A5066020211- Tuberculosis Research and Epidemiology
- COVID-19 diagnosis using AI
- Lung Cancer Diagnosis and Treatment
- Topic Modeling
- Neuroinflammation and Neurodegeneration Mechanisms
- Mycobacterium research and diagnosis
- S100 Proteins and Annexins
- Photovoltaic System Optimization Techniques
- Alzheimer's disease research and treatments
- Infectious Diseases and Mycology
- Elbow and Forearm Trauma Treatment
- Pleural and Pulmonary Diseases
- Radiomics and Machine Learning in Medical Imaging
- Chemokine receptors and signaling
- Amoebic Infections and Treatments
- Gallbladder and Bile Duct Disorders
- Shoulder Injury and Treatment
- Immune cells in cancer
- Orthopedic Surgery and Rehabilitation
- Solar Radiation and Photovoltaics
- Energy Load and Power Forecasting
- Smart Grid Energy Management
Seoul National University Hospital
2019-2025
Yonsei University
2023-2024
Seoul Metropolitan Government
2021
Severance Hospital
2021
Seoul National University Bundang Hospital
2021
Catholic University of Korea
2021
Pusan National University Yangsan Hospital
2021
Hyosung Corporation (South Korea)
2020
This study aimed to develop an open-source multimodal large language model (CXR-LLaVA) for interpreting chest X-ray images (CXRs), leveraging recent advances in models (LLMs) potentially replicate the image interpretation skills of human radiologists. For training, we collected 592,580 publicly available CXRs, which 374,881 had labels certain radiographic abnormalities (Dataset 1) and 217,699 provided free-text radiology reports 2). After pre-training a vision transformer with Dataset 1,...
Objectives This study aimed to develop a dual-input convolutional neural network (CNN)–based deep-learning algorithm that utilizes both anteroposterior (AP) and lateral elbow radiographs for the automated detection of pediatric supracondylar fracture in conventional radiography, assess its feasibility diagnostic performance. Materials Methods To model, 1266 pairs AP examined between January 2013 December 2017 at single institution were split into training set (1012 pairs, 79.9%) validation...
Background Determining the activity of pulmonary tuberculosis on chest radiographs is difficult. Purpose To develop a deep learning model to identify active radiographs. Materials and Methods Chest were retrospectively gathered from multicenter consecutive cohort with who successfully treated between 2011 2017, along normal enrich negative class. The pretreatment posttreatment labeled as positive classes, respectively. A neural network was trained those calculate probability versus healed...
The importance of neuroinflammation during the ischemic stroke has been extensively studied. role CD4+CD25+ regulatory T (Treg) cells recovery phase have shown infarct size reduction and functional improvement, possibly through mitigation inflammatory immune responses. We aimed to investigate molecular factors involved in microglia-Treg cell communication that result Treg trafficking. First, we observed migration patterns CD8+ (cytotoxic) then searched for chemokines released by activated...
As the relative importance of renewable energy in electric power systems increases, prediction photovoltaic (PV) generation has become a crucial technology, for improving stability operation next-generation systems, such as microgrid and virtual plants (VPP). In order to improve accuracy PV forecasting, fair amount research been applied weather forecast data (to learning process). Despite these efforts, problems forecasting remains challenging since existing methods show limited due...
Rationale: Imaging studies are widely performed when treating Mycobacterium avium complex pulmonary disease (MAC-PD); however, the clinical significance of post-treatment radiographic change is unknown. Objectives: To determine whether a deep neural network trained with tuberculosis could adequately score severity MAC-PD. Then, to examine relationships between and its from baseline long-term prognosis. Methods: We retrospectively collected chest radiographs adult patients MAC-PD treated for...
Purpose: This study aimed to develop an open-source multimodal large language model (CXR-LLAVA) for interpreting chest X-ray images (CXRs), leveraging recent advances in models (LLMs) potentially replicate the image interpretation skills of human radiologists Materials and Methods: For training, we collected 592,580 publicly available CXRs, which 374,881 had labels certain radiographic abnormalities (Dataset 1) 217,699 provided free-text radiology reports 2). After pre-training a vision...