A5030321168- Machine Learning in Healthcare
- Non-Invasive Vital Sign Monitoring
- Artificial Intelligence in Healthcare and Education
- Topic Modeling
- Blood Pressure and Hypertension Studies
- Hemodynamic Monitoring and Therapy
- ECG Monitoring and Analysis
- COVID-19 Clinical Research Studies
- Explainable Artificial Intelligence (XAI)
- Cardiac Structural Anomalies and Repair
- Cardiac Arrest and Resuscitation
- AI in cancer detection
- Cardiac, Anesthesia and Surgical Outcomes
- Synthesis of Indole Derivatives
- Long-Term Effects of COVID-19
- Context-Aware Activity Recognition Systems
- Mechanical Circulatory Support Devices
- Chemical Synthesis and Reactions
- COVID-19 and healthcare impacts
- Acute Myocardial Infarction Research
- Ionic liquids properties and applications
- Electronic Health Records Systems
- Heart Rate Variability and Autonomic Control
- Artificial Intelligence in Healthcare
- Cardiovascular Health and Risk Factors
University of Wisconsin–Madison
2023-2024
Texas A&M University
2019-2023
Mitchell Institute
2021
Abilene Christian University
2013-2014
Acute myocardial infarction (AMI) complicated by cardiogenic shock is associated with substantial morbidity and mortality. Although intravascular microaxial left ventricular assist devices (LVADs) provide greater hemodynamic support as compared intra-aortic balloon pumps (IABPs), little known about clinical outcomes LVAD use in practice.To examine among patients undergoing percutaneous coronary intervention (PCI) for AMI treated mechanical circulatory (MCS) devices.A propensity-matched...
Accurate prediction of adverse outcomes after acute myocardial infarction (AMI) can guide the triage care services and shared decision-making, novel methods hold promise for using existing data to generate additional insights.To evaluate whether contemporary machine learning facilitate risk by including a larger number variables identifying complex relationships between predictors outcomes.This cohort study used American College Cardiology Chest Pain-MI Registry identify all AMI...
<h3>Importance</h3> Mechanical circulatory support (MCS) devices, including intravascular microaxial left ventricular assist devices (LVADs) and intra-aortic balloon pumps (IABPs), are used in patients who undergo percutaneous coronary intervention (PCI) for acute myocardial infarction (AMI) complicated by cardiogenic shock despite limited evidence of their clinical benefit. <h3>Objective</h3> To examine trends the use MCS among underwent PCI AMI with shock, hospital-level variation, factors...
Severe acute respiratory syndrome virus (SARS-CoV-2) has infected millions of people worldwide. Our goal was to identify risk factors associated with admission and disease severity in patients SARS-CoV-2.
Large language models (LLMs) excel at answering knowledge-based questions. Many aspects of blood banking and transfusion medicine involve no direct patient care require only knowledge judgment. We hypothesized that public LLMs could perform such tasks with accuracy precision.We presented three sets to publicly-available (Bard, GPT-3.5, GPT-4). The first was review short case presentations then decide if a red cell indicated. second task answer set consultation questions common in clinical...
Artificial intelligence and large language models (LLMs) have emerged as potentially disruptive technologies in healthcare. In this study GPT-3.5, an accessible LLM, was assessed for its accuracy reliability performing guideline-based evaluation of neuraxial bleeding risk hypothetical patients on anticoagulation medication. The also explored the impact structured prompt guidance LLM's performance.
Severe acute respiratory syndrome virus (SARS-CoV-2) has infected millions of people worldwide. Our goal was to identify risk factors associated with admission and disease severity in patients SARS-CoV-2. This an observational, retrospective study based on real-world data for 7,995 SARS-CoV-2 from a clinical repository. Yale New Haven Health (YNHH) is five-hospital academic health system serving diverse patient population community teaching facilities both urban suburban areas. The included...
Blood pressure monitoring is an essential component of hypertension management and in the prediction associated comorbidities. a dynamic vital sign with frequent changes throughout given day. Capturing blood remotely frequently (also known as ambulatory monitoring) has traditionally been achieved by measuring at discrete intervals using inflatable cuff. However, there growing interest developing cuffless system to measure continuously. One such approach utilizing bioimpedance sensors build...
Cardiovascular disorders cause nearly one in three deaths the United States. Short- and long-term care for these is often determined short-term settings. However, decisions are made with minimal longitudinal data. To overcome this bias towards data from acute settings, improved monitoring cardiovascular patients needed. Longitudinal provides a more comprehensive picture of patient health, allowing informed decision making. This work surveys sensing machine learning field remote health...
An environmentally friendly method for alkylating aromatic compounds with simple alcohols in the presence of a catalytic amount indium(III) triflate is reported. Ionic liquids are used as solvents and energy-efficient heating provided by microwave radiation. Good yields obtained benzyl, secondary, tertiary alcohols. Simple primary not effective agents under these conditions. With alcohols, activated such toluene anisole must be to obtain good yields. The catalyst, which immobilized...
Activity recognition in wearable computing faces two key challenges: i) activity characteristics may be context-dependent and change under different contexts or situations; ii) unknown activities occur from time to time, requiring flexibility adaptability of the algorithm. We develop a context-aware mixture deep models termed {\alpha}-\b{eta} network coupled with uncertainty quantification (UQ) based upon maximum entropy enhance human performance. improve accuracy F score by 10% identifying...
Observational medical data present unique opportunities for analysis of outcomes and treatment decision making. However, because these datasets do not contain the strict pairing randomized control trials, matching techniques are to draw comparisons among patients. A key limitation such is verification that variables used model making also relevant in identifying risk major adverse events. This article explores a deep mixture experts approach jointly learn how match patients events Although...
This paper introduces a sparse embedding for electronic health record (EHR) data in order to predict hospital admission. We use k-sparse autoencoder embed the original registry into much lower dimension, with sparsity as goal. Then, t-SNE is used show of each patient's 2D plot. then demonstrate predictive accuracy different existing machine learning algorithms. Our performs competitively against and traditional vectors an AUROC 0.878. In addition, we expressive power our embedding, i.e....
Visual summarization of clinical data collected on patients contained within the electronic health record (EHR) may enable precise and rapid triage at time patient presentation to an emergency department (ED). The process is critical in appropriate allocation resources anticipating eventual disposition, typically admission hospital or discharge home. EHR are high-dimensional complex, but offer opportunity discover characterize underlying data-driven phenotypes. These phenotypes will...
In this study, Hurley et al [1][1] examined whether the use of GPT-3 artificial intelligence could improve ability clinicians to apply ASRA Pain Medicine anticoagulation guidelines hypothetical patients. The baseline model without context produced an area under receiver operating
Abstract Background Bleeding is a complication of percutaneous coronary intervention (PCI), leading to significant morbidity, mortality, and cost. Existing risk models produce single estimate bleeding anchored at point in time do not update estimates as clinical information emerges, despite the dynamic nature risk. Objective We sought develop that over time, incorporating evolving information, demonstrate updated predictive performance. Methods Using data available from National...
Clinical notes contain a large amount of clinically valuable information that is ignored in many clinical decision support systems due to the difficulty comes with mining information. Recent work has found success leveraging deep learning models for prediction outcomes using notes. However, these fail provide relevant and interpretable clinicians can utilize informed care. In this work, we augment popular convolutional model an attention mechanism apply it unstructured ICU readmission...
Cardiovascular disorders account for nearly 1 in 3 deaths the United States. Care these are often determined during visits to acute care facilities, such as hospitals. While length of stay settings represents just a small proportion patients' lives, they disproportionately large amount decision making. To overcome this bias towards data from settings, there is need longitudinal monitoring patients with cardiovascular disorders. Longitudinal can provide more comprehensive picture patient...
Cardiogenic shock (CS) is a deadly and complicated illness. Despite extensive research into its treatment, mortality remains high has not decreased over time. Patients suffering from CS are highly heterogeneous. Developing an understanding of phenotypes among these patients crucial for this disease appropriate treatments individual patients. In work, we develop deep mixture experts approach to jointly find with while simultaneously estimating their risk in-hospital mortality. This model...
Abstract In(O‐Tf) 3 immobilized in a water‐insoluble ionic liquid is shown to be highly efficient catalyst for the alkylation of aromatic hydrocarbons with benzyl and secondary alcohols.