A5100718929- Radiomics and Machine Learning in Medical Imaging
- AI in cancer detection
- Dental Radiography and Imaging
- Forensic Anthropology and Bioarchaeology Studies
- Breast Cancer Treatment Studies
- Natural Antidiabetic Agents Studies
- Diet, Metabolism, and Disease
- Orthodontics and Dentofacial Orthopedics
- Nanoparticle-Based Drug Delivery
- Nasolacrimal Duct Obstruction Treatments
- Robotics and Automated Systems
- Wound Healing and Treatments
- Winter Sports Injuries and Performance
- Diet and metabolism studies
- COVID-19 diagnosis using AI
- Cancer Treatment and Pharmacology
- Bat Biology and Ecology Studies
- Seaweed-derived Bioactive Compounds
- Spectroscopy Techniques in Biomedical and Chemical Research
- Dental Erosion and Treatment
- Multiple Myeloma Research and Treatments
- Surfactants and Colloidal Systems
- Cytomegalovirus and herpesvirus research
- Herpesvirus Infections and Treatments
- Colorectal Cancer Screening and Detection
Asan Medical Center
2020-2022
University of Ulsan
2020-2022
Ulsan College
2020-2022
Gwangju Institute of Science and Technology
2020
Hanpoong Pharm (South Korea)
2014
Seoul National University
2012
Sungkyunkwan University
2003
Abstract Purpose: The rationale for developing an alternative paclitaxel formulation concerns Cremophor EL-related side effects, and a novel delivery system might augment its therapeutic efficacy. Genexol-PM is polymeric micelle formulated free of EL. A phase I study was performed to determine the maximum tolerated dosage, dose-limiting toxicities, pharmacokinetic profile in patients with advanced, refractory malignancies. Experimental Design: Twenty-one were entered into study. i.v....
The human cytomegalovirus (HCMV) clinical strain Toledo and the attenuated AD169 exhibit a striking difference in pathogenic potential cell tropism. virulent genome contains 15-kb segment, which is present all strains but absent from genome. differences between 2 are thought to be associated with this additional segment. Cytokines induced during viral infection play major roles regulation of cellular interactions involving cells immune inflammatory systems consequently determine outcome...
Abstract Fast and accurate confirmation of metastasis on the frozen tissue section intraoperative sentinel lymph node biopsy is an essential tool for critical surgical decisions. However, diagnosis by pathologists difficult within time limitations. Training a robust deep learning model also owing to limited number datasets with high quality labels. To overcome these issues, we validated effectiveness transfer from CAMELYON16 improve performance convolutional neural network (CNN)-based...
Type 2 diabetes is a serious problem for developed countries. Prevention of prediabetes progression to type with the use natural products appears cost-effective solution. Previously we showed that enzymatically digested low molecular weight chitosan-oligosaccharide (MW) below 1,000 Da (GO2KA1) has potential hyperglycemia management. In this study evaluated effect long-term supplementation GO2KA1 on using db/db mice model. Additionally, sucrase and glucoamylase activities expression, same...
Recent MLLMs have shown emerging visual understanding and reasoning abilities after being pre-trained on large-scale multimodal datasets. Unlike pre-training, where receive rich visual-text alignment, instruction-tuning is often text-driven with weaker supervision, leading to the degradation of causing forgetting. Existing approaches, such as direct fine-tuning continual learning methods, fail explicitly address this issue, compressing representations prioritizing task alignment over...
Abstract The quality of cephalometric analysis depends on the accuracy delineating landmarks in orthodontic and maxillofacial surgery. Due to extensive number landmarks, each costs orthodontists considerable time per patient, leading fatigue inter- intra-observer variabilities. Therefore, we proposed a fully automated cephalometry with cascade convolutional neural net (CNN). One thousand x-ray images (2 k × 3 k) pixel were used. dataset was split into training, validation, test sets as...
Abstract Realistic image generation is valuable in dental medicine, but still challenging for generative adversarial networks (GANs), which require large amounts of data to overcome the training instability. Thus, we generated lateral cephalogram X-ray images using a deep-learning-based progressive growing GAN (PGGAN). The quality was evaluated by three methods. First, signal-to-noise ratios real/synthesized images, at posterior arch region first cervical vertebra, showed no statistically...
The purpose of this study was to investigate the accuracy one-step automated orthodontic diagnosis skeletodental discrepancies using a convolutional neural network (CNN) and lateral cephalogram images with different qualities from nationwide multi-hospitals.Among 2,174 cephalograms, 1,993 cephalograms two hospitals were used for training internal test sets 181 eight other an external set. They divided into three classification groups according anteroposterior skeletal (Class I, II, III),...
Purpose Assessing the status of metastasis in sentinel lymph nodes (SLNs) by pathologists is an essential task for accurate staging breast cancer. However, histopathological evaluation SLNs a pathologist not easy and tedious time-consuming task. The purpose this study to review challenge competition (HeLP 2018) develop automated solutions classification metastases hematoxylin eosinâstained frozen tissue sections cancer patients. Materials Methods A total 297 digital slides were obtained...
Assessing the metastasis status of sentinel lymph nodes (SLNs) for hematoxylin and eosin-stained frozen tissue sections by pathologists is an essential but tedious time-consuming task that contributes to accurate breast cancer staging. This study aimed review a challenge competition (HeLP 2019) development automated solutions classifying patients.A total 524 digital slides were obtained from SLN sections: 297 (56.7%) Asan Medical Center (AMC) 227 (43.4%) Seoul National University Bundang...
The purpose of this study was to analyze echolocation patterns a bat and apply the results identification using u1trasound data other ecologica1 areas bat.Rhinolophμs ferrumequinU1η， 뼈αis peα andPψ빼ellμs abramus are used in study.Captured bats Gyeongsangnarn-do 때d Gyeongsangbuk-do from August October, 2010.According analysis on by three species bats, Rh i.ηolophus ferrumequi.ηum'st따ts with FM signa1s ends sign따 s via CF sign떠s.Compared Myotis petax Pψ istrellus abramus, it has longer pulse...
Met-myoglobin is a major component related to meat discoloration, and it gradually accumulates over time after the slaughtered. Recently, studies have been conducted observe changes in composition of met-myoglobin along with its storage using Diffuse Reflectance Spectroscopy(DRS). DRS an optical technique that simple can estimate chromophores without damaging sample. However, since requires high resolution complicated fitting process, difficult apply mobile environment. Therefore, purpose...