Head-to-head comparison of proprietary PPG and single-lead ECG algorithms for atrial fibrillation detection

Abstract Funding Acknowledgements Type of funding sources: None. Background Population based screening for atrial fibrillation (AF) has been suggested to reduce stroke. Photoplethysmography (PPG) deriving smartphone apps and single-lead electrocardiography (ECG) tools are attractive devices for screening due to their low cost, convenience, and accessibility. Automated algorithm analysis can serve as pre-screening or remote monitoring for AF, while confirmation on an ECG trace >30s is required to establish the diagnosis. This work directly compares the performance of proprietary algorithms on PPG vs single-lead ECG for the detection of AF. Purpose To evaluate and compare the diagnostic performance of a PPG-deriving smartphone app and a single-lead ECG-deriving handheld device for AF detection. Methods Patients were recruited from the cardiology ward. After obtaining written informed consent, demographic and medical information were collected. Patients were instructed to perform one measurement using a pulse-deriving smartphone app and one via a single-lead ECG handheld device. A 12-lead electrocardiogram (ECG) was collected and interpreted by a cardiologist as gold standard. Patients with atrial flutter were excluded, with additional exclusions for insufficient quality measurements and unsuccessful measurements resulting due to technical errors. Unclassified single-lead ECG measurements were handled as test-negative. Sensitivity, specificity and accuracy were calculated with respect to the reference diagnosis. McNemar’s analysis was performed to compare the sensitivity and specificity of the proprietary PPG and single-lead ECG AF detection algorithms. Results The median age in the study population (n = 300) was 70 years (interquartile range: 51-78), 56.3% were men, and the median CHA2DS2-VASc was 3 (interquartile range: 1-4) with an AF-prevalence of 32.3%. PPG signal and single‑lead ECG quality was sufficient in 272/300 (91.0%) and 278/298 (93.3%) participants, respectively. After excluding atrial flutter patients (n = 25) and insufficient quality measurements, the sensitivity and specificity were 97.6% (95% CI 93.8 to 99.3) and 94.1% (95% CI 86.8 to 98.1) for the PPG signal versus 95.7% (95% CI 91.4 to 98.3) and 91.1% (95% CI 83.2 to 96.1) for the single‑lead ECG signal, respectively. Results demonstrated a 96.4% (95% CI 93.2 to 98.3) accuracy for PPG and 94.1% (95% CI 90.4 to 96.6) for single-lead ECG. No significant differences in sensitivity (P = 0.453) or specificity (P = 0.219) between the proprietary PPG and single-lead ECG algorithms were found. Conclusion This study demonstrated equivalent diagnostic performance of PPG and single-lead ECG proprietary AF detection algorithms in smartphone apps.

Download Full-text

Evaluation of the device independent nature of a photoplethysmography-deriving smartphone app

EP Europace ◽

10.1093/europace/euab116.525 ◽

2021 ◽

Vol 23 (Supplement_3) ◽

Author(s):

H Gruwez ◽

S Evens ◽

T Proesmans ◽

C Smeets ◽

P Haemers ◽

...

Keyword(s):

Atrial Flutter ◽

Sensitivity And Specificity ◽

Heart Rhythm ◽

Smartphone Application ◽

Detection Algorithm ◽

Interquartile Range ◽

Detection Algorithms ◽

Funding Sources ◽

Successful Measurement ◽

Sensitivity Specificity

Abstract Funding Acknowledgements Type of funding sources: None. Background Smartphone apps using photoplethysmography (PPG) technology enable digital heart rhythm monitoring through their built-in camera, without the need for additional, specific, or costly hardware. This may positively impact the availability and scalability of remote monitoring. However, the diversity of smartphone specifications on the consumer market may raise concerns regarding the robustness of AF detection algorithms between various devices. Purpose To study the device independency of AF detection performance by a PPG-based smartphone application. Methods Patients from the cardiology department were consecutively enrolled. Patients were handed 7 iOS models and 1 Android model and were asked to consecutively perform one PPG measurement per device. A 12-lead electrocardiogram (ECG) was collected during the same consultation and interpreted by a cardiologist as reference diagnosis. To allow an objective comparison across the devices, patients who failed to perform one successful measurement on each device were excluded. Additional exclusions were atrial flutter rhythms and insufficient quality results. Sensitivity, specificity and accuracy were calculated with respect to the reference diagnosis. McNemar’s analysis was used for the head-to-head comparison of the sensitivity and specificity of the proprietary algorithm on the different smartphone devices. Results A total of 150 patients participated in the study with a median CHA2DS2-VASc score of 3 (interquartile range: 1-5). The median age of the study population was 70 (interquartile range: 56-79) years. In total, 54.7% of the population was male and the AF-prevalence was 35.3%. After the exclusion of patients with atrial flutter (n = 14) and patients who did not successfully perform a PPG measurement on each device (n = 5), diagnostic-grade results of 131 patients were used to calculate the performance of the proprietary algorithm. The sensitivity and specificity of the AF detection algorithm ranged from 90.9% (95% CI 75.7-98.1) to 100.0% (95% CI 91.0-100) and 94.5% (95% CI 86.6-98.5) to 100.0% (95% CI 94.6-100), respectively. The overall accuracy across the devices ranged from 94.4% (95% CI 88.3-97.9) to 99.0% (95% CI 94.6-100). Head-to-head comparisons of the results did not reveal significant differences in sensitivity (P = 0.125-1.000) or specificity (P = 0.375-1.000) of the proprietary AF detection algorithm among the different devices. Conclusion This study demonstrated the device-independent nature of the PPG-deriving smartphone application with respect to 12-lead ECG diagnosis.

Download Full-text

A novel diagnostic algorithm for the identification of different supraventircular tachycardias

European Heart Journal ◽

10.1093/ehjci/ehaa946.0371 ◽

2020 ◽

Vol 41 (Supplement_2) ◽

Author(s):

M Solis Cancino ◽

A.D Pacheco Bouthillier ◽

L.A Moreno Ruiz

Keyword(s):

Atrial Fibrillation ◽

Atrial Flutter ◽

Sensitivity And Specificity ◽

Diagnostic Algorithm ◽

Atrioventricular Nodal Reentrant Tachycardia ◽

Supraventricular Arrhythmias ◽

Reentrant Tachycardia ◽

Clinical Tool ◽

Diagnostic Challenge ◽

Atrioventricular Reentrant Tachycardia

Abstract Background Supraventricular arrhythmias represent a diagnostic challenge. Its prevalence and causes are not well established, given the impossibility to differentiate between all different types of supraventricular tachycardias (SVT). There are several supraventricular arrhythmias, but we focus on: 1) Atrial tachycardia (AT) 2) Junctional tachycardia (JT) 3) Atrial fibrillation (AF) 4) Atrial flutter (AA) 5) Atrioventricular nodal reentrant tachycardia (AVNRT), and 6) Atrioventricular reentrant tachycardia (AVRT). The electrocardiographic diagnosis is based on the presence of P-waves, its morphology and relationship with the QRS complex, and the relationship between the atrial and ventricular frequency. Purpose The purpose of this study was to create a helpful clinical tool that could serve the physician as a guide to determine a diagnosis and initial treatment. Additionally, we wanted to establish the sensitivity and specificity of the algorithm. Methods It is a diagnostic test study. We include 190 electrocardiograms of different SVT of patients undergoing electrophysiological studies. The data consists of 760 observations from two different readings of the electrocardiograms. Results 104 of 112 AF, were correctly identified using the algorithm, with a sensitivity and specificity of 92.9% and 99.1%, respectively (95% CI: 0.86–0.96). 76 of 760 were AA, and 62 were correctly diagnosed, with a sensitivity and specificity of 81.6% and 95.5%, respectively (95% CI 0.71–0.88). 50 of the 72 AT were correctly classified, with a sensitivity of 69.4% and specificity of 97.4% (95% CI 0.58–0.78). 99 of 152 AVNRT were identified with a sensitivity and specificity of 64.5% and 87%, respectively (95% CI 0.84–0.89). 254 of 344 AVRT were diagnosed correctly with a sensitivity of 73.8% and specificity of 88.2% (95% CI 0.68–0.78). Finally, 1 of 4 JT were identified, with a sensitivity and specificity of 25% and 99.1% respectively (95% CI 0.04–0.69). Conclusion The algorithm is an excellent diagnostic tool to identify atrial flutter, atrial fibrillation and atrioventricular reentrant tachycardia. SVT algorithm Funding Acknowledgement Type of funding source: None

Download Full-text

P1457Preliminary results of the FLASH-AF: Validation of the device independent nature of a pulse deriving smartphone application for the detection of atrial fibrillation

European Heart Journal ◽

10.1093/eurheartj/ehz748.0222 ◽

2019 ◽

Vol 40 (Supplement_1) ◽

Author(s):

T Proesmans ◽

C Smeets ◽

P Dreesen ◽

J Vanhaeren ◽

P Vandervoort

Keyword(s):

Atrial Fibrillation ◽

Diagnostic Accuracy ◽

Medical Information ◽

Low Cost ◽

Smartphone Application ◽

Cardiac Monitoring ◽

Attractive Alternative ◽

Handheld Device ◽

The Mean ◽

Quality Measurements

Abstract Background Smartphone applications using photoplethysmography (PPG) technology through their camera are becoming an attractive alternative for atrial fibrillation (AF) screening due to their low cost, convenience, and broad accessibility. However, some important questions concerning their diagnostic accuracy, robustness and device independent nature remain to be answered. Purpose This study evaluated the diagnostic accuracy of a PPG-based pulse-deriving smartphone application with respect to handheld single-lead ECG and 12-lead ECG. In addition, the device dependent nature and robustness of the performance of the application was assessed. Methods 300 Patients who are scheduled for a regular consultation or procedure (i.e. ablation or cardioversion) will be recruited from the cardiology ward. Additionally, patients hospitalized for continuous cardiac monitoring will be recruited to enrich the database with AF measurements. After obtaining written informed consent, the patients fill in a questionnaire collecting demographic and medical information. The pulse-deriving application will be tested on total of 14 different smartphones, 7 iOS devices and 7 Android devices. In total, each device will be measured with 150 times. The patients will additionally perform a single-lead ECG measurement with a handheld device. Subsequently, a 12-lead ECG will be recorded to obtain the reference diagnosis. Results A total of 164 patients already participated in the study. The mean age was 64 (±19) years, 58% was male. The AF-prevalence was 37%. On average, patients in AF had a higher CHA2DS2-VASc score; 3.93 (±1.80) compared to 2.02 (±1.63) for non-AF patients. The amount of insufficient quality measurements recorded with the pulse-deriving smartphone application ranged from 4% (iOS) to 13% (Android). Averaged for all the smartphone devices, the pulse-deriving application scored 81.2% (±5%) sensitivity, 97.1% (±1%) specificity, 88.8% (±2%) NPV, 95.0% (±1%) PPV, and 90.9% (±2%) accuracy. The handheld single-lead ECG device had 78.2% sensitivity, 95.5% specificity, 87.6% NPV, 91.5% PPV, and 88.9% accuracy. The same calculations were preformed after excluding regular atrial flutter measurements. On average, the pulse-deriving application scored 90.1% (±2%) sensitivity, 97.1% (±1%) specificity, 95.2% (±1%) NPV, 94.0% (±1%) PPV, and 94.8% (±1%) accuracy. The handheld single-lead ECG device had 90.2% sensitivity, 97.7% specificity, 97.7% NPV, 95.1% PPV, and 96.9% accuracy. Conclusion The diagnostic accuracy of the pulse-deriving smartphone application and the handheld single-lead ECG device was strongly influenced by the presence of regular atrial flutters, stressing the importance of further thorough validation. For the pulse-deriving smartphone application, there was no significant influence from device type in terms of diagnostic accuracy for the detection of AF. Insufficient quality measurements were more frequently performed on Android devices.

Download Full-text

Smartphone applications for the detection of atrial fibrillation in primary care: a review (Preprint)

10.2196/preprints.26001 ◽

2020 ◽

Author(s):

Nicholas Fraser

Keyword(s):

Atrial Fibrillation ◽

Primary Care ◽

Screening Test ◽

Low Cost ◽

Cochrane Library ◽

Test Accuracy ◽

Smartphone Apps ◽

Cross Sectional ◽

Community Screening ◽

Detection Algorithms

BACKGROUND Smartphone technology continues to advance at a fervid pace. In the field of cardiology, traditional physical tests recommended to detect cardiac arrhythmias may soon be superseded by low-cost, convenient and reliable smartphone apps. We aimed to determine the screening test accuracy of smartphone apps in detecting atrial fibrillation (AF) in patients within primary care. OBJECTIVE The aim of this review was to determine the screening test accuracy of smartphone apps in detecting atrial fibrillation (AF) in patients within primary care. METHODS Systematically searched MEDLINE, PUBMED, Web of Science, CINAHL and Cochrane Library until 01 February 2019. Articles were screened, and evaluated before relevant data extracted and study quality appraised using the QUADAS-2 tool. The raw test accuracy data was constructed into a 2x2 contingency table and the test accuracy statistics were calculated, and organised in descriptive plots. RESULTS Seven cross-sectional studies tested one, or two smartphone apps including the AliveCor 1-lead 30 sec ECG (seven studies; n=16,359), and the Cardiio Rhythm PPG (one study; n=1,012). The prevalence of AF ranged from 1.17%-12.29%, with a mean of 2.65%. The AliveCor 1-lead device reported a sensitivity ranging from 0.92 to 0.99, and specificity from 0.99 to 1.00. Sensitivity and PPV showed the greatest heterogeneity, with results ranging from suboptimal to excellent. The Cardiio Rhythm PPG app recorded a sensitivity of 0.71 (95% CI 0.51-0.87) and specificity of 0.99 (95% CI 0.98-1.00). CONCLUSIONS Community screening for AF using smartphone electrocardiography (ECG) or photoplethysmography (PPG) is feasible. Smartphone apps that screen for AF in primary care demonstrate excellent specificity, but suboptimal sensitivity. Further optimisation of detection algorithms, to accommodate the spectrum of disease seen within the community, should be considered before such devices are used as a tool for systematic auto-screening.

Download Full-text