A5007370340- Acute Kidney Injury Research
- Machine Learning in Healthcare
- Artificial Intelligence in Healthcare
- Artificial Intelligence in Healthcare and Education
- Sepsis Diagnosis and Treatment
- Trauma, Hemostasis, Coagulopathy, Resuscitation
- Bone health and osteoporosis research
- Electrolyte and hormonal disorders
- Dialysis and Renal Disease Management
- Muscle and Compartmental Disorders
- Renal and Vascular Pathologies
- Chronic Kidney Disease and Diabetes
- Cardiac Arrest and Resuscitation
- Renal function and acid-base balance
- Bone health and treatments
- Hemodynamic Monitoring and Therapy
- Long-Term Effects of COVID-19
- Liver Disease and Transplantation
- Radiomics and Machine Learning in Medical Imaging
- Cell Image Analysis Techniques
- Machine Fault Diagnosis Techniques
- Organ Donation and Transplantation
- Organ Transplantation Techniques and Outcomes
- demographic modeling and climate adaptation
- Pharmacological Effects and Toxicity Studies
Fred Hutch Cancer Center
2024-2025
University of Kentucky
2020-2024
The definition of sepsis-associated acute kidney injury (SA-AKI) was updated in 2023. This study aims to describe the epidemiology SA-AKI using consensus and evaluate clinical outcomes. a retrospective cohort analysis conducted at two academic medical centers. Adult patients admitted intensive care units (ICU) between 2010 2022 were included categorized as SA-AKI, sepsis alone, or AKI alone. further classified by time onset (early < 2 days from diagnosis vs. late 2–7 following diagnosis)...
Hypophosphatemia is commonly observed in patients receiving continuous KRT. Patients who develop hypophosphatemia may be at risk of respiratory and neuromuscular dysfunction therefore subject to prolongation ventilator support. We evaluated the association phosphate-containing versus phosphate-free KRT solutions with dependence critically ill KRT.Our study was a single-center, retrospective, pre-post cohort adult mechanical ventilation during their intensive care unit stay. Zero-inflated...
<title>Abstract</title> <bold>Background</bold> Western blot (WB) is the gold standard for herpes simplex virus (HSV-1/HSV-2) serology but requires manual interpretation by trained laboratory personnel, which time-consuming and variable. Automating this process could improve efficiency diagnostic consistency. This study presents BlotDx, a deep learning tool automated of HSV WB assays. <bold>Methods</bold> BlotDx uses two-stage approach: (1) instance segmentation or object detection to...
Surviving critical illness and acute kidney injury (AKI) carries a burden of morbidity mortality that may be affected by post-discharge care. However, only 5% to 10% AKI survivors receive nephrology care, despite observational studies suggesting an association between early follow-up improved outcomes.1Siew E.D. Peterson J.F. Eden S.K. et al.Outpatient referral rates after injury.J Am Soc Nephrol. 2012; 23: 305-312Crossref PubMed Scopus (133) Google Scholar, 2Harel Z. Wald R. Bargman J.M....
Introduction Over the past decade, artificial intelligence (AI) through machine learning (ML) has been increasingly used in health care applications. The use of AI/ML to enhance provision continuous kidney replacement therapy (CKRT) is an area active investigation because growing clinical needs critically ill patients and abundance multimodal data from electronic records (EHRs) CKRT itself. Importantly, best practices have not sufficiently standardized, so there considerable heterogeneity...
Introduction: In 2017, the Centers for Medicare and Medicaid Services allowed survivors of hospitalized acute kidney injury requiring dialysis (AKI-D) who were ambulatory still dependent on hemodialysis (HD) to receive treatment in outpatient facilities. This policy change generated ongoing need improve AKI-D care setting. Methods: Quality improvement study adult patients admitted an HD unit with diagnosis AKI-D. We developed a protocol manage these that included: (a) multidisciplinary...
Since January 2017, patients with acute kidney injury requiring dialysis (AKI-D) can be discharged to outpatient centers for continued hemodialysis (HD) support. We aimed examine the rate of recovery, time and hospitalization-related clinical parameters associated recovery in AKI-D.Single-center prospective cohort study.111 adult who were admitted University Kentucky Hospital, experienced AKI-D, need HD.Hospitalization-related evaluated.Kidney as a composite being alive no longer HD or other...
Key Points IL-17A was higher in patients with AKI versus without during hospitalization and up to 1-year postdischarge. progression of kidney disease but not independently associated subsequent disease. Background is increased mortality new or progressive CKD. Inflammatory cells play an important role acute organ injury. We previously demonstrated that serum levels were significantly elevated critically ill hospital mortality. hypothesize are hospitalized at diagnosis, sustained elevation...
With the rapid accumulation of electronic health record (EHR) data, deep learning (DL) models have exhibited promising performance on patient risk prediction. Recent advances also demonstrated effectiveness knowledge graphs (KG) in providing valuable prior for further improving DL model performance. However, it is still unclear how KG can be utilized to encode high-order relations among clinical concepts and make full use encoded concept solve real-world healthcare problems interpret...
Rapid accumulation of temporal Electronic Health Record (EHR) data and recent advances in deep learning have shown high potential precisely timely predicting patients' risks using AI. However, most existing risk prediction approaches ignore the complex asynchronous irregular problems real-world EHR data. This paper proposes a novel approach called Knowledge-guIded Time-aware LSTM (KIT-LSTM) for continuous mortality predictions EHR. KIT-LSTM extends with two time-aware gates knowledge-aware...
Oral bisphosphonates are the primary medication for osteoporosis, but concerns exist regarding potential bone-quality changes or low-energy fractures. This cross-sectional study used artificial intelligence methods to analyze relationships among bisphosphonate treatment duration, a wide variety of parameters, and Fourier transform infrared spectroscopy histomorphometry quantified parameters in 67 osteoporotic women treated with oral 1 14 years. Artificial established two models relating...
ABSTRACT Piperacillin-tazobactam (TZP), cefepime (FEP), or meropenem (MEM) and vancomycin (VAN) are commonly used in combination for sepsis. Studies have shown an increased risk of acute kidney injury (AKI) with TZP VAN compared to FEP MEM. guidelines recommend area under the curve (AUC) monitoring over trough (Tr) minimize AKI. We investigated association AKI MAKE-30 two strategies when FEP/MEM. Adult patients between 2015 2019 > 72 hours were included. Patients prior within 48 baseline...
Abstract Background Fluid balance gap (FBgap—prescribed vs. achieved) is associated with hospital mortality. Downtime an important quality indicator for the delivery of continuous renal replacement therapy (CRRT). We examined association CRRT downtime FBgap and clinical outcomes including Methods This a retrospective cohort study critically ill adults receiving utilizing both electronic health records (EHR) machine data. was calculated as achieved minus prescribed fluid balance. Downtime, or...
Abstract Rapid accumulation of temporal Electronic Health Record (EHR) data and recent advances in deep learning have shown high potential precisely timely predicting patients’ risks using AI. However, most existing risk prediction approaches ignore the complex asynchronous irregular problems real-world EHR data. This paper proposes a novel approach called Knowledge-guIded Time-aware LSTM (KIT-LSTM) for continuous mortality predictions EHR. KIT-LSTM extends with two time-aware gates...
Many real-world Electronic Health Record (EHR) data contain a large proportion of missing values. Leaving substantial portion information unaddressed usually causes significant bias, leading to invalid conclusions be drawn. On the other hand, training machine learning model with much smaller nearly-complete subset can drastically impact reliability and accuracy inference. Data imputation algorithms that attempt replace meaningful values, inevitably increase variability effect estimates...
Many real-world Electronic Health Record (EHR) data contains a large proportion of missing values. Leaving substantial portion information unaddressed usually causes significant bias, which leads to invalid conclusion be drawn. On the other hand, training machine learning model with much smaller nearly-complete subset can drastically impact reliability and accuracy inference. Data imputation algorithms that attempt replace meaningful values inevitably increase variability effect estimates...
<b><i>Introduction:</i></b> The FDA authorized the emergency use of enhanced hemoadsorption with oXiris in critically ill adult COVID patients respiratory failure or severe disease to reduce inflammation. In this study, we evaluated acute kidney injury (AKI) who were exposed versus not during continuous renal replacement therapy (CRRT). <b><i>Methods:</i></b> Retrospective cohort study AKI requiring CRRT. Exposure was defined as receiving for...