Douglass, Erika J.

Abstract Background Cardiomyopathy is a leading cause of pregnancy-related mortality and represents the number one cause of death in the late postpartum period. Purpose The goal of this study was to prospectively evaluate the effectiveness of an artificial intelligence-based electrocardiogram (AI-ECG) and a point-of-care artificial intelligence-enhanced digital stethoscope for cardiomyopathy screening among pregnant and postpartum women. Methods We conducted a single-arm prospective study of pregnant and postpartum women (up to 12 months following delivery) enrolled at three sites between October 2021 and October 2022. Study participants completed a standard 12-lead ECG, ECG + phonocardiogram recordings with a digital stethoscope, a demographic questionnaire, and a transthoracic echocardiogram (TTE) within a 24-hour period. Diagnostic performance of the AI-ECG and digital stethoscope for cardiomyopathy detection, defined as left ventricular ejection fraction (LVEF) < 45%, was evaluated using the area under the curve (AUC) and standard measures of diagnostic performance. Results We included 100 consecutive pregnant and postpartum women who provided informed consent and had at least one ECG recorded, completed the questionnaire, and had an echocardiogram performed. All participants self-identified as female, and 41% were recruited from a federally qualified health center. The median age of participants was 31 years (Q1: 27, Q3: 34). Thirty-eight percent identified as non-Hispanic White, 32% as non-Hispanic Black, 21% as Hispanic or Latino, 6% as Asian, and 3% as Other (1 Native Hawaiian and 2 multiracial). Five percent (5/100) had LVEF <45%. The 12-lead AI-ECG model demonstrated close to perfect discrimination (accuracy 99%; AUC 1.00, sensitivity 100%) for detection of cardiomyopathy with 1 false positive screen (LVEF 49%), Figure 1A. The digital stethoscope (angled position) had an AUC of 0.99 (95% CI: 0.97, 1.00) for detection of LVEF <45%. Sensitivity, specificity, positive predictive value, and negative predictive value for the digital stethoscope were 60%, 98%, 60% and 98%, respectively for detection of LVEF <45% (Figure 1B & 1C). Conclusions In this small prospective study, we found that 5% of pregnant and postpartum women had cardiomyopathy, a potentially life-threatening condition, and that a rapid, inexpensive, ubiquitous, massively scalable point-of-care test - the ECG - showed promising results, identifying it with an AUC > 0.98 with the application of artificial intelligence.Figure 1

Abstract Aims Current non-invasive screening methods for cardiac allograft rejection have shown limited discrimination and are yet to be broadly integrated into heart transplant care. Given electrocardiogram (ECG) changes have been reported with severe cardiac allograft rejection, this study aimed to develop a deep-learning model, a form of artificial intelligence, to detect allograft rejection using the 12-lead ECG (AI-ECG). Methods and results Heart transplant recipients were identified across three Mayo Clinic sites between 1998 and 2021. Twelve-lead digital ECG data and endomyocardial biopsy results were extracted from medical records. Allograft rejection was defined as moderate or severe acute cellular rejection (ACR) based on International Society for Heart and Lung Transplantation guidelines. The extracted data (7590 unique ECG-biopsy pairs, belonging to 1427 patients) was partitioned into training (80%), validation (10%), and test sets (10%) such that each patient was included in only one partition. Model performance metrics were based on the test set (n = 140 patients; 758 ECG-biopsy pairs). The AI-ECG detected ACR with an area under the receiver operating curve (AUC) of 0.84 [95% confidence interval (CI): 0.78–0.90] and 95% (19/20; 95% CI: 75–100%) sensitivity. A prospective proof-of-concept screening study (n = 56; 97 ECG-biopsy pairs) showed the AI-ECG detected ACR with AUC = 0.78 (95% CI: 0.61–0.96) and 100% (2/2; 95% CI: 16–100%) sensitivity. Conclusion An AI-ECG model is effective for detection of moderate-to-severe ACR in heart transplant recipients. Our findings could improve transplant care by providing a rapid, non-invasive, and potentially remote screening option for cardiac allograft function.

Abstract An estimated 1 billion people worldwide have deficient or insufficient levels of vitamin D (VitD), while roughly 25% of individuals in the US are reported to have inadequate VitD levels. Considerable evidence indicates that VitD deficiency is associated with an increased risk of cardiovascular disease (CVD), yet it remains unclear whether low VitD is simply a biomarker of CVD or has a true pathologic role. The role of VitD deficiency in the pathogenesis of inflammatory myocarditis is unknown- a disease that is higher in men. We found that only 20% of myocarditis patients had low VitD levels. When we examined VitD receptor (VDR) deficient mice in a model of viral myocarditis, we found that VDR decreased myocarditis in females (p=0.007) but increased inflammation in males (p=0.006). Examination of cytokines in the heart found that VDR-/- male mice had significantly reduced IFN-γ and IL-1β and increased IL-17A, but no significant differences were found in females. Comparison of microarray data during myocarditis to published VitD response element genes revealed around 200 genes associated with proinflammatory (e.g. caspase-1) and profibrotic (e.g. TGF-β1, MMP) responses were increased in WT male mice, which was confirmed by qRT-PCR in VDR-/- mice. Our data confirms that VitD through the VDR promotes myocardial inflammation in males, providing a mechanism for how low VitD levels in men increase myocarditis. This work was funded by NIH R01 HL087033 and HL111938.

Abstract Background Screening for left ventricular (LV) systolic dysfunction (defined as ejection fraction ≤35%) based on data from a standard 12-lead electrocardiogram (ECG) has become well established when standard digital ECGs are available–8 independent leads sampled at least 250 hertz for 10 seconds. As the algorithm has been incorporated into various clinical scenarios and ancillary research projects, a limitation of the binary classification at 35% has become apparent. Purpose The objective of this study was to develop and validate a deep learning-based algorithm that would classify LVEF into three categories based on only the digital ECG input. Methods After IRB approval, native digital resting ECGs acquired between 1/1/2010 and 12/31/2021 on patients seen in Mayo Clinic in Jacksonville were extracted from the institutional electronic ECG database management system (MUSE, GE Healthcare). These ECGs were matched with transthoracic echocardiograms obtained up to four days prior or 30 days after the ECGs acquisition. A convolutional neural network consisting of 8 layers of convolutions, batch normalization and pooling was trained using Keras and Tensorflow with hyper-parameter optimization for L1 and L2 regularization, learning rate adjustments, and class weights to predict three classes of LVEF: ≤35%, 36–51%, and ≥52% based on clinical relevance. The primary measure of overall performance was the detection of LVEF ≤35%; however, the triad of model predictions was also considered in translating the model output to human interpretable findings. Results A total of 30,153 patients (60,169 ECG pairings; mean age 63 years; 48% male) were randomly split at the patient level into training (24,172 patients), validation (2,973 patients) and testing (3,008 patients). The trained model provided robust discrimination in the withheld testing data – AUROC of 0.941 (95% CI: 0.931 to 0.950). Using the optimal model threshold based on Youden's index from the validation data (0.186), sensitivity and specificity were estimated to be 87.9% (95% CI: 83.8% to 91.2%) and 86.3% (95% CI: 85.4% to 87.2%) in the testing data. In instances where discordant predictions were observed, the posterior distribution of model probabilities provide additional insights into the possible underlying value of LVEF (Figure 1). Conclusions The utilization of a multi-category deep learning classification model for the detection of reduced ejection fraction adds new dimensions to the use of AI technologies on digital ECGs. This work shows high discrimination can still be obtained when using three classes of LVEF. Funding Acknowledgement Type of funding sources: None.

Abstract Background Pregnancy related cardiomyopathy is a significant cause of maternal morbidity and mortality globally. A presumed overlap between normal pregnancy-associated symptoms and clinical symptoms of cardiomyopathy contributes to delays in diagnosis and increased risk of maternal mortality. Purpose We sought to evaluate the association between patient-reported cardiovascular symptoms and the presence of cardiomyopathy among pregnant and postpartum patients. We hypothesize that individual cardiovascular symptoms are unrelated to the presence of cardiomyopathy. We also evaluated the use of a novel adaptation of a validated health status questionnaire in relation to cardiomyopathy. Methods We enrolled 48 pregnant (>13 weeks) and postpartum (up to 12 months) participants in a prospective study between October 2021 and February 2022. All study participants completed a baseline questionnaire, which included current cardiovascular symptoms, an assessment of health status using an adapted version of the Kansas City Cardiomyopathy Questionnaire (KCCQ-12), followed by a resting transthoracic echocardiogram on the same day. We defined cardiomyopathy as a left ventricular ejection fraction (LVEF) <50% based on 2-D echocardiography. Fisher's exact and Wilcoxon rank-sum tests were employed to evaluate the association between reported cardiovascular symptoms, the adapted KCCQ-12 (KCC-A) score, and cardiomyopathy. Results At the time of enrollment, 67% were pregnant and 33% postpartum. Forty-eight percent identified as White, 31% as Black, 10% as Asian, and 10% as other race. The median age was 31 years (Q1: 28, Q3: 35) and 6% had an LVEF <50%. We found no statistically significant association between four reported cardiovascular symptoms (shortness of breath, orthopnea, fast breathing, and episodes of “asthma” that did not improve with inhalers or other treatment) and cardiomyopathy or medial E/e' ratio. KCC-A scores were low in the study population overall (median 52; Q1:40, Q3: 61). We demonstrated a significantly lower KCC-A score among women with LVEF <50% (median 24; Q1: 15, Q3: 44) compared to women with LVEF ≥50% (median 54; Q1: 44, Q3: 61) p=0.02. Conclusions We showed no significant association between individual cardiovascular symptoms and cardiomyopathy in an obstetric population. However, we demonstrate for the first time that an adapted KCCQ-12 questionnaire for health status assessment could potentially identify women with a high-likelihood of cardiomyopathy during the peripartum period who may benefit from additional evaluation including echocardiography. Larger studies are needed to validate this finding. Funding Acknowledgement Type of funding sources: Foundation. Main funding source(s): This study was funded by a research grant from the Miami Heart Research Institute, Florida Heart Foundation. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health

Abstract Background Atrial fibrillation (AF) is diagnosed in up to 50% of patients with heart failure. However, the prevalence of AF among patients with peripartum cardiomyopathy (PPCM) ranges from only 2–10%, with the lowest rates in Black women. An artificial intelligence enhanced electrocardiogram (AI-ECG) has previously been shown to be effective in detecting AF while in sinus rhythm, and for AF risk prediction in a population-based study. Purpose Our objective was to evaluate the use of an AI-ECG for AF risk stratification among women of reproductive age (18 to 49 years) with PPCM compared to other forms of cardiomyopathy. Methods We identified 59 reproductive age women with a diagnosis of PPCM between January 2007, and October 2018 and included matched controls in a 3:1 fashion. Matching was performed based on sex, age, race, and left ventricular ejection fraction. We excluded patients with a diagnosis of AF prior to cardiomyopathy diagnosis date. AI-ECG prediction probabilities were generated for ECGs performed within a 30-day window prior to the patient's first cardiomyopathy diagnosis date for the entire study cohort. Results A total of 236 patients were included in the final analysis (59 cases, 177 controls). Overall, the median age at cardiomyopathy diagnosis was 31.7 years (IQR: 18.5, 49.4), 76.3% were White, 8.5% were Black, and 15.3% represented other or unknown race. Over the period studied, 3.4% of women with PPCM developed AF compared to 5.6% of women with other cardiomyopathies. The frequency of positive AI-ECG predictions for AF was more common among women with other cardiomyopathies (40.7%) compared to women with PPCM (20.3%). The predicted odds ratio for AF development following a cardiomyopathy diagnosis based on AI-ECG results was 0.37 (95% CI: 0.18, 0.73) for PPCM compared to other cardiomyopathies (p=0.006). Conclusion We demonstrated that an AI-ECG model for AF prediction may play a potential role in arrhythmia risk stratification/prediction among young women with PPCM who have a demonstrable lower risk for AF compared to women with other cardiomyopathies. Mechanisms for lower AF risk among patients with PPCM remain unknown. Further studies evaluating mechanistic pathways will be essential. Funding Acknowledgement Type of funding sources: None.

Abstract Background Screening for left ventricular (LV) systolic dysfunction (defined as ejection fraction ≤35%) based on data from a standard 12-lead electrocardiogram (ECG) has become well established when standard digital ECGs are available–8 independent leads sampled at least 250 hertz for 10 seconds. As the algorithm has been incorporated into various clinical scenarios and ancillary research projects, a limitation of the binary classification at 35% has become apparent. Purpose The objective of this study was to develop and validate a deep learning-based algorithm that would classify LVEF into three categories based on only the digital ECG input. Methods After IRB approval, native digital resting ECGs acquired between 1/1/2010 and 12/31/2021 on patients seen in Mayo Clinic in Jacksonville were extracted from the institutional electronic ECG database management system (MUSE, GE Healthcare). These ECGs were matched with transthoracic echocardiograms obtained up to four days prior or 30 days after the ECGs acquisition. A convolutional neural network consisting of 8 layers of convolutions, batch normalization and pooling was trained using Keras and Tensorflow with hyper-parameter optimization for L1 and L2 regularization, learning rate adjustments, and class weights to predict three classes of LVEF: ≤35%, 36–51%, and ≥52% based on clinical relevance. The primary measure of overall performance was the detection of LVEF ≤35%; however, the triad of model predictions was also considered in translating the model output to human interpretable findings. Results A total of 30,153 patients (60,169 ECG pairings; mean age 63 years; 48% male) were randomly split at the patient level into training (24,172 patients), validation (2,973 patients) and testing (3,008 patients). The trained model provided robust discrimination in the withheld testing data – AUROC of 0.941 (95% CI: 0.931 to 0.950). Using the optimal model threshold based on Youden's index from the validation data (0.186), sensitivity and specificity were estimated to be 87.9% (95% CI: 83.8% to 91.2%) and 86.3% (95% CI: 85.4% to 87.2%) in the testing data. In instances where discordant predictions were observed, the posterior distribution of model probabilities provide additional insights into the possible underlying value of LVEF (Figure 1). Conclusions The utilization of a multi-category deep learning classification model for the detection of reduced ejection fraction adds new dimensions to the use of AI technologies on digital ECGs. This work shows high discrimination can still be obtained when using three classes of LVEF. Funding Acknowledgement Type of funding sources: None.

Abstract Background Current approaches utilizing non-invasive methods to screen for cardiac allograft rejection (gene expression profiling and cell free DNA) have yet to be broadly integrated into heart transplant management and have shown limited discrimination (AUCs of 0.68 and 0.77, respectively). Changes in the electrocardiogram (ECG) have been reported at the time of severe cardiac rejection, including low voltages and conduction abnormalities. It remains unknown if subtle ECG changes correlating with cardiac allograft rejection can be detected earlier using machine learning methods. Purpose We sought to develop an artificial intelligence (AI) model to detect cardiac allograft rejection based on the 12 lead ECG. Methods We identified all patients who underwent a heart transplant at 3 hospital sites within a single health system from Jan 1998 through Apr 2021 and extracted digital 12-lead ECG data as well as endomyocardial biopsy pathology results from the electronic medical record. We partitioned our data into a training (80%), validation (10%), and test set (10%) based on a group stratification sampling. Each patient was present in only one set and each set had a positivity rate of 2.6% with 6,074/758/758 ECGs belonging to 1,146/140/141 unique patients in each set respectively. Cardiac allograft rejection was defined as moderate or severe acute cellular rejection based on International Society for Heart and Lung Transplantation (ISHLT) guidelines. A convolutional neural network, using the 12-lead ECG data as input, was trained with hyperparameter optimization for regularization, learning rate adjustments, and class weights. Model performance metrics were based on the test data and estimated using the final model architecture. Results 1,587 heart transplant recipients who had at least one endomyocardial biopsy were evaluated for inclusion. We limited our sample to ECGs performed within 30 days of the biopsy date (7,590 ECGs, representing 1,425 unique patients). Our study population had a median age of 55.8 years and 28.7% were female. The median number of ECG-biopsy pairs per patient was 5. The majority of endomyocardial biopsy results were classified as none or mild rejection (97.1%), and 2.9% had moderate/severe rejection. The ECG-based AI model detected cardiac allograft rejection with an area under the receiver operative curve (AUC) of 0.84 in the test set. The sensitivity, specificity, positive and negative predictive values were 95%, 52.6%. 5.2% and 99.7% respectively. Conclusions An AI-ECG model appears to outperform novel non-invasive laboratory tests (gene expression profiling and cell free DNA) for detecting cardiac allograft rejection and does not require a blood draw or the additional complexities surrounding sample processing. This model relies on a readily available and relatively inexpensive test, the ECG. In addition, AI predictions can be made available within a few minutes following ECG acquisition. Funding Acknowledgement Type of funding sources: Private hospital(s). Main funding source(s): Mayo Clinic

As survival in breast cancer patients from newer therapies increases, concerns for chemotherapy-induced cardiotoxicity (CIC) have offset some of these benefits, manifesting as a decline in left ventricular ejection fraction (LVEF). Patients receiving anthracycline-based chemotherapy followed by trastuzumab are at risk for CIC. Previous research evaluating whether clinical biomarkers predict cardiotoxicity has been inconsistent. Recently, angiotensin II type 1 receptor (ATR1) and endothelin 1 (ET1) have been shown to play a role in breast tumor growth. We evaluated ATR1 and ET1 expression in breast cancer tissue and its association with CIC. A total of 33 paraffin-embedded breast tissue specimens from women with breast cancer treated with anthracycline-based chemotherapy and trastuzumab were analyzed by immunohistochemistry (IHC) and qRT-PCR. We found that ET1 expression was increased in patients with an LVEF ≤ 50% (p = 0.032) with a lower LVEF correlating with higher ET1 expression (r = 0.377, p = 0.031). In patients with a change in LVEF of greater than 10%, greater ET1 expression was noted compared to those without a change in LVEF (p = 0.017). Increased ET1 expression in breast tumor tissue is associated with reduced LVEF. Future studies need to examine whether ET1 may be a tissue biomarker that helps predict the risk of developing CIC in women with breast cancer.

Abstract Background Cardiovascular disease (CVD) has been identified as a major threat to maternal health in the US and UK with cardiomyopathy being one of the most common acquired CVD in the pregnant and postpartum period. Diagnosing cardiomyopathy in pregnancy is challenging due to an overlap of cardiovascular symptoms with normal pregnancy symptoms. Purpose The purpose of this study was to evaluate the effectiveness of an ECG based deep learning model in identifying cardiomyopathy among pregnant and postpartum women. Methods We utilized an ECG based deep learning model to detect cardiomyopathy in a cohort of pregnant or postpartum women seen at multiple hospital sites. Model performance was evaluated using area under the curve (AUC), accuracy, sensitivity, and specificity. We compared the diagnostic probabilities of the deep learning model with natriuretic peptides and a multivariable model consisting of demographic and clinical parameters. Results 1,807 women were included. 7%, 10% and 13% had LVEF ≤35%, <45% and <50% respectively. The ECG based deep learning model identified cardiomyopathy with an AUC of 0.92 for left ventricular ejection fraction (LVEF) ≤35%, 0.89 for LVEF <45% and 0.87 for LVEF <50%. For LVEF ≤35%, AUC was higher in Black (0.95) and Hispanic (0.98) women compared to white (0.91). Natriuretic peptides and the multivariable model had AUCs of 0.85 and 0.72 respectively. Conclusions A deep learning model effectively identifies cardiomyopathy in pregnant or postpartum women, outperforms natriuretic peptides and traditional clinical parameters with the potential to become a powerful initial screening tool for cardiomyopathy in the obstetric care setting. Funding Acknowledgement Type of funding sources: Other. Main funding source(s): This study was made possible using resources supported by the Mayo Clinic Women's Health Research Center and the Mayo Clinic Building Interdisciplinary Research Careers in Women's Health (BIRCWH) Program funded by the National Institutes of Health (NIH), grant number K12 HD065987. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.