López-Jiménez, Francisco

Abstract Background Moderate-severe mitral annulus calcification (MAC) and left ventricular (LV) diastolic dysfunction are both linked to increased risks of heart failure and mortality. Assessing LV diastolic function (LVDF) in moderate-severe-MAC by echocardiography is frequently challenging and unreliable. We recently validated an AI-ECG model for LVDF. We aimed to determine if this tool can predict long-term mortality in patients with moderate-severe-MAC and whether it can help establish MAC-specific echocardiographic LVDF parameter cutoffs. Methods This retrospective study included all patients from the AI-ECG LVDF study test cohort who underwent comprehensive transthoracic echocardiography confirming moderate-severe-MAC and ECG within fourteen days of each other at Mayo Clinic between 09/2001 and 06/2022. AI-ECG LVDF was determined based on the index ECG. Patients were categorized based on AI-ECG into groups representing normal/grade-1 LVDF, grade-2 LVDF, or grade-3 LVDG (DF-1, DF-2, and DF-3, respectively). Results Of 2,506 patients with moderate-severe-MAC [mean age 77, 64.1% females], 709 (28.3%), 1,510 (60.3%), and 287 (11.5%) were classified as DF-1, DF-2, and DF-3, respectively. Worse LVDF grade was associated with older age, male sex, lower LV ejection fraction, and higher rates of noncardiovascular and cardiovascular comorbidities. Over a median follow-up of 2.8 years, 1,059 (42.3%) patients died. In multivariable survival analysis, DF-2 [adjusted hazard ratio (aHR) 1.43, 95% confidence interval (95%CI) 1.21-1.68] and DF-3 (aHR 1.54, 95%CI 1.22-1.93) were independently associated with a higher risk of death (figure). Findings were also consistent after adjusting to mitral inflow gradients. When comparing LVDF echocardiographic parameters in controls without MAC matched by age, sex, left ventricular ejection fraction, and AI-ECG DF grade, significant differences were observed. Patients with moderate to severe MAC generally exhibited lower e' velocities, higher TR velocities, elevated E/e' ratios, and increased LAVi. Interestingly, the E/A ratio between MAC patients and their matched controls across AI-ECG LVDF grades remained consistent. Conclusion AI-ECG LVDF predicts mortality risk in patients with moderate-severe-MAC, regardless of mitral inflow gradients. Utilizing AI-ECG LVDF might aid in the risk stratification of patients with significant MAC.

Abstract Background AI-enhanced electrocardiogram (AI-ECG) is a cost-effective tool for left ventricular dysfunction screening. However, its cost-effectiveness for other forms of structural heart disease (SHD) is unknown. While AI-ECG is inexpensive, the relatively low positive predictive value (PPV) of many models can lead to high costs from unnecessary follow-up tests like transthoracic echocardiography (TTE). Therefore, strategies to improve the yield of AI-ECG-based screening are needed. Purpose To evaluate the cost savings of a stepwise approach to SHD screening using AI-ECG followed by point-of-care ultrasound (POCUS) compared to AI-ECG alone. Methods 286 adult outpatients undergoing AI-ECG were randomly selected to undergo same-day POCUS. All participants were required to have a recent TTE, our gold standard for SHD diagnosis, for comparison purposes. We evaluated four SHDs: aortic stenosis, cardiac amyloidosis, hypertrophic cardiomyopathy, and left ventricular dysfunction (defined as left ventricular ejection fraction <50%). The costs of AI-ECG ($75/test) and TTE ($1,305/test) were sourced from Healthcare Bluebook, while the cost of POCUS ($100/test) was independently estimated. We compared the costs of screening with AI-ECG alone to stepwise screening with AI-ECG followed by POCUS. Costs were analyzed for two paradigms: screening for all forms of SHD in parallel and screening for each SHD individually. Results AI-ECG alone identified potential SHD in 125 out of 286 patients. When compared to TTE, only 39 out of 125 were true positives (31% PPV). Among the AI-ECG-positive patients, POCUS demonstrated findings of SHD in 52 out of 125 cases. When compared to TTE, this stepwise approach yielded 32 true positives and 20 false positives (62% PPV). In the parallel screening paradigm, the cost per patient diagnosed with SHD was $4,733 with AI-ECG alone, which decreased to $3,182 with stepwise screening (33% cost savings). For the individual SHD screening paradigm, cost reductions were as follows: aortic stenosis ($18,724 to $6,315, 66% savings), cardiac amyloidosis ($21,023 to $12,230, 42% savings), hypertrophic cardiomyopathy ($9,883 to $6,175, 38% savings), and left ventricular dysfunction ($4,019 to $3,582, 11% savings). Conclusions Stepwise SDH screening with AI-ECG followed by POCUS markedly reduces costs compared to AI-ECG alone. The greatest cost savings for individual SHD screening were observed for aortic stenosis and cardiac amyloidosis, conditions where AI-ECG models have lower PPVs. Parallel screening for multiple SHDs is also less costly than screening for any individual SHD alone. Therefore, a parallel screening approach may maximize SHD detection while minimizing screening costs. Stepwise Screening Approach for SHD SDH Screening Costs

AbstractBackgroundThere is increasing need for noninvasive biomarkers of Alzheimer’s Disease (AD) neuropathologic change for early detection and intervention through risk‐factor modification and disease‐modifying therapies. One such biomarker is the prediction of chronological age from routine clinical tests such as an electrocardiogram (EKG) to discriminate between observed biological age from chronological age for healthy aging. Deviation of true age from predicted age has been associated with heart failure, hypertension, and coronary heart disease. However, the association between EKG‐based age and underlying neuropathology has not been studied.MethodWe used a cohort of 336 participants from the Mayo Clinic Study of Aging, a population‐based prospective study of residents in Olmsted County, Minnesota, with antemortem EKG and post‐mortem brain autopsy (Table 1 & 2). Age was predicted from EKG using a validated convolutional neural network algorithm. We used ordinal logistic regression to identify associations between lower neuropathologic scores of AD (Thal phase, neuritic plaque score, Braak stage, and overall AD neuropathologic change) and the Age‐Gap, defined as EKG‐predicted age minus chronological age. Results were controlled for true age, sex, and APOE e4 carrier status. Neuropathologic scales were grouped into none, low, intermediate, and high categories. Observations were weighted by the inverse of years between last EKG and time of death to account for the time difference between acquisition of EKG and autopsy.ResultWe found that an increasing Age‐Gap (i.e. older predicted age) was inversely associated with lower overall AD neuropathologic change (OR 0.93 [0.89, 0.97], p = 0.002). Investigation of the components of AD neuropathologic change showed that an increasing Age‐Gap was also inversely associated with lower amyloid metrics such as Thal phase (OR 0.85 [0.80, 0.89], p < 0.001) and neuritic plaque scores (OR 0.91 [0.87, 0.96], p < 0.001), but not tau metrics like Braak staging (OR 1.05 [1.00, 1.10], p = 0.05).ConclusionIn this work, we show that older predicted age on EKG, a noninvasive biomarker associated with vascular and biological aging, is associated with higher amyloid burden and neuropathologic change of AD.

Background: Risk stratification for suspected pulmonary emboli (PE) remains an area of active research. Current risk stratification algorithms have moderate discriminatory capabilities and are guideline-recommended, but they are poorly utilized in practice, with most clinicians relying predominantly on their clinical training and personal “gestalt.” Our group recently derived an Artificial Intelligence (AI) algorithm using electrocardiograms (ECG) to predict the presence of PE (AUC 0.69) on computed tomography scans of the pulmonary arteries (CTPA) in a cohort of 79K patients. Aim: Validate the AI-ECG algorithm in a contemporary dataset of patients with suspected PE. Methods: Emergency Department (ED) visits between 9/1/2022 and 12/31/2023 across the Mayo Clinic Enterprise were searched using a unified data platform. ED visits in patients over 18 years of age with a discretely documented Wells' PE or PERC Rule, a D-dimer result, or a CTPA were included. Only CTPA imaging and D-dimers performed within 48 hours of ED admission were eligible to be included. Patients were included in the final cohort if an ECG was performed within 6 hours of ED admission, D-dimer testing, or CTPA. Patients were considered negative for PE if CTPA was negative, and when this wasn't performed, a negative D-dimer (high sensitivity) or PERC rule of 0 was considered negative. A previously derived Artificial Neural Network algorithm to analyze the ECG to predict PE was applied to the dataset for validation. Results: A total of 27,028 ED visits in 24,288 unique patients meet inclusion criteria. Among this group, the first ED visit with a CTPA and an ECG within 6 hours was included, leaving 18,936 patients for the final validation cohort. The mean age of the cohort was 54.7 (SD 18.7), and 57% were male. Among the entire cohort, 13,573 underwent D-dimer testing for which 4,895 (36%) were positive at standard thresholds. A total of 10,486 patients underwent CTPA (55% of patients evaluated), of which 455 were positive for PE (4.3%). Using the AI-ECG model from the derivation cohort applied to this validation cohort resulted in an AUC of 0.69, identical to the derivation cohort. Using ECG model estimates alone, low (0.74%, n=2447, NPV 99.3%), elevated (2.27%, n=15404, NPV, 97.72%), and high-risk (8.02%, n=1085, NPV 91.98%) categories for PE can be created. A negative D-dimer in the elevated risk group increased the negative predictive value (NPV) to &gt;99.9% and could be used to reduce imaging in this group further. Using the proposed ECG to D-dimer cascade risk stratification could further reduce the need for PE imaging for an overall NPV of 99.9% and sensitivity of 95.2%. We then used additional clinical parameters to train a new algorithm with additional clinical data and the AI-ECG algorithm estimates. Multiple machine learning algorithms were then evaluated in a dataset divided into training (2/3) and testing (1/3) datasets, after downsampling (to address class imbalance). The best results were obtained with AI-ECG results and D-dimer: Ada Boost (AUC 0.93, F1 score 0.895) compared to standard logistic regression (AUC 0.79, F1 score 0.746). The addition of age, sex, and oxygen saturation did not improve model performance. Conclusion: We have now validated the previously derived AI-ANN ECG algorithm for PE in a contemporary and independent cohort and demonstrate potential for integration into clinical PE assessment algorithms. Unlike prior risk stratification processes, ECGs are invariably performed in patients with cardiopulmonary symptoms and have the potential to prevent missed or delayed PE diagnosis by proactively identifying high-risk patients. Additionally, we demonstrate that the results of this algorithm can be combined with D-dimer values, resulting in a simple, powerful, predictive algorithm (AUC 0.93) that pairs cardiovascular electrophysiology with laboratory biomarkers and could lead to a new paradigm in clinical risk stratification for PE.

Abstract Aims The increasing need for diagnostic echocardiography tests presents challenges in preserving the quality and promptness of reports. While Large Language Models (LLMs) have proven effective in summarizing clinical texts, their application in echo remains underexplored. Methods and results Adult echocardiography studies, conducted at the Mayo Clinic from 1 January 2017 to 31 December 2017, were categorized into two groups: development (all Mayo locations except Arizona) and Arizona validation sets. We adapted open-source LLMs (Llama-2, MedAlpaca, Zephyr, and Flan-T5) using In-Context Learning and Quantized Low-Rank Adaptation fine-tuning (FT) for echo report summarization from ‘Findings’ to ‘Impressions.’ Against cardiologist-generated Impressions, the models’ performance was assessed both quantitatively with automatic metrics and qualitatively by cardiologists. The development dataset included 97 506 reports from 71 717 unique patients, predominantly male (55.4%), with an average age of 64.3 ± 15.8 years. EchoGPT, a fine-tuned Llama-2 model, outperformed other models with win rates ranging from 87% to 99% in various automatic metrics, and produced reports comparable to cardiologists in qualitative review (significantly preferred in conciseness (P &lt; 0.001), with no significant preference in completeness, correctness, and clinical utility). Correlations between automatic and human metrics were fair to modest, with the best being RadGraph F1 scores vs. clinical utility (r = 0.42) and automatic metrics showed insensitivity (0–5% drop) to changes in measurement numbers. Conclusion EchoGPT can generate draft reports for human review and approval, helping to streamline the workflow. However, scalable evaluation approaches dedicated to echo reports remains necessary.

Background: We previously developed deep-learning algorithms that identify coronary artery disease (CAD) risk based on (i) coronary artery calcium (CAC), (ii) obstructive CAD by angiography, and (iii) left ventricular akinesis in ≥1 segment by echocardiogram, using a 12-lead electrocardiogram (CAD ECG-AI). We tested the hypothesis that those with increased probability of CAD by CAD ECG-AI algorithms will have an increased risk of atherosclerotic cardiovascular disease (ASCVD) events and would refine the AHA Predicting Risk of Cardiovascular Disease Events (PREVENT TM ) equations’ predictive capabilities. Methods: We assessed a group of consecutive patients who sought primary care in Olmsted County, MN, between 1997 and 2003. Passive follow-up was conducted using the Rochester Epidemiology Project's record linkage system. Patients included met the same criteria of the original PREVENT TM equations. The probability output of each CAD ECG-AI algorithm was used to predict ASCVD (fatal and non-fatal myocardial infarction and ischemic stroke) and ASCVD-Plus [further including percutaneous coronary intervention (PCI), coronary artery bypass grafting (CABG), and all-cause mortality]. Events were validated in duplicate. Cox proportional hazard models adjusted for age, sex, and risk factors while modeling each CAD ECG-AI algorithm as an individual and additive predictor (see footnote). PREVENT TM risk categories stratified analysis to evaluate the effect of the CAD ECG-AI models on predicted ASCVD risk. Results: We included 21,193 patients, with mean ± SD age 51.6 ± 12.0, 54% women and 95% white. After 14.2±12.4 years follow-up, 1,351 (6.4%) developed ASCVD, and 2,808 (13.2%) had ASCVD-Plus. The risk of ASCVD and ASCVD-Plus increased with positivity for the CAD ECG-AI algorithms after adjustments for age, sex, and PREVENT TM equations factors, all p for trend <0.001 (data not shown). In additive analyses, the risk of ASCVD and ASCVD-Plus increased with increased CAD ECG-AI factors, all p for trend <0.001 ( Fig. A-B ). Furthermore, the CAD ECG-AI algorithm enhanced the predictive capabilities of PREVENT TM equations across most risk subgroups ( Fig. C-E ). Conclusions: The deep-learning CAD ECG-AI algorithms displayed an independent and additive association with long-term ASCVD and ASCVD-Plus events in the community and improved the PREVENT TM predicted ASCVD risk. The CAD ECG-AI algorithms could help identify individuals at risk in primary prevention of cardiovascular events.

Background: Sodium-glucose cotransporter-2 inhibitors (SGLT-2i) and glucagon-like peptide-1 receptor agonists (GLP1-RA) have known cardiovascular benefits. We aimed to assess the efficacy of GLP1-RA or SGLT2i initiation in reducing mortality or cardiovascular events in a community-based cohort of adults with ischemic stroke. Methods: Patients ≥18 years old admitted for acute ischemic stroke between Jan 2000 and June 2022 were identified using the Rochester Epidemiology Project. Subjects were considered exposed to SGLT-2i or GLP1-RA if they were treated with either medication after the index stroke and before the primary (all-cause mortality) or secondary outcomes (MI, recurrent ischemic stroke or composite). Kaplan-Meier curves and Cox models are shown in the Figures . Results: We included 7044 subjects, mean age 72±14 yr, 52% male, 94% white, over a median follow-up of 3 [1-7] years, 416 subjects (6%) had a second stroke, 426 (6%) had an incident MI and 3738 (53%) died. Those on SGLT2i/GLP1-RA had a decreased risk for mortality, incident MI, and composite in the univariate analysis. These associations remained significant after multivariable adjustment and were not affected by minimum time of exposure Figure 1&2 . SGLT2i/GLP1-RA use was associated with recurrent ischemic stroke in the multivariate analysis Figure 2C . All associations remained significant in most sub-analyses of individual medications except for the association between GLP1-RA and recurrent stroke Figure 3 Conclusions: SGLT2i or GLP-1 RA use was associated with decreased mortality and risk of myocardial infarction and recurrent stroke in patients with ischemic stroke.

Background: Artificial intelligence (AI) can be used to predict a person’s age from the electrocardiogram (ECG) (AI-ECG age) and has been proposed as a measurement of biological age. The difference between AI-ECG age and chronological age (defined as delta-age) is an independent predictor of mortality in the general population. This study assessed the relationship between AI ECG estimated age and mortality among adults living with congenital heart disease (CHD). Methods: A previously validated neural network to estimate AI-ECG age was used to analyze standard digital 12-lead ECGs in a cohort of subjects aged >18 years seen in the adult CHD (ACHD) clinic between 1992 to 2023. A single ECG, collected during the year of the first visit to the ACHD clinic was analyzed to compute the delta-age.. A positive delta-age indicates an AI ECG estimated age higher than the subject’s chronological age. The relationship between the delta-age and mortality was evaluated using Cox proportional hazard models adjusting for influential clinical factors including chronologic age, sex, ACHD severity, NYHA class and history of arrhythmia. Results: Of 5,780 subjects tested (50% females), the mean chronological age was 39.1 ± 15.0 years. AI ECG estimated age was 52.3 ± 16.6 years. Complexity of CHD was classified as mild, moderate, and severe in 7.4%, 73.9% and 18.7% of patients respectively. Patients with severe CHD had the highest delta-age of 17.1 (±18.2) years followed by moderate 12.7 (±13.6) y and simple 7.7 (±11.5) y. Patients with single ventricle had the highest deviation in AI estimated age [median delta-age 21.0 (IQR 28.6) y] followed by those with a systemic right ventricle [median delta-age of 16.3 (IQR 29.3) y]. During a median follow-up of 6.4 y (25th – 75th centile 1.58 – 13.7 y), mortality occurred in 839 (14.5%) patients. After adjusting for chronological age, CHD severity, and other clinical variables, delta-age gap was associated with increased mortality risk (HR 1.06 (1.03 – 1.09) per 5-year increment in delta-age, p<0.005), Table 1. Conclusion: The difference between AI ECG and chronological age is an independent predictor of all-cause mortality in ACHD patients.

Background: Our team recently developed an Artificial Intelligence model that enables the identification of increased left ventricular filling pressures using a single-lead electrocardiogram (AI-ECG) with excellent performance (AUC 0.91). However, its ability to predict incident heart failure (HF) in the community is yet to be evaluated. Hypothesis: We tested the hypothesis that the output from the AI-ECG will predict the long-term incidence of heart failure (HF) and HF with preserved ejection fraction (HFpEF), and would refine the HF predictive capabilities of the AHA Predicting Risk of Cardiovascular Disease EVENTs (PREVENT TM ) Equations. Methods: Leveraging the resources of the Rochester Epidemiology Project (REP) we analyzed data from patients who sought primary care in Olmsted County, MN, between 1997-2003 with passive follow-up until April 2024. We included those with a digital ECG at baseline and excluded those who were part of the original algorithm development or history of heart failure. The diagnosis of heart failure was ascertained by using ICD codes (ICD9-428,ICD10-I.50) and a word search for the terms related to heart failure. HFpEF was ascertained using the Ejection Fraction (EF) data closest to the heart failure diagnosis date, using EF more than 40% as the criterion. We tested the association between AI-ECG predicted probabilities and new-onset HF and HFpEF using Cox proportional hazard models. These models were also adjusted for cardiovascular risk factors and were stratified to evaluate the effect of the AI-ECG output on PREVENT TM -predicted HF at 10 years. Results: We included 20,276 subjects, mean age 55±14.03, 52% women, 92% non-Hispanic white. After 18.2±8.5 years of follow-up, 4,008 (19.76%) were diagnosed with HF and 3,812 (18.8%) HFpEF. Risks of HF and HFpEF significantly increased in patients with increased left ventricular filling pressures by AI-ECG, independent of risk factors, all p for trend <0.001, see Fig A-B , respectively. AI-ECG enhanced the prediction capabilities of the PREVENT TM score across all risk groups, see Fig-C. Conclusions: AI-ECG-Derived increased left ventricular filling pressure is associated with new-onset HF and HFpEF in the community and was associated with incidental heart failure at any PREVENT TM risk category. Those results suggest a potential role of AI ECG in predicting heart failure.

Introduction: Calcium scoring computed tomography (CAC) is a widely available and inexpensive modality of screening for atherosclerosis. This study aims to improve the predictive value of CAC by incorporating pericardial fat characteristics in a CatBoost model. Methods: A sample of 7,000 non-contrast clinically indicated CAC scans were randomly selected from a cohort of consecutive Olmsted County MN residents. Known history of atherosclerotic cardiovascular disease or impaired images were excluded. A pericardial fat segmentation model, trained on 191 images of the same dataset with a dice similarity score of 94%, was applied to 6,194 CT scans; Volume and mean Hounsfield unit (mHU) of the output masks were added to age at imaging, radiologist-measured calcium score, male gender, diabetes, and smoking (Figure 1). A 70,20 and 10% split was used to produce the training, validation and test set, respectively. Features were inputted to CatBoost classifier, as the best model tested, to predict four outcomes: acute coronary syndrome (ACS), heart failure (HF), ischemic stroke, and all-cause mortality. Model was evaluated by area under the ROC curve (AUC) on the unseen test set. Results: Mean age was 55.2 (8.8), 55.2 (8.8) and 54.8 (8.5) years in the training, validation and test sets, respectively, and 65%, 67% and 66% of the populations were male. The mean AUC, obtained through bootstrapping with 100 iterations, was 0.81 [0.77-0.85], 0.86 [0.60-0.92], 0.80 [0.73-0.87] and 0.87 [0.81-0.91] for HF, ACS, stroke and death, respectively; The same analysis excluding pericardial fat measurements showed a 3% reduction in the AUC for all of the investigated outcomes (Figure 2). Conclusion: Our study presents a robust risk prediction model for ACS, HF, stroke, and mortality based on CAC scans. The improved performance of the model by incorporating pericardial fat characteristics, highlights the potentials of deep learning in timely inexpensive risk prediction in large populations.