A5032681013- Biomedical Text Mining and Ontologies
- Genomics and Rare Diseases
- Electronic Health Records Systems
- Blood Pressure and Hypertension Studies
- Statistical Methods in Clinical Trials
- Artificial Intelligence in Healthcare
- Health Systems, Economic Evaluations, Quality of Life
- Machine Learning in Healthcare
- Genetics, Bioinformatics, and Biomedical Research
- Blood transfusion and management
- Cardiovascular Function and Risk Factors
- Venous Thromboembolism Diagnosis and Management
- Platelet Disorders and Treatments
- Peripheral Artery Disease Management
- IoT and Edge/Fog Computing
- Acute Ischemic Stroke Management
- Semantic Web and Ontologies
- Heart Failure Treatment and Management
- Genetic Associations and Epidemiology
- Meta-analysis and systematic reviews
- Metabolomics and Mass Spectrometry Studies
- Data Quality and Management
- Gene expression and cancer classification
- Topic Modeling
- Lipoproteins and Cardiovascular Health
Mass General Brigham
2016-2025
Political Research Associates
2024-2025
Brigham and Women's Hospital
2024
Importance Blood pressure (BP) and cholesterol control remain challenging. Remote care can deliver more effective outside of traditional clinician-patient settings but scaling ensuring access to among diverse populations remains elusive. Objective To implement evaluate a remote hypertension management program across health network. Design, Setting, Participants Between January 2018 July 2021, 20 454 patients in large integrated network were screened; 18 444 approached, 10 803 enrolled...
BackgroundScreening participants in clinical trials is an error-prone and labor-intensive process that requires significant time resources. Large language models such as generative pretrained transformer 4 (GPT-4) present opportunity to enhance the screening with advanced natural processing. This study evaluates utility of a Retrieval-Augmented Generation (RAG)–enabled GPT-4 system improve accuracy, efficiency, reliability for trial involving patients symptomatic heart failure.MethodsThe...
This randomized clinical trial compares the efficiency of prescreening patients with heart failure using an artificial intelligence (AI) large language model for inclusion in a vs manual prescreening.
Large language models (LLMs) hold promise for improving literature review of variants in clinical genetic testing. We analyzed the performance, nondeterminism, and drift Generative Pretrained Transformer 4 (GPT-4) series to assess their present suitability use complex laboratory processes. optimized a chained, two-prompt GPT-4 sequence automated classification functional evidence using training set 45 article–variant pairs. The initial prompt asked supply all given article variant interest...
Hypertension is a modifiable risk factor for numerous comorbidities and treating hypertension can greatly improve health outcomes. We sought to increase the efficiency of virtual management program through workflow automation processes.We developed customer relationship (CRM) solution at our institution purpose improving processes describe here development, implementation, initial experience this CRM system.Notable system features include task automation, patient data capture, multi-channel...
Academic medical centers require many interconnected systems to fully support genetic testing processes. We provide an overview of the end-to-end that has been established surrounding a laboratory within our environment, including both and clinician facing infrastructure. explain key functions we have found useful in supporting systems. also consider ways this infrastructure could be enhanced enable deeper assessment test results clinic.
The conventional approach for clinical studies is to identify a cohort of potentially eligible patients and then screen enrollment. In an effort reduce the cost manual involved in screening process, several have leveraged electronic health records (EHR) refine cohorts better match eligibility criteria, which referred as phenotyping. We extend this dynamically by repeating phenotyping alternation with screening.Our consists multiple cycles. At start each cycle, algorithm used from EHR,...
Background. Large Language Models (LLMs) hold promise for improving genetic variant literature review in clinical testing. We assessed Generative Pretrained Transformer 4's (GPT-4) performance, nondeterminism, and drift to inform its suitability use complex processes. Methods. A 2-prompt process classification of functional evidence was optimized using a development set 45 articles. The prompts asked GPT-4 supply all data present an article related or indicate that no is present. For...
Clinical variant interpretation is a multi-step process relying on accurate integration of diverse types evidence, including genetic and functional data that are primarily collected through literature search review. Functional studies aim at providing evidence supporting or arguing against pathogenicity critical for the classification variants. However, manual curation these time-consuming, leading to delays in classification. To better scale genomic testing, it essential decrease time,...
Calculated panel reactive antibody (cPRA) scoring is used to assess whether platelet refractoriness mediated by human leukocyte antigen (HLA) antibodies in the recipient. cPRA testing uses a national sample of US kidney donors estimate population frequency HLA antigens, which may be different than frequencies within local inventories. We aimed determine impact on patient scores using derived from typing donations rather frequencies.We built an open-source web service calculate based or...
Analysis of health data typically requires development queries using structured query language (SQL) by a data-analyst. As the SQL are manually created, they prone to errors. In addition, accurate implementation depends on effective communication with clinical experts, that further makes analysis error prone. potential resolution, we explore an alternative approach wherein graphical interface automatically generates is used perform analysis. The latter allows experts directly complex data,...
The i2b2 platform is used at major academic health institutions and research consortia for querying electronic data. However, a obstacle wider utilization of the complexity data loading that entails steep curve learning platform's complex schemas. To address this problem, we have developed i2b2-etl package simplifies process, which will facilitate deployment platform.We implemented as Python application imports ontology patient using simplified input file schemas provides inbuilt record...
<sec> <title>BACKGROUND</title> While usage of EHRs has substantially increased over the past decade, building clinical applications remains challenging for multiple reasons, including security and privacy considerations, data integration, application distribution, deployment. </sec> <title>OBJECTIVE</title> We discuss importance platforms health IT development, specifically describe “igia,” an open-source collection software packages that enable developers to build deploy more efficiently....