardiogram interpretation using variational auto-encoders.

Aims: Deep neural networks (DNNs) perform excellently in interpreting electrocardiograms (ECGs), both for conventional ECG interpretation and for novel applications such as detection of reduced ejection fraction (EF). Despite these promising developments, implementation is hampered by the lack of trustworthy techniques to explain the algorithms to clinicians. Especially, currently employed heatmap-based methods have shown to be inaccurate. Methods and results: We present a novel pipeline consisting of a variational auto-encoder (VAE) to learn the underlying factors of variation of the median beat ECG morphology (the FactorECG), which are subsequently used in common and interpretable prediction models. As the ECG factors can be made explainable by generating and visualizing ECGs on both the model and individual level, the pipeline provides improved explainability over heatmap-based methods. By training on a database with 1.1 million ECGs, the VAE can compress the ECG into 21 generative ECG factors, most of which are associated with physiologically valid underlying processes. Performance of the explainable pipeline was similar to 'black box' DNNs in conventional ECG interpretation [area under the receiver operating curve (AUROC) 0.94 vs. 0.96], detection of reduced EF (AUROC 0.90 vs. 0.91), and prediction of 1-year mortality (AUROC 0.76 vs. 0.75). Contrary to the 'black box' DNNs, our pipeline provided explainability on which morphological ECG changes were important for prediction. Results were confirmed in a population-based external validation dataset. Conclusions: Future studies on DNNs for ECGs should employ pipelines that are explainable to facilitate clinical implementation by gaining confidence in artificial intelligence and making it possible to identify biased models.

Deep neural networks reveal novel sex-specific electrocardiographic features relevant for mortality risk.

Aims: Incorporation of sex in study design can lead to discoveries in medical research. Deep neural networks (DNNs) accurately predict sex based on the electrocardiogram (ECG) and we hypothesized that misclassification of sex is an important predictor for mortality. Therefore, we first developed and validated a DNN that classified sex based on the ECG and investigated the outcome. Second, we studied ECG drivers of DNN-classified sex and mortality. Methods and results: A DNN was trained to classify sex based on 131 673 normal ECGs. The algorithm was validated on internal (68 500 ECGs) and external data sets (3303 and 4457 ECGs). The survival of sex (mis)classified groups was investigated using time-to-event analysis and sex-stratified mediation analysis of ECG features. The DNN successfully distinguished female from male ECGs {internal validation: area under the curve (AUC) 0.96 [95% confidence interval (CI): 0.96, 0.97]; external validations: AUC 0.89 (95% CI: 0.88, 0.90), 0.94 (95% CI: 0.93, 0.94)}. Sex-misclassified individuals (11%) had a 1.4 times higher mortality risk compared with correctly classified peers. The ventricular rate was the strongest mediating ECG variable (41%, 95% CI: 31%, 56%) in males, while the maximum amplitude of the ST segment was strongest in females (18%, 95% CI: 11%, 39%). Short QRS duration was associated with higher mortality risk. Conclusion: Deep neural networks accurately classify sex based on ECGs. While the proportion of ECG-based sex misclassifications is low, it is an interesting biomarker. Investigation of the causal pathway between misclassification and mortality uncovered new ECG features that might be associated with mortality. Increased emphasis on sex as a biological variable in artificial intelligence is warranted.

Discovering and Visualizing Disease-Specific Electrocardiogram Features Using Deep Learning: Proof-of-Concept in Phospholamban Gene Mutation Carriers.

BACKGROUND: ECG interpretation requires expertise and is mostly based on physician recognition of specific patterns, which may be challenging in rare cardiac diseases. Deep neural networks (DNNs) can discover complex features in ECGs and may facilitate the detection of novel features which possibly play a pathophysiological role in relatively unknown diseases. Using a cohort of PLN (phospholamban) p.Arg14del mutation carriers, we aimed to investigate whether a novel DNN-based approach can identify established ECG features, but moreover, we aimed to expand our knowledge on novel ECG features in these patients. METHODS: A DNN was developed on 12-lead median beat ECGs of 69 patients and 1380 matched controls and independently evaluated on 17 patients and 340 controls. Differentiating features were visualized using Guided Gradient Class Activation Mapping++. Novel ECG features were tested for their diagnostic value by adding them to a logistic regression model including established ECG features. RESULTS: The DNN showed excellent discriminatory performance with a c-statistic of 0.95 (95% CI, 0.91-0.99) and sensitivity and specificity of 0.82 and 0.93, respectively. Visualizations revealed established ECG features (low QRS voltages and T-wave inversions), specified these features (eg, R- and T-wave attenuation in V2/V3) and identified novel PLN-specific ECG features (eg, increased PR-duration). The logistic regression baseline model improved significantly when augmented with the identified features (P<0.001). CONCLUSIONS: A DNN-based feature detection approach was able to discover and visualize disease-specific ECG features in PLN mutation carriers and revealed yet unidentified features. This novel approach may help advance diagnostic capabilities in daily practice.

Efficacy of multi-electrode linear irreversible electroporation.

AIMS: We investigated the efficacy of linear multi-electrode irreversible electroporation (IRE) ablation in a porcine model. METHODS AND RESULTS: The study was performed in six pigs (weight 60-75 kg). After median sternotomy and opening of the pericardium, a pericardial cradle was formed and filled with blood. A linear seven polar 7-Fr electrode catheter with 2.5 mm electrodes and 2.5 mm inter-electrode spacing was placed in good contact with epicardial tissue. A single IRE application was delivered using 50 J at one site and 100 J at two other sites, in random sequence, using a standard monophasic defibrillator connected to all seven electrodes connected in parallel. The pericardium and thorax were closed and after 3 weeks survival animals were euthanized. A total of 82 histological sections from all 18 electroporation lesions were analysed. A total of seven 50 J and fourteen 100 J epicardial IRE applications were performed. Mean peak voltages at 50 and 100 J were 1079.2 V ± 81.1 and 1609.5 V ± 56.8, with a mean peak current of 15.4 A ± 2.3 and 20.2 A ± 1.7, respectively. Median depth of the 50 and 100 J lesions were 3.2 mm [interquartile range (IQR) 3.1-3.6] and 5.5 mm (IQR 4.6-6.6) (P < 0.001), respectively. Median lesion width of the 50 and 100 J lesions was 3.9 mm (IQR 3.7-4.8) and 5.4 mm (IQR 5.0-6.3), respectively (P < 0.001). Longitudinal sections showed continuous lesions for 100 J applications. CONCLUSION: Epicardial multi-electrode linear application of IRE pulses is effective in creating continuous deep lesions.