Accuracy of mHealth Devices for Atrial Fibrillation Screening: Systematic Review

H. Lip G, Thomas GN, J. Moore D, Nirantharakumar K, Kumarendran B, Subaschandran K, Kanesamoorthy S, Uruthirakumar P, Guruparan M. Diagnostic accuracy of digital technologies compared with 12-lead ECG in the diagnosis of atrial fibrillation in adults: A protocol for a systematic review

BACKGROUND

Atrial fibrillation (AF) is the most prevalent cardiac arrhythmia in the world. AF increases the risk of stroke 5-fold, though the risk can be reduced with appropriate treatment. Therefore, early diagnosis is imperative but remains a global challenge. In low-and middle-income countries (LMICs), a lack of diagnostic equipment and under-resourced healthcare systems generate further barriers. The rapid development of digital technologies that are capable of diagnosing AF remotely and cost-effectively could prove beneficial for LMICs. However, evidence is lacking on what digital technologies exist and how they compare in regards to diagnostic accuracy. We aim to systematically review the diagnostic accuracy of all digital technologies capable of AF diagnosis.

METHODS

MEDLINE, Embase and Web of Science will be searched for eligible studies. Free text terms will be combined with corresponding index terms where available and searches will not be limited by language nor time of publication. Cohort or cross-sectional studies comprising adult (≥18 years) participants will be included. Only studies that use a 12-lead ECG as the reference test (comparator) and report outcomes of sensitivity, specificity, the diagnostic odds ratio (DOR) or the positive and negative predictive value (PPV and NPV) will be included (or if they provide sufficient data to calculate these outcomes). Two reviewers will independently assess articles for inclusion, extract data using a piloted tool and assess risk of bias using the QUADAS-2 tool. The feasibility of a meta-analysis will be determined by assessing heterogeneity across the studies, grouped by index device, diagnostic threshold and setting. If a meta-analysis is feasible for any index device, pooled sensitivity and specificity will be calculated using a random effect model and presented in forest plots.

DISCUSSION

The findings of our review will provide a comprehensive synthesis of the diagnostic accuracy of available digital technologies capable for diagnosing AF. Thus, this review will aid in the identification of which devices could be further trialed and implemented, particularly in a LMIC setting, to improve the early diagnosis of AF.

TRIAL REGISTRATION

Systematic review registration: PROSPERO registration number is CRD42021290542. https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42021290542.

Adherence to a handheld device-based atrial fibrillation screening protocol is associated with clinical outcomes

OBJECTIVE

To evaluate adherence and adherence consistency to the handheld ECG device-based screening protocol and their association with adverse cerebral and cardiovascular outcomes in two systematic atrial fibrillation (AF) screening programmes.

METHODS

In 2012 (Systematic ECG Screening for Atrial Fibrillation Among 75-Year Old Subjects in the Region of Stockholm and Halland, Sweden (STROKESTOP) study) and 2016 (Stepwise mass screening for atrial fibrillation using N-terminal pro b-type natriuretic peptide (STROKESTOP II) study), half of all 75- and 76-year-old inhabitants of up to two Swedish regions were invited to participate in a systematic AF screening programme. Participants were instructed to perform 30-second measurements twice daily in STROKESTOP and four times daily in STROKESTOP II for 2 weeks. Adherence was defined as the number of measurements performed divided by the number of measurements asked, whereas adherence consistency was defined as the number of days with complete registrations.

RESULTS

In total, 6436 participants (55.7% female) from STROKESTOP and 3712 (59.8% female) from STROKESTOP II were included. Median adherence and adherence consistency were 100 (92-100)% and 12 (11-13) days in STROKESTOP and 90 (75-98)% and 8 (3-11) days in STROKESTOP II. Female sex and lower education were factors associated with both optimal adherence and adherence consistency in both studies. In STROKESTOP, low adherence and adherence consistency were associated with higher risk of adverse cerebral and cardiovascular outcomes (HR for composite primary endpoint 1.30 (1.11 to 1.51), p=0.001), including stroke (HR 1.68 (1.22 to 2.32), p=0.001) and dementia (1.67 (1.27 to 2.19), p<0.001).

CONCLUSIONS

Adherence to twice daily handheld ECG measurements in STROKESTOP was higher than to four times daily measurements in STROKESTOP II. Female sex and lower educational attainment were associated with ≥100% adherence and adherence consistency. Low adherence and adherence consistency were associated with a higher risk of adverse outcomes.

Identification of Atrial Fibrillation With Single-Lead Mobile ECG During Normal Sinus Rhythm Using Deep Learning

BACKGROUND

The acquisition of single-lead electrocardiogram (ECG) from mobile devices offers a more practical approach to arrhythmia detection. Using artificial intelligence for atrial fibrillation (AF) identification enhances screening efficiency. However, the potential of single-lead ECG for AF identification during normal sinus rhythm (NSR) remains under-explored. This study introduces a method to identify AF using single-lead mobile ECG during NSR.

METHODS

We employed three deep learning models: recurrent neural network (RNN), long short-term memory (LSTM), and residual neural networks (ResNet50). From a dataset comprising 13,509 ECGs from 6,719 patients, 10,287 NSR ECGs from 5,170 patients were selected. Single-lead mobile ECGs underwent noise filtering and segmentation into 10-second intervals. A random under-sampling was applied to reduce bias from data imbalance. The final analysis involved 31,767 ECG segments, including 15,157 labeled as masked AF and 16,610 as Healthy.

RESULTS

ResNet50 outperformed the other models, achieving a recall of 79.3%, precision of 65.8%, F1-score of 71.9%, accuracy of 70.5%, and an area under the receiver operating characteristic curve (AUC) of 0.79 in identifying AF from NSR ECGs. Comparative performance scores for RNN and LSTM were 0.75 and 0.74, respectively. In an external validation set, ResNet50 attained an F1-score of 64.1%, recall of 68.9%, precision of 60.0%, accuracy of 63.4%, and AUC of 0.68.

CONCLUSION

The deep learning model using single-lead mobile ECG during NSR effectively identified AF at risk in future. However, further research is needed to enhance the performance of deep learning models for clinical application.

Effectiveness of single-lead ECG devices for detecting atrial fibrillation: An overview of systematic reviews

BACKGROUND

Individuals with atrial fibrillation (AF) are at an increased risk for stroke. Early detection of undiagnosed AF by screening is recommended. Single-lead electrocardiogram (ECG) is the most widely used technology in AF detection. Several systematic reviews on the diagnostic accuracy of single-lead ECG devices for AF detection have been performed but have yielded inconclusive results.

AIMS

The aim of this study was to synthesize the available evidence on the effectiveness of single-lead ECG devices in detecting AF.

METHODS

An overview of systematic reviews was conducted. Five English databases (Cochrane Database of Systematic Reviews, PubMed, Embase, Ovid, and Web of Science) and two Chinese databases (Wanfang and CNKI) were searched from inception to July 31, 2021. Systematic reviews that examined the accuracy of tools based on single-lead ECG technology for detecting AF were included. A narrative data synthesis was performed.

RESULTS

Eight systematic reviews were finally included. Systematic reviews with meta-analysis showed that single-lead ECG-based devices had good sensitivity and specificity (both ≥90%) in detecting AF. According to subgroup analysis, the sensitivities of tools used in populations with a history of AF were all >90%. However, among handheld and thoracic placed single-lead ECG devices, large variations in diagnostic performance were observed.

LINKING EVIDENCE TO ACTION

Single-lead ECG devices can potentially be used for AF detection. Due to the heterogeneity in the study population and tools, future studies are warranted to explore the suitable circumstances in which each tool could be applied for AF screening in an effective and cost-effective manner.

Video-Based Activity Recognition for Automated Motor Assessment of Parkinson's Disease

Over the last decade, video-enabled mobile devices have become ubiquitous, while advances in markerless pose estimation allow an individual's body position to be tracked accurately and efficiently across the frames of a video. Previous work by this and other groups has shown that pose-extracted kinematic features can be used to reliably measure motor impairment in Parkinson's disease (PD). This presents the prospect of developing an asynchronous and scalable, video-based assessment of motor dysfunction. Crucial to this endeavour is the ability to automatically recognise the class of an action being performed, without which manual labelling is required. Representing the evolution of body joint locations as a spatio-temporal graph, we implement a deep-learning model for video and frame-level classification of activities performed according to part 3 of the Movement Disorder Society Unified PD Rating Scale (MDS-UPDRS). We train and validate this system using a dataset of n = 7310 video clips, recorded at 5 independent sites. This approach reaches human-level performance in detecting and classifying periods of activity within monocular video clips. Our framework could support clinical workflows and patient care at scale through applications such as quality monitoring of clinical data collection, automated labelling of video streams, or a module within a remote self-assessment system.

Prevalence and risk factors for atrial fibrillation in a semi-rural sub-Saharan African population: The hEart oF ethiopia: Focus on Atrial Fibrillation (TEFF-AF) Study

Background

There is a scarcity of reported data on the prevalence of atrial fibrillation (AF) in sub-Saharan Africa.

Objectives

To undertake AF screening in semi-rural Ethiopia.

Methods

The TEFF-AF (The hEart oF Ethiopia: Focus on Atrial Fibrillation) study conducted AF screening using a single-lead electrocardiogram device (KardiaMobile) on willing community participants at the Soddo Christian Hospital, Ethiopia. Participants' clinical parameters and medical history were obtained to characterize their risk factor profile, including calculation of CHARGE-AF (Cohorts for Heart and Aging Research in Genomic Epidemiology Atrial Fibrillation) score.

Results

A total of 3000 Ethiopians (median 31 [interquartile range 25-41] years of age; 65% men) were screened. The participants were generally well educated, from the local region and with a low burden of cardiovascular risk factors. A total of 50 participants had a CHARGE-AF score (5-year AF risk) of ≥2%. AF was detected in 13 (0.43%) individuals (median 50 [interquartile range 36-60] years of age; n = 7 men). The prevalence among participants over 40 years of age was 1% (n = 9 of 930). AF prevalence was higher for older age groups, with ≥70 years of age reaching 6.67% (n = 3 of 45). Population prevalence was estimated to be 234 (95% confidence interval 7-460) per 10,000 persons for ≥60 years of age. Four (31%) of the 13 participants with AF had a CHA₂DS₂-VASc (congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, prior stroke or transient ischemic attack or thromboembolism, vascular disease, age 65-74 years, sex category) score of ≥2, and others likely had rheumatic valvular AF, but only 2 of the 13 participants with AF were on oral anticoagulation therapy.

Conclusion

In this semi-rural Ethiopian community of relatively younger participants, AF prevalence was found to be low but increased with increasing age. Mobile single-lead electrocardiogram technology can be used effectively for AF screening in low-resource settings.

Regional implementation of atrial fibrillation screening: benefits and pitfalls

Aims

Despite general awareness that screening for atrial fibrillation (AF) could reduce health hazards, large-scale implementation is lagging behind technological developments. As the successful implementation of a screening programme remains challenging, this study aims to identify facilitating and inhibiting factors from healthcare providers' perspectives.

Methods and results

A mixed-methods approach was used to gather data among practice nurses in primary care in the southern region of the Netherlands to evaluate the implementation of an ongoing single-lead electrocardiogram (ECG)-based AF screening programme. Potential facilitating and inhibiting factors were evaluated using online questionnaires (N = 74/75%) and 14 (of 24) semi-structured in-depth interviews (58.3%). All analyses were performed using SPSS 26.0. In total, 16 682 screenings were performed on an eligible population of 64 000, and 100 new AF cases were detected. Facilitating factors included 'receiving clear instructions' (mean ± SD; 4.12 ± 1.05), 'easy use of the ECG-based device' (4.58 ± 0.68), and 'patient satisfaction' (4.22 ± 0.65). Inhibiting factors were 'time availability' (3.20 ± 1.10), 'insufficient feedback to the practice nurse' (2.15 ± 0.89), 'absence of coordination' (54%), and the 'lack of fitting policy' (32%).

Conclusion

Large-scale regional implementation of an AF screening programme in primary care resulted in a low participation of all eligible patients. Based on the perceived barriers by healthcare providers, future AF screening programmes should create preconditions to fit the intervention into daily routines, appointing an overall project lead and a General Practitioner (GP) as a coordinator within every GP practice.

Cost-effectiveness of Screening for Atrial Fibrillation Using Wearable Devices

Question

Is population-based atrial fibrillation (AF) screening using wearable devices cost-effective?

Findings

In this economic evaluation of 30 million simulated individuals with an age, sex, and comorbidity profile matching the US population aged 65 years or older, AF screening using wearable devices was cost-effective, with the overall preferred strategy identified as wearable photoplethysmography, followed conditionally by wearable electrocardiography with patch monitor confirmation (incremental cost-effectiveness ratio, $57 894 per quality-adjusted life-year). The cost-effectiveness of screening was consistent across multiple scenarios, including strata of sex, screening at earlier ages, and with variation in the association of anticoagulation with risk of stroke associated with screening-detected AF.

Meaning

This study suggests that contemporary AF screening using wearable devices may be cost-effective.

Importance

Undiagnosed atrial fibrillation (AF) is an important cause of stroke. Screening for AF using wrist-worn wearable devices may prevent strokes, but their cost-effectiveness is unknown.

Objective

To evaluate the cost-effectiveness of contemporary AF screening strategies, particularly wrist-worn wearable devices.

Design, Setting, and Participants

This economic evaluation used a microsimulation decision-analytic model and was conducted from September 8, 2020, to May 23, 2022, comprising 30 million simulated individuals with an age, sex, and comorbidity profile matching the US population aged 65 years or older.

Interventions

Eight AF screening strategies, with 6 using wrist-worn wearable devices (watch or band photoplethysmography, with or without watch or band electrocardiography) and 2 using traditional modalities (ie, pulse palpation and 12-lead electrocardiogram) vs no screening.

Main Outcomes and Measures

The primary outcome was the incremental cost-effectiveness ratio, defined as US dollars per quality-adjusted life-year (QALY). Secondary measures included rates of stroke and major bleeding.

Results

In the base case analysis of this model, the mean (SD) age was 72.5 (7.5) years, and 50% of the individuals were women. All 6 screening strategies using wrist-worn wearable devices were estimated to be more effective than no screening (range of QALYs gained vs no screening, 226-957 per 100 000 individuals) and were associated with greater relative benefit than screening using traditional modalities (range of QALYs gained vs no screening, −116 to 93 per 100 000 individuals). Compared with no screening, screening using wrist-worn wearable devices was associated with a reduction in stroke incidence by 20 to 23 per 100 000 person-years but an increase in major bleeding by 20 to 44 per 100 000 person-years. The overall preferred strategy was wearable photoplethysmography, followed conditionally by wearable electrocardiography with patch monitor confirmation, which had an incremental cost-effectiveness ratio of $57 894 per QALY, meeting the acceptability threshold of $100 000 per QALY. The cost-effectiveness of screening was consistent across multiple scenarios, including strata of sex, screening at earlier ages (eg, ≥50 years), and with variation in the association of anticoagulation with risk of stroke in the setting of screening-detected AF.

Conclusions and Relevance

This economic evaluation of AF screening using a microsimulation decision-analytic model suggests that screening using wearable devices is cost-effective compared with either no screening or AF screening using traditional methods.

Comparison of Apple Watch Series 4 vs. KardiaMobile: A Tale of Two Devices

Background

The Apple Watch Series 4 (AW4) and the KardiaMobile single bipolar lead model (KM) are 2 of the most popular US Food & Drug Administration (FDA)-approved commercial heart trackers. However, a lack of knowledge remains regarding their rhythm-detection accuracy in real-life clinical situations. This paper aims to determine the practicality of using an AW4 or a KM in modern medical practice, by assessing the accuracy of each in identifying heart rhythms and heart rate.

Methods

Participants from the Toronto Heart Centre clinic were enrolled from January 2019 to December 2019. They had a 12-lead electrocardiogram (ECG), followed by wearing the AW4 watch (OS 5.3), and pressing on the KM electrode plates, within the span of 5 minutes of one another. Each session involved a 12-lead ECG, an ECG from each device, and AW4’s photoplethysmography function (APPG).

Results

Of 200 participants, 162 (81%) were in sinus rhythm, and 38 (19%) had atrial fibrillation. The rhythm-detection accuracy for sinus rhythm was 100% for the AW4, and 99.03% for the KM. For atrial fibrillation, accuracy was 90.48% for the AW4, and 100% for the KM. The heart rate accuracy for sinus rhythm was 94.39% for the KM, 90.65% for the APPG, and 96.26% for the Apple ECG function. The heart rate accuracy for atrial fibrillation was 91.30% for the KM, 82.61% for the APPG, and 86.96% for the Apple ECG function.

Conclusions

Both the AW4 and the KM could reliably detect rhythm and heart rate in real-life clinical situations. However, a nonsignificant trend occurred toward better rhythm detection and accuracy with KM, compared with AW4. The difference is mainly due to artifacts (eg, tremors) and the fit of the strap for AW4. The findings have important implications for how these consumer devices can be used in real-life clinical settings.

Continuous mHealth Patch Monitoring for the Algorithm-Based Detection of Atrial Fibrillation: Feasibility and Diagnostic Accuracy Study

Background

The detection of atrial fibrillation (AF) is a major clinical challenge as AF is often paroxysmal and asymptomatic. Novel mobile health (mHealth) technologies could provide a cost-effective and reliable solution for AF screening. However, many of these techniques have not been clinically validated.

Objective

The purpose of this study is to evaluate the feasibility and reliability of artificial intelligence (AI) arrhythmia analysis for AF detection with an mHealth patch device designed for personal well-being.

Methods

Patients (N=178) with an AF (n=79, 44%) or sinus rhythm (n=99, 56%) were recruited from the emergency care department. A single-lead, 24-hour, electrocardiogram-based heart rate variability (HRV) measurement was recorded with the mHealth patch device and analyzed with a novel AI arrhythmia analysis software. Simultaneously registered 3-lead electrocardiograms (Holter) served as the gold standard for the final rhythm diagnostics.

Results

Of the HRV data produced by the single-lead mHealth patch, 81.5% (3099/3802 hours) were interpretable, and the subject-based median for interpretable HRV data was 99% (25th percentile=77% and 75th percentile=100%). The AI arrhythmia detection algorithm detected AF correctly in all patients in the AF group and suggested the presence of AF in 5 patients in the control group, resulting in a subject-based AF detection accuracy of 97.2%, a sensitivity of 100%, and a specificity of 94.9%. The time-based AF detection accuracy, sensitivity, and specificity of the AI arrhythmia detection algorithm were 98.7%, 99.6%, and 98.0%, respectively.

Conclusions

The 24-hour HRV monitoring by the mHealth patch device enabled accurate automatic AF detection. Thus, the wearable mHealth patch device with AI arrhythmia analysis is a novel method for AF screening.

Trial Registration

ClinicalTrials.gov NCT03507335; https://clinicaltrials.gov/ct2/show/NCT03507335

Comparison Between the 24-hour Holter Test and 72-hour Single-Lead Electrocardiogram Monitoring With an Adhesive Patch-Type Device for Atrial Fibrillation Detection: Prospective Cohort Study

Background

There is insufficient evidence for the use of single-lead electrocardiogram (ECG) monitoring with an adhesive patch-type device (APD) over an extended period compared to that of the 24-hour Holter test for atrial fibrillation (AF) detection.

Objective

In this paper, we aimed to compare AF detection by the 24-hour Holter test and 72-hour single-lead ECG monitoring using an APD among patients with AF.

Methods

This was a prospective, single-center cohort study. A total of 210 patients with AF with clinical indications for the Holter test at cardiology outpatient clinics were enrolled in the study. The study participants were equipped with both the Holter device and APD for the first 24 hours. Subsequently, only the APD continued ECG monitoring for an additional 48 hours. AF detection during the first 24 hours was compared between the two devices. The diagnostic benefits of extended monitoring using the APD were evaluated.

Results

A total of 200 patients (mean age 60 years; n=141, 70.5% male; and n=59, 29.5% female) completed 72-hour ECG monitoring with the APD. During the first 24 hours, both monitoring methods detected AF in the same 40/200 (20%) patients (including 20 patients each with paroxysmal and persistent AF). Compared to the 24-hour Holter test, the APD increased the AF detection rate by 1.5-fold (58/200; 29%) and 1.6-fold (64/200; 32%) with 48- and 72-hour monitoring, respectively. With the APD, the number of newly discovered patients with paroxysmal AF was 20/44 (45.5%), 18/44 (40.9%), and 6/44 (13.6%) at 24-, 48-, and 72-hour monitoring, respectively. Compared with 24-hour Holter monitoring, 72-hour monitoring with the APD increased the detection rate of paroxysmal AF by 2.2-fold (44/20).

Conclusions

Compared to the 24-hour Holter test, AF detection could be improved with 72-hour single-lead ECG monitoring with the APD.

mCardia: A Context-Aware ECG Collection System for Ambulatory Arrhythmia Screening

This article presents the design, technical implementation, and feasibility evaluation of mCardia —a context-aware, mobile electrocardiogram (ECG) collection system for longitudinal arrhythmia screening under free-living conditions. Along with ECG, mCardia also records active and passive contextual data, including patient-reported symptoms and physical activity. This contextual data can provide a more accurate understanding of what happens before, during, and after an arrhythmia event, thereby providing additional information in the diagnosis of arrhythmia. By using a plugin-based architecture for ECG and contextual sensing, mCardia is device-agnostic and can integrate with various wireless ECG devices and supports cross-platform deployment. We deployed the mCardia system in a feasibility study involving 24 patients who used the system over a two-week period. During the study, we observed high patient acceptance and compliance with a satisfactory yield of collected ECG and contextual data. The results demonstrate the high usability and feasibility of mCardia for longitudinal ambulatory monitoring under free-living conditions. The article also reports from two clinical cases, which demonstrate how a cardiologist can utilize the collected contextual data to improve the accuracy of arrhythmia analysis. Finally, the article discusses the lessons learned and the challenges found in the mCardia design and the feasibility study.

Cardiac Rhythm Monitoring Using Wearables for Clinical Guidance before and after Catheter Ablation

Mobile health technologies are gaining importance in clinical decision-making. With the capability to monitor the patient’s heart rhythm, they have the potential to reduce the time to confirm a diagnosis and therefore are useful in patients eligible for screening of atrial fibrillation as well as in patients with symptoms without documented symptom rhythm correlation. Such is crucial to enable an adequate arrhythmia management including the possibility of a catheter ablation. After ablation, wearables can help to search for recurrences, in symptomatic as well as in asymptomatic patients. Furthermore, those devices can be used to search for concomitant arrhythmias and have the potential to help improving the short- and long-term patient management. The type of wearable as well as the adequate technology has to be chosen carefully for every situation and every individual patient, keeping different aspects in mind. This review aims to describe and to elaborate a potential workflow for the role of wearables for cardiac rhythm monitoring regarding detection and management of arrhythmias before and after cardiac electrophysiological procedures.

Mobile health technology in atrial fibrillation

INTRODUCTION

Mobile health (mHealth) solutions in atrial fibrillation (AF) are becoming widespread, thanks to everyday life devices such as smartphones. Their use is validated both in monitoring and in screening scenarios. In the published literature, the diagnostic accuracy of mHealth solutions wide differs, and their current clinical use is not well established in principal guidelines.

AREAS COVERED

mHealth solutions have progressively built an AF-detection chain to guide patients from the device's alert signal to the health care practitioners' (HCPs) attention. This review aims to critically evaluate the latest evidence regarding mHealth devices and the future possible patient's uses in everyday life.

EXPERT OPINION

The patients are the first to be informed of the rhythm anomaly, leading to the urgency of increasing the patients' AF self-management. Furthermore, HCPs need to update themselves about mHealth devices use in clinical practice. Nevertheless, these are promising instruments in specific populations, such as post-stroke patients, to promote an early arrhythmia diagnosis in the post-ablation/cardioversion period, allowing checks on the efficacy of the treatment or intervention.

The Potential and Limitations of Mobile Health and Insertable Cardiac Monitors in the Detection of Atrial Fibrillation in Cryptogenic Stroke Patients: Preliminary Results From the REMOTE Trial

Aim

This paper presents the preliminary results from the ongoing REMOTE trial. It aims to explore the opportunities and hurdles of using insertable cardiac monitors (ICMs) and photoplethysmography-based mobile health (PPG-based mHealth) using a smartphone or smartwatch to detect atrial fibrillation (AF) in cryptogenic stroke and transient ischemic attack (TIA) patients.

Methods and Results

Cryptogenic stroke or TIA patients (n = 39) received an ICM to search for AF and were asked to use a blinded PPG-based mHealth application for 6 months simultaneously. They were randomized to smartphone or smartwatch monitoring. In total, 68,748 1-min recordings were performed using PPG-based mHealth. The number of mHealth recordings decreased significantly over time in both smartphone and smartwatch groups (p < 0.001 and p = 0.002, respectively). Insufficient signal quality was more frequently observed in smartwatch (43.3%) compared to smartphone recordings (17.8%, p < 0.001). However, when looking at the labeling of the mHealth recordings on a patient level, there was no significant difference in signal quality between both groups. Moreover, the use of a smartwatch resulted in significantly more 12-h periods (91.4%) that were clinically useful compared to smartphone users (84.8%) as they had at least one recording of sufficient signal quality. Simultaneously, continuous data was collected from the ICMs, resulting in approximately 6,660,000 min of data (i.e., almost a 100-fold increase compared to mHealth). The ICM algorithm detected AF and other cardiac arrhythmias in 10 and 19 patients, respectively. However, these were only confirmed after adjudication by the remote monitoring team in 1 (10%) and 5 (26.3%) patients, respectively. The confirmed AF was also detected by PPG-based mHealth.

Conclusion

Based on the preliminary observations, our paper illustrates the potential as well as the limitations of PPG-based mHealth and ICMs to detect AF in cryptogenic stroke and TIA patients in four elements: (i) mHealth was able to detect AF in a patient in which AF was confirmed on the ICM; (ii) Even state-of-the-art ICMs yielded many false-positive AF registrations; (iii) Both mHealth and ICM still require physician revision; and (iv) Blinding of the mHealth results impairs compliance and motivation.

Will Smartphone Applications Replace the Insertable Cardiac Monitor in the Detection of Atrial Fibrillation? The First Comparison in a Case Report of a Cryptogenic Stroke Patient

Background and Case

This case report exemplifies the clinical application of non-invasive photoplethysmography (PPG)-based rhythm monitoring in the awakening mobile health (mHealth) era to detect symptomatic and asymptomatic paroxysmal atrial fibrillation (AF) in a cryptogenic stroke patient. Despite extensive diagnostic workup, the etiology remains unknown in one out of three ischemic strokes (i.e., cryptogenic stroke). Prolonged cardiac monitoring can reveal asymptomatic atrial fibrillation in up to one-third of this population. This case report describes a cryptogenic stroke patient who received prolonged cardiac monitoring with an insertable cardiac monitor (ICM) as standard of care. In the context of a clinical study, the patient simultaneously monitored his heart rhythm with a PPG-based smartphone application. AF was detected simultaneously on both the ICM and smartphone application after three days of monitoring. Similar AF burden was detected during follow-up (five episodes, median duration of 28 and 34 h on ICM and mHealth, respectively, p = 0.5). The detection prompted the initiation of oral anticoagulation and AF catheter ablation procedure.

Conclusion

This is the first report of the cryptogenic stroke patient in whom PPG-based mHealth was able to detect occurrence and burden of the symptomatic and asymptomatic paroxysmal AF episodes with similar precision as ICM. It accentuates the potential role of PPG-based mHealth in prolonged cardiac rhythm monitoring in cryptogenic stroke patients.

Data Collection Mechanisms in Health and Wellness Apps: Review and Analysis

Background

There has been a steady rise in the availability of health wearables and built-in smartphone sensors that can be used to collect health data reliably and conveniently from end users. Given the feature overlaps and user tendency to use several apps, these are important factors impacting user experience. However, there is limited work on analyzing the data collection aspect of mobile health (mHealth) apps.

Objective

This study aims to analyze what data mHealth apps across different categories usually collect from end users and how these data are collected. This information is important to guide the development of a common data model from current widely adopted apps. This will also inform what built-in sensors and wearables, a comprehensive mHealth platform should support.

Methods

In our empirical investigation of mHealth apps, we identified app categories listed in a curated mHealth app library, which was then used to explore the Google Play Store for health and medical apps that were then filtered using our selection criteria. We downloaded these apps from a mirror site hosting Android apps and analyzed them using a script that we developed around the popular AndroGuard tool. We analyzed the use of Bluetooth peripherals and built-in sensors to understand how a given app collects health data.

Results

We retrieved 3251 apps meeting our criteria, and our analysis showed that 10.74% (349/3251) of these apps requested Bluetooth access. We found that 50.9% (259/509) of the Bluetooth service universally unique identifiers to be known in these apps, with the remainder being vendor specific. The most common health-related Bluetooth Low Energy services using known universally unique identifiers were Heart Rate, Glucose, and Body Composition. App permissions showed the most used device module or sensor to be the camera (669/3251, 20.57%), closely followed by location (598/3251, 18.39%), with the highest occurrence in the staying healthy app category.

Conclusions

We found that not many health apps used built-in sensors or peripherals for collecting health data. The small number of the apps using Bluetooth, with an even smaller number of apps using standard Bluetooth Low Energy services, indicates a wider use of proprietary algorithms and custom services, which restrict the device use. The use of standard profiles could open this ecosystem further and could provide end users more options for apps. The relatively small proportion of apps using built-in sensors along with a high reliance on manual data entry suggests the need for more research into using sensors for data collection in health and fitness apps, which may be more desirable and improve end user experience.

Machine Learning for Cardiovascular Outcomes From Wearable Data: Systematic Review From a Technology Readiness Level Point of View

BACKGROUND

Wearable technology has the potential to improve cardiovascular health monitoring by using machine learning. Such technology enables remote health monitoring and allows for the diagnosis and prevention of cardiovascular diseases. In addition to the detection of cardiovascular disease, it can exclude this diagnosis in symptomatic patients, thereby preventing unnecessary hospital visits. In addition, early warning systems can aid cardiologists in timely treatment and prevention.

OBJECTIVE

This study aims to systematically assess the literature on detecting and predicting outcomes of patients with cardiovascular diseases by using machine learning with data obtained from wearables to gain insights into the current state, challenges, and limitations of this technology.

METHODS

We searched PubMed, Scopus, and IEEE Xplore on September 26, 2020, with no restrictions on the publication date and by using keywords such as "wearables," "machine learning," and "cardiovascular disease." Methodologies were categorized and analyzed according to machine learning-based technology readiness levels (TRLs), which score studies on their potential to be deployed in an operational setting from 1 to 9 (most ready).

RESULTS

After the removal of duplicates, application of exclusion criteria, and full-text screening, 55 eligible studies were included in the analysis, covering a variety of cardiovascular diseases. We assessed the quality of the included studies and found that none of the studies were integrated into a health care system (TRL<6), prospective phase 2 and phase 3 trials were absent (TRL<7 and 8), and group cross-validation was rarely used. These issues limited these studies' ability to demonstrate the effectiveness of their methodologies. Furthermore, there seemed to be no agreement on the sample size needed to train these studies' models, the size of the observation window used to make predictions, how long participants should be observed, and the type of machine learning model that is suitable for predicting cardiovascular outcomes.

CONCLUSIONS

Although current studies show the potential of wearables to monitor cardiovascular events, their deployment as a diagnostic or prognostic cardiovascular clinical tool is hampered by the lack of a realistic data set and proper systematic and prospective evaluation.

Mobile health technologies in the diagnosis and management of atrial fibrillation

PURPOSE OF REVIEW

Atrial fibrillation is the most common sustained rhythm abnormality and is associated with stroke, heart failure, cognitive decline, and premature death. Digital health technologies using consumer-grade mobile technologies (i.e. mHealth) capable of recording heart rate and rhythm can now reliably detect atrial fibrillation using single lead or multilead ECG or photoplethysmography (PPG). This review will discuss how these developments are being used to detect and manage atrial fibrillation.

RECENT FINDINGS

Studies have established the accuracy of mHealth devices for atrial fibrillation detection. The feasibility of using mHealth technology to screen for atrial fibrillation has also been established, though the utility of screening is controversial. In addition to screening, key aspects of atrial fibrillation management can also be performed remotely and effectively using mHealth, though with some important limitations.

SUMMARY

mHealth technologies have proven disruptive in the diagnosis and management of atrial fibrillation. Healthcare providers can leverage these advances to better care for their atrial fibrillation patients whenever necessary.

Accuracy of Detecting Atrial Fibrillation: A Systematic Review and Meta-Analysis of Wrist-Worn Wearable Technology

Atrial fibrillation (AF) is the most commonly diagnosed arrhythmia, and ECG remains the gold standard for diagnosing AF. Wrist-worn technologies are appealing for their ability to passively process near-continuous pulse signals. The clinical application of wearable devices is controversial. Our systematic review and meta-analysis qualitatively and quantitatively analyze available literature on wrist-worn wearable devices (Apple Watch, Samsung, and KardiaBand) and their sensitivity and specificity in detecting AF compared to conventional methods. Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were followed, yielding nine studies (n = 1,581). Observational studies assessing the sensitivity and specificity of wrist-worn wearables in detecting AF in patients with and without a history of AF were included and analyzed using a fixed-effect model with an inverse-variance method. In patients with a history of AF, the overall sensitivity between device groups did not significantly differ (96.83%; P = 0.207). Specificity significantly differed between Apple, Samsung, and KardiaBand (99.61%, 81.13%, and 97.98%, respectively; P<0.001). The effect size for this analysis was highest in the Samsung device group. Two studies (n = 796) differentiated cohorts to assess device sensitivity in patients with known AF and device specificity in patients with normal sinus rhythm (NSR) (sensitivity: 96.02%; confidence intervals (CI) 93.85%-97.59% and specificity: 98.82%; CI:97.46%-99.57%). Wrist-worn wearable devices demonstrate promising results in detecting AF in patients with paroxysmal AF. However, more rigorous prospective data is needed to understand the limitations of these devices in regard to varying specificities which may lead to unintended downstream medical testing and costs.

Applicability of a Novel Wearable Wireless Electrocardiogram Monitoring Device (Spyder) for Arrhythmia Detection in Patients with Suspected Cardiac Arrhythmias

Introduction

In clinical practice, many cardiovascular symptoms can be caused by arrhythmias that are not detected by electrocardiography (ECG) or 24–48 h Holter ECG monitoring.

Aims

To describe the efficacy and applicability of a new device (Spyder) in detecting cardiac arrhythmias with midterm ECG monitoring.

Methods

A descriptive, prospective study was performed on 26 consecutive patients who underwent midterm ECG monitoring with the novel ECG patch device (Spyder). The study was conducted over a 6-month period from August 2020 to February 2021.

Results

Twenty-six patients (mean age, 57.8 ± 12.5 years; men, 77%) wearing a Spyder wireless ECG-monitoring device were recruited. The mean wearing time was 84 hours. The main indications for using the device were detection of recurrent atrial fibrillation after radiofrequency ablation (30.7%) and screening for atrial fibrillation after cryptogenic stroke (23.1%). All ECG monitor recordings obtained during the study period were of good quality. The device detected 12 episodes of atrial fibrillation in eight patients, one episode of ventricular tachycardia, one supraventricular tachycardia event, one case of paroxysmal third-degree atrioventricular block, and five cases of frequent premature ventricular contraction. The time to detection of the first episodes of atrial fibrillation and ventricular and supraventricular tachycardia was 28.8 and 47 hours, respectively.

Conclusions

The new wearable wireless ECG-monitoring device (Spyder) is a feasible and effective method for the detection of cardiac arrhythmias.

Accuracy of pulsatile photoplethysmography applications or handheld devices vs. 12-lead ECG for atrial fibrillation screening: a systematic review and meta-analysis

BACKGROUND

The relative accuracy of pulsatile photoplethysmography applications (PPG) or handheld (HH) devices compared with the gold standard 12-lead electrocardiogram (ECG) for the diagnosis of atrial fibrillation is unknown.

METHODS

Digital databases were searched to identify relevant articles. Raw data were pooled using a bivariate model to calculate diagnostic accuracy measures and estimate Hierarchical Summary Receiver Operating Characteristic (HSROC).

RESULTS

A total of 10 articles comprising 4296 patients (mean age 68.9 years, with 56% males) were included in the analysis. Compared with EKG, the pooled sensitivity of PPG and HH devices in AF detection was 0.93 (95% CI 0.87-0.96; p < 0.05) and 0.87 (95% CI. 0.74-0.94; p < 0.05), respectively. The pooled specificity of PPG and HH devices in AF detection was 0.91 (95% CI 0.88-0.94; p < 0.05) and 0.96 (95% CI 0.90-0.98; p < 0.05), respectively. The diagnostic odds ratio was 129 and 144 for PPG and HH devices, respectively. Fagan's nomogram showed the probability of a patient having AF and normal rhythm on PPG or HH devices was 2-3%, while the post-test probability of having AF with an irregular R-R interval on PPG or HH devices was 73% and 82%, respectively. The scatter plot of positive and negative likelihood ratio showed high confirmation of AF and reliability of exclusion of absence of irregular R-R intervals (positive likelihood ratio > 10, and negative likelihood ratio < 0.1) on HH devices while PPG was used as confirmation only.

CONCLUSIONS

The PPG or HH devices can serve as a reliable alternative for the detection of AF.

Automatic Mobile Health Arrhythmia Monitoring for the Detection of Atrial Fibrillation: Prospective Feasibility, Accuracy, and User Experience Study

Background

Atrial fibrillation (AF) is the most common tachyarrhythmia and associated with a risk of stroke. The detection and diagnosis of AF represent a major clinical challenge due to AF’s asymptomatic and intermittent nature. Novel consumer-grade mobile health (mHealth) products with automatic arrhythmia detection could be an option for long-term electrocardiogram (ECG)-based rhythm monitoring and AF detection.

Objective

We evaluated the feasibility and accuracy of a wearable automated mHealth arrhythmia monitoring system, including a consumer-grade, single-lead heart rate belt ECG device (heart belt), a mobile phone application, and a cloud service with an artificial intelligence (AI) arrhythmia detection algorithm for AF detection. The specific aim of this proof-of-concept study was to test the feasibility of the entire sequence of operations from ECG recording to AI arrhythmia analysis and ultimately to final AF detection.

Methods

Patients (n=159) with an AF (n=73) or sinus rhythm (n=86) were recruited from the emergency department. A single-lead heart belt ECG was recorded for 24 hours. Simultaneously registered 3-lead ECGs (Holter) served as the gold standard for the final rhythm diagnostics and as a reference device in a user experience survey with patients over 65 years of age (high-risk group).

Results

The heart belt provided a high-quality ECG recording for visual interpretation resulting in 100% accuracy, sensitivity, and specificity of AF detection. The accuracy of AF detection with the automatic AI arrhythmia detection from the heart belt ECG recording was also high (97.5%), and the sensitivity and specificity were 100% and 95.4%, respectively. The correlation between the automatic estimated AF burden and the true AF burden from Holter recording was >0.99 with a mean burden error of 0.05 (SD 0.26) hours. The heart belt demonstrated good user experience and did not significantly interfere with the patient’s daily activities. The patients preferred the heart belt over Holter ECG for rhythm monitoring (85/110, 77% heart belt vs 77/109, 71% Holter, P=.049).

Conclusions

A consumer-grade, single-lead ECG heart belt provided good-quality ECG for rhythm diagnosis. The mHealth arrhythmia monitoring system, consisting of heart-belt ECG, a mobile phone application, and an automated AF detection achieved AF detection with high accuracy, sensitivity, and specificity. In addition, the mHealth arrhythmia monitoring system showed good user experience.

Trial Registration

ClinicalTrials.gov NCT03507335; https://clinicaltrials.gov/ct2/show/NCT03507335

Comparative Clinical Effectiveness of Population-Based Atrial Fibrillation Screening Using Contemporary Modalities: A Decision-Analytic Model

Background Atrial fibrillation (AF) screening is endorsed by certain guidelines for individuals aged ≥65 years. Yet many AF screening strategies exist, including the use of wrist-worn wearable devices, and their comparative effectiveness is not well-understood. Methods and Results We developed a decision-analytic model simulating 50 million individuals with an age, sex, and comorbidity profile matching the United States population aged ≥65 years (ie, with a guideline-based AF screening indication). We modeled no screening, in addition to 45 distinct AF screening strategies (comprising different modalities and screening intervals), each initiated at a clinical encounter. The primary effectiveness measure was quality-adjusted life-years, with incident stroke and major bleeding as secondary measures. We defined continuous or nearly continuous modalities as those capable of monitoring beyond a single time-point (eg, patch monitor), and discrete modalities as those capable of only instantaneous AF detection (eg, 12-lead ECG). In total, 10 AF screening strategies were effective compared with no screening (300-1500 quality-adjusted life-years gained/100 000 individuals screened). Nine (90%) effective strategies involved use of a continuous or nearly continuous modality such as patch monitor or wrist-worn wearable device, whereas 1 (10%) relied on discrete modalities alone. Effective strategies reduced stroke incidence (number needed to screen to prevent a stroke: 3087-4445) but increased major bleeding (number needed to screen to cause a major bleed: 1815-4049) and intracranial hemorrhage (number needed to screen to cause intracranial hemorrhage: 7693-16 950). The test specificity was a highly influential model parameter on screening effectiveness. Conclusions When modeled from a clinician-directed perspective, the comparative effectiveness of population-based AF screening varies substantially upon the specific strategy used. Future screening interventions and guidelines should consider the relative effectiveness of specific AF screening strategies.

Is mHealth a Useful Tool for Self-Assessment and Rehabilitation of People with Multiple Sclerosis? A Systematic Review

The development of mobile technology and mobile Internet offers new possibilities in rehabilitation and clinical assessment in a longitudinal perspective for multiple sclerosis management. However, because the mobile health applications (mHealth) have only been developed recently, the level of evidence supporting the use of mHealth in people with multiple sclerosis (pwMS) is currently unclear. Therefore, this review aims to list and describe the different mHealth available for rehabilitation and self-assessment of pwMS and to define the level of evidence supporting these interventions for functioning problems categorized within the International Classification of Functioning, Disability and Health (ICF). In total, 36 studies, performed with 22 different mHealth, were included in this review, 30 about rehabilitation and six for self-assessment, representing 3091 patients. For rehabilitation, most of the studies were focusing on cognitive function and fatigue. Concerning the efficacy, we found a small but significant effect of the use of mHealth for cognitive training (Standardized Mean Difference (SMD) = 0.28 [0.12; 0.45]) and moderate effect for fatigue (SMD = 0.61 [0.47; 0.76]). mHealth is a promising tool in pwMS but more studies are needed to validate these solutions in the other ICF categories. More replications studies are also needed as most of the mHealth have only been assessed in one single study.

[Interpretation of photoplethysmography: a step-by-step guide]

By applying photoplethysmography (PPG), the camera of the mobile phone can be used to remotely assess heart rate and rhythm, which was widely used in conjunction with teleconsultations within the TeleCheck-AF project during the coronavirus disease 2019 (COVID-19) pandemic. Herein, we provide an educational, structured, stepwise practical guide on how to interpret PPG signals. A better understanding of PPG recordings is critical for the implementation of this widely available technology into clinical practice.

Enhancing Healthcare Access-Smartphone Apps in Arrhythmia Screening: Viewpoint

Atrial fibrillation is the most commonly reported arrhythmia and, if undiagnosed or untreated, may lead to thromboembolic events. It is therefore desirable to provide screening to patients in order to detect atrial arrhythmias. Specific mobile apps and accessory devices, such as smartphones and smartwatches, may play a significant role in monitoring heart rhythm in populations at high risk of arrhythmia. These apps are becoming increasingly common among patients and professionals as a part of mobile health. The rapid development of mobile health solutions may revolutionize approaches to arrhythmia screening. In this viewpoint paper, we assess the availability of smartphone and smartwatch apps and evaluate their efficacy for monitoring heart rhythm and arrhythmia detection. The findings obtained so far suggest they are on the right track to improving the efficacy of early detection of atrial fibrillation, thus lowering the risk of stroke and reducing the economic burden placed on public health.

Problems and Barriers related to the Use of Digital Health Applications: A Scoping Review Protocol (Preprint)

Background

The use of mobile health (mHealth) apps is increasing rapidly worldwide. More and more institutions and organizations develop regulations and guidelines to enable an evidence-based and safe use. In Germany, mHealth apps fulfilling predefined criteria (Digitale Gesundheitsanwendungen [DiGA]) can be prescribed and are reimbursable by the German statutory health insurance scheme. Due to the increasing distribution of DiGA, problems and barriers should receive special attention.

Objective

This study aims to identify the relevant problems and barriers related to the use of mHealth apps fulfilling the criteria of DiGA.

Methods

This scoping review will follow published methodological frameworks and the PRISMA-Scr (Preferred Reporting Items for Systematic Reviews and Meta-analyses Extension for Scoping Reviews) criteria. Electronic databases (MEDLINE, EMBASE, PsycINFO, and JMIR), reference lists of relevant articles, and grey literature sources will be searched. Two reviewers will assess the eligibility of the articles by a two-stage (title and abstract as well as full text) screening process. Only problems and barriers related to mHealth apps fulfilling the criteria of DiGA are included for this research. The identified studies will be categorized and analyzed with MAXQDA.

Results

This scoping review gives an overview of the available evidence and identifies research gaps regarding problems and barriers related to DiGA. The results are planned to be submitted to an indexed, peer-reviewed journal in the first quarter of 2022.

Conclusions

This is the first review to identify the problems and barriers related to the use of mHealth apps fulfilling the German definition of DiGA. Nevertheless, the findings can be applied to other contexts and health care systems as well.

International Registered Report Identifier (IRRID)

DERR1-10.2196/32702

The comparison of Kardia Mobile and Hartmann Veroval 2 in 1 in detecting first diagnosed atrial fibrillation

BACKGROUND

Atrial fibrillation (AF) is the leading cause of stroke. The European Society of Cardiology (ESC) advises opportunistic AF screening among patients aged ≥ 65 years. Considering this, the aim herein, was compare the feasibility of two different systems of smartphone-based electrocardiogram (ECG) recordings to identify AF among those without a previous arrhythmia history.

METHODS

Prospective AF screening was conducted at six pharmacies using Kardia Mobile and Hartmann Veroval 2 in 1. A single-lead ECG was acquired by the placement of fingers on the pads. A cardiologist evaluated findings from both devices.

RESULTS

Atrial fibrillation was identified in 3.60% and previously unknown AF was detected in 1.92% of the study participants. Sensitivity and specificity of the Kardia application in detecting AF were 66.7% (95% confidence interval [CI] 38.4-88.2%) and 98.5% (95% CI 96.7-99.5%), and for Veroval 10.0% (95% CI 0.23-44.5%) and 94.96% (95% CI 92.15-96.98%), accordingly. Inter-rater agreement was k = 0.088 (95% CI 1.59-16.1%).

CONCLUSIONS

Mobile devices can detect AF, but each finding must be verified by a professional. The Kardia application appeared to be more user-friendly than Veroval. Cardiovascular screening using mobile devices is feasible at pharmacies. Hence it might be considered for routine use.

Mobile health applications for the detection of atrial fibrillation: a systematic review

AIMS

Atrial fibrillation (AF) is the most common sustained arrhythmia and an important risk factor for stroke and heart failure. We aimed to conduct a systematic review of the literature and summarize the performance of mobile health (mHealth) devices in diagnosing and screening for AF.

METHODS AND RESULTS

We conducted a systematic search of MEDLINE, Embase, and the Cochrane Central Register of Controlled Trials. Forty-three studies met the inclusion criteria and were divided into two groups: 28 studies aimed at validating smart devices for AF diagnosis, and 15 studies used smart devices to screen for AF. Evaluated technologies included smartphones, with photoplethysmographic (PPG) pulse waveform measurement or accelerometer sensors, smartbands, external electrodes that can provide a smartphone single-lead electrocardiogram (iECG), such as AliveCor, Zenicor and MyDiagnostick, and earlobe monitor. The accuracy of these devices depended on the technology and the population, AliveCor and smartphone PPG sensors being the most frequent systems analysed. The iECG provided by AliveCor demonstrated a sensitivity and specificity between 66.7% and 98.5% and 99.4% and 99.0%, respectively. The PPG sensors detected AF with a sensitivity of 85.0-100% and a specificity of 93.5-99.0%. The incidence of newly diagnosed arrhythmia ranged from 0.12% in a healthy population to 8% among hospitalized patients.

CONCLUSION

Although the evidence for clinical effectiveness is limited, these devices may be useful in detecting AF. While mHealth is growing in popularity, its clinical, economic, and policy implications merit further investigation. More head-to-head comparisons between mHealth and medical devices are needed to establish their comparative effectiveness.

Returning Cardiac Rhythm Data to Patients: Opportunities and Challenges

Spurred by federal legislation, professional organizations, and patients themselves, patient access to data from electronic cardiac devices is increasingly transparent. Patients can collect data through consumer devices and access data traditionally shared only with health care providers. These data may improve screening, self-management, and shared decision-making for cardiac arrhythmias, but challenges remain, including patient comprehension, communication with providers, and sustained engagement. Ways to address these challenges include leveraging visualizations that support comprehension, involving patients in designing and developing patient-facing digital tools, and establishing clear practices and goals for data exchange with health care providers.

Detection and classification of arrhythmia using an explainable deep learning model

BACKGROUND

Early detection and intervention is the cornerstone for appropriate treatment of arrhythmia and prevention of complications and mortality. Although diverse deep learning models have been developed to detect arrhythmia, they have been criticized due to their unexplainable nature. In this study, we developed an explainable deep learning model (XDM) to classify arrhythmia, and validated its performance using diverse external validation data.

METHODS

In this retrospective study, the Sejong dataset comprising 86,802 electrocardiograms (ECGs) was used to develop and internally variate the XDM. The XDM based on a neural network-backed ensemble tree was developed with six feature modules that are able to explain the reasons for its decisions. The model was externally validated using data from 36,961 ECGs from four non-restricted datasets.

RESULTS

During internal and external validation of the XDM, the average area under the receiver operating characteristic curves (AUCs) using a 12‑lead ECG for arrhythmia classification were 0.976 and 0.966, respectively. The XDM outperformed a previous simple multi-classification deep learning model that used the same method. During internal and external validation, the AUCs of explainability were 0.925-0.991.

CONCLUSION

Our XDM successfully classified arrhythmia using diverse formats of ECGs and could effectively describe the reason for the decisions. Therefore, an explainable deep learning methodology could improve accuracy compared to conventional deep learning methods, and that the transparency of XDM can be enhanced for its application in clinical practice.

The photoplethysmography dictionary: practical guidance on signal interpretation and clinical scenarios from TeleCheck-AF

Aims

Within the TeleCheck-AF project, numerous centres in Europe used on-demand photoplethysmography (PPG) technology to remotely assess heart rate and rhythm in conjunction with teleconsultations. Based on the TeleCheck-AF investigator experiences, we aimed to develop an educational structured stepwise practical guide on how to interpret PPG signals and to introduce typical clinical scenarios how on-demand PPG was used.

Methods and results

During an online conference, the structured stepwise practical guide on how to interpret PPG signals was discussed and further refined during an internal review process. We provide the number of respective PPG recordings (FibriCheck^®) and number of patients managed within a clinical scenario during the TeleCheck-AF project. To interpret PPG recordings, we introduce a structured stepwise practical guide and provide representative PPG recordings. In the TeleCheck-AF project, 2522 subjects collected 90 616 recordings in total. The majority of these recordings were classified by the PPG algorithm as sinus rhythm (57.6%), followed by AF (23.6%). In 9.7% of recordings, the quality was too low to interpret. The most frequent clinical scenarios where PPG technology was used in the TeleCheck-AF project was a follow-up after AF ablation (1110 patients) followed by heart rate and rhythm assessment around (tele)consultation (966 patients).

Conclusion

We introduce a newly developed structured stepwise practical guide on PPG signal interpretation developed based on presented experiences from TeleCheck-AF. The present clinical scenarios for the use of on-demand PPG technology derived from the TeleCheck-AF project will help to implement PPG technology in the management of AF patients.

Distinguishing atrial fibrillation from sinus rhythm using commercial pulse detection systems: The non-interventional BAYathlon study

Objective

Early diagnosis of atrial fibrillation (AFib) is a priority for stroke prevention. We sought to test four commercial pulse detection systems (CPDSs) for ability to distinguish AFib from normal sinus rhythm using a published algorithm (Zhou et al., PLoS One 2015;10:e0136544), compared with visual diagnosis by electrocardiogram inspection.

Methods

BAYathlon was a prospective, non-interventional, single-centre study. Adult cardiology patients with documented AFib or sinus rhythm who were due to have a routine 5-min electrocardiogram were randomized to undergo a parallel 5-min pulse assessment with a Polar V800, eMotion Faros 360, TomTom heart rate monitor, or Adidas miCoach Smart Run.

Results

144 patients (73 with AFib, 71 with sinus rhythm (based on electrocardiograms); median age: 73 years; 53.5% male) were analysed. Algorithm sensitivities (primary endpoint) and specificities for AFib when applied to CPDS recordings were 93.3% and 94.1% with the Polar V800, 90.0% and 84.2% with the eMotion Faros 360, and 0% and 100% with the other CPDSs (analysis period: 127 heart rate signals + 2 min). When applied to routine electrocardiograms, the algorithm correctly detected AFib in 71/73 patients. Different analysis periods (127 heart rate signals +1 or 3 min) only slightly changed the sensitivities with the Polar V800 and eMotion Faros 360 and had no effect on the sensitivities with the other CPDSs.

Conclusion

AFib screening using the applied algorithm is feasible with the Polar V800 and eMotion Faros 360 (which provide RR interval data) but not with the other CPDSs (which provide pre-processed heart rate time series).

ClinicalTrials.gov identifier: NCT02875106

Performance of a Mobile Single-Lead Electrocardiogram Technology for Atrial Fibrillation Screening in a Semirural African Population: Insights From "The Heart of Ethiopia: Focus on Atrial Fibrillation" (TEFF-AF) Study

Background

Atrial fibrillation (AF) screening using mobile single-lead electrocardiogram (ECG) devices has demonstrated variable sensitivity and specificity. However, limited data exists on the use of such devices in low-resource countries.

Objective

The goal of the research was to evaluate the utility of the KardiaMobile device’s (AliveCor Inc) automated algorithm for AF screening in a semirural Ethiopian population.

Methods

Analysis was performed on 30-second single-lead ECG tracings obtained using the KardiaMobile device from 1500 TEFF-AF (The Heart of Ethiopia: Focus on Atrial Fibrillation) study participants. We evaluated the performance of the KardiaMobile automated algorithm against cardiologists’ interpretations of 30-second single-lead ECG for AF screening.

Results

A total of 1709 single-lead ECG tracings (including repeat tracing on 209 occasions) were analyzed from 1500 Ethiopians (63.53% [953/1500] male, mean age 35 [SD 13] years) who presented for AF screening. Initial successful rhythm decision (normal or possible AF) with one single-lead ECG tracing was lower with the KardiaMobile automated algorithm versus manual verification by cardiologists (1176/1500, 78.40%, vs 1455/1500, 97.00%; P<.001). Repeat single-lead ECG tracings in 209 individuals improved overall rhythm decision, but the KardiaMobile automated algorithm remained inferior (1301/1500, 86.73%, vs 1479/1500, 98.60%; P<.001). The key reasons underlying unsuccessful KardiaMobile automated rhythm determination include poor quality/noisy tracings (214/408, 52.45%), frequent ectopy (22/408, 5.39%), and tachycardia (>100 bpm; 167/408, 40.93%). The sensitivity and specificity of rhythm decision using KardiaMobile automated algorithm were 80.27% (1168/1455) and 82.22% (37/45), respectively.

Conclusions

The performance of the KardiaMobile automated algorithm was suboptimal when used for AF screening. However, the KardiaMobile single-lead ECG device remains an excellent AF screening tool with appropriate clinician input and repeat tracing.

Trial Registration

Australian New Zealand Clinical Trials Registry ACTRN12619001107112; https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=378057&isReview=true

Association of Wearable Device Use With Pulse Rate and Health Care Use in Adults With Atrial Fibrillation

Importance

Increasingly, individuals with atrial fibrillation (AF) use wearable devices (hereafter wearables) that measure pulse rate and detect arrhythmia. The associations of wearables with health outcomes and health care use are unknown.

Objective

To characterize patients with AF who use wearables and compare pulse rate and health care use between individuals who use wearables and those who do not.

Design, Setting, and Participants

This retrospective, propensity-matched cohort study included 90 days of follow-up of patients in a tertiary care, academic health system. Included patients were adults with at least 1 AF-specific International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) code from 2017 through 2019. Electronic medical records were reviewed to identify 125 individuals who used wearables and had adequate pulse-rate follow-up who were then matched using propensity scores 4 to 1 with 500 individuals who did not use wearables. Data were analyzed from June 2020 through February 2021.

Exposure

Using commercially available wearables with pulse rate or rhythm evaluation capabilities.

Main Outcomes and Measures

Mean pulse rates from measures taken in the clinic or hospital and a composite health care use score were recorded. The composite outcome included evaluation and management, ablation, cardioversion, telephone encounters, and number of rate or rhythm control medication orders.

Results

Among 16 320 patients with AF included in the analysis, 348 patients used wearables and 15 972 individuals did not use wearables. Prior to matching, patients using wearables were younger (mean [SD] age, 64.0 [13.0] years vs 70.0 [13.8] years; P < .001) and healthier (mean [SD] CHA2DS2-VASc [congestive heart failure, hypertension, age ≥ 65 years or 65-74 years, diabetes, prior stroke/transient ischemic attack, vascular disease, sex] score, 3.6 [2.0] vs 4.4 [2.0]; P < .001) compared with individuals not using wearables, with similar gender distribution (148 [42.5%] women vs 6722 women [42.1%]; P = .91). After matching, mean pulse rate was similar between 125 patients using wearables and 500 patients not using wearables (75.01 [95% CI, 72.74-77.27] vs 75.79 [95% CI, 74.68-76.90] beats per minute [bpm]; P = .54), whereas mean composite use score was higher among individuals using wearables (3.55 [95% CI, 3.31-3.80] vs 3.27 [95% CI, 3.14-3.40]; P = .04). Among measures in the composite outcome, there was a significant difference in use of ablation, occurring in 22 individuals who used wearables (17.6%) vs 37 individuals who did not use wearables (7.4%) (P = .001). In the regression analyses, mean composite use score was 0.28 points (95% CI, 0.01 to 0.56 points) higher among individuals using wearables compared with those not using wearables and mean pulse was similar, with a -0.79 bpm (95% CI -3.28 to 1.71 bpm) difference between the groups.

Conclusions and Relevance

This study found that follow-up health care use among individuals with AF was increased among those who used wearables compared with those with similar pulse rates who did not use wearables. Given the increasing use of wearables by patients with AF, prospective, randomized, long-term evaluation of the associations of wearable technology with health outcomes and health care use is needed.

Arrhythmia in an athlete diagnosed by smartphone electrocardiogram: a case report

Background

This is the first case report demonstrating the use of a smartphone device, enabling the diagnosis of an arrhythmia in the sports cardiology literature.

Case summary

A 17-year-old semi-professional rugby player presented with recurrent episodes of palpitations terminated by vagal manoeuvres. The rugby player’s resting 12-lead electrocardiogram (ECG), echocardiogram, and exercise stress test were normal. Due to his suggestive history and an ECG trace from a smartphone device, demonstrating a narrow complex tachycardia, an electrophysiological study was arranged. The study demonstrated a slow-fast atrioventricular nodal re-entrant tachycardia which was successfully ablated.

Discussion

The ambulatory use of a smartphone ECG device assisted in the timely diagnosis and management of an undiagnosed paroxysmal arrhythmia in a rugby player. This resulted in an expedited return to play.

Validation of Adhesive Single-Lead ECG Device Compared with Holter Monitoring among Non-Atrial Fibrillation Patients

There are few reports on head-to-head comparisons of electrocardiogram (ECG) monitoring between adhesive single-lead and Holter devices for arrhythmias other than atrial fibrillation (AF). This study aimed to compare 24 h ECG monitoring between the two devices in patients with general arrhythmia. Twenty-nine non-AF patients with a workup of pre-diagnosed arrhythmias or suspicious arrhythmic episodes were evaluated. Each participant wore both devices simultaneously, and the cardiac rhythm was monitored for 24 h. Selective ECG parameters were compared between the two devices. Two cardiologists independently compared the diagnoses of each device. The two most frequent monitoring indications were workup of premature atrial contractions (41.4%) and suspicious arrhythmia-related symptoms (37.9%). The single-lead device had a higher noise burden than the Holter device (0.04 ± 0.05% vs. 0.01 ± 0.01%, p = 0.024). The number of total QRS complexes, ventricular ectopic beats, and supraventricular ectopic beats showed an excellent degree of agreement between the two devices (intraclass correlation coefficients = 0.991, 1.000, and 0.987, respectively). In addition, the minimum/average/maximum heart rates showed an excellent degree of agreement. The two cardiologists made coherent diagnoses for all 29 participants using both monitoring methods. In conclusion, the single-lead adhesive device could be an acceptable alternative for ambulatory ECG monitoring in patients with general arrhythmia.

Impact of Mobile Health Devices for the Detection of Atrial Fibrillation: Systematic Review

Background

Atrial fibrillation (AF) is the most common arrhythmia, and its prevalence is increasing. Early diagnosis is important to reduce the risk of stroke. Mobile health (mHealth) devices, such as single-lead electrocardiogram (ECG) devices, have been introduced to the worldwide consumer market over the past decade. Recent studies have assessed the usability of these devices for detection of AF, but it remains unclear if the use of mHealth devices leads to a higher AF detection rate.

Objective

The goal of the research was to conduct a systematic review of the diagnostic detection rate of AF by mHealth devices compared with traditional outpatient follow-up. Study participants were aged 16 years or older and had an increased risk for an arrhythmia and an indication for ECG follow-up—for instance, after catheter ablation or presentation to the emergency department with palpitations or (near) syncope. The intervention was the use of an mHealth device, defined as a novel device for the diagnosis of rhythm disturbances, either a handheld electronic device or a patch-like device worn on the patient’s chest. Control was standard (traditional) outpatient care, defined as follow-up via general practitioner or regular outpatient clinic visits with a standard 12-lead ECG or Holter monitoring. The main outcome measures were the odds ratio (OR) of AF detection rates.

Methods

Two reviewers screened the search results, extracted data, and performed a risk of bias assessment. A heterogeneity analysis was performed, forest plot made to summarize the results of the individual studies, and albatross plot made to allow the P values to be interpreted in the context of the study sample size.

Results

A total of 3384 articles were identified after a database search, and 14 studies with a 4617 study participants were selected. All studies but one showed a higher AF detection rate in the mHealth group compared with the control group (OR 1.00-35.71), with all RCTs showing statistically significant increases of AF detection (OR 1.54-19.16). Statistical heterogeneity between studies was considerable, with a Q of 34.1 and an I² of 61.9, and therefore it was decided to not pool the results into a meta-analysis.

Conclusions

Although the results of 13 of 14 studies support the effectiveness of mHealth interventions compared with standard care, study results could not be pooled due to considerable clinical and statistical heterogeneity. However, smartphone-connectable ECG devices provide patients with the ability to document a rhythm disturbance more easily than with standard care, which may increase empowerment and engagement with regard to their illness. Clinicians must beware of overdiagnosis of AF, as it is not yet clear when an mHealth-detected episode of AF must be deemed significant.

Redefining the Standard for Atrial Fibrillation: A Patient-centric Report

A roundtable discussion with three European clinical experts in AF and one expert patient diagnosed and treated for AF was conducted in London in October 2019. The panel discussed the implications of AF for patients, current patient pathways, what treatment outcomes were relevant for patients and how the recommendations for the management of AF may change in the future, based on the outcomes of recently published and on-going clinical trials. This article summarises the discussion, and draws upon wider sources to detail best practice and optimal patient treatment pathways.

The Implementation and Acceptability of a Mobile Application for Screening for Atrial Fibrillation at Home

Introduction: Although patients are able to easily record electrocardiograms using consumer devices, these are typically not shared with their clinicians. This article discusses the development and acceptability of a mobile application (app) that integrates with the electronic health record to facilitate screening for atrial fibrillation (AF). Methods: After app development and implementation, we compared workflows with and without the mobile app. Seven older adults used it during a prospective twice-daily 2-week home-based AF screening protocol and completed an acceptability survey with Likert scale responses. Results: Compliance with the screening protocol was 82%. Acceptability and usability was favorable. Patients reported confidence in the connection between the app and their medical record. Discussion: The availability of apps to capture data and facilitate a connection with health systems is critical. The app developed is a feasible solution for older patients with AF to self-monitor and report results to their health provider.

2021 ISHNE/HRS/EHRA/APHRS Expert Collaborative Statement on mHealth in Arrhythmia Management: Digital Medical Tools for Heart Rhythm Professionals: From the International Society for Holter and Noninvasive Electrocardiology/Heart Rhythm Society/European Heart Rhythm Association/Asia-Pacific Heart Rhythm Society

Supplemental Digital Content is available in the text.

This collaborative statement from the International Society for Holter and Noninvasive Electrocardiology/Heart Rhythm Society/European Heart Rhythm Association/Asia-Pacific Heart Rhythm Society describes the current status of mobile health technologies in arrhythmia management. The range of digital medical tools and heart rhythm disorders that they may be applied to and clinical decisions that may be enabled are discussed. The facilitation of comorbidity and lifestyle management (increasingly recognized to play a role in heart rhythm disorders) and patient self-management are novel aspects of mobile health. The promises of predictive analytics but also operational challenges in embedding mobile health into routine clinical care are explored.

2021 ISHNE/HRS/EHRA/APHRS Collaborative Statement on mHealth in Arrhythmia Management: Digital Medical Tools for Heart Rhythm Professionals

This collaborative statement from the International Society for Holter and Noninvasive Electrocardiology/Heart Rhythm Society/European Heart Rhythm Association/Asia Pacific Heart Rhythm Society describes the current status of mobile health ("mHealth") technologies in arrhythmia management. The range of digital medical tools and heart rhythm disorders that they may be applied to and clinical decisions that may be enabled are discussed. The facilitation of comorbidity and lifestyle management (increasingly recognized to play a role in heart rhythm disorders) and patient self-management are novel aspects of mHealth. The promises of predictive analytics but also operational challenges in embedding mHealth into routine clinical care are explored.

2021 ISHNE/ HRS/ EHRA/ APHRS collaborative statement on mHealth in Arrhythmia Management: Digital Medical Tools for Heart Rhythm Professionals: From the International Society for Holter and Noninvasive Electrocardiology/Heart Rhythm Society/European Heart Rhythm Association/Asia Pacific Heart Rhythm Society

This collaborative statement from the International Society for Holter and Noninvasive Electrocardiology/ Heart Rhythm Society/ European Heart Rhythm Association/ Asia Pacific Heart Rhythm Society describes the current status of mobile health ("mHealth") technologies in arrhythmia management. The range of digital medical tools and heart rhythm disorders that they may be applied to and clinical decisions that may be enabled are discussed. The facilitation of comorbidity and lifestyle management (increasingly recognized to play a role in heart rhythm disorders) and patient self‐management are novel aspects of mHealth. The promises of predictive analytics but also operational challenges in embedding mHealth into routine clinical care are explored.

2021 ISHNE/HRS/EHRA/APHRS collaborative statement on mHealth in Arrhythmia Management: Digital Medical Tools for Heart Rhythm Professionals: From the International Society for Holter and Noninvasive Electrocardiology/Heart Rhythm Society/European Heart Rhythm Association/Asia Pacific Heart Rhythm Society

This collaborative statement from the International Society for Holter and Noninvasive Electrocardiology/Heart Rhythm Society/European Heart Rhythm Association/Asia Pacific Heart Rhythm Society describes the current status of mobile health (“mHealth”) technologies in arrhythmia management. The range of digital medical tools and heart rhythm disorders that they may be applied to and clinical decisions that may be enabled are discussed. The facilitation of comorbidity and lifestyle management (increasingly recognized to play a role in heart rhythm disorders) and patient self‐management are novel aspects of mHealth. The promises of predictive analytics but also operational challenges in embedding mHealth into routine clinical care are explored.

2021 ISHNE / HRS / EHRA / APHRS Collaborative Statement on mHealth in Arrhythmia Management: Digital Medical Tools for Heart Rhythm Professionals: From the International Society for Holter and Noninvasive Electrocardiology / Heart Rhythm Society / European Heart Rhythm Association / Asia Pacific Heart Rhythm Society

This collaborative statement from the International Society for Holter and Noninvasive Electrocardiology / Heart Rhythm Society / European Heart Rhythm Association / Asia Pacific Heart Rhythm Society describes the current status of mobile health ("mHealth") technologies in arrhythmia management. The range of digital medical tools and heart rhythm disorders that they may be applied to and clinical decisions that may be enabled are discussed. The facilitation of comorbidity and lifestyle management (increasingly recognized to play a role in heart rhythm disorders) and patient self-management are novel aspects of mHealth. The promises of predictive analytics but also operational challenges in embedding mHealth into routine clinical care are explored.

Over-fitting suppression training strategies for deep learning-based atrial fibrillation detection

Nowadays, deep learning-based models have been widely developed for atrial fibrillation (AF) detection in electrocardiogram (ECG) signals. However, owing to the inevitable over-fitting problem, classification accuracy of the developed models severely differed when applying on the independent test datasets. This situation is more significant for AF detection from dynamic ECGs. In this study, we explored two potential training strategies to address the over-fitting problem in AF detection. The first one is to use the Fast Fourier transform (FFT) and Hanning-window-based filter to suppress the influence from individual difference. Another is to train the model on the wearable ECG data to improve the robustness of model. Wearable ECG data from 29 patients with arrhythmia were collected for at least 24 h. To verify the effectiveness of the training strategies, a Long Short-Term Memory (LSTM) and Convolution Neural Network (CNN)-based model was proposed and tested. We tested the model on the independent wearable ECG data set, as well as the MIT-BIH Atrial Fibrillation database and PhysioNet/Computing in Cardiology Challenge 2017 database. The model achieved 96.23%, 95.44%, and 95.28% accuracy rates on the three databases, respectively. Pertaining to the comparison of the accuracy rates on each training set, the accuracy of the model trained in conjunction with the proposed training strategies only reduced by 2%, while the accuracy of the model trained without the training strategies decreased by approximately 15%. Therefore, the proposed training strategies serve as effective mechanisms for devising a robust AF detector and significantly enhanced the detection accuracy rates of the resulting deep networks.

The 2020 Canadian Cardiovascular Society/Canadian Heart Rhythm Society Comprehensive Guidelines for the Management of Atrial Fibrillation

The Canadian Cardiovascular Society (CCS) atrial fibrillation (AF) guidelines program was developed to aid clinicians in the management of these complex patients, as well as to provide direction to policy makers and health care systems regarding related issues. The most recent comprehensive CCS AF guidelines update was published in 2010. Since then, periodic updates were published dealing with rapidly changing areas. However, since 2010 a large number of developments had accumulated in a wide range of areas, motivating the committee to complete a thorough guideline review. The 2020 iteration of the CCS AF guidelines represents a comprehensive renewal that integrates, updates, and replaces the past decade of guidelines, recommendations, and practical tips. It is intended to be used by practicing clinicians across all disciplines who care for patients with AF. The Grading of Recommendations, Assessment, Development and Evaluations (GRADE) system was used to evaluate recommendation strength and the quality of evidence. Areas of focus include: AF classification and definitions, epidemiology, pathophysiology, clinical evaluation, screening and opportunistic AF detection, detection and management of modifiable risk factors, integrated approach to AF management, stroke prevention, arrhythmia management, sex differences, and AF in special populations. Extensive use is made of tables and figures to synthesize important material and present key concepts. This document should be an important aid for knowledge translation and a tool to help improve clinical management of this important and challenging arrhythmia.