Publications by authors named "Phyllis M Thangaraj"
Background: Assessing the generalizability of randomized clinical trials (RCTs) to real-world patients remains challenging. We propose a multidimensional metric to quantify the representativeness of an RCT cohort in an electronic health record (EHR) population and estimate real-world effects based on individualized treatment effects observed in the RCT.
Methods: We identified 65 clinical prerandomization characteristics of patients with heart failure with preserved ejection fraction within the TOPCAT (Treatment of Preserved Cardiac Function Heart Failure with an Aldosterone Antagonist Trial) and extracted those features in similar patients in EHR data from 4 hospitals in the Yale New Haven Health System.
Article Synopsis
- Serial assessments of functional status play a crucial role in managing heart failure but are typically documented in a way that limits their use for quality improvement and clinical trials.
- The study aimed to create and validate a deep learning model using natural language processing (NLP) to extract functional status data from unstructured clinical notes related to heart failure patients.
- Results showed that the NLP model performed excellently in identifying New York Heart Association (NYHA) classifications and heart failure symptoms from patient documentation, achieving very high accuracy metrics across different healthcare sites.
Article Synopsis
- Digital twins are virtual representations of individuals and their environments used in cardiovascular medicine to improve decision-making and risk prediction.
- They integrate various data types to create accurate models of heart conditions, enhancing diagnosis and treatment planning.
- The review discusses advancements in AI that strengthen digital twin capabilities, while also addressing ethical and societal challenges in their implementation for personalized care.
Article Synopsis
- This study compares the cardiovascular effectiveness of different second-line antihyperglycemic agents (SGLT2 inhibitors, GLP-1 receptor agonists, DPP-4 inhibitors, and sulfonylureas) in patients with type 2 diabetes and cardiovascular disease.
- Using data from over 1.4 million patients across multiple databases, the researchers analyzed the risk of major adverse cardiovascular events (MACE) over a follow-up period of several years.
- Results indicated that SGLT2 inhibitors and GLP-1 receptor agonists had significantly lower risks of MACE compared to DPP-4 inhibitors and sulfonylureas, pointing to their potential superiority as treatment options for
Article Synopsis
- Randomized clinical trials (RCTs) provide a framework for evidence-based disease management, but their applicability to real-world patients is often unclear and difficult to assess.
- The study aimed to create an algorithm that helps quantify how similar patients in electronic health records (EHRs) are to those in a specific RCT (TOPCAT), with the goal of estimating treatment effects in real-world scenarios.
- The analysis involved comparing patient data from the TOPCAT trial to a larger cohort from the Yale New Haven Hospital System, revealing significant differences in patient characteristics that may affect treatment outcomes.
Article Synopsis
- Randomized clinical trials (RCTs) help inform medical practice, but their applicability to different populations can be unclear.
- The RCT-Twin-GAN model was developed to create digital twins of RCTs that simulate treatment effects using data from varying patient populations.
- The model successfully reproduced treatment effects from two notable studies, SPRINT and ACCORD, demonstrating its potential to bridge gaps in understanding how different populations might respond to medical interventions.
Article Synopsis
- * The study analyzed data from over 1.4 million patients treated with various second-line diabetes medications, using advanced statistical methods to compare outcomes and risks of heart issues.
- * Findings indicated that both SGLT2 inhibitors and GLP-1 receptor agonists reduce the risk of cardiovascular events compared to DPP-4 inhibitors and sulfonylureas, but no significant differences were found between SGLT2is and GLP1-RAs themselves regarding heart risks.
Article Synopsis
- Randomized clinical trials (RCTs) often fail to represent key populations adequately, making it challenging to assess their real-world effectiveness; the study proposes using digital twins of RCTs to better understand these effects using electronic health records (EHR).
- A new model called RCT-Twin-GAN generates RCT-like datasets by simulating data from EHRs, allowing researchers to analyze the impact of treatments like spironolactone on heart failure patients more accurately.
- The results showed that the simulated RCT-Twin data closely mirrored real RCT results, with a balanced representation of covariates and similar treatment effects, suggesting that this approach could enhance the relevance of RCT findings in broader clinical contexts.
Article Synopsis
- * The study proposes using machine learning to create adaptive predictive models, termed computational trial phenomaps, which can help optimize participant selection based on predicted treatment benefits.
- * Results from simulations of two cardiovascular trials show that this approach could have significantly reduced trial sizes while maintaining similar treatment effects, suggesting improved efficiency in RCT enrollment.
Article Synopsis
- Randomized controlled trials (RCTs) are crucial for evidence-based medicine but require significant resources, prompting the need for more efficient enrollment methods.
- The study proposes a machine learning approach to improve RCT enrollment by using adaptive predictive enrichment and computational trial phenomaps to identify candidates based on their potential treatment benefits.
- Simulations of two large cardiovascular trials showed this method could reduce trial sizes by approximately 15-18% while maintaining the integrity of study outcomes, indicating a more efficient way to conduct RCTs.
Accelerating chest pain evaluation with machine learning.
Eur Heart J Acute Cardiovasc Care
November 2023
Article Synopsis
- The study investigates how machine learning can automate the identification of acute ischemic stroke (AIS) patients using electronic health records (EHRs), which simplifies traditional, labor-intensive methods of cohort identification.
- Researchers developed and tested various machine learning models on patient data from a hospital to identify AIS, achieving a high detection success rate with minimal data processing involved.
- The validation process showed that the models effectively identified AIS patients even without diagnosis codes, indicating that machine learning is a promising, efficient tool for AIS patient identification in clinical settings.
Article Synopsis
- - Understanding how SARS-CoV-2 interacts with the host helps develop better treatments and public health responses, particularly targeting immune pathways linked to complement and coagulation systems.
- - A study found that individuals with a history of macular degeneration or coagulation disorders are at higher risk for severe COVID-19 outcomes, regardless of age, sex, or smoking history.
- - Genetic analysis identified specific variants linked to complement and coagulation functions, suggesting that these factors influence COVID-19 severity and highlighting the need for comprehensive research methods to assess immunity and disease susceptibility.
Article Synopsis
- - Understanding how SARS-CoV-2 interacts with the body can help develop better therapies and public health strategies, focusing on immune pathways related to complement and coagulation systems.
- - A study of over 11,000 patients found that pre-existing conditions related to these systems, like macular degeneration and coagulation disorders, increase risks of severe illness and death from COVID-19, independent of other factors like age or smoking.
- - Genetic analysis revealed specific genetic markers linked to immune response that could help predict COVID-19 outcomes, illustrating the importance of combining various research methods to understand disease susceptibility.
Article Synopsis
- The article discusses the increasing use of electronic health record (EHR) data for biomedical research and highlights the challenges of understanding the EHR structure and data science needed for effective usage, referencing the OMOP common data model (CDM) for standardization.
- The authors developed an R package called ROMOP, which simplifies the analysis and exploration of EHR data using the OMOP CDM, allowing users to extract and summarize clinical and demographic information more efficiently.
- ROMOP is open-source under the MIT license and can be downloaded from GitHub, with additional resources including setup instructions and a public sandbox for testing out the package and OMOP data.
Article Synopsis
- Electronic health records (EHRs) are widely used in healthcare, but issues with interoperability and technical requirements hinder their application in research, prompting the need for more accessible tools for researchers.
- PatientExploreR is a user-friendly application built on the R/Shiny framework that allows researchers to interact with EHR data, creating dynamic reports and visualizations without needing programming skills.
- The software is open-source and can be downloaded from GitHub, providing researchers with easy access to EHR data and a synthesized data sandbox for those without direct EHR access.