Diagnostic Accuracy of a Novel Mobile Phone Application for the Detection and Monitoring of Atrial Fibrillation

doi:10.1016/j.amjcard.2018.01.035

The American Journal of Cardiology

Volume 121, Issue 10, 15 May 2018, Pages 1187-1191

https://doi.org/10.1016/j.amjcard.2018.01.035 Get rights and content

Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia in adults, associated with significant morbidity, increased mortality, and rising health-care costs. Simple and available tools for the accurate detection of arrhythmia recurrence in patients after electrical cardioversion (CV) or ablation procedures for AF can help to guide therapeutic decisions. We conducted a prospective, single-center study to evaluate the accuracy of Cardiio Rhythm Mobile Application (CRMA) for AF detection. Patients >18 years of age who were scheduled for elective CV for AF were enrolled in the study. CRMA finger pulse recordings, utilizing an iPhone camera, were obtained before (pre-CV) and after (post-CV) the CV. The findings were validated against surface electrocardiograms. Ninety-eight patients (75.5% men), mean age of 67.7 ± 10.5 years, were enrolled. No electrocardiogram for validation was available in 1 case. Pre-CV CRMA readings were analyzed in 97 of the 98 patients. Post-CV CRMA readings were analyzed for 92 of 93 patients who underwent CV. One patient left before the recording was obtained. The Cardiio Rhythm Mobile Application correctly identified 94 of 101 AF recordings (93.1%) as AF and 80 of 88 non-AF recordings (90.1%) as non-AF. The sensitivity was 93.1% (95% confidence interval [CI] = 86.9% to 97.2%) and the specificity was 90.9% (95% CI = 82.9% to 96.0%). The positive predictive value was 92.2% (95% CI = 85.8% to 95.8%) and the negative predictive value was 92.0% (95% CI = 94.8% to 95.9%). In conclusion, the CRMA demonstrates promising potential in accurate detection and discrimination of AF from normal sinus rhythm in patients with a history of AF.

Section snippets

Methods

This prospective, single-center study was approved by the institutional review board at the Massachusetts General Hospital (MGH) and is compliant with the Health Insurance Portability and Accountability Act. Consecutive patients with a diagnosis of AF who were scheduled for elective direct current cardioversion (DCCV) at MGH were eligible to participate. Patients younger than 18 years of age were excluded. Informed consent was obtained from patients who agreed to participate in the study before

Results

A total of 100 subjects arriving for elective DCCV at MGH were screened for participation in the study. In 1 case, the patient was not scheduled to undergo CV and was mistakenly recruited. In another case, we were unable to obtain the subject's CRMA recordings because of a technical issue. A total of 98 subjects, 73 men and 24 women, with a mean age 67.7 ± 10.5 years (range 40 to 92 years), were enrolled in the study. Table 1 provides a summary of the patient demographics. At least 1 set of

Discussion

AF continues to be challenging to screen, diagnose, and monitor, especially when the arrhythmia is paroxysmal and/or asymptomatic. Camm et al performed 24-hour ambulatory monitoring in 106 asymptomatic elderly subjects and found 10.5% to have AF.⁹ In the Framingham Heart Study, 228 of the 562 subjects (40%) with AF were reported to be asymptomatic.¹⁰

Tools other than the standard 12-lead ECG are also used for the detection of AF. Taggar et al analyzed 21 studies that investigated 39 different

Disclosures

Drs. Yukkee and Ming-Zher Poh are employees of Cardiio, Inc. and have an ownership stake in the company. Dr Ming-Zher Poh has a patent for the AF detection algorithm described here. There are no other potential conflicts of interest relevant to this study.

References (20)

S. Colilla et al.
Estimates of current and future incidence and prevalence of atrial fibrillation in the U.S. adult population
Am J Cardiol
(2013)
A.J. Camm et al.
Usefulness of continuous electrocardiographic monitoring for atrial fibrillation
Am J Cardiol
(2012)
A.J. Camm et al.
The rhythm of the heart in active elderly subjects
Am Heart J
(1980)
P.D. Ziegler et al.
Comparison of continuous versus intermittent monitoring of atrial arrhythmias
Heart Rhythm
(2006)
S.S. Chugh et al.
Worldwide epidemiology of atrial fibrillation: a Global Burden of Disease 2010 Study
Circulation
(2014)
M.Z. Poh et al.
Non-contact, automated cardiac pulse measurements using video imaging and blind source separation
Opt Express
(2010)
M.Z. Poh et al.
Advancements in noncontact, multiparameter physiological measurements using a webcam
IEEE Trans Biomed Eng
(2011)
P.H. Chan et al.
Diagnostic performance of a smartphone-based photoplethysmographic application for atrial fibrillation screening in a primary care setting
J Am Heart Assoc
(2016)
G.S. Stergiou et al.
Diagnostic accuracy of a home blood pressure monitor to detect atrial fibrillation
J Hum Hypertens
(2009)
J. Wiesel et al.
Detection of atrial fibrillation using a modified Microlife blood pressure monitor
Am J Hypertens
(2009)

There are more references available in the full text version of this article.

Cited by (37)

Reducing inappropriate and unnecessary implantable cardioverter-defibrillator therapy: Is patient confirmation via a mobile application the solution?
2023, Heart Rhythm O2
Comparison of 2 Smart Watch Algorithms for Detection of Atrial Fibrillation and the Benefit of Clinician Interpretation: SMART WARS Study
2022, JACC: Clinical Electrophysiology
Smart watches and wearable technology capable of heart rhythm assessment have increased in use in the general population. The Apple Watch Series 4 (AW4) and KardiaBand (KB) are devices capable of obtaining single-lead electrocardiographic recordings, presenting a novel opportunity for the detection of paroxysmal arrhythmias.
The aim of this study was to assess the diagnostic utility of the AW4 and KB in an elderly outpatient population.
Consecutive recordings were taken from patients attending cardiology outpatient clinic from the AW4 and KB concurrently with 12-lead electrocardiography. Automated diagnoses and blinded single-lead electrocardiographic tracing interpretations by 2 cardiologists were analyzed. Analysis was also conducted to assess the effect of combined device and clinician interpretation.
One hundred twenty-five patients were prospectively recruited (mean age 76 ± 7 years, 62% men). The accuracy of the automated rhythm assessment was higher with the KB than the AW4 (74% vs 65%). For the detection of atrial fibrillation, the sensitivity and negative predictive value of the KB were 89% and 97%, respectively, and of the AW4 were 19% and 82%, respectively. Using hybrid automated and clinician interpretation, the overall accuracy of the KB and AW4 was 91% and 87%, respectively.
The KB automated algorithm outperformed the AW4 in its accuracy and sensitivity for detecting atrial fibrillation in the outpatient setting. Clinician assessment of the single-lead electrocardiogram improved accuracy. These findings suggest that although these devices’ tracings are of sufficient quality, automated diagnosis alone is not sufficient for making clinical decisions about atrial fibrillation diagnosis and management.
A novel multi-scale convolutional network with attention-based bidirectional gated recurrent unit for atrial fibrillation discrimination
2021, Biocybernetics and Biomedical Engineering
Citation Excerpt :
Also, it should be noted that the above CNN deep models including [21,22,24–27,47] all use single-scale convolution kernel as the main building block and their performances of these models are lower than that of our proposed multi-scale convolution kernel model. Besides, instead of ECG signals, Aschbacher et al. [42] proposed a deep model based on the features of photoplethysmography (PPG) signals and LSTM with the specificity of 92.1% and the sensitivity of 81.0%, and Bashar et al. [43] developed a methodology using several PPG features such as entropy for AF detection as well as the similar model strategies include the methods of Bonomi et al. [44] and Rozen et al. [45], but they still need to extract additional features to further improve performance and the application cost of PPG signal is relatively high. Therefore, based on the above analysis, it is clear that the proposed model is more superior to these state-of-the-art methods and the major contributions and novelties can be shown as follows:
Atrial fibrillation (AF) is a major cardiovascular disease that has affected thousands of individuals worldwide. The electrocardiogram (ECG) is the most extensively applied approach to detect AF at present, while the traditional detection strategy based on the visual observation of ECG data is often laborious and inefficient. In this work, we specially designed an intelligent recognition system based on a novel convolutional neural network that utilizes the multi-scale convolution kernel and bidirectional gated recurrent unit with attention mechanism for AF detection. Also, two standard control groups using 10-fold cross-validation were performed to assess the validity of the proposed model. The empirical results not only demonstrate the high efficiency of multi-scale convolution kernel, but also show that the model has a more superior classification performance to several state-of-the-art methods with an accuracy of 98.3% and 97.7% on two public databases, respectively. Due to its high performance, we plan to develop the model into portable devices to benefit more individuals such as the elderly and athletes.
Diagnostic accuracy of smart gadgets/wearable devices in detecting atrial fibrillation: A systematic review and meta-analysis
2021, Archives of Cardiovascular Diseases
Recently, smart devices have been used for medical purposes, particularly to screen for atrial fibrillation. However, current data on the diagnostic performance of these devices are scarce.
We performed a systemic review and meta-analysis to assess the accuracy of atrial fibrillation diagnosis by smart gadgets/wearable devices.
We comprehensively searched the MEDLINE, EMBASE and Cochrane databases for all works since the inception of each database until January 2020. Included in this review were published observational studies of the diagnostic accuracy of smartphones or smartwatches in detecting atrial fibrillation. We calculated the area under the summary receiver operating characteristic curves and pooled sensitivities and specificities.
Participants in our study were from the general population or were patients with underlying atrial fibrillation. In the overall analyses, the areas under the summary receiver operating characteristic curves were 0.96 and 0.94 for smartphones and smartwatches, respectively. Smartphones had a sensitivity of 94% and a specificity of 96%, and smartwatches showed similar diagnostic accuracy, with a specificity of 94% and a sensitivity of 93%. In subgroup analyses, we found no difference in diagnostic accuracy between photoplethysmography and single-lead electrocardiography.
This study suggests that smart devices have similar diagnostic accuracies. Regarding atrial fibrillation detection methods, there was also no difference between photoplethysmography and single-lead electrocardiography. However, further studies are warranted to determine their clinical implications in atrial fibrillation management.
Récemment, des appareils intelligents ont été utilisés à des fins médicales, en particulier pour dépister la fibrillation atriale. Cependant, les données actuelles sur le diagnostic et les performances de ces appareils sont limitées.
Nous avons effectué une revue systémique et une méta-analyse pour évaluer la précision diagnostique dans la détection de la fibrillation atriale par des gadgets intelligents/appareils portables (smartphones & smartwatches).
Nous avons effectué une recherche complète dans MEDLINE, EMBASE et Cochrane, bases de données incluant tous les travaux publiés depuis la création de chaque base de données et ce jusqu’en janvier 2020. Cette revue a inclus des études observationnelles publiées sur la précision diagnostique des téléphones (smartphones) et montres intelligentes (smartwatches) pour détecter la fibrillation atriale.
Les participants à notre étude provenaient de la population générale ou étaient des patients avec fibrillation atriale sous-jacente. Nous avons calculé les aires sous les courbes ROC, 0.96 pour les smartphones et 0.94 pour les smartwatches, respectivement. Les smartphones ont une sensibilité de 94% et une spécificité de 96% pour détecter une fibrillation atriale et les smartwatches ont une précision diagnostique similaire avec une spécificité de 94% et une sensibilité de 93%. Dans les analyses en sous-groupes, nous n’avons trouvé aucune différence dans la précision diagnostique entre la photopléthysmographie et l’électrocardiogramme à une seule dérivation.
Cette étude suggère que les appareils intelligents (smartphones & smartwatches) ont une précision diagnostique similaire. Concernant les méthodes de détection de la fibrillation atriale, il n’y avait pas non plus de différence entre la photopléthysmographie et l’électrocardiogramme à une seule dérivation. Cependant, des études sont nécessaires pour déterminer les implications cliniques de ces résultats dans la prise en charge de la fibrillation atriale.
Big data and new information technology: what cardiologists need to know
2021, Revista Espanola de Cardiologia
La continua progresión tecnológica que experimenta la medicina se produce cada vez a mayor velocidad, lo que exige una actualización constante del profesional de la salud. La nueva ola de tecnologías que está abriéndose camino en la práctica clínica incluye: a) salud asistida por el móvil (mHealth) o dispositivos miniaturizados que permiten la detección constante de parámetros biológicos, a cualquier hora y en cualquier lugar, de cientos de miles de pacientes a la vez; b) inteligencia artificial impulsada por nuevas técnicas de aprendizaje profundo que están batiendo a médicos expertos en su propio campo (pruebas de imagen o electrocardiografía); c) impresión tridimensional que permite vislumbrar un mundo de prótesis cardiovasculares adaptadas a cada paciente; d) medicina de sistemas, que apoyándose en el big data abrirá las puertas a la medicina personalizada, aunando en modelos matemáticos de gran complejidad datos genéticos, epigenéticos, ambientales, clínicos y sociales para diseñar tratamientos de precisión. Esta revisión pretende resumir la evidencia sobre los últimos avances tecnológicos basados en tecnologías de la información y ciencias de la computación aplicados a la cardiología y esbozar un mapa que de un solo vistazo permita tener una impresión general del horizonte hacia el que va a progresar la cardiología en los próximos años.
Technological progress in medicine is constantly garnering pace, requiring that physicians constantly update their knowledge. The new wave of technologies breaking through into clinical practice includes the following: a) mHealth, which allows constant monitoring of biological parameters, anytime, anyplace, of hundreds of patients at the same time; b) artificial intelligence, which, powered by new deep learning techniques, are starting to beat human experts at their own game: diagnosis by imaging or electrocardiography; c) 3-dimensional printing, which may lead to patient-specific prostheses; d) systems medicine, which has arisen from big data, and which will open the way to personalized medicine by bringing together genetic, epigenetic, environmental, clinical and social data into complex integral mathematical models to design highly personalized therapies. This state-of-the-art review aims to summarize in a single document the most recent and most important technological trends that are being applied to cardiology, and to provide an overall view that will allow readers to discern at a glance the direction of cardiology in the next few years.
Performance of an automated photoplethysmography-based artificial intelligence algorithm to detect atrial fibrillation
2020, Cardiovascular Digital Health Journal

View all citing articles on Scopus

: See page 1190 for disclosure information.

¹: These authors contributed equally to this study and to the manuscript preparation.

View full text

Diagnostic Accuracy of a Novel Mobile Phone Application for the Detection and Monitoring of Atrial Fibrillation

Section snippets

Methods

Results

Discussion

Disclosures

Am J Cardiol

Am J Cardiol

Am Heart J

Heart Rhythm

Worldwide epidemiology of atrial fibrillation: a Global Burden of Disease 2010 Study

Circulation

Non-contact, automated cardiac pulse measurements using video imaging and blind source separation

Opt Express

Advancements in noncontact, multiparameter physiological measurements using a webcam

IEEE Trans Biomed Eng

Diagnostic performance of a smartphone-based photoplethysmographic application for atrial fibrillation screening in a primary care setting

J Am Heart Assoc

Diagnostic accuracy of a home blood pressure monitor to detect atrial fibrillation

J Hum Hypertens

Detection of atrial fibrillation using a modified Microlife blood pressure monitor

Am J Hypertens