A5043198554- ECG Monitoring and Analysis
- Cardiovascular Function and Risk Factors
- Cardiovascular Health and Risk Factors
- Cardiac Imaging and Diagnostics
- COVID-19 diagnosis using AI
- Lipoproteins and Cardiovascular Health
- Blood Pressure and Hypertension Studies
- Diabetes Treatment and Management
- Phonocardiography and Auscultation Techniques
- Diabetes Management and Research
- Cardiomyopathy and Myosin Studies
- Machine Learning in Healthcare
- Artificial Intelligence in Healthcare
- Misinformation and Its Impacts
- Mobile Health and mHealth Applications
- Acute Myocardial Infarction Research
- Heart Failure Treatment and Management
- Cardiac Valve Diseases and Treatments
- Cardiovascular Effects of Exercise
- Heart Rate Variability and Autonomic Control
- Body Composition Measurement Techniques
- Cardiac electrophysiology and arrhythmias
- Antiplatelet Therapy and Cardiovascular Diseases
- Cardiac Health and Mental Health
- Cardiovascular Disease and Adiposity
Yale University
2021-2025
Yale New Haven Hospital
2021-2023
Cardiovascular Research Center
2023
All India Institute of Medical Sciences
2019-2022
Indraprastha Institute of Information Technology Delhi
2020-2022
Importance Wearable devices may be able to improve cardiovascular health, but the current adoption of these could skewed in ways that exacerbate disparities. Objective To assess sociodemographic patterns use wearable among adults with or at risk for disease (CVD) US population 2019 2020. Design, Setting, and Participants This population-based cross-sectional study included a nationally representative sample from Health Information National Trends Survey (HINTS). Data were analyzed June 1...
Left ventricular (LV) systolic dysfunction is associated with a >8-fold increased risk of heart failure and 2-fold premature death. The use ECG signals in screening for LV limited by their availability to clinicians. We developed novel deep learning-based approach that can images the dysfunction.
Artificial intelligence (AI) can detect left ventricular systolic dysfunction (LVSD) from electrocardiograms (ECGs). Wearable devices could allow for broad AI-based screening but frequently obtain noisy ECGs. We report a novel strategy that automates the detection of hidden cardiovascular diseases, such as LVSD, adapted single-lead ECGs obtained on wearable and portable devices. use 385,601 development standard noise-adapted model. For model, are augmented during training with random...
Abstract Background and Aims Current heart failure (HF) risk stratification strategies require comprehensive clinical evaluation. In this study, artificial intelligence (AI) applied to electrocardiogram (ECG) images was examined as a strategy predict HF risk. Methods Across multinational cohorts in the Yale New Haven Health System (YNHHS), UK Biobank (UKB), Brazilian Longitudinal Study of Adult (ELSA-Brasil), individuals without baseline were followed for first hospitalization. An AI-ECG...
The COVID-19 pandemic has sparked unprecedented public health and social measures (PHSM) by national local governments, including border restrictions, school closures, mandatory facemask use stay at home orders. Quantifying the effectiveness of these interventions in reducing disease transmission is key to rational policy making response current future pandemics. In order estimate interventions, detailed descriptions their timelines, scale scope are needed. Health Intervention Tracking for...
The COVID-19 pandemic has revealed the power of internet disinformation in influencing global health. deluge information travels faster than epidemic itself and is a threat to health millions across globe. Health apps need leverage machine learning for delivering right while constantly misinformation trends deliver these effectively vernacular languages order combat infodemic at grassroot levels general public. Our application, WashKaro, multi-pronged intervention that uses conversational...
Risk stratification strategies for cancer therapeutics-related cardiac dysfunction (CTRCD) rely on serial monitoring by specialized imaging, limiting their scalability. We aimed to examine an application of artificial intelligence (AI) ECG images as a surrogate imaging risk biomarkers and its association with early CTRCD. Across US-based health system (2013-2023), we identified 1550 patients (aged, 60 [interquartile range, 51-69] years, 1223 [78.9%] women) without cardiomyopathy who received...
ABSTRACT Background Risk stratification strategies for cancer therapeutics-related cardiac dysfunction (CTRCD) rely on serial monitoring by specialized imaging, limiting their scalability. Objectives To examine an artificial intelligence (AI)-enhanced electrocardiographic (AI-ECG) surrogate imaging risk biomarkers, and its association with CTRCD. Methods Across a five-hospital U.S.-based health system (2013-2023), we identified patients breast or non-Hodgkin lymphoma (NHL) who received...
BACKGROUND: While universal screening for Lp(a; lipoprotein[a]) is increasingly recommended, <0.5% of patients undergo Lp(a) testing. Here, we assessed the feasibility deploying Algorithmic Risk Inspection Screening Elevated ARISE), a validated machine learning tool, to health system electronic records increase yield METHODS: We randomly sampled 100 000 from Yale-New Haven Health System evaluate ARISE deployment. also evaluated Lp(a)-tested populations in (n=7981) and Vanderbilt...
An algorithmic strategy for anatomical vs. functional testing in suspected coronary artery disease (CAD) (Anatomical Stress teSting decIsion Support Tool; ASSIST) is associated with better outcomes than random selection. However, the real world, this decision rarely random. We explored agreement between a provider-driven simulated approach to cardiac and its association across multinational cohorts.
Proactive detection of hemodynamic shock can prevent organ failure and save lives. Thermal imaging is a non-invasive, non-contact modality to capture body surface temperature with the potential reveal underlying perfusion disturbance in shock. In this study, we automate early prediction using machine learning upon thermal images obtained pediatric intensive care unit tertiary hospital. 539 were recorded out which 253 had concomitant measurement continuous intra-arterial blood pressure, gold...
Hypertrophic cardiomyopathy (HCM) affects 1 in every 200 individuals and is the leading cause of sudden cardiac death young adults. HCM can be identified using an electrocardiogram (ECG) raw voltage data deep learning approaches, but their point-of-care application limited by inaccessibility these signal data. We developed a learning-based approach that overcomes this limitation detects from images 12-lead ECGs across layouts. patients with features present on magnetic resonance imaging...
Timely and accurate assessment of electrocardiograms (ECGs) is crucial for diagnosing, triaging, clinically managing patients. Current workflows rely on a computerized ECG interpretation using rule-based tools built into the signal acquisition systems with limited accuracy flexibility. In low-resource settings, specialists must review every single such decisions, as these interpretations are not available. Additionally, high-quality even more essential in settings there higher burden...
Current risk stratification strategies for heart failure (HF) require either specific blood-based biomarkers or comprehensive clinical evaluation. In this study, we evaluated the use of artificial intelligence (AI) applied to images electrocardiograms (ECGs) predict HF risk.
Serial functional status assessments are critical to heart failure (HF) management but often described narratively in documentation, limiting their use quality improvement or patient selection for clinical trials.
Abstract Background Rich data in cardiovascular diagnostic testing are often sequestered unstructured reports, limiting their use. Methods We sequentially deployed generative and interpretative open-source large language models (LLMs; Llama2-70b, Llama2-13b). Using we generated varying formats of transthoracic echocardiogram (TTE) reports from 3000 real-world with paired structured elements. prompt-based supervised training, fine-tuned Llama2-13b using larger batches TTE as inputs, to...
Abstract Background and Aims AI-enhanced 12-lead ECG can detect a range of structural heart diseases (SHDs) but has limited role in community-based screening. We developed externally validated noise-resilient single-lead AI-ECG algorithm that SHD predict the risk their development using wearable/portable devices. Methods Using 266,740 ECGs from 99,205 patients with paired echocardiographic data at Yale New Haven Hospital, we ADAPT-HEART, noise-resilient, deep-learning algorithm, to lead I...
Abstract Background: The COVID-19 pandemic has uncovered the potential of digital misinformation in shaping health nations. deluge unverified information that spreads faster than epidemic itself is an unprecedented phenomenon put millions lives danger. Mitigating this ‘Infodemic’ requires robust messaging systems are engaging, vernacular, scalable, effective, and continuously learn new patterns. Objective: We created WashKaro, a multi-pronged intervention for mitigating through...
Smartphone-based health applications are increasingly popular, but their real-world use for cardiovascular risk management remains poorly understood.
Objective To assess the uptake of second line antihyperglycaemic drugs among patients with type 2 diabetes mellitus who are receiving metformin. Design Federated pharmacoepidemiological evaluation in LEGEND-T2DM. Setting 10 US and seven non-US electronic health record administrative claims databases Observational Health Data Sciences Informatics network eight countries from 2011 to end 2021. Participants 4.8 million (≥18 years) across based had received metformin monotherapy initiated...