Publications by authors named "Andreas Coppi"
Aims: Artificial intelligence (AI)-enhanced 12-lead electrocardiogram (ECG) can detect a range of structural heart diseases (SHDs); however, it has a limited role in community-based screening. We developed and externally validated a noise-resilient single-lead AI-ECG algorithm that can detect SHDs and predict the risk of their development using wearable/portable devices.
Methods And Results: Using 266 740 ECGs from 99 205 patients with paired echocardiographic data at Yale New Haven Hospital, we developed AI Deep learning for Adapting Portable Technology in HEART disease detection (ADAPT-HEART), a noise-resilient, deep learning algorithm, to detect SHDs using lead I ECG.
Background And Aims: The diagnosis of transthyretin amyloid cardiomyopathy (ATTR-CM) requires advanced imaging, precluding large-scale preclinical testing. Artificial intelligence (AI)-enabled transthoracic echocardiography (TTE) and electrocardiography (ECG) may provide a scalable strategy for preclinical monitoring.
Methods: This was a retrospective analysis of individuals referred for nuclear cardiac amyloid testing at the Yale-New Haven Health System (YNHHS, internal cohort) and Houston Methodist Hospitals (HMH, external cohort).
Nirmatrelvir-ritonavir versus placebo-ritonavir in individuals with long COVID in the USA (PAX LC): a double-blind, randomised, placebo-controlled, phase 2, decentralised trial.
Lancet Infect Dis
August 2025
Background: The substantial burden of post-COVID-19 condition (also known as long COVID) underscores the need for effective pharmacological interventions. Given that viral persistence has been hypothesised as a potential cause of long COVID, antiviral therapy might offer a promising approach to alleviating long COVID symptoms. We therefore investigated the efficacy, safety, and tolerability of nirmatrelvir-ritonavir for treating long COVID.
View Article and Find Full Text PDFBackground: Identifying structural heart diseases (SHDs) early can change the course of the disease, but their diagnosis requires cardiac imaging, which is limited in accessibility.
Objectives: The purpose of this study was to leverage images of 12-lead electrocardiograms (ECGs) for automated detection and prediction of multiple SHDs using an ensemble deep learning approach.
Methods: We developed a series of convolutional neural network models for detecting a range of individual SHDs from images of ECGs with SHDs defined by transthoracic echocardiograms performed within 30 days of the ECG at the Yale New Haven Hospital (YNHH).
Objectives: Direct electronic access to multiple electronic health record (EHR) systems through patient portals offers a novel avenue for decentralized research. Given the critical value of patient characterization, we sought to compare computable evaluation of health conditions from patient-portal EHR against the traditional self-report.
Materials And Methods: In the nationwide Innovative Support for Patients with SARS-CoV-2 Infections Registry (INSPIRE) study, which linked self-reported questionnaires with multiplatform patient-portal EHR data, we compared self-reported health conditions across different clinical domains against computable definitions based on diagnosis codes, medications, vital signs, and laboratory testing.
Background: Point-of-care ultrasonography (POCUS) enables cardiac imaging at the bedside and in communities but is limited by abbreviated protocols and variation in quality. We aimed to develop and test artificial intelligence (AI) models to screen for under-diagnosed cardiomyopathies from cardiac POCUS.
Methods: In a development set of 290 245 transthoracic echocardiographic videos across the Yale-New Haven Health System (YNHHS), we used augmentation approaches, and a customised loss function weighted for view quality to derive a POCUS-adapted, multi-label, video-based convolutional neural network that discriminates hypertrophic cardiomyopathy and transthyretin amyloid cardiomyopathy from controls without known disease.
Background: The lack of automated tools for measuring care quality limits the implementation of a national program to assess guideline-directed care in heart failure with reduced ejection fraction (HFrEF).
Objectives: The authors aimed to automate the identification of patients with HFrEF at hospital discharge, an opportunity to evaluate and improve the quality of care.
Methods: The authors developed a novel deep-learning language model for identifying patients with HFrEF from discharge summaries of hospitalizations with heart failure at Yale New Haven Hospital during 2015 to 2019.
Development and Multinational Validation of an Ensemble Deep Learning Algorithm for Detecting and Predicting Structural Heart Disease Using Noisy Single-lead Electrocardiograms.
medRxiv
October 2024
Background And Aims: AI-enhanced 12-lead ECG can detect a range of structural heart diseases (SHDs) but has a limited role in community-based screening. We developed and externally validated a noise-resilient single-lead AI-ECG algorithm that can detect SHD and predict the risk of 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 developed ADAPT-HEART, a noise-resilient, deep-learning algorithm, to detect SHD using lead I ECG.
Background: Identifying structural heart diseases (SHDs) early can change the course of the disease, but their diagnosis requires cardiac imaging, which is limited in accessibility.
Objective: To leverage images of 12-lead ECGs for automated detection and prediction of multiple SHDs using an ensemble deep learning approach.
Methods: We developed a series of convolutional neural network models for detecting a range of individual SHDs from images of ECGs with SHDs defined by transthoracic echocardiograms (TTEs) performed within 30 days of the ECG at the Yale New Haven Hospital (YNHH).
Article Synopsis
- Researchers explored using artificial intelligence (AI) to improve the diagnosis of transthyretin amyloid cardiomyopathy (ATTR-CM) through echocardiograms (TTE) and electrocardiograms (ECG), potentially allowing for earlier detection of the disease.
- They trained deep learning models to identify ATTR-CM patterns, achieving high accuracy in recognizing these signatures from cardiac data in two large patient groups.
- The study found that AI can effectively predict the likelihood of ATTR-CM in individuals up to three years before a formal diagnosis, suggesting that it could help identify patients who might benefit from early treatment options.
Artificial Intelligence-Enhanced Risk Stratification of Cancer Therapeutics-Related Cardiac Dysfunction Using Electrocardiographic Images.
Circ Cardiovasc Qual Outcomes
January 2025
Article Synopsis
- Researchers evaluated the potential of artificial intelligence (AI) applied to electrocardiograms (ECG) to predict cardiac dysfunction related to cancer treatments, aiming to develop a more scalable risk stratification method.
- In a study involving 1,550 patients treated with anthracyclines and/or trastuzumab, the AI model classified patients into low, intermediate, and high-risk groups based on their baseline ECG images.
- The findings revealed that patients in the high-risk group had significantly higher incidents of cardiac dysfunction within a year post-treatment, highlighting the effectiveness of AI-ECG in identifying those at greater risk for complications.
Article Synopsis
- - The PAX LC trial is a digital and decentralized study aimed at evaluating the effectiveness and safety of nirmatrelvir/ritonavir for long COVID symptoms in 100 adults across the U.S.
- - Participants will be randomly assigned to receive either the medication or a placebo, with data collected through digital diaries and home blood draws to enhance accessibility and engagement.
- - The trial focuses on measuring changes in physical and mental health, alongside exploring biomarkers for long COVID, with a goal of identifying potential treatments for this condition.
Article Synopsis
- Aortic stenosis (AS) presents a serious public health issue, and the new Digital AS Severity index (DASSi) leverages AI to improve detection of severe AS through simple echocardiography, without needing complex Doppler analysis.
- The study focused on patients with no severe AS initially (from Yale and Cedars-Sinai hospitals) to assess how well DASSi could predict AS onset and progression, with results analyzed from August 2023 to February 2024.
- Findings revealed that a higher DASSi score was linked to a quicker increase in peak aortic valve velocity and significantly elevated risks for aortic valve replacement, especially for scores above 0.2
Article Synopsis
- This study explores an AI-enhanced electrocardiogram (AI-ECG) as a tool for risk stratification in patients at risk of cancer therapeutics-related cardiac dysfunction (CTRCD) rather than relying solely on specialized imaging techniques.
- Patients with breast cancer or non-Hodgkin lymphoma who received certain chemotherapy agents were analyzed, revealing that higher AI-ECG predictions of left ventricular systolic dysfunction (LVSD) correlated with worse heart function metrics and increased risk of future cardiac issues.
- The findings suggest that using AI on ECGs can effectively identify patients at greater risk for CTRCD associated with specific cancer treatments, potentially allowing for better management of these patients' heart health.
Article Synopsis
- Point-of-care ultrasonography (POCUS) is a bedside cardiac imaging technique, but its effectiveness is hampered by inconsistent protocols and image quality, prompting the development of AI models to enhance cardiomyopathy diagnosis.
- Researchers utilized a massive dataset of transthoracic echocardiographic videos to create an AI model that identifies hypertrophic cardiomyopathy (HCM) and transthyretin amyloid cardiomyopathy (ATTR-CM) from POCUS without prior disease knowledge.
- The AI model demonstrated high accuracy in screening for HCM and ATTR-CM, detecting cases about 2 years prior to clinical diagnoses and showing significant prognostic potential for individuals without known cardiomyopathy.
Article Synopsis
- Accurate assessment of ECGs is vital for patient diagnosis and care, but current automated systems lack flexibility and reliability, especially in low-resource areas where specialists review each ECG manually.
- AI systems show promise for improved accuracy but often have limitations in the variety of conditions they can assess and require raw data not typically available to doctors.
- The ECG-GPT model was developed to generate expert-level diagnosis directly from ECG images, demonstrating strong performance across diverse healthcare settings and providing an accessible web application for accurate triage of patients.
Article Synopsis
- Aortic stenosis (AS) is a significant public health issue with limited current biomarkers for personalized monitoring, prompting the investigation of a new AI-based method, the Digital AS Severity index (DASSi), for assessing AS severity through echocardiography.* -
- This study followed two patient groups from Yale-New Haven Health System and Cedars-Sinai Medical Center who had no severe AS at the start, using DASSi scores to predict the development and worsening of AS over several years.* -
- Results indicated that higher baseline DASSi scores correlated with faster increases in aortic valve velocity, significantly raising the likelihood of patients needing aortic valve replacement, demonstrating DASSi's potential for improving AS patient management
Article Synopsis
- The study addresses the challenge of assessing care quality for heart failure patients, specifically those with reduced ejection fraction (HFrEF), due to a lack of automated measurement tools at hospital discharge.
- Researchers developed a deep learning language model that identifies HFrEF patients from discharge summaries using a semi-supervised approach, validated with hospital data from Yale New Haven Hospital and external institutions.
- The model demonstrated high performance, achieving AUROC values of up to 0.97 in detecting HFrEF, effectively improving the identification of these patients and potentially enhancing care quality.
Objective: We aimed to discover computationally-derived phenotypes of opioid-related patient presentations to the ED via clinical notes and structured electronic health record (EHR) data.
Methods: This was a retrospective study of ED visits from 2013-2020 across ten sites within a regional healthcare network. We derived phenotypes from visits for patients ≥18 years of age with at least one prior or current documentation of an opioid-related diagnosis.
Article Synopsis
- Over the past century, pandemics like COVID-19 highlight the need for prepared and coordinated responses to disease outbreaks.
- A new machine learning model was created to predict COVID-19 severity and hospital stays by analyzing plasma data from patients and healthy individuals, which identifies key biomarkers for triage.
- Significant findings include that higher eosinophil levels are linked to worse outcomes and lower serotonin levels are seen in critical cases; this model could be adapted for future viruses to improve resource allocation and patient care.
Article Synopsis
- Researchers created a smart computer program that can help doctors find a serious heart condition called aortic stenosis (AS) just by looking at ultrasound videos of the heart.
- They trained this program using a lot of heart videos and tested it with different groups of patients to make sure it works well.
- The program correctly identified severe AS almost 98% of the time and can be used easily in clinics to check patients without needing extra complicated tools.
Article Synopsis
- Artificial intelligence can effectively identify left ventricular systolic dysfunction (LVSD) using electrocardiograms (ECGs), even with the challenges posed by noisy signals from wearable devices.* -
- A new approach was developed that enhances AI models to better detect cardiovascular diseases by training them with augmented noisy ECG data that mimics real-world conditions.* -
- The noise-adapted AI model outperformed the standard model on ECGs affected by device noise, achieving better accuracy (AUROC of 0.87 vs. 0.72), showcasing its potential for improving remote cardiovascular monitoring.*
Background: The benefit of primary and booster vaccination in people who experienced a prior Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) infection remains unclear. The objective of this study was to estimate the effectiveness of primary (two-dose series) and booster (third dose) mRNA vaccination against Omicron (lineage BA.1) infection among people with a prior documented infection.
View Article and Find Full Text PDFBackground: The impact variant-specific immune evasion and waning protection have on declining coronavirus disease 2019 (COVID-19) vaccine effectiveness (VE) remains unclear. Using whole-genome sequencing (WGS), we examined the contribution these factors had on the decline that followed the introduction of the Delta variant. Furthermore, we evaluated calendar-period-based classification as a WGS alternative.
View Article and Find Full Text PDF