Search

[Figure: see text].

Artificial intelligence (AI) and machine learning (ML) in medicine are currently areas of intense exploration, showing potential to automate human tasks and even perform tasks beyond human capabilities. Literacy and understanding of AI/ML methods are becoming increasingly important to researchers and clinicians. The first objective of this review is to provide the novice reader with literacy of AI/ML methods and provide a foundation for how one might conduct an ML study. We provide a technical overview of some of the most commonly used terms, techniques, and challenges in AI/ML studies, with reference to recent studies in cardiac electrophysiology to illustrate key points. The second objective of this review is to use examples from recent literature to discuss how AI and ML are changing clinical practice and research in cardiac electrophysiology, with emphasis on disease detection and diagnosis, prediction of patient outcomes, and novel characterization of disease. The final objective is to highlight important considerations and challenges for appropriate validation, adoption, and deployment of AI technologies into clinical practice.

Background: Optical sensors on wearable devices can detect irregular pulses. The ability of a smartwatch application (app) to identify atrial fibrillation during typical use is unknown., Methods: Participants without atrial fibrillation (as reported by the participants themselves) used a smartphone (Apple iPhone) app to consent to monitoring. If a smartwatch-based irregular pulse notification algorithm identified possible atrial fibrillation, a telemedicine visit was initiated and an electrocardiography (ECG) patch was mailed to the participant, to be worn for up to 7 days. Surveys were administered 90 days after notification of the irregular pulse and at the end of the study. The main objectives were to estimate the proportion of notified participants with atrial fibrillation shown on an ECG patch and the positive predictive value of irregular pulse intervals with a targeted confidence interval width of 0.10., Results: We recruited 419,297 participants over 8 months. Over a median of 117 days of monitoring, 2161 participants (0.52%) received notifications of irregular pulse. Among the 450 participants who returned ECG patches containing data that could be analyzed - which had been applied, on average, 13 days after notification - atrial fibrillation was present in 34% (97.5% confidence interval [CI], 29 to 39) overall and in 35% (97.5% CI, 27 to 43) of participants 65 years of age or older. Among participants who were notified of an irregular pulse, the positive predictive value was 0.84 (95% CI, 0.76 to 0.92) for observing atrial fibrillation on the ECG simultaneously with a subsequent irregular pulse notification and 0.71 (97.5% CI, 0.69 to 0.74) for observing atrial fibrillation on the ECG simultaneously with a subsequent irregular tachogram. Of 1376 notified participants who returned a 90-day survey, 57% contacted health care providers outside the study. There were no reports of serious app-related adverse events., Conclusions: The probability of receiving an irregular pulse notification was low. Among participants who received notification of an irregular pulse, 34% had atrial fibrillation on subsequent ECG patch readings and 84% of notifications were concordant with atrial fibrillation. This siteless (no on-site visits were required for the participants), pragmatic study design provides a foundation for large-scale pragmatic studies in which outcomes or adherence can be reliably assessed with user-owned devices. (Funded by Apple; Apple Heart Study ClinicalTrials.gov number, NCT03335800.)., (Copyright © 2019 Massachusetts Medical Society.)

Computerized electrocardiogram (ECG) interpretation plays a critical role in the clinical ECG workflow 1 . Widely available digital ECG data and the algorithmic paradigm of deep learning 2 present an opportunity to substantially improve the accuracy and scalability of automated ECG analysis. However, a comprehensive evaluation of an end-to-end deep learning approach for ECG analysis across a wide variety of diagnostic classes has not been previously reported. Here, we develop a deep neural network (DNN) to classify 12 rhythm classes using 91,232 single-lead ECGs from 53,549 patients who used a single-lead ambulatory ECG monitoring device. When validated against an independent test dataset annotated by a consensus committee of board-certified practicing cardiologists, the DNN achieved an average area under the receiver operating characteristic curve (ROC) of 0.97. The average F 1 score, which is the harmonic mean of the positive predictive value and sensitivity, for the DNN (0.837) exceeded that of average cardiologists (0.780). With specificity fixed at the average specificity achieved by cardiologists, the sensitivity of the DNN exceeded the average cardiologist sensitivity for all rhythm classes. These findings demonstrate that an end-to-end deep learning approach can classify a broad range of distinct arrhythmias from single-lead ECGs with high diagnostic performance similar to that of cardiologists. If confirmed in clinical settings, this approach could reduce the rate of misdiagnosed computerized ECG interpretations and improve the efficiency of expert human ECG interpretation by accurately triaging or prioritizing the most urgent conditions.

This white paper, prepared by members of the Cardiac Safety Research Consortium (CSRC), discusses important issues regarding scientific and clinical aspects of long-term electrocardiographic safety monitoring during clinical drug development. To promote multistakeholder discussion of this topic, a Cardiac Safety Research Consortium-sponsored Think Tank was held on 2 December 2015 at the American College of Cardiology's Heart House in Washington, DC. The goal of the Think Tank was to explore how and under what circumstances new and evolving ambulatory monitoring technologies could be used to improve and streamline drug development. This paper provides a detailed summary of discussions at the Think Tank: it does not represent regulatory guidance., (Copyright © 2017 Elsevier Inc. All rights reserved.)

Background: The significance of early repolarization, particularly regarding the morphology of the R-wave downslope, has come under question., Methods and Results: We evaluated 29 281 resting ambulatory ECGs from the VA Palo Alto Health Care System. With PR interval as the isoelectric line and amplitude criteria ≥0.1 mV, ST-segment elevation is defined at the end of the QRS, J wave as an upward deflection, and slur as a conduction delay on the QRS downstroke. Associations of ST-segment elevation patterns, J waves, and slurs with cardiovascular mortality were analyzed with Cox analysis. With a median follow-up of 7.6 years, there were 1995 cardiac deaths. Of 29 281 subjects, 87% were male (55±14 years) and 13% were female (56±17 years); 13% were black, 6% were Hispanic, and 81% were white or other. Six hundred sixty-four (2.3%) had inferior or lateral ST-segment elevation: 185 (0.6%) in inferior leads and 479 (1.6%) in lateral leads, 163 (0.6%) in both, and 0.4% had global elevation. A total of 4041 ECGs were analyzed with enhanced display, and 583 (14%) had J waves or slurring, which were more prevalent in those with than in those without ST-segment elevation (61% versus 13%; P<0.001). ST-segment elevation occurred more in those with than in those without J waves or slurs (12% versus 1.3%; P<0.001). Except when involving only inferior leads, all components of early repolarization were more common in young individuals, male subjects, blacks, and those with bradycardia. All patterns and components of early repolarization were associated with decreased cardiovascular mortality, but this was not significant after adjustment for age., Conclusions: We found no significant association between any components of early repolarization and cardiac mortality.

The technological evolution and widespread availability of wearables and handheld ECG devices capable of screening for atrial fibrillation (AF), and their promotion directly to consumers, has focused attention of health care professionals and patient organizations on consumer-led AF screening. In this Frontiers review, members of the AF-SCREEN International Collaboration provide a critical appraisal of this rapidly evolving field to increase awareness of the complexities and uncertainties surrounding consumer-led AF screening. Although there are numerous commercially available devices directly marketed to consumers for AF monitoring and identification of unrecognized AF, health care professional-led randomized controlled studies using multiple ECG recordings or continuous ECG monitoring to detect AF have failed to demonstrate a significant reduction in stroke. Although it remains uncertain if consumer-led AF screening reduces stroke, it could increase early diagnosis of AF and facilitate an integrated approach, including appropriate anticoagulation, rate or rhythm management, and risk factor modification to reduce complications. Companies marketing AF screening devices should report the accuracy and performance of their products in high- and low-risk populations and avoid claims about clinical outcomes unless improvement is demonstrated in randomized clinical trials. Generally, the diagnostic yield of AF screening increases with the number, duration, and temporal dispersion of screening sessions, but the prognostic importance may be less than for AF detected by single-time point screening, which is largely permanent, persistent, or high-burden paroxysmal AF. Consumer-initiated ECG recordings suggesting possible AF always require confirmation by a health care professional experienced in ECG reading, whereas suspicion of AF on the basis of photoplethysmography must be confirmed with an ECG. Consumer-led AF screening is unlikely to be cost-effective for stroke prevention in the predominantly young, early adopters of this technology. Studies in older people at higher stroke risk are required to demonstrate both effectiveness and cost-effectiveness. The direct interaction between companies and consumers creates new regulatory gaps in relation to data privacy and the registration of consumer apps and devices. Although several barriers for optimal use of consumer-led screening exist, results of large, ongoing trials, powered to detect clinical outcomes, are required before health care professionals should support widespread adoption of consumer-led AF screening.

Early identification of atrial fibrillation (AF) can reduce the risk of stroke, heart failure, and other serious cardiovascular outcomes. However, paroxysmal AF may not be detected even after a two-week continuous monitoring period. We developed a model to quantify the risk of near-term AF in a two-week period, based on AF-free ECG intervals of up to 24 h from 459,889 patch-based ambulatory single-lead ECG (modified lead II) recordings of up to 14 days. A deep learning model was used to integrate ECG morphology data with demographic and heart rhythm features toward AF prediction. Observing a 1-day AF-free ECG recording, the model with deep learning features produced the most accurate prediction of near-term AF with an area under the curve AUC = 0.80 (95% confidence interval, CI = 0.79–0.81), significantly improving discrimination compared to demographic metrics alone (AUC 0.67; CI = 0.66–0.68). Our model was able to predict incident AF over a two-week time frame with high discrimination, based on AF-free single-lead ECG recordings of various lengths. Application of the model may enable a digital strategy for improving diagnostic capture of AF by risk stratifying individuals with AF-negative ambulatory monitoring for prolonged or recurrent monitoring, potentially leading to more rapid initiation of treatment. [ABSTRACT FROM AUTHOR]

The Coronavirus Disease 2019 (COVID-19) pandemic curtailed clinical trial activity. Decentralized clinical trials (DCTs) can expand trial access and reduce exposure risk but their feasibility remains uncertain. We evaluated DCT feasibility for atrial fibrillation (AF) patients on oral anticoagulation (OAC). DeTAP (Decentralized Trial in Afib Patients, NCT04471623) was a 6-month, single-arm, 100% virtual study of 100 AF patients on OAC aged >55 years, recruited traditionally and through social media. Participants enrolled and participated virtually using a mobile application and remote blood pressure (BP) and six-lead electrocardiogram (ECG) sensors. Four engagement-based primary endpoints included changes in pre- versus end-of-study OAC adherence (OACA), and % completion of televisits, surveys, and ECG and BP measurements. Secondary endpoints included survey-based nuisance bleeding and patient feedback. 100 subjects (mean age 70 years, 44% women, 90% White) were recruited in 28 days (traditional: 6 pts; social media: 94 pts in 12 days with >300 waitlisted). Study engagement was high: 91% televisits, 85% surveys, and 99% ECG and 99% BP measurement completion. OACA was unchanged at 6 months (baseline: 97 ± 9%, 6 months: 96 ± 15%, p = 0.39). In patients with low baseline OACA (<90%), there was significant 6-month improvement (85 ± 16% to 96 ± 6%, p < 0.01). 86% of respondents (69/80) expressed willingness to continue in a longer trial. The DeTAP study demonstrated rapid recruitment, high engagement, and physiologic reporting via the integration of digital technologies and dedicated study coordination. These findings may inform DCT designs for future cardiovascular trials. [ABSTRACT FROM AUTHOR]

Background: The risk of recurrence after an initial ischemic stroke or transient ischemic attack (TIA) may be impacted by undiagnosed atrial fibrillation (AF). We therefore assessed the impact of AF diagnosis and timing on stroke/TIA recurrence rates in a large real-world sample of patients.Methods and Results: Using commercial claims data (Truven Health Analytics MarketScan), we performed a retrospective cohort study of patients with an index stroke or TIA event recorded in years 2008 through 2011. Patients were characterized by baseline oral anticoagulation, CHADS2 and CHA2DS2-VASc scores, AF diagnosis and timing with respect to the index stroke, and presence or absence of post-index ambulatory cardiac monitoring. The primary outcome was the recurrence of an ischemic stroke or TIA. Of 179 160 patients (age 67±16.2 years; 53.7% female), the Kaplan-Meier estimate for stroke/TIA recurrence within 1 year was 10.6%. Not having oral anticoagulation prescribed at baseline and having AF first diagnosed >7 days post-stroke (late AF) was highly associated with recurrent stroke/TIA (hazard ratio, 2.0; 95% confidence interval, 1.9-2.1). Among patients with at least 1 year of follow-up, only 2.6% and 9.7% had ambulatory ECG monitoring in the 7 days and 12 months post-stroke, respectively.Conclusions: AF diagnosed after stroke is an important hallmark of recurrent stroke risk. Increasing the low utilization of cardiac monitoring after stroke could identify undiagnosed AF earlier, leading to appropriate oral anticoagulation treatment and a reduction in stroke/TIA recurrence. [ABSTRACT FROM AUTHOR]

Atrial fibrillation (AF) is a very common arrhythmia and significantly increases stroke risk. This risk can be mitigated with oral anticoagulation, but AF is often asymptomatic, or occult, preventing timely detection and treatment. Accordingly, occult AF may cause stroke before it is clinically diagnosed. Currently, guidelines for the early detection and treatment of occult AF are limited. This review addresses recent advancements in occult AF detection methods, identification of populations at high risk for occult AF, the treatment of occult AF with oral anticoagulation, as well as ongoing trials that may answer critically important questions regarding occult AF screening. [ABSTRACT FROM AUTHOR]

Although extending the duration of ambulatory electrocardiographic monitoring beyond 24 to 48 hours can improve the detection of arrhythmias, lead-based (Holter) monitors might be limited by patient compliance and other factors. We, therefore, evaluated compliance, analyzable signal time, interval to arrhythmia detection, and diagnostic yield of the Zio Patch, a novel leadless, electrocardiographic monitoring device in 26,751 consecutive patients. The mean wear time was 7.6 ± 3.6 days, and the median analyzable time was 99% of the total wear time. Among the patients with detected arrhythmias (60.3% of all patients), 29.9% had their first arrhythmia and 51.1% had their first symptom-triggered arrhythmia occur after the initial 48-hour period. Compared with the first 48 hours of monitoring, the overall diagnostic yield was greater when data from the entire Zio Patch wear duration were included for any arrhythmia (62.2% vs 43.9%, p <0.0001) and for any symptomatic arrhythmia (9.7% vs 4.4%, p <0.0001). For paroxysmal atrial fibrillation (AF), the mean interval to the first detection of AF was inversely proportional to the total AF burden, with an increasing proportion occurring after 48 hours (11.2%, 10.5%, 20.8%, and 38.0% for an AF burden of 51% to 75%, 26% to 50%, 1% to 25%, and <1%, respectively). In conclusion, extended monitoring with the Zio Patch for ≼14 days is feasible, with high patient compliance, a high analyzable signal time, and an incremental diagnostic yield beyond 48 hours for all arrhythmia types. These findings could have significant implications for device selection, monitoring duration, and care pathways for arrhythmia evaluation and AF surveillance. [ABSTRACT FROM AUTHOR]