A5012511062- Sepsis Diagnosis and Treatment
- Machine Learning in Healthcare
- Hemodynamic Monitoring and Therapy
- Respiratory Support and Mechanisms
- COVID-19 diagnosis using AI
- Healthcare Technology and Patient Monitoring
- Heart Rate Variability and Autonomic Control
- Electronic Health Records Systems
- Intensive Care Unit Cognitive Disorders
- ECG Monitoring and Analysis
- Non-Invasive Vital Sign Monitoring
- Neonatal Respiratory Health Research
- Time Series Analysis and Forecasting
- Lung Cancer Diagnosis and Treatment
- Radiomics and Machine Learning in Medical Imaging
- Artificial Intelligence in Healthcare and Education
- Asthma and respiratory diseases
- Chronic Obstructive Pulmonary Disease (COPD) Research
- Parkinson's Disease Mechanisms and Treatments
- Data Visualization and Analytics
- Immune Response and Inflammation
- EEG and Brain-Computer Interfaces
- Thermal Regulation in Medicine
- Emergency and Acute Care Studies
- Clusterin in disease pathology
Emory University
2019-2025
Duke University
2024-2025
Georgia Institute of Technology
2020-2025
Children's Healthcare of Atlanta
2025
Collaborative Research Group
2021-2024
Cohort (United Kingdom)
2021-2024
Siemens Healthcare (United States)
2024
The Wallace H. Coulter Department of Biomedical Engineering
2022-2024
Royal Brompton Hospital
2023
Chelsea and Westminster Hospital
2023
The National COVID Cohort Collaborative (N3C) is a centralized, harmonized, high-granularity electronic health record repository that the largest, most representative COVID-19 cohort to date. This multicenter data set can support robust evidence-based development of predictive and diagnostic tools inform clinical care policy.
<h3>Importance</h3> Discrepancies in oxygen saturation measured by pulse oximetry (Spo<sub>2</sub>), when compared with arterial (Sao<sub>2</sub>) blood gas (ABG), may differentially affect patients according to race and ethnicity. However, the association of these disparities health outcomes is unknown. <h3>Objective</h3> To examine racial ethnic discrepancies between Sao<sub>2</sub>and Spo<sub>2</sub>measures their associations clinical outcomes. <h3>Design, Setting, Participants</h3> This...
We used artificial intelligence to develop a novel algorithm using physiomarkers predict the onset of severe sepsis in critically ill children.Observational cohort study.PICU.Children age between 6 and 18 years old.None.Continuous minute-by-minute physiologic data were available for total 493 children admitted tertiary care PICU over an 8-month period, 20 whom developed sepsis. Using alert time stamp generated by electronic screening as reference point, we studied up 24 prior hours...
Atrial fibrillation (AF) is a major cause of hospitalization and death in the United States. Moreover, as average age individuals increases around world, early detection diagnosis AF become even more pressing. In this paper, we introduce novel deep learning architecture for normal sinus rhythm, AF, other abnormal rhythms, noise.We have demonstrated through systematic approach many hyperparameters, input sets, optimization methods that yielded influence both training time performance...
Established guidelines describe minimum requirements for reporting algorithms in healthcare; it is equally important to objectify the characteristics of ideal that confer maximum potential benefits patients, clinicians, and investigators. We propose a framework algorithms, including 6 desiderata: explainable (convey relative importance features determining outputs), dynamic (capture temporal changes physiologic signals clinical events), precise (use high-resolution, multimodal data aptly...
Objectives: To develop the first core Critical Care Data Dictionary (C2D2) with common data elements (CDEs) to characterize critical illness and injuries. Design: Group consensus process using modified Delphi approach. Setting: Electronic surveys in-person meetings. Subjects: A multidisciplinary workgroup of clinicians researchers expertise in care critically ill injured. Interventions: The was divided into domain CDE portions each composed two item generation rounds one reduction/refinement...
A complicated clinical course for critically ill patients admitted to the intensive care unit (ICU) usually includes multiorgan dysfunction and subsequent death. Owing heterogeneity, complexity, unpredictability of disease progression, ICU patient is challenging. Identifying predictors courses mortality at early stages recognizing trajectory from vast array longitudinal quantitative data difficult. Therefore, we attempted perform a meta-analysis previously published gene expression datasets...
We present an interpretable machine learning algorithm called ‘eARDS’ for predicting ARDS in ICU population comprising COVID-19 patients, up to 12-hours before satisfying the Berlin clinical criteria. The analysis was conducted on data collected from Intensive care units (ICU) at Emory Healthcare, Atlanta, GA and University of Tennessee Health Science Center, Memphis, TN Cerner ® Facts Deidentified Database, a multi-site EMR database. participants consisted adults over 18 years age. Clinical...
Abstract With the recent COVID-19 pandemic, healthcare systems all over world are struggling to manage massive increase in emergency department (ED) visits. This has put an enormous demand on medical professionals. Increased wait times ED increases risk of infection transmission. In this work we present open-source, low cost, off-body system assist automatic triage patients based widely available hardware. The initially focuses two symptoms fever and cyanosis. use visible far-infrared...
This paper presents a novel method for hierarchical analysis of machine learning algorithms to improve predictions at risk patients, thus further enabling prompt therapy. Specifically, we develop multi-layer approach analyze continuous, high-frequency data. We illustrate the capabilities this early identification patients sepsis, potentially life-threatening complication an infection, using (minute-by-minute) physiological data collected from bedside monitors. In our cohort 586 model...
Heart failure (HF) exacerbations, characterized by pulmonary congestion and breathlessness, require frequent hospitalizations, often resulting in poor outcomes. Current methods for tracking lung fluid respiratory distress are unable to produce continuous, holistic measures of cardiopulmonary health. We present a multimodal sensing system that captures bioimpedance spectroscopy (BIS), multi-channel sounds from four contact microphones, multi-frequency impedance pneumography (IP), temperature,...
Detection of atrial fibrillation (AF) from electrocardiogram (ECG) recordings is one the prevailing challenges in field cardiac computing.The task PhysioNet/Computing Cardiology 2017 challenge to distinguish AF rhythms non-AF using a short single lead ECG recording.In this study, we analyzed 62 time and frequency-domain, linear, nonlinear features discriminate four classes, viz., normal sinus rhythm, AF, noisy, or other rhythm.The feature space dimension was reduced 37 Genetic Algorithm...
Real-time analysis of streaming physiological data to identify earlier abnormal conditions is an important aspect precision medicine. However, open-source systems supporting this workflow are lacking. In paper, we present PhysOnline, a pipeline built on the Apache Spark platform ingest for online feature extraction and machine learning. We consider scalability factors horizontal deployment support growing requirements. further integrate real-time extraction, including pattern recognition...
Sepsis is a life-threatening condition with high mortality rates. Early detection and treatment are critical to improving outcomes. Our primary objective was develop artificial intelligence capable of predicting sepsis earlier using minimal set streaming physiological data in real time.A total 29,552 adult patients were admitted the intensive care unit across five regional hospitals Memphis, Tenn, over 18 months from January 2017 July 2018. From these, 5,958 selected after filtering for...
The SARS-CoV-2 virus has caused tremendous healthcare burden worldwide. Our focus was to develop a practical and easy-to-deploy system predict the severe manifestation of disease in patients with COVID-19 an aim assist clinicians triage treatment decisions. proposed predictive algorithm is trained artificial intelligence-based network using 8,427 patient records from four systems. model provides severity risk score along likelihoods various clinical outcomes, namely ventilator use mortality....
Multiple organ dysfunction syndrome (MODS) disproportionately drives morbidity and mortality among critically ill patients. However, we lack a comprehensive understanding of its pathobiology. Identification genes associated with persistent MODS trajectory may shed light on underlying biology allow for accurate prediction those at-risk.
Abstract Objective Common data models provide a standard means of describing for artificial intelligence (AI) applications, but this process has never been undertaken medications used in the intensive care unit (ICU). We sought to develop common model (CDM) ICU standardize medication features needed support future AI efforts. Materials and Methods A 9-member, multi-professional team clinicians experts conducted 5-round modified Delphi employing conference calls, web-based communication,...
Abstract Septic shock is a devastating health condition caused by uncontrolled sepsis. Advancements in high-throughput sequencing techniques have increased the number of potential genetic biomarkers under review. Multiple markers and functional pathways play part development progression pediatric septic shock. We identified 53 differentially expressed using gene expression data sampled from 181 patients admitted to intensive care unit within first 24 hours their admission. The signatures...
Sepsis is a life-threatening condition, caused by the body’s extreme response to an infection. In United States, 1.7 million cases of sepsis occur annually, resulting in 265,000 deaths. Delayed diagnosis and treatment are associated with higher mortality rates. An exponential rise availability medical data has allowed for development sophisticated machine learning algorithms predict earlier than onset. However, these models often underperform, as training retrospective do not fully capture...
Point-of-care screening tools are essential to expedite patient care and decrease reliance on slow diagnostic (e.g., microbial cultures) identify pathogens their associated antibiotic resistance. Analysis of volatile organic compounds (VOC) emitted from biological media has seen increased attention in recent years as a potential non-invasive procedure. This work explores the use solid phase micro-extraction (SPME) ambient plasma ionization mass spectrometry (MS) rapidly acquire VOC...
Abstract Background Identifying patterns within ICU medication regimens may help artificial intelligence algorithms to better predict patient outcomes; however, machine learning methods incorporating medications require further development, including standardized terminology. The Common Data Model for Intensive Care Unit (ICU) Medications (CDM-ICURx) provide important infrastructure clinicians and researchers support analysis of medication-related outcomes healthcare costs. Using an...