A5024988228- Electronic Health Records Systems
- Colorectal Cancer Screening and Detection
- Health disparities and outcomes
- Food Security and Health in Diverse Populations
- Data Quality and Management
- Chronic Disease Management Strategies
- Healthcare Policy and Management
- Biomedical Text Mining and Ontologies
- Mobile Health and mHealth Applications
- Topic Modeling
- Machine Learning in Healthcare
- Natural Language Processing Techniques
- Parathyroid Disorders and Treatments
- Emergency and Acute Care Studies
- COVID-19 and healthcare impacts
- COVID-19 Clinical Research Studies
- Vaccine Coverage and Hesitancy
- Healthcare Systems and Technology
- COVID-19 diagnosis using AI
- Radiomics and Machine Learning in Medical Imaging
- Medical Coding and Health Information
- Digital Radiography and Breast Imaging
- COVID-19 epidemiological studies
- Digital Mental Health Interventions
- Cardiovascular Health and Risk Factors
University of Arkansas for Medical Sciences
2020-2024
Multimorbidity, having a diagnosis of two or more chronic conditions, increases as people age. It is predictor used in clinical decision-making, but underdiagnosis underserved populations produces bias the data that support algorithms healthcare processes. Artificial intelligence (AI) systems could produce inaccurate predictions if patients have multiple unknown conditions. Rural are likely to be and also In this study, collected during course care centrally located academic hospital,...
OBJECTIVES Addressing parental/caregivers’ coronavirus disease 2019 (COVID-19) vaccine hesitancy is critical to improving uptake in children. Common concerns have been previously reported through online surveys, but qualitative data from KII and focus groups may add much-needed context. Our objective was examine factors impacting pediatric COVID-19 decision-making Black, Spanish-speaking, rural white parents/caregivers inform the content design of a mobile application improve uptake. METHODS...
Social determinants of health (SDOH) impact 80% outcomes from acute to chronic disorders, and attempts are underway provide these data elements clinicians. It is, however, difficult collect SDOH through (1) surveys, which inconsistent incomplete data, or (2) aggregates at the neighborhood level. Data sources is not sufficiently accurate, complete, up-to-date. To demonstrate this, we have compared Area Deprivation Index (ADI) purchased commercial consumer individual-household The ADI composed...
Abstract With the increasing need for timely submission of data to state and national public health registries, current manual approaches acquisition are insufficient. In clinical practice, federal regulations now mandating use messaging standards, i.e., Health Level Seven (HL7 ® ) Fast Healthcare Interoperability Resources (FHIR standard, facilitate electronic exchange (patient) data. both research we can also leverage FHIR ‒ infrastructure already in place supporting practice enable...
An estimated one-third of patients experience post-traumatic stress disorder (PTSD) or depression in the year following a traumatic injury. The American College Surgeons requires postinjury PTSD and screening trauma centers, although implementation has been limited. Tech-based solutions have proposed to improve uptake mental health screening. goals this pilot study were assess usability acceptability Blueprint, tech-based platform, explore attitudes toward intervention.
Colonoscopy is a screening and diagnostic procedure for detection of colorectal carcinomas with specific quality metrics that monitor improve adenoma rates. These are stored in disparate documents i.e., colonoscopy, pathology, radiology reports. The lack integrated standardized documentation impeding cancer research. Clinical concept extraction using Natural Language Processing (NLP) Machine Learning (ML) techniques an alternative to manual data abstraction. Contextual word embedding models...
Abstract Introduction Primary hyperparathyroidism (PHPT) is a condition in which one or more parathyroid glands secrete excess amounts of hormone (PTH). In short, PHPT characterized by hypercalcemia/hypercalciuria with concurrent elevated PTH levels. This known to increase the risk cardiovascular disease, osteoporosis, psychiatric disturbances, and renal complications. As now, disease typically runs long course before being identified treated. At present, surgery only viable treatment option...
Named Entity Recognition (NER) aims to identify and classify entities into predefined categories is a critical pre-processing task in Natural Language Processing (NLP) pipeline. Readily available off-the-shelf NER algorithms or programs are trained on general corpus often need be retrained when applied different domain. The end model’s performance depends the quality of named generated by these models used NLP task. To improve model accuracy, researchers build domain-specific corpora for...
Children with asthma and obesity are more likely to have lower vitamin D levels, but the optimal replacement dose is unknown in this population. The objective of study identifying a children obesity-related that safely achieves serum levels ≥ 40 ng/mL. This prospective multisite randomized controlled trial recruited children/adolescents body mass index 85% for age/sex. Part 1 (dose finding), evaluated 4 oral regimens 16 weeks identify achieved 2 compared calculated from part (50,000 IU...
To prove the concept of diagnosing primary hyperparathyroidism (pHPT) without calcium and parathyroid hormone (PTH) values identifying potential risk factors for pHPT.Data were extracted from clinical data warehouse (CDW) at University Arkansas Medical Sciences (UAMS) Epic EHR (2014-2019).1737 patients with over 185 000 rows provided in a relational structure processed/flattened to facilitate modeling. Phenotype elements identified pHPT advance knowledge PTH levels. The area under curve...
Abstract Social determinants of health (SDOH) impact 80% outcomes from acute to chronic disorders, and attempts are underway provide these data elements clinicians. It is, however, difficult collect SDOH through (1) surveys, which inconsistent incomplete data, or (2) aggregates at the neighborhood level. Data sources is not sufficiently accurate, complete, up-to-date. To demonstrate this, we have compared Area Deprivation Index (ADI) purchased commercial consumer individual-household The ADI...
Endoscopy procedures are often performed with either moderate or deep sedation. While sedation is costly, not always well tolerated resulting in patient discomfort, and aborted. Due to lack of clear guidelines, the decision utilize anesthesia for a procedure made by providers, leading high variability clinical practice. The objective this study was build Machine Learning (ML) model that predicts if colonoscopy can be successfully completed based on patients' demographics, comorbidities,...
Clinical named entity recognition (NER) is an essential building block for many downstream natural language processing (NLP) applications such as information extraction and de-identification. Recently, deep learning (DL) methods that utilize word embeddings have become popular in clinical NLP tasks. However, there has been little work on evaluating combining the trained from different domains. The goal of this study to improve performance NER discharge summaries by developing a DL model...
Social and behavioral aspects of our lives significantly impact health, yet minimal social determinants health (SDOH) data elements are collected in the healthcare system.In this proof-of-concept study we developed a repeatable SDOH enrichment integration process to incorporate dynamically evolving domain concepts from consumers into clinical data. This included mapping, linking compiled consumer patient records Electronic Health Records, quality analysis preprocessing, storage.Consumer...
The ongoing COVID-19 pandemic has become the most impactful of past century. SARS-CoV-2 virus spread rapidly across globe affecting and straining global health systems. More than 2 million people have died from (as 30 January 2021). To lessen pandemic's impact, advanced methods such as Artificial Intelligence models are proposed to predict mortality, morbidity, disease severity, other outcomes sequelae. We performed a rapid scoping literature review identify deep learning techniques that...
In this study we examined the correlation of COVID-19 positivity with area deprivation index (ADI), social determinants health (SDOH) factors based on a consumer and electronic medical record (EMR) data population density in patient from tertiary healthcare system Arkansas. was significantly associated density, age, race, household size. Understanding disparities SDOH can add value to creation trustable AI.
Although colonoscopy is the most frequently performed endoscopic procedure, lack of standardized reporting impeding clinical and translational research. Inadequacies in data extraction from raw, unstructured text electronic health records (EHR) pose an additional challenge to procedure quality metric reporting, as vital details related are stored disparate documents. Currently, there no EHR workflow that links these documents specific making process error prone. We hypothesize extracting...
Colonoscopy plays a critical role in screening of colorectal carcinomas (CC). Unfortunately, the data related to this procedure are stored disparate documents, colonoscopy, pathology, and radiology reports respectively. The lack integrated standardized documentation is impeding accurate reporting quality metrics clinical translational research. Natural language processing (NLP) has been used as an alternative manual abstraction. Performance Machine Learning (ML) based NLP solutions heavily...
Studies examining disparities in hospital mortality for patients hospitalized with COVID-19 are mixed. Findings from within academic medical centers (AMCs) lacking, but important given the role of AMCs improving health equity. This study assessed whether minority National COVID Cohort Collaborative (N3C) institutions, which consist predominantly AMCs, have higher rates relative to White patients. A retrospective analysis was performed. Logistic regression used test primary hypothesis a...