Electrocardiographic T-wave morphology and risk of mortality

Reliable peak detection and feature extraction for wireless electrocardiograms

The electrocardiogram (ECG) is a vital device to examine the electrical activities of the heart. It is useful for diagnosing cardiovascular diseases, which often manifest themselves through alterations in the ECG signals' characteristics. These alterations are primarily observed in the signals' key components: the Q, R, S, T, and P peaks. At present, cardiologists typically rely on manual inspection of ECG measurements taken in controlled environments, such as hospitals and clinics, but most cardiac conditions reveal themselves outside clinical settings, when patients freely move and exert. In this paper, we dynamically identify and extract prominent ECG features in measurements taken outside clinical settings by subjects who have no medical training. The activities we consider are typical activities cardiac patients carry out in residential and rehabilitation environments, such as sitting, climbing up and down stairs, and standing up. To achieve accurate feature extraction, we employ adaptive thresholding and localization techniques. Our approach achieves promising results, with an average% for R peak detection and 92% for Q and S peaks detection. Similarly, our approach enables the detection of T and P peaks with an average accuracy of 87% and 84%, respectively.

Short QT intervals in African lions

The cardiac conduction system in large carnivores, such as the African lion (Panthera leo), represents a significant knowledge gap in both veterinary science and in cardiac electrophysiology. Short QT intervals have been reported from zoo-kept, anaesthetized lions, and our goal was to record the first ECGs from wild, conscious lions roaming freely, and compare them to zoo-kept lions under the hypothesis that short QT is unique to zoo-kept lions. Macroscopic and histological examinations were performed on heart tissue removed from nine healthy zoo lions. ECGs were recorded from the nine anaesthetized zoo-kept lions, and from 15 anaesthetized and conscious wild lions in Africa. Our histological and topographical description of the lion's heart matched what has previously been published. In conscious lions, the ECG recordings revealed a mean heart rate of 70 ± 4 beats/min, with faster heart rates during the night. PQ and QT intervals were heart rate dependent in the conscious lions. Interestingly, QT intervals recorded in wild lions were markedly longer than QT intervals from zoo lions (398 ± 40 vs. 297 ± 9 ms, respectively; P < 0.0001). Anaesthesia or heart rate did not account for this difference. We provide a comprehensive description of the cardiac anatomy and electrophysiology of wild and zoo-kept lions. QT intervals were significantly shorter in zoo lions, suggesting functional disparities in cardiac electrophysiology between wild and zoo-kept lions, potentially related to physical fitness. These findings underscore the plasticity of cardiac electrophysiology and may be of value when reintroducing endangered species into the wild and when managing lions in human care.

Decoding 2.3 million ECGs: interpretable deep learning for advancing cardiovascular diagnosis and mortality risk stratification

Aims

Electrocardiogram (ECG) is widely considered the primary test for evaluating cardiovascular diseases. However, the use of artificial intelligence (AI) to advance these medical practices and learn new clinical insights from ECGs remains largely unexplored. We hypothesize that AI models with a specific design can provide fine-grained interpretation of ECGs to advance cardiovascular diagnosis, stratify mortality risks, and identify new clinically useful information.

Methods and results

Utilizing a data set of 2 322 513 ECGs collected from 1 558 772 patients with 7 years follow-up, we developed a deep-learning model with state-of-the-art granularity for the interpretable diagnosis of cardiac abnormalities, gender identification, and hypertension screening solely from ECGs, which are then used to stratify the risk of mortality. The model achieved the area under the receiver operating characteristic curve (AUC) scores of 0.998 (95% confidence interval (CI), 0.995-0.999), 0.964 (95% CI, 0.963-0.965), and 0.839 (95% CI, 0.837-0.841) for the three diagnostic tasks separately. Using ECG-predicted results, we find high risks of mortality for subjects with sinus tachycardia (adjusted hazard ratio (HR) of 2.24, 1.96-2.57), and atrial fibrillation (adjusted HR of 2.22, 1.99-2.48). We further use salient morphologies produced by the deep-learning model to identify key ECG leads that achieved similar performance for the three diagnoses, and we find that the V1 ECG lead is important for hypertension screening and mortality risk stratification of hypertensive cohorts, with an AUC of 0.816 (0.814-0.818) and a univariate HR of 1.70 (1.61-1.79) for the two tasks separately.

Conclusion

Using ECGs alone, our developed model showed cardiologist-level accuracy in interpretable cardiac diagnosis and the advancement in mortality risk stratification. In addition, it demonstrated the potential to facilitate clinical knowledge discovery for gender and hypertension detection which are not readily available.

Prevalence and prognostic value of electrocardiographic abnormalities in hypokalemia: A multicenter cohort study

BACKGROUND

Hypokalemia is common in hospitalized patients and associated with ECG abnormalities. The prevalence and prognostic value of ECG abnormalities in hypokalemic patients are, however, not well established.

METHODS

The study was a multicentered cohort study, including all ault patients with an ECG and potassium level <4.4 mmol/L recorded at arrival to four emergency departments in Denmark and Sweden. Using computerized measurements from ECGs, we investigated the relationship between potassium levels and heart rate, QRS duration, corrected QT (QTc) interval, ST-segment depressions, T-wave flattening, and T-wave inversion using cubic splines. Within strata of potassium levels, we further estimated the hazard ratio (HR) for 7-day mortality, admission to the intensive care unit (ICU), and diagnosis of ventricular arrhythmia or cardiac arrest, comparing patients with and without specific ECG abnormalities matched 1:2 on propensity scores.

RESULTS

Among 79,599 included patients, decreasing potassium levels were associated with a concentration-dependent increase in all investigated ECG variables. ECG abnormalities were present in 40% of hypokalemic patients ([K+ ] <3.5 mmol/L), with T-wave flattening, ST-segment depression, and QTc prolongation occurring in 27%, 16%, and 14%. In patients with mild hypokalemia ([K+ ] 3.0-3.4 mmol/L), a heart rate >100 bpm, ST-depressions, and T-wave inversion were associated with increased HRs for 7-day mortality and ICU admission, whereas only a heart rate >100 bpm predicted both mortality and ICU admission among patients with [K+ ] <3.0 mmol/L. HR estimates were, however, similar to those in eukalemic patients. The low number of events with ventricular arrhythmia limited evaluation for this outcome.

CONCLUSIONS

ECG abnormalities were common in hypokalemic patients, but they are poor prognostic markers for short-term adverse events under the current standard of care.

T-wave morphology abnormalities in the STREAM stage 1 trial

BACKGROUND

Shorter regimens for drug-resistant tuberculosis (DR-TB) have non-inferior efficacy compared with longer regimens, but QT prolongation is a concern. T-wave morphology abnormalities may be a predictor of QT prolongation.

RESEARCH DESIGN AND METHODS

STREAM Stage 1 was a randomized controlled trial in rifampicin-resistant TB, comparing short and long regimens. All participants had regular ECGs. QT/QTcF prolongation (≥500 ms or increase in ≥60 ms from baseline) was more common on the short regimen which contained high-dose moxifloxacin and clofazimine. Blinded ECGs were selected from the baseline, early (weeks 1-4), and late (weeks 12-36) time points. T-wave morphology was categorized as normal or abnormal (notched, asymmetric, flat-wave, flat peak, or broad). Differences between groups were assessed using Chi-Square tests (paired/unpaired, as appropriate).

RESULTS

Two-hundred participants with available ECGs at relevant times were analyzed (QT prolongation group n = 82; non-prolongation group n = 118). At baseline, 23% (45/200) of participants displayed abnormal T-waves, increasing to 45% (90/200, p < 0.001) at the late time point. Abnormalities were more common in participants allocated the Short regimen (75/117, 64%) than the Long (14/38, 36.8%, p = 0.003); these occurred prior to QT/QTcF ≥500 ms in 53% of the participants (Long 2/5; Short 14/25).

CONCLUSIONS

T-wave abnormalities may help identify patients at risk of QT prolongation on DR-TB treatment.

TRIAL REGISTRATION

The trial is registered at ClinicalTrials.gov (CT.gov identifier: NCT02409290). Current Controlled Trial number, ISRCTN78372190.

Synchronous recording of magnetocardiographic and electrocardiographic signals

We present a system for simultaneous recording of the electrocardiogram and the magnetocardiogram. The measurement system contained of printed carbon electrodes and SERF magnetometer. The use of this system confirms that the position of the end of the magnetic T wave extends further than the electric T wave, which is an important indicator for the diagnosis of cardiological patients and for drug arrhythmogenicity. We analyze this phenomenon in depth, and demonstrate, that it originates from the fundamental difference between electric and magnetic measurements. The measured value is always bipolar since the electric measurements require two electrodes. We demonstrate how the dual electric and magnetic measuring system adds a new information to the commonly used electrocardiographic diagnosis. The ECG should be interpreted as the spatial asymmetry of the electric cardiac potential, and not as the potential itself. The results seem to prove, that the relation between the magnetic and the electric imaging of neural activities may be broadly applied for the benefit of medical diagnosis in cardiology and many other fields, where the neural activity is measured. This is a pilot study which requires further confirmation at the clinical level.

The Heart of Rett Syndrome: A Quantitative Analysis of Cardiac Repolarization

Background

Rett syndrome (RTT) is a developmental encephalopathy disorder that is associated with a high incidence of sudden death presumably from cardiorespiratory etiologies. Electrocardiogram (ECG) abnormalities, such as prolonged heart-rate corrected QT (QTc) interval, are markers of cardiac repolarization and are associated with potentially lethal ventricular arrhythmias. This study investigates the cardiac repolarization characteristics of RTT patients, including QTc and T-wave morphology characteristics.

Methods

A retrospective quantitative analysis on 110 RTT patients and 124 age and sex-matched healthy controls was conducted.

Results

RTT patients had longer QTc, more abnormal T-wave morphology, and greater heterogeneity of cardiac repolarization parameters compared to controls. Even RTT patients without prolonged QTc had more abnormal ECG and T-wave characteristics than controls. Among RTT patients, MECP2 patients had prolonged QTc compared to CDKL5 and FOXG1 patients. A subset of five RTT patients who died had normal QTc, but more abnormal T-wave morphology than the remaining RTT patients.

Conclusions

Cardiac repolarization abnormalities are present in RTT patients, even without long QTc. T-wave morphology is related to RTT genotype and may be predictive of mortality. These findings could be used to help the management and monitoring of RTT patients.

Mortality risk prediction of the electrocardiogram as an informative indicator of cardiovascular diseases

Background

The electrocardiogram (ECG) may be the most popular test in the management of cardiovascular disease (CVD). Although wide applications of artificial intelligence (AI)-enabled ECG have been developed, an integrating indicator for CVD risk stratification was not investigated. Since mortality may be the most important global outcome, this study aimed to develop a survival deep learning model (DLM) to establish a critical ECG value and explore the associations with various CVD events.

Methods

We trained a DLM with 451,950 12-lead resting ECGs obtained from 210,552 patients, for whom 23,592 events occurred. The internal validation set included 27,808 patients with one ECG for each patient. The external validations were performed in a community hospital with 33,047 patients and two transnational data sets with 233,647 and 1631 ECGs. We distinguished the cause of mortality and additionally investigated CVD-related outcomes, including new-onset acute myocardial infarction (AMI), stroke (STK), and heart failure (HF).

Results

The DLM achieved C-indices of 0.858/0.836 in internal/external validation sets by using ECG over a 10-year period. The high-mortality-risk group identified by the proposed DLM presented a hazard ratio (HR) of 14.16 (95% confidence interval (CI): 11.33-17.70) compared to the low-risk group in the internal validation and presented a higher risk of cardiovascular (CV) mortality (HR: 18.50, 95% CI: 9.82-34.84), non-CV mortality (HR: 13.68, 95% CI: 10.76-17.38), AMI (HR: 4.01, 95% CI: 2.24-7.17), STK (HR: 2.15, 95% CI: 1.70-2.72), and HF (HR: 6.66, 95% CI: 4.54-9.77), which was consistent in an independent community hospital. The transnational validation also revealed HRs of 4.91 (95% CI: 2.63-9.16) and 2.29 (95% CI: 2.15-2.44) for all-cause mortality in the SaMi-Trop and Clinical Outcomes in Digital Electrocardiography 15% (CODE15) cohorts.

Conclusions

The mortality risk by AI-enabled ECG may be applied in passive electronic-health-record-based CVD risk screening, which may identify more asymptomatic and unaware high-risk patients.

Celebrities in the heart, strangers in the pancreatic beta cell: Voltage-gated potassium channels Kv 7.1 and Kv 11.1 bridge long QT syndrome with hyperinsulinaemia as well as type 2 diabetes

Voltage‐gated potassium (K_v) channels play an important role in the repolarization of a variety of excitable tissues, including in the cardiomyocyte and the pancreatic beta cell. Recently, individuals carrying loss‐of‐function (LoF) mutations in KCNQ1, encoding K_v7.1, and KCNH2 (hERG), encoding K_v11.1, were found to exhibit post‐prandial hyperinsulinaemia and episodes of hypoglycaemia. These LoF mutations also cause the cardiac disorder long QT syndrome (LQTS), which can be aggravated by hypoglycaemia. Interestingly, patients with LQTS also have a higher burden of diabetes compared to the background population, an apparent paradox in relation to the hyperinsulinaemic phenotype, and KCNQ1 has been identified as a type 2 diabetes risk gene. This review article summarizes the involvement of delayed rectifier K⁺ channels in pancreatic beta cell function, with emphasis on K_v7.1 and K_v11.1, using the cardiomyocyte for context. The functional and clinical consequences of LoF mutations and polymorphisms in these channels on blood glucose homeostasis are explored using evidence from pre‐clinical, clinical and genome‐wide association studies, thereby evaluating the link between LQTS, hyperinsulinaemia and type 2 diabetes.

Effect of hydroxychloroquine on the cardiac ventricular repolarization: A randomized clinical trial

AIMS

Hydroxychloroquine has been suggested as possible treatment for severe acute respiratory syndrome-coronavirus-2. Studies reported an increased risk of QTcF-prolongation after treatment with hydroxychloroquine. The aim of this study was to analyse the concentration-dependent effects of hydroxychloroquine on the ventricular repolarization, including QTcF-duration and T-wave morphology.

METHODS

Twenty young (≤30 y) and 20 elderly (65-75 y) healthy male subjects were included. Subjects were randomized to receive either a total dose of 2400 mg hydroxychloroquine over 5 days, or placebo (ratio 1:1). Follow-up duration was 28 days. Electrocardiograms (ECGs) were recorded as triplicate at baseline and 4 postdose single recordings, followed by hydroxychloroquine concentration measurements. ECG intervals (RR, QRS, PR, QTcF, J-Tpc, Tp-Te) and T-wave morphology, measured with the morphology combination score, were analysed with a prespecified linear mixed effects concentration-effect model.

RESULTS

There were no significant associations between hydroxychloroquine concentrations and ECG characteristics, including RR-, QRS- and QTcF-interval (P = .09, .34, .25). Mean ΔΔQTcF-interval prolongation did not exceed 5 ms and the upper limit of the 90% confidence interval did not exceed 10 ms at the highest measured concentrations (200 ng/mL). There were no associations between hydroxychloroquine concentration and the T-wave morphology (P = .34 for morphology combination score). There was no significant effect of age group on ECG characteristics.

CONCLUSION

In this study, hydroxychloroquine did not affect ventricular repolarization, including the QTcF-interval and T-wave morphology, at plasma concentrations up to 200 ng/mL. Based on this analysis, hydroxychloroquine does not appear to increase the risk of QTcF-induced arrhythmias.

Explaining deep neural networks for knowledge discovery in electrocardiogram analysis

Deep learning-based tools may annotate and interpret medical data more quickly, consistently, and accurately than medical doctors. However, as medical doctors are ultimately responsible for clinical decision-making, any deep learning-based prediction should be accompanied by an explanation that a human can understand. We present an approach called electrocardiogram gradient class activation map (ECGradCAM), which is used to generate attention maps and explain the reasoning behind deep learning-based decision-making in ECG analysis. Attention maps may be used in the clinic to aid diagnosis, discover new medical knowledge, and identify novel features and characteristics of medical tests. In this paper, we showcase how ECGradCAM attention maps can unmask how a novel deep learning model measures both amplitudes and intervals in 12-lead electrocardiograms, and we show an example of how attention maps may be used to develop novel ECG features.