Friedman, Paul A.

Introduction: Biologic sex and hormonal concentrations shape ECG parameters due to sex hormone effects on cardiac function. AI-enabled ECGs have previously been shown to accurately predict patient sex in adults and correlate with sex hormone levels. Aims: We aimed to test the ability of AI-enabled ECGs to predict pediatric patient sex and explore the influence of pubertal development. Methods: Two AI-enabled ECG models were created using a convolutional neural network trained on pediatric 10-second, 12-lead surface ECGs. The first model was trained de novo using pediatric data. The second model used transfer learning from a previously validated adult data-derived algorithm. We analyzed the first ECG from 89,063 pediatric patients (aged <19 years) recorded from 1988-2022, and divided the cohort into training, validation, and testing datasets. Subgroup analysis was performed on prepubertal (0-7 years), pubertal (8-14 years), and post-pubertal (15-18 years) patients. Results: The cohort consisted of 46.6% males, with 21,434 prepubertal, 26,300 pubertal, and 41,329 post-pubertal children. No significant performance difference was observed between the two models. Both models demonstrated 81% accuracy and an overall AUC of 0.91 in the testing cohorts, with high discriminatory ability in post-pubertal teenagers (AUC = 0.98). (Figure 1) Conclusions: AI-ECG predicts pediatric patient sex with varying performance across pubertal stages. The highest discriminatory ability is observed post-puberty, decreasing in pubertal and prepubertal children. The results suggest hormonal and physiological changes during puberty may influence cardiac electrophysiology measurably. More research is required to interpret CNN-predicted sex discordance with actual sex and its implications for future cardiovascular risk.

Introduction: Over 100 FDA-approved medications, electrolyte perturbations, and many disease states can prolong the QT interval in up to 10% of patients. In contrast, approximately 1 in 2,000 people have congenital long QT syndrome (LQTS) hallmarked by pathological QT prolongation secondary to LQTS-causative mutations. Hypothesis: an artificial intelligence (AI) deep neural network (DNN) analysis of the 12-lead ECG can distinguish patients with LQTS from those with acquired QT prolongation. Methods: The study cohort included 1599 patients with genetically confirmed LQTS and over 2.5 million controls from Mayo Clinic’s ECG data vault. Every patient/control with ≥ 1 ECG above age- and sex- specific 99 th percentile values for QTc [> 460 ms for all patients (male/female) < 13 years of age, or > 470 ms for men and > 480 ms for women above this age] was included. An AI-DNN involving a multi-layer convolutional neural network was developed. To simulate screening conditions, patients were matched at a ratio of 1:2,000 (incidence of LQTS) or 1:200 (balance of LQTS vs acquired QT prolongation in a tertiary referral center). Results: Among the 1,599 patients with LQTS, 808 (> 50%) had ≥ 1 ECG with QTc above the aforementioned QTc thresholds (2,987 ECGs) compared to 361,069/2.5M controls (14% of Mayo Clinic patients getting an ECG, 989,313 ECGs). Following age- and sex- matching and splitting, the model successfully distinguished LQTS from those with acquired QT prolongation at 1:2,000 matching with an AUC of 0.937 (accuracy 89%, sensitivity 81%, specificity 90%, PPV 0.05, NPV 0.99). Furthermore, when matching at a rate that genetically-mediated QT prolongation would be encountered in clinic (1:200), the model still successfully separated the two groups (AUC 0.912, accuracy 88%, sensitivity 78%, specificity 89%, PPV 0.1, NPV 0.99). Conclusion: For patients with a QTc exceeding its 99 th percentile values seen in health, this novel AI-DNN 12-lead ECG distinguishes abnormal QT prolongation stemming from LQTS versus acquired QT prolongation with high performance characteristics (AUC > 0.93). Even when scaled to referral center and LQTS incidence ratios, a negative AI-DNN signal rules out the presence of a LQTS disease-causative mutation with 99% negative predictive value.

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.

Background: Recently, our team developed an end-to-end artificial intelligence (AI) framework that can successfully estimate left ventricular ejection fraction (LVEF) from echocardiogram videos. However, this framework requires specific views (A2C and PLAX), and cannot estimate LVEF for studies from which these views are missing. Goal: To develop a deep learning model to estimate LVEF from any type and number of echocardiogram video(s). Methods: We built a deep learning model for B-mode echocardiogram videos of the left ventricle (A2C, A3C, A4C, or PLAX views), using retrospective transthoracic echocardiography (TTE) data. The model was evaluated with two test datasets: one with random valid (i.e., showing the left ventricle) views, and one with preselected views (both having 433 patients). Later, the model was tested with a prospective patient cohort that had both TTE and point-of-care ultrasound (POCUS) data collected simultaneously (393 patients). Results: On the retrospective TTE test set, we observed excellent performance (five random valid views per study: RMSE of 6.87% and Pearson correlation (PC) of 0.82 on LVEF estimation, and an AUC of 0.96 (95% CI: of 0.94-0.98) on low LVEF classification (<=40%); preselected views: RMSE of 6.75%, PC of 0.83, and an AUC of 0.96 (95% CI: 0.94-0.98)). For both TTE and POCUS data from the prospective cohort, we also observed strong performance (TTE: RMSE of 5.62%, PC of 0.87, and an AUC of 0.97; POCUS: RMSE of 6.12%, PC of 0.84, and an AUC of 0.96). For the prospective test sets as well as for the retrospective TTE test set with preselected views, results reflect the estimated LVEF average of A2C, A3C, A4C, and PLAX videos from each patient. Conclusions: For both TTE and POCUS data, the current model showed strong performance which was comparable to that of our previously established view-specific model. Alongside this comparable performance, the current model surpasses our previous model in terms of flexibility, as it isn't dependent on view classification and allows a broader selection in terms of the number and types of videos per study or patient. This innovative approach could prove particularly beneficial in point-of-care situations in which it may not be feasible to obtain high-quality, view-specific videos.

Introduction: Hyperkalemia has been associated with increased mortality in cardiac intensive care unit (CICU) patients. An artificial intelligence (AI) augmented electrocardiogram (ECG) can predict hyperkalemia, and other AI-ECG algorithms have predicted mortality in CICU patients. Hypothesis: We hypothesized that the AI-ECG hyperkalemia algorithm could stratify mortality risk beyond the serum potassium measurement. Methods: We included 11 234 unique Mayo Clinic CICU patients admitted from 2007 to 2018 who had both a 12 lead ECG and blood potassium (K) level obtained at admission with K>=5 mEq/L defining hyperkalemia. ECGs underwent AI evaluation for probability of hyperkalemia (probability >=0.5 defined as positive). Hospital mortality was analyzed using logistic regression and survival to one year was estimated using Kaplan-Meier and Cox analysis. Results: The mean age was 70 years, 37.8% were females, and 92.4% were white. Chronic kidney disease was present in 20%. The mean K value was 4.2 mEq/L, and 13% (n=1460) had hyperkalemia. AI-ECG predicted hyperkalemia in 34% (n=3820). In-hospital mortality was higher in patients with laboratory and/or AI-ECG hyperkalemia, and these patients had lower survival out to 1 year. Mortality was lowest in patients with normal K and negative AI-ECG and highest in patients with both laboratory-confirmed hyperkalemia and AI-ECG predicted hyperkalemia. Patients with isolated laboratory hyperkalemia (false-negative AI-ECG) had higher mortality than those with isolated AI-ECG hyperkalemia (false-positive AI-ECG). Conclusions: AI-ECG probability of hyperkalemia, even with a normal laboratory K, was associated with higher short-term and long-term mortality in CICU patients. The laboratory K value and AI-ECG probability of hyperkalemia are complementary risk factors. The AI algorithm can detect prognostically important ECG abnormalities which may reflect the effects of K levels on the myocardium.

Introduction: Accurate pre-procedure localization of the premature ventricular contractions (PVCs) is critical in preparing for the ablation. So far, no tools exist for artificial intelligence guided PVC-localization. Hypothesis: Trained deep learning model will be able to predict the site of origin (SOO) of PVCs using 12 lead electrocardiogram (ECG). Methods: We analyzed 707 ECG-ablation pairs from 2403 patients who had at least one ECG flagged for PVCs in the 6 months prior to their procedure. First, we used a cohort of 100,000 ECGs without PVCs and 10,000 ECGs with PVCs to train a convolutional neural network (CNN) to isolate the ECG segments with PVCs in them from a 12-lead ECG (Figure 1). We then trained two models, one utilizing a 10-second 12-lead ECG, and the other with just the isolated PVC beats as input. The models were trained using the Keras framework with TensorFlow (Google). Results: The mean age of the patients in the cohort was 61.2 ± 15.0 years. 66.3 percent of patients underwent left ventricle ablation, 54.6 percent had outflow ablation, and 55.3 percent had ablation of sites with leftward axis (septal location for midline and left ventricle PVCs). The model trained on 10-second ECGs yielded validation and test AUCs of 0.941 and 0.931 for outflow, 0.898 and 0.891 for ventricle, and 0.863 and 0.846 for leftward axis (Figure 1A). The model trained on isolated segments containing PVC beats yielded validation and test AUCs of 0.966 and 0.957 for outflow, 0.879 and 0.907 for ventricle, and 0.846 and 0.829 for SOO for leftward axis (Figure 1B). Conclusions: We have trained an accurate convolutional neural network that can predict SOO using the surface ECG. Further fine-tuning would be needed to refine the model and develop prediction of specific sites.

Introduction: Left ventricular diastolic function and filling pressure (FP) can be assessed by echocardiography and cardiac catheterization. Recently, a novel artificial intelligence (AI)-enabled ECG has proven to be effective in identifying increased FP determined by echocardiographic assessment of diastolic function. Aims: We aimed to compare a prognostic value of AI-ECG determination of FP compared to echocardiography and catheterization. Methods: This study included 4,213 patients who had a transthoracic echocardiogram and ECG within 30 days before hemodynamic cardiac catheterization. Increased FP was defined as pulmonary artery wedge pressure >15 mmHg by catheterization and grade 2 or 3 diastolic dysfunction by echocardiography based on diastolic function guidelines. Increased FP by ECG was identified using previously developed AI-ECG. AI-ECG performance was evaluated using the area under the curve (AUC) of the receiver operating curve. We assessed the risk of all-cause mortality using Kaplan-Meier estimate and multivariable Cox proportional-hazards models. Results: AI-ECG had AUCs of 0.8 and 0.7 for predicting increased FP determined by echocardiography and catheterization, respectively. Death was observed in 2,776 (34%) patients over a median follow-up of 7.7 years (IQR 7.1, 8.3). Survival of patients with normal and increased FP by AI-ECG, Echocardiography, and catheterization are shown in Figure 1a. Concordant diagnoses of normal and increased FP by AI-ECG and catheterization or echocardiography had the highest and lowest survival rates, respectively. When FP by AI-ECG was discordant, survival was better predicted by AI-ECG compared to echocardiography and catheterization (Figure 1b). Conclusion: AI-ECG for FP has a potential to discriminate prognosis further compared to FP determined by echocardiography or catheterization.

Background: Early diagnosis of heart failure with reduced ejection fraction (HFrEF) can lead to reduced healthcare costs and improved outcomes. In this study, we evaluated the economic effectiveness of an AI algorithm for detecting low ejection fraction using 12-lead ECG data. Aims: The primary aim was to estimate the health economic impact of the AI-ECG algorithm intervention to detect low ejection fraction. Methods: The study population consisted of 113,719 individuals without a prior history of HFrEF who visited a multistate health system between 2010-2018 with 12-lead ECG data and an echocardiogram within 1 year. The AI-ECG algorithm was run retrospectively, and projected costs were estimated assuming each patient incurred the average cost based on their age group and AI-detected heart failure stage. Hospitalization costs were estimated using Medicare reimbursement rates, and economic benefits from increased survival were estimated using a willingness to pay of $50,000 per quality-adjusted life year (QALY) gained. Results: At the optimal model threshold, the total cost savings per patient (including benefits from QALYs gained) for the AI-ECG algorithm vs. standard-of-care on the overall study population were projected to be $79.11 (95% CI: [$53.77, $104.44]) and $124.93 (95% CI: [$83.99, $165.88]) for the 2 and 4 year periods, respectively. For patients over 50 years old, the projected total cost savings were $85.56 (95% CI: [$56.35, $114.78]) and $135.28 (95% CI: [$88.18, $182.38]) per patient for the 2 and 4 year periods, respectively. See Figure. Conclusions: The AI-ECG algorithm to detect low ejection fraction is a cost-effective early screening tool for HFrEF.

Background: Depression is prevalent in almost half of patients with cardiovascular disease (CVD) and is associated with CVD and mortality. We used an artificial intelligence algorithm using ECG that predicts physiological age (ECG-Age) that has been linked to increased total and CVD mortality and CVD risk factors. We hypothesized that depressive symptoms would be associated with physiologic aging, as measured by AI-ECG Methods: A previously validated algorithm was applied to determine ECG-Age to 105,732 consecutive ECGs of individuals age ≥18 who sought care at Mayo Clinic between June 2019 and January 2022 and filled the Patient Health Questionnaire-9 (PHQ-9) within a month of ECG. PHQ-9 ascertains severity and frequency of depressive symptoms with values from 0 - 27 and were categorized as (None≤4, Mild 5-9, Moderate 10-14, Moderately severe 15-19 and Severe ≥20). Delta-Age was defined as ECG-Age minus chronological age and accelerated aging was defined as a Delta-Age >1SD above the mean. Logistic regression analyses associated PHQ-9 scores with positive Delta-Age while adjusting for CVD risk factors Results: Mean chronological age was 57±17 yr and mean ECG-Age 58±14 yr (R2 0.85, P < 0.0001), Delta-Age 0.71±9.40 yr, 58% women, average PHQ-9 5±6, 61% had none, 20% mild, 10 % moderate, 5%moderately severe and 4% had severe depressive symptoms. Delta-Age was 2.10 years higher in those with depressive symptoms, reflecting accelerated aging. Compared to those with no depressive symptoms, those with severe symptoms were 1.5 times more likely to have Delta-Age >1SD (OR 1.52 CI 95%1.73-1.33), independently of CV risk factors and the Charlson comorbidity index. Figure. There is significant attenuation of the association once accounting for age, sex and CVD risk factors Conclusions: Depressive symptoms are associated with physiologic aging, highlighting the role of depression in physiologic aging and the potential of preventive strategies in patients with depression and CVD

Introduction: The effect of obesity on physiologic aging has not been completely established. We previously developed an artificial intelligence-enabled electrocardiogram (AI-ECG) algorithm that predicts age and determines that the difference between AI-ECG age and chronological age (Age-Gap) may reflect accelerated aging, as it is associated with increased total and cardiovascular (CV) mortality. Hypothesis: Obesity is associated with an increased Age-Gap reflecting morbidity and accelerated aging. Methods: We included adults undergoing a wellness evaluation including, body fat percentage (BF%) determination by air displacement plethysmography (Bod-Pod®) and a 12-lead ECG within 2 years. We excluded those with coronary artery disease, congestive heart failure, stroke or being underweight defined as having a BMI<18.5 kg/m2. Logistic regression models tested the association between obesity by BMI (≥30 kg/m2) or BF% quartiles (≥35% for men and ≥25 in women) and being older by AI-ECG. Results: We included 1295 subjects (70% women), mean chronological age 46.7 ± 13.1 years. Mean Age-gap positive difference compared to those with BMI<25 was 1.39, 2.06 and 2.83 for those with BMI 25-29.9, 30-34.9, and >35 respectively (p<0.05). In the BF% group, Age-gap difference compared to those in the 1st quartile was 0.008, 0.37 and 1.52 for those in the 2nd, 3rd, and 4th quartile (p<0.05). Those>3 years older by ECG were more likely to have obesity by BMI and BF%, see figure. Conclusions: Obesity, either by BMI or BF%, is associated with increased physiologic age as measured by AI-ECG, highlighting the importance of obesity as a health threat.