Risultati per: Le campagne di screening come strumento di prevenzione oncologica tra dubbi e certezze

Circulation, Volume 150, Issue Suppl_1, Page A4112869-A4112869, November 12, 2024. Introduction:Sitosterolemia, a hereditary disorder marked by elevated plant sterol levels, presents diagnostic challenges due to its similarity to other lipid disorders. The development of the Sitosterolemia Risk Prediction Scale (SRPS) aims to address this by synthesising genetic, clinical, and dietary data into a coherent risk assessment model.Research Question:We propose that a structured risk scale, integrating diverse factors known to affect sitosterolemia, can significantly improve the accuracy of predicting the disorder. The SRPS is hypothesised to facilitate early detection and inform targeted interventions.Aim:The primary aim is to conceptualise and outline the SRPS, which categorises individuals into risk categories based on a point system reflecting genetic predispositions, clinical symptoms, dietary habits, and response to treatments. This scale seeks to enhance the clinical identification of sitosterolemia, promoting timely and personalised management strategies.Methods:A detailed table was generated to present the SRPS, categorising risk factors into genetic, clinical, dietary, and response to treatment. This innovative method allowed for the efficient synthesis and visualisation of complex data.Results:The SRPS table methodically organizes risk factors into low (0-2 points), moderate (3-5 points), and high (6+ points) categories. This stratification guides further diagnostic actions, ranging from exploring alternative causes of hyperlipidemia to necessitating comprehensive genetic and lipid analyses.Conclusion:The SRPS represents an innovative framework for assessing sitosterolemia risk, highlighting the potential benefits of integrating genetic, clinical, and dietary information. It further underscores the importance of a multifactorial approach in the early detection and management of sitosterolemia.

Circulation, Volume 150, Issue Suppl_1, Page A4144973-A4144973, November 12, 2024. Background:Cardiac calcium, which includes coronary and extra-coronary calcification, is often incidentally found in chest CT scans performed for various reasons. Despite its prognostic value, manual quantification of cardiac calcium in non-gated chest CT images is labor-intensive.Goals:This retrospective study aims to perform automatic quantification and scoring of cardiac calcium in non-contrast-enhanced chest CTs. The objective is to determine associations between automatic calcium scoring and outcomes such as all-cause mortality, non-fatal myocardial infarction (MI), and non-fatal stroke.Methods:We conducted a nationwide cohort study using the Taiwan National Health Insurance Research Database (NHIRD) from 2016 to 2022. Patients under 20 years old, with a diagnosis of malignancy, or with outcome events before the CT acquisition were excluded. HeaortaNet 1.0, a validated AI model, was used for cardiac calcium scoring. Comorbidities were determined using ICD diagnostic codes for ≥2 consecutive outpatient visits within the year before the index date. Outcomes were censored at the first occurrence of mortality or relevant ICD codes for MI or stroke.Results:The retrospective cohort included 279,415 patients (56.37% male, mean age 60.31±16.54). All-cause mortality occurred in 12.82% of patients within a 3-year follow-up. The 3-year incidence rates of non-fatal MI and non-fatal stroke were 0.86% and 2.07%, respectively. Multivariate-adjusted Cox hazard ratios (95% confidence intervals) for any composite outcome were 1.51 (1.46-1.57), 2.09 (2.01-2.17), 2.63 (2.53-2.74), and 3.37 (3.24-3.50) for cardiac calcium scores of 1-100, 101-400, 401-1000, and >1000, compared to a score of 0. Adjusted Cox hazard ratios for all-cause mortality were 1.62 (1.56-1.69), 2.29 (2.19-2.39), 2.91 (2.78-3.04), and 3.80 (3.64-3.96) for scores of 1-100, 101-400, 401-1000, and >1000, compared to a score of 0.Conclusion:AI-enabled opportunistic screening of non-contrast chest CT for cardiac calcium scoring is associated with all-cause mortality and cardiovascular events. This is the first large-scale cohort study to use an AI model for comprehensive cardiac calcium screening.

Circulation, Volume 150, Issue Suppl_1, Page A4115235-A4115235, November 12, 2024. Intro:The AHA endorses screening youth athletes to identify risk for sudden cardiac arrest (SCA). Rates of SCA can be predicted by social determinants of health (SDOH) such as education level and proportion of Black residents in ZIP Code. The Child Opportunity Index (COI) quantifies neighborhood factors that influence health and development. The link between COI and youth cardiac screening findings and outcomes remains unclear.Hypothesis:Cardiac screening data will differ significantly by COI.Aims:To identify differences in cardiac screening data in children of varying COI.Methods:The HeartBytes Database, including sports exams, self-reported physical activity (PA), and zip codes from Simon’s Heart screenings was augmented with COI index zip code data. Chi-squared and logistic regression were used to analyze demographics, cardiac risk factors, and screening results.Data:Screening data of 11,431 youth athletes (median age 14.3 (IQR = 3), BMI 20.6 (4.8), 53.7% male, 70.6% White) was analyzed. The majority of children had very high overall COI (Figure 1). Hypertension, hyperlipidemia, Kawasaki disease, and heart infection were similar across COI levels (p > 0.05). Levels of physical activity varied significantly across levels of overall COI, with the highest levels reported in the lowest COI group (50.4% with >10 hours PA/week) (Chi-Squared; p = 0.007). Positive screening rates varied significantly by level of COI (p = 0.013) (Figure 2). The overall level of education, health environment, and socioeconomic COI did not predict positive screening outcomes in logistic regression analysis (all p >0.05).Conclusion:Prevalence of cardiac risk factors did not vary significantly across COI levels, however, positive screening rates were highest in moderate and very low COI levels. Simon’s Heart engaged communities across the COI spectrum; however, a majority of children had high or very high COI. Further efforts are needed to expand access to underserved populations of lower COI.

Circulation, Volume 150, Issue Suppl_1, Page A4144730-A4144730, November 12, 2024. Background:A personalized, non-invasive assessment approach for evaluating the risk of obstructive coronary artery disease (CAD) is crucial for patients with an intermediate or low clinical likelihood of CAD before undergoing invasive coronary angiography (ICA). This method allows clinicians to effectively rule out the presence of obstructive CAD without the need for ICA or to determine if a referral for ICA is warranted. Emerging lipidomics biomarkers may be valuable in this process. However, technological challenges in detecting structurally similar lipids and the requirement for advanced computational tools have so far impeded the clinical application of lipidomics research.Hypothesis:Our study aims to develop an innovative non-invasive diagnostic test utilizing novel lipidomics biomarkers, potentially revolutionizing current risk classification schemes for CAD.Methods:In this post-hoc analysis of the CorLipid trial (NCT04580173), we employed extreme gradient boosting (XGBoost) machine learning to assess the predictive power of a lipidomics panel for obstructive CAD risk. Liquid chromatography-mass spectrometry analyzed lipid profiles from 146 individuals undergoing ICA. SYNTAX Score (SS) was used to define obstructive CAD as SS >0 versus non-obstructive CAD (SS=0).Results:Of the 146 participants (25% female, mean age: 61 ±11 years old), 55% had obstructive CAD (SS >0). Lipidome changes [phosphatidylinositols, (lyso-)phosphatidylethanolamine, (lyso-)phosphatidylcholine, triglycerides, diglycerides, and sphingomyelins] were investigated to identify lipids potentially associated with the phenotype and complexity of CAD. Using this information, 290 quantified serum lipid species were utilized to develop an XGBoost algorithm with 17 serum biomarkers ( consisting of sphingolipids, glycerophospholipids, triacylglycerols, galectin-3, glucose, low-density lipoprotein, and lactate dehydrogenase) with very good discriminative ability [ROC AUC: 0.875 (95%CI: 0.867-0.883)], excellent sensitivity (100%) but moderate specificity (62.1%) for the prediction of obstructive CAD.Conclusions:These findings indicate that a deep-learning-based non-invasive diagnostic test, using lipidomics serum biomarkers, could reliably rule-out obstructive CAD without necessitating ICA. To enhance generalizability, these results should be validated in larger and similar cohorts. Further research, particularly leveraging machine learning, is promising for refining risk stratification.

Circulation, Volume 150, Issue Suppl_1, Page A4121454-A4121454, November 12, 2024. Introduction:Recent research has shown that AI is able to assess biological aging and cardiovascular disease (CVD) risk using chest radiographs. However, the lack of explainability of such deep learning algorithms hinders clinical utility and adoption. This motivates the current study which searches for and tests the use of machine extractable quantitative features in chest radiographs to predict CVD risk in population screening.Method:Chest radiograph measurements characterizing cardiomediastinal geometry, aortic calcification and tortuosity were handpicked for development of a segmentation-based feature extraction algorithm. The algorithm was applied on the PLCO lung screening dataset for analysis. The association between measurement-based imaging features, clinical characteristics (age, sex, BMI, smoking status, hypertension, diabetes, liver disease) with CVD mortality and 10-year major adverse cardiovascular events (MACE) were analysed by using proportional hazard regression, with feature selection done by LASSO.Result:Of 29,453 eligible subjects, 5693 subjects from a single study centre were used for fitting of all models. The median follow-up time was 19 years. A total of 32 imaging features were extracted and analysed. For both 10-year MACE and CVD mortality, model using imaging features, age, and sex performed similarly to model using conventional risk factors, and a deep learning chest radiograph CVD risk model. Two imaging features, mediastinal width at valve-level [HR 1.36 (1.23-1.50)] and maximal lateral displacement of descending aorta [HR 1.29 (1.18-1.42)] were found to be prognostic. To the best of our knowledge, these features have not been reported previously.Conclusion:Quantitative imaging features can predict CVD risk in chest radiograph similar to deep learning models while providing feature interpretability and explainability. Two novel imaging features prognostic of CVD risk were found and shown to be complementary to conventional risk factors.

Circulation, Volume 150, Issue Suppl_1, Page A4139026-A4139026, November 12, 2024. Background:Familial hypercholesteremia (FH) leads to elevated low-density lipoprotein cholesterol (LDL-C) and atherosclerotic cardiovascular disease (ASCVD). Although treatable, FH is underdiagnosed. Lipid lowering therapy may mask diagnostic pretreatment LDL-C levels. Participants of ASCVD trials may be enriched for FH, so ASCVD trial enrollment may be a unique contact point to opportunistically diagnose FH.Hypothesis:The population of the ACCELERATE trial of evacetrapib and ASCVD outcomes is enriched for FH.Methods:ACCELERATE is a phase 3 cardiovascular outcomes trial which randomized 12,092 patients with high-risk vascular disease to receive evacetrapib or placebo. FH was not reported. Using participant-level data, we estimated pretreatment LDL-c using validated corrections based on type and dose of statin therapy. We defined severe hypercholesterolemia as pretreatment LDL-C ≥ 190 mg/dl and FH as severe hypercholesterolemia with total cholesterol > 290 mg/dL in a first or second degree relative, consistent with Simon Broome register criteria. We compared trial prevalence to general prevalence (severe hypercholesterolemia ~7%, FH ~0.4%). We evaluated the adjusted association of severe hypercholesterolemia with the primary trial endpoint of ASCVD events using multivariable Cox proportional hazards regression.Results:Data were available for 11,993 participants (99%). The prevalence of severe hypercholesteremia was 15% (1809/11993). The prevalence of FH was 2.1% (255/11993). Pretreatment LDL-C ≥ 190 mg/dL, as compared with pretreatment LDL-C < 190 mg/dL, was significantly associated with a higher incidence of the primary ASCVD trial endpoint (15% vs 13.5% respectively, adjusted hazard ratio 1.19; 95% CI 1.03-1.38, P=0.021;Figure).Conclusion:In a participant-level analysis of a rigorous, independently adjudicated ASCVD outcomes trial, severe hypercholesterolemia and FH were more prevalent in the trial population than the general population based on pretreatment LDL-C calculation. Severe hypercholesterolemia was significantly associated with higher ASCVD incidence. ASCVD trial enrollment may be a novel high-yield contact point for index FH case identification using simple pretreatment LDL-C calculation.

Circulation, Volume 150, Issue Suppl_1, Page A4138955-A4138955, November 12, 2024. Introduction:Rheumatic heart disease (RHD) is the most common acquired heart disorder in children and adolescents worldwide. We developed and validated an automated artificial intelligence (AI)-guided RHD screening algorithm adapted for point-of-care ultrasonography (POCUS) in school-aged children.Methods:We employed a cross-domain transfer learning approach, in which a 3D convolutional neural network (CNN) was first trained to detect structural RHD deformation of the mitral or aortic valves in 244,523 videos, representing all views from 5,614 adult transthoracic echocardiograms (1:5 age and sex-matched cases and controls; median age 69 [58-80] years, 76.4% female) in a large US health system. The model was fine-tuned for stage ≥B (“definite”) RHD in 21,472 POCUS videos (2D parasternal and apical acquisitions) from 5,525 studies (75% training, 25% validation) in a pediatric screening program (median age 11 [IQR 10-13] years, 54.6% female) in Brazilian low-income schools. Testing was performed in a held-out set of 1,966 parasternal long-axis (PLAX) videos from 1,138 studies in Brazil (14 [1.2%] with stage ≥B RHD) as well as in an external pediatric screening set in Uganda consisting of 249 videos from 96 studies (34 [35.4%] with stage ≥B RHD) (Fig. 1).Results:Our model (Fig. 2) achieved a study-level AUROC (area under the receiver operating characteristic curve) of 0.88 across the held-out/external testing sets for identifying stage ≥B RHD from cardiac POCUS (Fig. 3A). On a video-level the model learned a continuous spectrum of phenotypes on PLAX acquisitions spanning stage ≥B (“definite”) and stage A (“borderline”) cases, ranging from a median video-level AI probability of 0.13 [0.01-0.73] for stage ≥B to 0.00 [0.00-0.01] for non-RHD POCUS (Fig. 3B). At the threshold that maximized Youden’s J in the held-out Brazil set, our algorithm’s performance in the set from Uganda showed 97% recall (sensitivity), a positive predictive value (precision) of 46%, and a negative predictive value of 95%.Conclusions:A transfer learning approach that employs multi-view learning achieves excellent performance for RHD on single-view two-dimensional cardiac POCUS without Doppler. Our study suggests a scalable approach to AI-enabled RHD detection with images that can be acquired by individuals with modest training.

Circulation, Volume 150, Issue Suppl_1, Page A4137986-A4137986, November 12, 2024. Background:Clinical trial screening is labor-intensive, time-consuming, and error prone. We have developed RECTIFIER, an AI-based clinical trial screening tool, to enhance the efficiency and accuracy of patient recruitment. This study aims to evaluate RECTIFIER’s effectiveness compared to manual screening in a randomized implementation study.Methods:This study was designed as an implementation study as part of an active heart failure trial named COPILOT-HF (NCT05734690). Potential eligible patients were identified via a structured electronic medical record query and randomized to be screened for clinical trial eligibility either by RECTIFIER or manually by clinical staff. The outcome measures included the number of patients contacted, and the number of patients reached for clinical trial enrollment. Data was collected over a period of 3 months.Results:A total of 3834 patients were included in the study, with 1919 patients randomized to the RECTIFIER group and 1915 patients to the manual screening group (Figure). Study staff could manually screen only 1367 patients at the end of the 3-month period. RECTIFIER identified more eligible patients compared to manual screening (833[43.4%] vs. 284[14.8%], p

Circulation, Volume 150, Issue Suppl_1, Page A4145524-A4145524, November 12, 2024. Introduction:White-coat hypertension (WCH) and masked hypertension (MH) complicate accurate blood pressure (BP) monitoring. While ambulatory BP monitoring (ABPM) is effective, its high cost and limited availability are significant barriers.Hypothesis:We hypothesized that a machine learning (ML) model using clinical data from a single outpatient visit could accurately predict WCH and MH.Aims:This study aimed to develop and validate ML-based prediction models for WCH and MH using accessible clinical data to improve diagnostic efficiency and accessibility.Methods:We enrolled patients from two hypertension cohorts, after excluding those with incomplete data. Patients were classified by office BP and ABPM readings per American Heart Association guidelines. ML models, including Multi-layer Perceptron （MLP）, Support Vector Machine (SVM), and Tabular Prior-Data Fitted Network (Tab-PFN), were developed. Input parameters included demographic data (age, gender, height, weight, smoker), and office BP (OBP) and heart rate measurements. Principal Component Analysis (PCA), kernel PCA (kPCA), or t-distributed stochastic neighbor embedding (t-SNE) were used to improve class separability.Results:The study population comprised 1481 participants with a mean age of 47.6 years (SD 13.6), 65% of whom were male and 20.1% were smokers. OBP measurements showed a mean systolic BP (SBP) of 128.7 mmHg (SD 15.4) and a mean diastolic BP (DBP) of 84.2 mmHg (SD 11.6). ABPM showed a mean 24-hour systolic BP of 122.5 mmHg (SD 11.8) and diastolic BP of 79.3 mmHg (SD 10.1). The inclusion of demographic and OBP data, along with advanced resampling and dimensionality reduction techniques, significantly improved the model’s predictive ability. The final TabPFN model achieved the best performance with recall, precision, F1 score, and accuracy of 0.747, 0.931, 0.829, and 0.807 for WCH, and 0.713, 0.954, 0.816, and 0.907 for MH.Conclusion:Our ML-based model effectively predicts WCH and MH using accessible clinical data, offering a cost-effective alternative before applying ABPM.

Circulation, Volume 150, Issue Suppl_1, Page A4144083-A4144083, November 12, 2024. Background:The AI-CVD initiative aims to extract all useful opportunistic screening information from coronary artery calcium (CAC) scans and combines them with traditional risk factors to create a stronger predictor of cardiovascular diseases (CVD). These measurements include cardiac chambers volumes (left atrium (LA), left ventricle (LV), right atrium (RA), right ventricle (RV), and left ventricular mass (LVM)), aortic wall and valvular calcification, aorta and pulmonary artery volumes, torso visceral fat, emphysema score, thoracic bone mineral density, and fatty liver score. We have previously reported that the automated cardiac chambers volumetry component of AI-CVD predicts incident atrial fibrillation (AF), heart failure (HF), and stroke in the Multi-Ethnic Study of Atherosclerosis (MESA). In this report, we examine the contribution of other AI-CVD components for all coronary heart disease (CHD), AF, HF, stroke plus transient ischemic attack (TIA), all-CVD, and all-cause mortality.Methods:We applied AI-CVD to CAC scans of 5830 individuals (52.2% women, age 61.7±10.2 years) without known CVD that were previously obtained for CAC scoring at MESA baseline examination. We used 10-year outcomes data and assessed hazard ratios for AI-CVD components plus CAC score and known CVD risk factors (age, sex, diabetes, smoking, LDL-C, HDL-C, systolic and diastolic blood pressure, hypertension medication). AI-CVD predictors were modeled per standard deviation (SD) increase using Cox proportional hazards regression.Results:Over 10 years of follow-up, 1058 CVD (550 AF, 198 HF, 163 stroke, 389 CHD) and 628 all-cause mortality events accrued with some cases having multiple events. Among AI-CVD components, CAC score and chamber volumes were the strongest predictors of different outcomes. Expectedly, age was the strongest predictor for all outcomes except HF where LV volume and LV mass were stronger predictors than age. Figure 1 shows contribution of each predictor for various outcomes.Conclusion:AI-enabled opportunistic screening of useful information in CAC scans contributes substantially to CVD and total mortality prediction independently of CAC score and CVD risk factors. Further studies are warranted to evaluate the clinical utility of AI-CVD.

Circulation, Volume 150, Issue Suppl_1, Page A4144283-A4144283, November 12, 2024. Introduction:Pulmonary arterial hypertension (PAH) remains an underrecognized, fatal disease. Limited awareness, non-specific symptoms, and late referral to accredited PH centers all contribute to an overall poor prognosis. The previously validated Virtual Echocardiography Screening Tool (VEST) uses 3 routine transthoracic echocardiogram (TTE) parameters (left atrial size, transmitral E:e’ and systolic interventricular septal flattening) to recognize a high PAH likelihood. A positive VEST score has been shown to have 80% sensitivity and 76% specificity for PAH hemodynamics, while a VEST score of +3 has 92.7% specificity for PAH hemodynamics with a positive predictive value of 88.0%.Aim:We aimed to implement a novel algorithm via our electronic medical record (EMR) as an automated VEST calculator to identify patients with a high likelihood of PAH.Methods:An automated EMR VEST calculator was applied retrospectively to 4,952 patients who underwent TTE with TR velocity >/= 2.9 m/s at an accredited PH center from 12/2021-8/2023. Automated EMR VEST scores were validated by comparison to 60 manually scored echocardiograms. Those with VEST score of +3 (highest risk for PAH) underwent chart review to identify whether they were seen by a PH specialist.Results:There was 100% correlation between the automated EMR VEST scores and the manual results.Of the 4,952 patients, 1,655 had a positive automated EMR VEST score, and 355 had a score of +3, predicting the highest likelihood of PAH and warranting urgent referral to an accredited PH center. Of those patients with a +3 score, 103 (29.0%) were never seen by a PH specialist (Fig 1).Conclusion:VEST is a validated, noninvasive and accessible screening tool for identification of patients with a high likelihood of PAH likely to benefit from early referral to a PH center. We present a novel, accurate, and automated EMR algorithm for determination of the VEST score to prompt urgent referral for PH expert evaluation and timely initiation of complex medical therapies. These findings highlight the potential of future artificial intelligence and machine-learning applications for improved recognition of life-threatening PAH.

Circulation, Volume 150, Issue Suppl_1, Page A4131622-A4131622, November 12, 2024. Introduction:Chronic liver disease affects more than 1.5 billion adults worldwide, but the majority of cases are asymptomatic and undiagnosed. Echocardiography is broadly performed and visualizes the liver; however, this information is not diagnostically leveraged.Hypothesis and Aims:We hypothesized that a deep-learning algorithm can detect chronic liver diseases using subcostal echocardiography images that contains hepatic tissue. To develop and evaluate a deep learning algorithm on subcostal echocardiography videos to enable opportunistic screening for chronic liver disease.Methods:We identified adult patients who received echocardiography and abdominal imaging (either abdominal ultrasound or abdominal magnetic resonance imaging) with ≤30 days between tests. A convolutional neural network pipeline was developed to predict the presence of cirrhosis or steatotic liver disease (SLD) using echocardiogram images. The model performance was evaluated in a held-out test dataset, dataset in which diagnosis was made by magnetic resonance imaging, and external dataset.Results:A total of 2,083,932 echocardiography videos (51,608 studies) from Cedars-Sinai Medical Center (CSMC) were used to develop EchoNet-Liver, an automated pipeline that identifies high quality subcostal images from echocardiogram studies and detects presence of cirrhosis or SLD. In a total of 11,419 quality-controlled subcostal videos from 4,849 patients, a chronic liver disease detection model was able to detect the presence of cirrhosis with an AUC of 0.837 (0.789 – 0.880) and SLD with an AUC of 0.799 (0.758 – 0.837). In a separate test cohort with paired abdominal MRIs, cirrhosis was detected with an AUC of 0.726 (0.659-0.790) compared to MR elastography and SLD was detected with an AUC of 0.704 (0.689-0.718). In the external test cohort of 66 patients (n = 130 videos), the model detected cirrhosis with an AUC of 0.830 (0.738 – 0.909) and SLD with an AUC of 0.768 (0.652 – 0.875).Conclusions:Deep learning assessment of clinically indicated echocardiography enables opportunistic screening of SLD and cirrhosis. Application of this algorithm may identify patients who may benefit from further diagnostic testing and treatment for hepatic disease.

Circulation, Volume 150, Issue Suppl_1, Page A4147292-A4147292, November 12, 2024. Background:Tissue hypoxia and chronic anemia associated with sickle cell disease (SCD) leads to structural and physiological alterations in the heart. Early detection of heart failure (HF) in patients with SCD can assist with timely interventions, but current methods (e.g., echocardiogram and heart MRI) are not easily accessible in resource-deprived settings. The integration of artificial intelligence (AI)-powered tools utilizing low-cost ECG data to increase the power to detect more patients eligible for early treatment, thus improving patient outcomes, and needs to be validated.Hypothesis:We hypothesize that ECG-AI models developed to detect incident HF in the general population can detect HF in SCD patients.Methods/Approach:We previously developed an ECG-AI model employing convolutional neural networks to classify patients with HF using a large ECG-repository at Wake Forest Baptist Health (WFBH). This model was developed using 1,078,198 digital ECGs from 165,243 patients, 73% White, 19% Black, and 52% female individuals, with a mean age (SD) of 58 (15) years. The hold-out AUC of this previous model in distinguishing ECGs of HF patients from controls was 0.87. In this study, we externally validated this ECG-AI model using SCD patients’ data from the University of Tennessee Health Science Center (UTHSC). Additionally, a logistic regression (LR) model was constructed in the UTHSC cohort by incorporating other simple demographic variables with the outcome of ECG-AI model.Results/Data:The UTHSC external validation cohort included data from 2,107 SCD patients (188 HF and 1,919 SCD patients with no HF), 98% were Black, 72% were female, with a mean age of 39 (14) years. Despite demographic differences between the validation (more Blacks) and derivation cohorts (lower age), our ECG-AI model accurately identified HF with an AUC of 0.80 (0.77-0.82) in the UTHSC SCD cohort. When incorporating ECG-AI outcome (an ECG-based risk value between 0 and 1), age, sex, and race in a LR model, the AUC significantly improved (DeLong Test, p

Circulation, Volume 150, Issue Suppl_1, Page A4141112-A4141112, November 12, 2024. Background:It is estimated that Peripheral Artery Disease (PAD) affects between 8.5 and 12 million Americans and its prevalence among adults over 40 years of age is increasing. PAD disproportionately affects Black Americans who, at any age, are twice as likely to experience PAD as their white counterparts but are less likely to be screened and benefit from early diagnosis and treatment.Research Questions/Hypothesis:Despite the high prevalence of PAD and the importance of early intervention, screening for PAD remains limited and/or underutilized particularly in primary care settings where most cases of PAD can be identified. This study sought to understand provider knowledge of PAD, associated risk factors, treatment, understanding of disparities in PAD and barriers and facilitators of PAD screening. It was hypothesized that limited resources, lack of awareness on the part of providers and patients, limitations of training in vascular medicine, and other issues are contributing to PAD morbidity and mortality, particularly among Black and Hispanic populations.Methods:Because no current PAD survey was found in the literature, a survey for providers to determine their knowledge, attitude, and beliefs about PAD and the importance and process of PAD screening for patients at risk was developed. The survey was administered to CommonSpirit Health providers in Sacramento, CA between December 2023- January 2024. Specialties engaged in the survey (N=145) included primary care, endocrine, nephrology, cardiology and podiatry providers.Results:Response rate was 21%. Of those responding, primary care was the specialty most represented(69%). A total of 65% of respondents identified medical treatment of risk factors as the primary way to treat PAD, 32% rated their knowledge of risk reduction therapies in PAD as below average, and 88% of respondents were either somewhat or not familiar with racial disparities in PAD. 24% of respondents identified the ‘lack of knowledge of PAD management guidelines’ as the most important barrier to their patients with PAD not receiving risk reduction therapies.Conclusions:Initial survey of providers identifies lack of knowledge as a key indicator of PAD screening practices, including knowledge on racial disparities in PAD. These identified gaps can inform targeted interventions to improve screening, early detection and treatment of PAD.

Circulation, Volume 150, Issue Suppl_1, Page A4124675-A4124675, November 12, 2024. Introduction:The prevalence of hypertrophic cardiomyopathy (HCM) in the UK Biobank based on ICD-10 codes (.07%) is lower than global estimates of disease prevalence (0.2 – 0.5%). Prior studies using this data have remarked on the limitations of findings given likely underdiagnosis. The availability of cardiac MRI scans on a fraction of the participants offers an opportunity to identify missed diagnoses.Aims:This study seeks to utilize a generalizable deep learning model to detect likely cases of undiagnosed hypertrophic cardiomyopathy from cardiac MRIs in the UK Biobank.Methods:The foundational model was trained on a multi-institutional dataset of 14,073 cardiac MRIs via a self-supervised contrastive learning approach that sought to minimize the divergence between scans and their associated radiology reports. The pre-trained model was fine-tuned to diagnose hypertrophic cardiomyopathy on a distinct cohort of 4,870 MRIs with 368 cases of HCM, achieving an AUC of 0.94. The fine-tuned model was applied to the UK Biobank cardiac MRI dataset to ascertain predicted probabilities of HCM. Cases exceeding a threshold of 95% – correlating to the top 0.5% of cases (expected specificity of 97% and sensitivity of 60%) – were screened in for manual reading. In a blinded fashion, a board-certified radiologist was tasked with diagnosing HCM on a sample of cases composed of high and low predicted probabilities.Results:Of the 43,017 patients with cardiac MRIs, only 9 (.02%) had an ICD diagnosis of HCM. 266 cardiac MRIs were manually reviewed: 216 had greater than 95% predicted probability of HCM; 50 negative controls were randomly selected amongst cases with predicted probability less than 10%. The radiologist concurred with an HCM diagnosis for 115 cases (sensitivity 53%, specificity 98%), 112 of which were previously undiagnosed. The prevalence of hypertension and aortic stenosis did not significantly differ between the cohort of true positives (69.2%) and false positives (76.6%). The corrected prevalence of HCM in the UK BioBank MRI cohort is estimated at 0.28%.Conclusions:The findings of this study illustrate the remarkable ability of a generalizable deep learning model to detect undiagnosed cases of a rare disease process from cardiac MRIs. This is an important milestone that may allow for widespread screening of hypertrophic cardiomyopathy while minimizing demand for radiologist labor, and thereby allow patients to reap the substantial benefits of earlier treatment.

Circulation, Volume 150, Issue Suppl_1, Page A4140494-A4140494, November 12, 2024. Background:Primary care after pregnancy is recommended, especially for individuals with recent adverse pregnancy outcomes (APOs, such as preeclampsia or gestational diabetes), who are at increased risk for future heart disease. Health-related social needs (HRSNs) are recognized barriers to care, yet their pregnancy-related prevalence and associations with care are unknown. We sought to (1) describe the pregnancy-related prevalence of HRSNs, and (2) assess associations between pregnancy-related HRSNs and subsequent linkage to primary care.Methods:We analyzed electronic health record data for individuals with prenatal care and delivery (2018-2021) at our urban safety-net hospital. HRSNs were assessed via a routine screener, and we summarized individual responses during pregnancy through 6 weeks post partum as: any positive, all negative, or never screened. Postpartum linkage to primary care was defined as a completed primary care visit after 6 weeks through 1 year post partum. We analyzed the prevalence of HRSNs and their associations with linkage to primary care, using adjusted log-linked binomial regression models. In stratified models we assessed for effect modification by APO history and other variables.Results:Of 4941 individuals in our sample, 53% identified as Black non-Hispanic and 21% as Hispanic, 68% were publicly insured, and 93% completed ≥1 HRSN screening. Nearly 1 in 4 screened positive for any HRSN, most often food insecurity (14%) or housing instability (12%), and 53% linked to primary care. Compared with those who screened negative for all HRSNs (n=3491), linkage to primary care was similar among those who screened positive for any HRSNs (n=1079; adjusted risk ratio, aRR 1.04, 95% confidence interval, CI: 0.98-1.10) and lower among those never screened (n=371; aRR 0.77, 95% CI: 0.68-0.86). We found no evidence of effect modification by APO history, race/ethnicity, insurance, language, or Covid-19 pandemic exposure.Conclusions:In this diverse postpartum sample, we identified a 24% prevalence of pregnancy-related HRSNs and 53% subsequent linkage to primary care. Linkage to primary care was not associated with HRSN screening result (positive versus negative) but was significantly negatively associated with being missed by HRSN screening. Further research is needed to better understand HRSN screening practices and who is missed by screening, and to identify modifiable barriers to postpartum primary care especially after APOs.