A5026474928- Artificial Intelligence in Healthcare and Education
- Radiomics and Machine Learning in Medical Imaging
- Acupuncture Treatment Research Studies
- Sleep and related disorders
- Pharmacological Effects of Natural Compounds
- Complementary and Alternative Medicine Studies
- Cardiac, Anesthesia and Surgical Outcomes
- COVID-19 diagnosis using AI
- Hemodynamic Monitoring and Therapy
- Healthcare and Venom Research
- Traditional Chinese Medicine Studies
- Heart Rate Variability and Autonomic Control
- Non-Invasive Vital Sign Monitoring
- Pain Management and Placebo Effect
- Machine Learning in Healthcare
- Intensive Care Unit Cognitive Disorders
- Scientific Computing and Data Management
- Phytochemistry and Biological Activities
- Rheumatoid Arthritis Research and Therapies
- Anesthesia and Neurotoxicity Research
- Radiology practices and education
- Chemotherapy-induced organ toxicity mitigation
- Research Data Management Practices
- Medicinal Plants and Bioactive Compounds
- Sleep and Work-Related Fatigue
University of Duisburg-Essen
2025
TU Dortmund University
2025
Seoul National University Hospital
2022-2024
Seoul National University
2024
Republic of Korea Army
2022
Kyung Hee University
2012-2022
Predicting in-hospital cardiac arrest in patients admitted to an intensive care unit (ICU) allows prompt interventions improve patient outcomes. We developed and validated a machine learning-based real-time model for predictions using electrocardiogram (ECG)-based heart rate variability (HRV) measures. The HRV measures, including time/frequency domains nonlinear were calculated from 5 min epochs of ECG signals ICU patients. A light gradient boosting (LGBM) algorithm was used develop the...
We present the INSPIRE dataset, a publicly available research dataset in perioperative medicine, which includes approximately 130,000 surgical operations at an academic institution South Korea over ten-year period between 2011 and 2020. This comprehensive patient characteristics such as age, sex, American Society of Anesthesiologists physical status classification, diagnosis, procedure code, department, type anaesthesia. The also vital signs operating theatre, general wards, intensive care...
Foundation Models that are capable of processing and generating multi-modal data have transformed AI's role in medicine. However, a key limitation their reliability is hallucination, where inaccurate or fabricated information can impact clinical decisions patient safety. We define medical hallucination as any instance which model generates misleading content. This paper examines the unique characteristics, causes, implications hallucinations, with particular focus on how these errors...
Background The COVID-19 pandemic has limited daily activities and even contact between patients primary care providers. This makes it more difficult to provide adequate services, which include connecting an appropriate medical specialist. A smartphone-compatible artificial intelligence (AI) chatbot that classifies patients’ symptoms recommends the specialty could a valuable solution. Objective In order establish contactless method of recommending specialty, this study aimed construct deep...
O6-methylguanine-DNA methyl transferase (MGMT) methylation prediction models were developed using only small datasets without proper external validation and achieved good diagnostic performance, which seems to indicate a promising future for radiogenomics. However, the performance was not reproducible numerous research teams when larger dataset in RSNA-MICCAI Brain Tumor Radiogenomic Classification 2021 challenge. To our knowledge, there has been no study regarding of MGMT large-scale...
Bee venom (BV) is one of the alternative medicines that have been widely used in treatment chronic inflammatory diseases. We previously demonstrated BV induces immune tolerance by increasing population regulatory T cells (Tregs) disorders. However, major component and how it regulates response not elucidated. investigated whether bee phospholipase A2 (bvPLA2) exerts protective effects are mediated via Tregs OVA-induced asthma model. bvPLA2 was administered intraperitoneal injection into...
ChatGPT (Open AI) is a state-of-the-art artificial intelligence model with potential applications in the medical fields of clinical practice, research, and education.This study aimed to evaluate as an educational tool college acupuncture programs, focusing on its ability support students learning point selection, treatment planning, decision-making.We collected case studies published Acupuncture Medicine between June 2022 May 2023. Both ChatGPT-3.5 ChatGPT-4 were used generate suggestions...
Artificial intelligence (AI) has become commonplace in solving routine everyday tasks. Because of the exponential growth medical imaging data volume and complexity, workload on radiologists is steadily increasing. AI been shown to improve efficiency image generation, processing, interpretation, various such models have developed across research laboratories worldwide. However, very few these, if any, find their way into clinical use, a discrepancy that reflects divide between successful...
Stemona tuberosa has long been used in Korean and Chinese medicine to ameliorate various lung diseases such as pneumonia bronchitis. However, it not yet proven that positive effects on inflammation. extract (ST) was orally administered C57BL/6 mice 2 hr before exposure CS for weeks. Twenty-four hours after the last exposure, were sacrificed investigate changes expression of cytokines tumor necrosis factor-alpha (TNF-α) interleukin-6 (IL-6), chemokines keratinocyte-derived chemokine (KC)...
Ventilation should be assisted without asynchrony or cardiorespiratory instability during anesthesia emergence until sufficient spontaneous ventilation is recovered. In this multicenter cohort study, we develop and validate a reinforcement learning-based Artificial Intelligence model for control Emergence (AIVE) from general anesthesia. Ventilatory hemodynamic parameters 14,306 surgical cases at an academic hospital between 2016 2019 are used training internal testing of the model. The...
Purpose: Sleep quality among military service members is important for enhancing their capabilities and preventing psychiatric problems. We aimed to explore the association of dietary behaviors with poor sleep increased risk obstructive apnea (OSA) in men on active duty. Patients Methods: A large-scale multi-site cross-sectional survey was conducted five units Republic Korea's army. Poor OSA were defined using Pittsburgh index (PSQI) Berlin Questionnaire, respectively. Information behaviors,...
Abstract Delirium can result in undesirable outcomes including increased length of stays and mortality patients admitted to the intensive care unit (ICU). Dexmedetomidine has emerged for delirium prevention these patients; however, optimal dosing is challenging. A reinforcement learning-based Artificial Intelligence model (AID) proposed optimize dexmedetomidine dosing. The was developed internally validated using 2416 (2531 ICU admissions) externally on 270 (274 admissions). estimated...
Pattern identification (PI) is a diagnostic method used in Traditional East Asian medicine (TEAM) to select appropriate and personalized acupuncture points herbal medicines for individual patients. Developing reproducible PI model using clinical information important as it would reflect the actual setting improve effectiveness of TEAM treatment. In this paper, we suggest novel deep learning-based with feature extraction autoencoder k-means clustering through cross-sectional study sleep...
Background: Use of endotracheal tubes (ETTs) with appropriate size and depth can help minimize intubation-related complications in pediatric patients. Existing age-based formulae for selecting the optimal ETT present several inaccuracies. We developed a machine learning model that predicts ETTs patients using demographic data, enabling clinical applications. Methods: Data from 37,057 younger than 12 years who underwent general anesthesia intubation were retrospectively analyzed. Gradient...
Medical Decision-Making (MDM) is a multi-faceted process that requires clinicians to assess complex multi-modal patient data patient, often collaboratively. Large Language Models (LLMs) promise streamline this by synthesizing vast medical knowledge and health data. However, single-agent are ill-suited for nuanced contexts requiring adaptable, collaborative problem-solving. Our MDAgents addresses need dynamically assigning collaboration structures LLMs based on task complexity, mimicking...