ECG Algorithm Research Articles

AI-ECG sex-estimation discordances predict cardiovascular events

Abstract Background Cardiovascular disease remains the leading cause of death and burden globally, warranting the need to explore novel predictors of major adverse cardiovascular events (MACE) using new technologies. Artificial Intelligence (AI)-enhanced ECG (AI-ECG) algorithms can provide information on cardiovascular health of individuals, independent of conventional risk factors. Purpose The current study aimed to understand whether a previously described AI-ECG sex estimation algorithm could be applied for risk stratification of major adverse cardiovascular events (MACE), with a focus on a population without previous history of MACE. Methods Patients having 12-lead ECGs between 2018-2019 were included and followed up until August 2023. Exclusion criterion was a prior history of MACE. AI-ECG sex estimation was determined using a previously described algorithm predicting sex based on 12-lead ECGs. The algorithm generates outputs ranging from 0 to 1, representing the estimated probability of being male. Outputs above 0.5 were considered as AI-ECG-estimated male, and below as AI-ECG-estimated female. This was compared with biological sex to determine whether the two matched. Inconsistency between AI-ECG estimated sex and biological sex was defined as discordance. MACE was defined as myocardial infarction, coronary revascularization, non-fatal stroke, and all-cause death. Results Among 80,578 subjects (mean age 59.1 ± 16.1 years, 52.4% female), 9.7% exhibited sex discordance in AI-ECG estimation. Discordance was associated with a 56% higher risk of MACE (HR, 1.56, 95% CI 1.46-1.65, p <0.001), and remained significant in multivariable analysis with traditional risk factors (HR 1.35, 95% CI 1.27-1.43, p <0.001) (Figure 1). Sex-specific analysis showed increased MACE risk in both discordant males and females (Figure 2). Conclusion In patients with no history of MACE, discordance in AI-ECG sex estimation is associated with increased risk of MACE. These findings suggest the potential of AI-ECG sex estimation algorithm for risk assessment, especially in patients with no history of MACE.Figure 1.Cox proportional hazard modelFigure 2.Kaplan-Meier of MACE by sex

Evaluation of Previous ECG Algorithms for Identifying the Culprit Artery in Inferior STEMI

Abstract Background ST elevation myocardial infarction (STEMI) is the leading cause of cardiovascular related mortality. In the context of Inferior STEMI, the culprit artery is whether the right coronary artery (RCA) or left circumflex coronary artery (LCx). An accurate prediction of culprit artery prior to primary percutaneous coronary intervention (PCI) could improve the door-to-balloon time and could prevent contrast-induced-nephropathy. However algorithms established for that purpose were not evaluated within the Tunisian population. We aimed then to assess their performance in our population Methods We conducted a single-center hostorical cohort study from January 2014 to December 2022 on patients admitted at Sahloul university hospital Sousse Tunisia for PCI following an inferior STEMI. We analyzed the ST segment deviations at the J-point and 80 ms after the J-point then evaluated the performance of previous algorithms. Results We collected data from 156 patients, of which 130 (83.3%) hada RCA occlusion, and 26 (16.7%) a LCx occlusion. We evaluated 19 published algorithms. The criterion of zimetbaum et al (ST elevation in III> II and I and/or VL <-1 mm) had the highest AUC 0.77[0.66 - 0.88] YI = 52 accuracy = 0.81. The criterion of Herz et al (ST III > II) had AUC ROC of 0.71 [0.58 - 0.83] YI = 0.41 accuracy = 0.78, the algorithm of Tierala et al: AUC = 0.70[0.58 - 0.83] YI = 0.40 accuracy = 0.78. The algorithm og Fiol et al: AUC = 0.79 [0.58-0.83] YI = 0.59 accuracy = 0.73. Conclusions The Zimetbaum criterion had the highest performance in predicting the culprit artery and Fiol’s algorithm was the most specific to the LCx occlusion in our population. Further study on the impact of these algorithms on patients outcome is necessary. Key messages • The prediction of the culprit artery in inferior STEMI is challenging. • The criterion of Zimetbaum et al had a good performance for that regard.

Use of artificial intelligence-powered ECG to differentiate between cardiac and pulmonary pathologies in patients with acute dyspnoea in the emergency department

BackgroundAcute dyspnoea is common in acute care settings. However, identifying the origin of dyspnoea in the emergency department (ED) is often challenging. We aimed to investigate whether our artificial intelligence...

Enhancing cardiac postoperative care: a smartwatch-integrated remote telemonitoring platform for health screening with ECG analysis.

The integration of smartwatches into postoperative cardiac care transforms patient monitoring, systematically tracking vital signs and delivering real-time data to a centralized platform. This study focuses on developing a platform for seamless integration, assessing reliability, and evaluating the impact on post-cardiac surgery. The goal is to establish a robust foundation for understanding the efficacy and dependability of smartwatch-based telemonitoring, enhancing care for this population. A total of 108 cardiac surgery patients were divided into telemonitoring (TLM) and control (CTL) groups. The TLM group utilized smartwatches for continuous monitoring of vital parameters (SpO2, HR, BP, ECG) over 30 ± 3 days. Statistical analyses (Pearson, Intraclass Correlation, Bland-Altman, Tost Test) were employed to compare smartwatch measurements with traditional methods. Significant correlations and concordance were observed, particularly in HR and BP measurements. Challenges were noted in SpO2 measurement. The ECG algorithm exhibited substantial agreement with cardiologists (Kappa: 0.794; p > 0.001), highlighting its reliability. The telemonitoring platform played a crucial role in early detection of clinical changes, including prompt Emergency Department (ED) visits, contributing significantly to preventing outcomes that could lead to mortality, such as asymptomatic Atrioventricular block. Positive patient responses affirmed technological efficacy, especially in identifying cardiac arrhythmias like atrial fibrillation. The integration of smartwatches into remote telemonitoring for postoperative cardiac care demonstrates substantial potential, improving monitoring and early complication detection, thereby enhancing patient outcomes. The FAPO-X Study (Assisted Digital Telemonitoring with Wearables in Patients After Cardiovascular Surgery; NCT05966857) underscores the promising role of telemonitoring in postoperative cardiac care.

Activation pattern of the coronary sinus facilitates the differentiation for ventricular outflow tract arrhythmias.

The accuracy of surface ECG algorithms for predicting the origin of outflow tract ventricular arrhythmias (OT-VAs) might be questioned. Intracardiac electrograms recorded at anatomic landmarks could provide new predictive insights. We aim to evaluate the efficacy of a novel criterion utilizing the activation pattern of the coronary sinus (CS) in localizing OT-VAs, including VAs originating from the right ventricular outflow tract (RVOT), endocardial left ventricular outflow tract (Endo-LVOT), and epicardial left ventricular outflow tract (Epi-LVOT). We measured the ventricular activation time of the mitral annulus (MA) from the onset of the earliest QRS complex of VAs to the initial deflection over the isoelectric line at local signals, namely the QRS-MA interval. The activation at 3 and 12 o'clock of the MA was recorded as the QRS-MA3 and QRS-MA12 intervals, respectively. Their predictive values were compared to previous ECG algorithms. A total of 68 patients with OT-VAs were enrolled (51 for development and 17 for validation). From early to late, the ventricular activation sequences at MA12 were as follows: Epi-LVOT, Endo-LVOT, and RVOT. In LBBB morphology OT-VAs, the QRS-MA12 interval was significantly earlier for LVOT origins than RVOT origins. In the combined cohort of development and validation cohort, a cut-off value of ≤10 ms predicted the LVOT origin with a sensitivity of 100% and specificity of 78%. The QRS-MA12 interval ≤ -24 ms additionally predicted epicardial LVOT sites of origin. The QRS-MA interval could accurately differentiate the OT-VAs localization.

Assessing the efficacy of AI-enhanced ECG algorithms with prospective evaluation in a virtual environment: An innovative strategy

Assessing the efficacy of AI-enhanced ECG algorithms with prospective evaluation in a virtual environment: An innovative strategy

12-Lead ECG fiducials extraction using hybrid bidirectional LSTM networks and commercial ECG algorithm

12-Lead ECG fiducials extraction using hybrid bidirectional LSTM networks and commercial ECG algorithm

Development Model of Basic ECG Algorithm Interpretation based on Android Application to Increase Ability of Nurses

Introduction: ECG interpretation ability by nurses is still not appropriate, thus affecting the accuracy and speed of the next action that will be given to patients to reduce mortality. This study aimed to develop the use of an Android system based ECG interpretation algorithm application to improve the ability of nurse. Methods: Research development design to develop the Android system based ECG interpretation algorithm application. The sampling method used purposive sampling as many as 90 nurses in Universitas Airlangga Hospital, Surabaya, Indonesia. The independent variables were interpretation skills (rhythm, frequency, pacemaker, axis, extrasistole, block, enlargement of heart muscle, coronary heart disease, conclusion). The instruments that are used are ECG recording, ECG interpretation sheet, and focus group disscusion. The analysis used an independent t-test with a significance level of α ≤ 0.05. Results: The results of developing an ECG interpretation algorithm application based on the Android system have received recommendations from experts and feasibility tests have been carried out. There was a relationship between basic ECG interpretation capabilities with the use of an android system based ECG interpretation algorithm application. Increasing the ability of nurses had p value of 0,000. Conclusion: The application of an ECG interpretation algorithm based on an android system can improve nurses' abilities in interpreting basic ECG. For future researchers, it is necessary to develop and refine the Android system-based ECG interpretation algorithm application and this can be applied on a case-by-case basis.

Predictive Value of Artificial Intelligence-Enabled Electrocardiography in Patients With Takotsubo Cardiomyopathy.

Recent studies have indicated high rates of future major adverse cardiovascular events in patients with Takotsubo cardiomyopathy (TC), but there is no well-established tool for risk stratification. This study sought to evaluate the prognostic value of several artificial intelligence-augmented ECG (AI-ECG) algorithms in patients with TC. This study examined consecutive patients in the prospective and observational Mayo Clinic Takotsubo syndrome registry. Several previously validated AI-ECG algorithms were used for the estimation of ECG- age, probability of low ejection fraction, and probability of atrial fibrillation. Multivariable models were constructed to evaluate the association of AI-ECG and other clinical characteristics with major adverse cardiac events, defined as cardiovascular death, recurrence of TC, nonfatal myocardial infarction, hospitalization for congestive heart failure, and stroke. In the final analysis, 305 patients with TC were studied over a median follow-up of 4.8 years. Patients with future major adverse cardiac events were more likely to be older, have a history of hypertension, congestive heart failure, worse renal function, as well as high-risk AI-ECG findings compared with those without. Multivariable Cox proportional hazards analysis indicated that the presence of 2 or 3 high-risk findings detected by AI-ECG remained a significant predictor of major adverse cardiac events in patients with TC after adjustment by conventional risk factors (hazard ratio, 4.419 [95% CI, 1.833-10.66], P=0.001). The combined use of AI-ECG algorithms derived from a single 12-lead ECG might detect subtle underlying patterns associated with worse outcomes in patients with TC. This approach might be beneficial for stratifying high-risk patients with TC.

A machine learning approach to differentiate wide QRS tachycardia: distinguishing ventricular tachycardia from supraventricular tachycardia.

Differential diagnosis of wide QRS tachycardia (WQCT) has been a challenging issue. Published algorithms to distinguish ventricular tachycardia (VT) and supraventricular tachycardia (SVT) have limited diagnostic capabilities. A total of 278 patients with WQCT from January 2010 to March 2022 were enrolled. The electrophysiological study confirmed SVT in 154 patients and VT in 65 ones. Two hundred nineteen WQCT 12-lead ECGs were randomly divided into development cohort (n = 165) and testing cohort (n = 54) data sets. The development cohort was split into a training group (n = 115) and an internal validation group (n = 50). Forty ECG features extracted from the 219 WQCT ECGs are fed into 9 iteratively trained ML algorithms. This novel ML algorithm was also compared with four published algorithms. In the development cohort, the Gradient Boosting Machine (GBM) model displayed the maximum area under curve (AUC) (0.91, 95% confidence interval (CI) 0.81-1.00). In the testing cohort, the GBM model had a higher AUC of 0.97 compared to 4 validated ECG algorithms, namely, Brugada (0.68), avR (0.62), RWPTII (0.72), and LLA algorithms (0.70). Accuracy, sensitivity, specificity, negative predictive value, and positive predictive value of the GBM model were 0.94, 0.97, 0.90, 0.94, and 0.95, respectively. A GBM ML model contributes to distinguishing SVT from VT based on surface ECG features. In addition, we were able to identify important indicators for distinguishing WQCT.

Deep learning-based NT-proBNP prediction from the ECG for risk assessment in the community.

The biomarker N-terminal pro B-type natriuretic peptide (NT-proBNP) has predictive value for identifying individuals at risk for cardiovascular disease (CVD). However, it is not widely used for screening in the general population, potentially due to financial and operational reasons. This study aims to develop a deep-learning model as an efficient means to reliably identify individuals at risk for CVD by predicting serum levels of NT-proBNP from the ECG. A deep convolutional neural network was developed using the population-based cohort study Hamburg City Health Study (HCHS, n=8,253, 50.9 % women). External validation was performed in two independent population-based cohorts (SHIP-START, n=3,002, 52.1 % women, and SHIP-TREND, n=3,819, 51.2 % women). Assessment of model performance was conducted using Pearson correlation (R) and area under the receiver operating characteristics curve (AUROC). NT-proBNP was predictable from the ECG (R, 0.566 [HCHS], 0.642 [SHIP-START-0], 0.655 [SHIP-TREND-0]). Across cohorts, predicted NT-proBNP (pNT-proBNP) showed good discriminatory ability for prevalent and incident heart failure (HF) (baseline: AUROC 0.795 [HCHS], 0.816 [SHIP-START-0], 0.783 [SHIP-TREND-0]; first follow-up: 0.669 [SHIP-START-1, 5 years], 0.689 [SHIP-TREND-1, 7.3 years]), comparable to the discriminatory value of measured NT-proBNP. pNT-proBNP also demonstrated comparable results for other incident CVD, including atrial fibrillation, stroke, myocardial infarction, and cardiovascular death. Deep learning ECG algorithms can predict NT-proBNP concentrations with high diagnostic and predictive value for HF and other major CVD and may be used in the community to identify individuals at risk. Long-standing experience with NT-proBNP can increase acceptance of such deep learning models in clinical practice.

Diagnostic performance of bedside transthoracic echocardiography in the management of patients with suspected myocardial infarction

Abstract Introduction The diagnostic management of patients with symptoms indicative of acute myocardial infarction (MI) is largely based on the clinical assessment of symptoms, ECG and diagnostic algorithms based on high-sensitivity cardiac troponin (hs-cTn). Additionally, guidelines recommend transthoracic echocardiography (TTE) for patients neither eligible for rule-out nor rule-in. Purpose To assess the diagnostic performance of TTE in the diagnostic management of patients with suspected MI. Methods We conducted a prospective, observational cohort study enrolling consecutive patients presenting with symptoms suggestive of MI to the emergency department of a tertiary care hospital. Management was at the discretion of the treating physician. The final diagnosis was adjudicated by two independent cardiologists according to the 4th Universal Definition of MI. Evaluation in the emergency department included TTE with an assessment of left-ventricular (LV) function and regional wall motion abnormalities (WMA). We calculated diagnostic performance parameters (sensitivity, specificity, and negative and positive predictive value [NPV, PPV] with 95% confidence intervals) to rule-out or rule-in MI, respectively, for the presence of WMA. Performance was calculated for patients stratified to the observe and rule-in groups after application the ESC 0/1-hour algorithm using hs-cTnI. Patients without available TTE and those with STEMI were excluded. WMA were only reported if the acoustic window allowed a reasonable evaluation. Variables associated with the conduction of coronary angiography were studied using logistic regression. Results Overall, 2,163 patients with available TTE were included, of those 1,707 (78.9%) patients had data on WMA available, with 360 (21.1%) having any WMA (hypokinesia or akinesia) detected. Median age was 64 (51, 75) years, 1,383 (63.9%) were males and 383 (17.7%) were diagnosed with MI. Application of the ESC 0/1-hour algorithm resulted in assignment of 729 (42.9%) and 308 (18.6%) patients to the rule-out and -in groups, respectively, with 599 (38.5%) patients remaining in the observe group. Of these, 45 (7.5%) had MI. Performance of the ESC 0/1-hour algorithm was good and is shown in Figure 1. In the observe group, detection of WMA poorly identified patients with MI (PPV 12.7%, Figure 1). Absence of WMA resulted in moderate performance to rule-out MI (NPV 94.0%, Figure 1). Adjusted for age, sex, symptoms, baseline hs-cTnI, presence of ischemic signs on ECG and cardiovascular risk factors, the presence of WMA was significantly associated with the conduction of coronary angiography at index presentation (OR 2.7 [1.8, 4.0], p < 0.001). Conclusion The diagnostic value of bedside TTE is limited for the rule-out or -in of MI. Still, patients with prevalent WMA were significantly more likely to undergo coronary angiography.Performance of TTE in ESC 0/1-hour alg.

The R'' wave in V1 and the terminal QRS vector in aVF combine to a novel 12-lead ECG algorithm to identify slow conducting anatomical isthmus 3 in patients with tetralogy of Fallot

Abstract Background Patients with repaired tetralogy of Fallot (rTOF) have an increased risk for ventricular tachycardia (VT), with slow conducting anatomical isthmus (SCAI) 3 as the dominant VT substrate. Right bundle branch block (RBBB) is common in these patients. SCAI 3 can lead to additional activation delay affecting the terminal RV activation.(1) We hypothesize that these alterations can be detected by sinus rhythm (SR) 12-lead ECG. Purpose To develop a novel algorithm to non-invasively identify SCAI 3 with the SR 12-lead ECG in rTOF patients with RBBB. Methods Consecutive rTOF patients aged ≥16 years old with RBBB who underwent electroanatomical mapping (EAM) at our institution between 2017-2022, and 2010-2016, comprised the derivation, and validation cohort, respectively. 12-lead ECGs in SR were analyzed for presence of R’’ wave in V1 (any positive deflection after R’ in V1) and for duration of a negative terminal QRS portion (NTP) in aVF (ms between QRS complex traversing the isoelectric line from positive to negative vector until QRS offset) (Figure, A). Results Forty-six rTOF patients (aged 40±15 years, 70% male) comprised the derivation cohort. Patients were repaired at a median age of 3.3 [IQR 1.2-6.6] years via a ventricular incision in 21 (55%). Mean QRS duration was 165±23ms and 12 (26%) had a QRS duration of ≥180ms. SCAI 3 was detected in 31 (67%) patients and 15 (33%) had a normal conducting AI 3. Monomorphic sustained VT was inducible in 16 (35%). Among patients with SCAI 3, 17 (55%), and 16 (52%), had an R’’ in V1, and NTP ≥80ms in aVF, respectively, compared to only 1 (7%), and 1 (7%) patients with normal conducting AIs. Of the 24 patients with an R’’ in V1 and/or NTP ≥80ms in aVF, 22 (92%) had a SCAI 3. In univariable analysis an R’’ in V1 had an odds ratio (OR) of 17.0 (p=0.010), a NTP ≥80ms in aVF had an OR of 14.9 (p=0.014), and QRS duration had an OR of 1.04 (per ms, p=0.030) in identifying SCAI 3. After correction for QRS duration, R’’ in V1 and NTP ≥80ms in aVF both remained independently predictive of SCAI 3 (OR 15.4, p=0.017, and OR 10.0, p=0.042, respectively). Combining R’’ in V1 and NTP ≥80ms in aVF into a diagnostic algorithm (Figure, B) yielded a sensitivity of 71% and specificity of 87% in detecting SCAI 3. In the validation cohort (n=34, age 44±14 years, 19 (56%) with SCAI 3 and 15 (44%) with normal conducting AI 3) the diagnostic algorithm had a sensitivity of 84% and specificity of 80% for a SCAI. Conclusion A diagnostic algorithm including R’’ in V1 and NTP ≥80ms in aVF had excellent sensitivity and specificity in detecting SCAI 3 in rTOF patients with RBBB in SR. These ECGs characteristics likely represent a delayed terminal RV activation directed towards the lateral RVOT caused by slow infundibular conduction. Whether this algorithm can contribute to non-invasive risk stratification for VT in a general rTOF population requires additional studies.Diagnostic algorithm to identify SCAI 3

Abstract 14208: New Artificial Intelligence Algorithms of Electrocardiogram Detection for Patients With Heart Failure

Introduction: Heart failure (HF) is a global pandemic with an increasing prevalence. The burden of HF-related hospitalizations and costs are increasing. Artificial intelligence(AI) algorithms applied to electrocardiograms have shown promise in diagnosing HF, but they require a large amount of training data, which is scarce and costly. Transfer learning addresses this challenge by utilizing knowledge from previous tasks, resulting in superior performance with limited data compared to conventional AI algorithms Hypothesis: AI algorithms using transfer learning can be accurately diagnose heart failure with reduced ejection fraction(HFrEF) using electrocardiograms compared to conventional AI algorithms. Methods: All patients aged 18 years older from the two hospitals with at least one ECG were included in the study. Electrocardiogram, transthoracic Echocardiogram (TTE), and demographic data were collected. The AI algorithm consisted of two phases: conventional AI algorithms, pre-training with a large-scale ECG dataset and transfer learning for HFrEF detection. Performance evaluation was conducted using various metrics, including AUROC, sensitivity, specificity, accuracy, F1 score. Results: The dataset consisted of 687,911 ECGs from 259,943 patients. Testing the conventional model on the HFrEF resulted in a sensitivity of 63.8%, specificity of 96.2%, accuracy of 93.6%, area under the receiver operating curve (AUC) of 0.93 (95% CI, 0.92 - 0.94), and F1 score of 61.6%. Testing the transfer learning model on the HFrEF increased the sensitivity of 75.1%, accuracy of 93.7%, area under the receiver operating curve (AUC) of 0.952 (95% CI, 0.93 - 0.96), and F1 score of 65.6%. but lowered the specificity of 95.3% Conclusions: A New AI-enabled ECG algorithm for identifying patients with HFrEF can be more accurately diagnosed than conventional AL algorithms.

Abstract 18941: Pediatric Electrocardiogram-Based Deep Learning to Predict Left Ventricular Dysfunction and Remodeling

Background: Artificial intelligence-enhanced ECG algorithms show promise to detect a range of adult phenotypes, inclusive of ventricular dysfunction and remodeling. However, it remains unclear whether similar approaches provide meaningful and generalizable predictions in children. Methods: We trained a convolutional neural network on paired ECG-echos (≤ 2 days apart) from patients ≤ 18 years old without major congenital heart disease to detect human expert classified greater than mild left ventricular (LV) dysfunction, hypertrophy, and dilation (individually and as a composite outcome). Model performance was assessed by calculating area under the receiver operating curve (AUROC) in an internal test set and external setting (emergency department). Results: The main cohort comprised of 119,787 ECG-echo pairs (61,722 patients; median age 8.3 [IQR, 2.9-13.8] years; 55% male). Training, internal test, and external test sets comprised of 92,377, 24,343, and 3,067 ECG-echo pairs with 8.2%, 7.7%, and 10.0% composite outcome rates, respectively. During internal testing, AUROCs of 0.84 [95% CI, 0.83-0.85], 0.87 [95% CI, 0.85-0.88], 0.86 [95% CI, 0.84-0.87], and 0.84 [95% CI, 0.83-0.86] were achieved for the LV composite outcome, dysfunction, hypertrophy, and dilation, respectively (Figure). During external testing, AUROCs of 0.80 [95% CI, 0.77-0.83], 0.84 [95% CI, 0.80-0.88], 0.77 [95% CI, 0.72-0.82], and 0.86 [95% CI, 0.82-0.89] were achieved, respectively (Figure). Similar model performance was achieved when using quantitative z-score cutoffs for each outcome. Exploratory saliency mapping provides insight into notable ECG components that influence model predictions. Conclusions: Our artificial intelligence-enhanced pediatric ECG algorithm shows promise to inexpensively screen for and diagnose LV dysfunction or remodeling in children. Prospective trials to guide model implementation for supporting clinical decision making are warranted.

Abstract 17189: Association Between an Artificial Intelligence-Enabled ECG Algorithm for Detection of Atrial Fibrillation While in Sinus Rhythm, and Hospitalization for Recurrent Ischemic Stroke in the Community

Introduction: Artificial Intelligence-enabled electrocardiogram (AI-ECG) can estimate the probability of having atrial fibrillation (AFib) while in sinus rhythm. Recurrent stroke in subjects without known AFib may be due to undiagnosed AFib. We evaluated the association between AI-ECG AFib and recurrent ischemic stroke and with incident AFib Methods: Consecutive adults admitted to a Mayo Clinic or Mayo Clinic Health System hospital or to Olmsted Community Hospital for ischemic stroke between January,1,2000 and June,30,2022 were identified using the Rochester Epidemiology Project. Recurrent ischemic strokes were identified after reviewing medical records and imaging studies. We excluded those with known AFib or without ECGs within 3 years prior to the first ischemic stroke. The AI-ECG network was previously developed using 649,931 ECGs from > 180,000 subjects. The primary endpoint was recurrent ischemic stroke; secondary endpoint was incident AFib. Cox model adjusted for age, sex, and CHA2DS2-VASc2 score variables. Results: We included 4,816 subjects, mean age 70±15 yr, 54% male, 92% white, median CHA2DS2-VASc2 score 4 [3-5], median follow-up 3 [1-7] yr. 280 subjects (6%) had a second stroke and 366 (8%) were diagnosed with incident AFib over follow-up. Those with a positive AI-ECG AFib had an increased risk for recurrent ischemic stroke(HR 1.54, 95% CI 1.21-1.96, P<0.001 Figure A) This association remained significant after adjustment for CHA2DS2-VASc2 score(HR 1.37 95% CI 1.07-1.76, P=0.01). These associations were not affected by anticoagulation use over follow-up. AI-ECG AFib was also associated with incident AFib(HR 2.42 95% CI 1.96-2.98, P<0.001 Figure B) Conclusions: AI-ECG AFib predicts recurrent ischemic stroke in the community, possibly reflecting undiagnosed AFib that can be detected by AI-ECG, as it also predicts incident AFib. These results underscore the potential utility of AI-ECG for risk stratification of patients with ischemic stroke.

Performance of Single-Lead Handheld Electrocardiograms for Atrial Fibrillation Screening in Primary Care: The VITAL-AF Trial

BackgroundHandheld single-lead electrocardiographic (1L ECG) devices are increasingly used for atrial fibrillation (AF) screening, but their real-world performance is not well understood. ObjectivesThe purpose of this study was to quantify the diagnostic test characteristics of 1L ECG automated interpretations for prospective AF screening. MethodsWe calculated the diagnostic test characteristics of the AliveCor KardiaMobile 1L ECG (AliveCor, US) algorithm using unblinded cardiologist overread as the gold standard using single 30s tracings administered by medical assistants among individuals aged ≥65 years participating in the VITAL-AF trial (NCT03515057) of population-based AF screening embedded within routine primary care. ResultsA total of 14,230 individuals (mean age 74 ± 7 years, 60% women, 82% White) had 31,376 tracings reviewed by 13 cardiologists. A total of 24,906 (79.6%) tracings had an AliveCor interpretation of normal, 5,046 (16.1%) were unclassified, 797 (2.5%) were possible AF, and 573 (1.8%) were no analysis. Cardiologists read 808 (2.6%) tracings as AF. AliveCor possible AF had a PPV of 51.7% (95% CI: 47.8%-55.6%). AliveCor normal had an NPV of 99.8% (95% CI: 99.7%-99.8%). The AliveCor algorithm had an overall sensitivity of 51.0% (95% CI: 47.1%-54.9%) and a specificity of 98.7% (95% CI: 98.6%-98.9%). AliveCor tracings interpreted as unclassified (PPV 5.9%, 95% CI: 5.1%-6.7%) and no analysis (PPV 6.5%, 95% CI: 4.6%-8.9%) had low predictive values for AF and were increasingly prevalent at older ages (13.7% for age 65-69 years to 28.1% for age ≥85 years, P < 0.01). ConclusionsIn an older primary care population undergoing AF screening with handheld 1L ECGs, automated algorithm interpretations were sufficiently accurate to exclude the presence of AF but not to establish an AF diagnosis.

A novel ECG algorithm to differentiate between ventricular arrhythmia from right versus left ventricular outflow tract.

The aim of this study was to evaluate the accuracy of the diagnostic criteria for determining the origin of outflow tract ventricular arrhythmia (OTVA) and develop an ECG algorithm to predict its origin. We analyzed the ECGs of 100 patients with OTVA who underwent successful ablation. The QRS complex was measured during sinus rhythm and ventricular arrhythmia. After the ECG algorithm was developed, it was validated in an additional 100 patients from two different hospitals. In this retrospective study, among the parameters without restrictions in the transition lead, the V2S/V3R index (AUC = 0.96) was significantly better in predicting ventricular arrhythmia originating from the right ventricular outflow tract (RVOT). Further, the larger initial r wave surface area (ISA) in V1 and V2 (AUC = 0.06) was significantly better in predicting ventricular arrhythmias originating from the left ventricular outflow tract (LVOT). Among the parameters with the transition lead in V3, the V2S/V3R index (AUC = 0.82) was significantly better in predicting VAs originating from the RVOT. On the contrary, the V3 R-wave deflection interval (AUC = 0.19) was significantly better in predicting ventricular arrhythmias originating from the LVOT. The algorithm combining the V2S/V3R index and the larger ISA in V1 and V2 could predict OTVA origin with an accuracy of 95.00%, a sensitivity of 87.18%, a specificity of 100.00%, a positive predictive value (PPV) of 100.00%, and a negative predictive value (NPV) of 92.42%. In the validation study, the algorithm exhibited excellent accuracy (95.00%) and AUC (AUC = 0.95), with a sensitivity of 94.12%, a specificity of 95.45%, a PPV of 91.43%, and an NPV of 96.92%. Our developed algorithm can reliably predict OTVA origin without restrictions in the transition lead.

Validation of a clinical model for predicting left versus right ventricular outflow tract origin of idiopathic ventricular arrhythmias.

Prediction of the chamber of origin in patients with outflow tract ventricular arrhythmias (OTVA) remains challenging. A clinical risk score based on age, sex and presence of hypertension was associated with a left ventricular outflow tract (LVOT) origin. We aimed to validate this clinical score to predict an LVOT origin in patients with OTVA. In a two-center observational cohort study, unselected patients undergoing catheter ablation (CA) for OTVA were enrolled. All procedures were performed using an electroanatomical mapping system. Successful ablation was defined as a ≥80% reduction of the initial overall PVC burden after 3 months of follow-up. Patients with unsuccessful ablation were excluded from this analysis. We included 187 consecutive patients with successful CA of idiopathic OTVA. Mean age was 52±15 years, 102 patients (55%) were female, and 74 (40%) suffered from hypertension. A LVOT origin was found in 64 patients (34%). A score incorporating age, sex and presence of hypertension reached 73% sensitivity and 67% specificity for a low (0-1) and high (2-3) score, to predict an LVOT origin. The combination of one ECG algorithm (V2 S/V3 R-index) with the clinical score resulted in a sensitivity and specificity of 81% and 70% for PVCs with R/S transition at V3 . The published clinical score yielded a lower sensitivity and specificity in our cohort. However, for PVCs with R/S transition at V3, the combination with an existing ECG algorithm can improve the predictability of LVOT origin.

Paced P-wave morphology templates to guide atrial tachycardia localization: A derivation and validation study.

Surface ECG is a useful tool to guide mapping of focal atrial tachycardia (AT). We aimed to construct 12-lead ECG templates for P-wave morphology (PWM) during endocardial pacing from different sites in both atria in patients with no apparent structural heart disease (derivation cohort), with the goal of creating a localization algorithm, which could subsequently be validated in a cohort of patients undergoing catheter ablation of focal AT (validation cohort). We prospectively enrolled consecutive patients who underwent electrophysiology study, had no structural heart disease and no atrial enlargement. Atrial pacing, at twice diastolic threshold, was carried out at different anatomical sites in both atria. Paced PWM and duration were assessed. An algorithm was generated from the constructed templates of each pacing site. The algorithm was applied on a retrospective series of successfully ablated AT patients. Overall and site-specific accuracy were determined. Derivation cohort included 65 patients (25 men, age 37±13 years). Atrial pacing was performed in 1025 sites in 61 patients (95%) in RA and in 15 patients (23%) in LA. The validation cohort included 71 patients (28 men, age 52±19 years). AT were right atrial in 66.2%. The algorithm successfully predicted AT origin in 91.5% of patients (100% in LA and 87.2% in RA). It was off by one adjacent segment in the remaining 8.5%. A simple ECG algorithm based on paced PWM templates was highly accurate in localizing site of origin of focal AT in patients with structurally normal hearts.