A5026864108- Advanced Radiotherapy Techniques
- Medical Imaging Techniques and Applications
- Radiation Therapy and Dosimetry
- Physics of Superconductivity and Magnetism
- Advanced X-ray and CT Imaging
- Radiation Dose and Imaging
- Radiomics and Machine Learning in Medical Imaging
- ZnO doping and properties
- Breast Cancer Treatment Studies
- Medical Imaging and Analysis
- Magnetic and transport properties of perovskites and related materials
- Analog and Mixed-Signal Circuit Design
- Semiconductor materials and devices
- Ultrasound and Hyperthermia Applications
- Radiation Detection and Scintillator Technologies
- Wireless Networks and Protocols
- IPv6, Mobility, Handover, Networks, Security
- Magnetic properties of thin films
- PARP inhibition in cancer therapy
- Ovarian cancer diagnosis and treatment
- Orthodontics and Dentofacial Orthopedics
- Effects of Radiation Exposure
- Advanced Condensed Matter Physics
- Nonmelanoma Skin Cancer Studies
- Advanced Neural Network Applications
Yonsei University
2007-2025
Samsung (South Korea)
2002-2024
Stanford University
2005-2024
Seoul National University
2002-2024
Korea University
2024
Korea Information System (South Korea)
2024
Kyungpook National University
2021-2023
Gangnam Severance Hospital
2022
LG (South Korea)
2022
Asan Medical Center
2017-2020
Abstract Segmentation of normal organs is a critical and time-consuming process in radiotherapy. Auto-segmentation abdominal has been made possible by the advent convolutional neural network. We utilized U-Net, 3D-patch-based network, added graph-cut algorithm-based post-processing. The inputs were CT images consisting 64 × voxels designed to produce 3D multi-label semantic representing liver, stomach, duodenum, right/left kidneys. datasets for training, validating, testing consisted 80, 20,...
The compressed sensing (CS) technique has been employed to reconstruct CT/CBCT images from fewer projections as it is designed recover a sparse signal highly under-sampled measurements. Since the CT image itself cannot be sparse, variety of transforms were developed make sufficiently sparse. total-variation (TV) transform with local gradient in L1-norm was adopted most cases. This approach, however, which utilizes very information and penalizes weight at constant rate regardless different...
A new self-microemulsifying drug delivery system (SMEDDS) was developed to increase the dissolution rate, solubility, and, ultimately, bioavailability of a poorly water soluble drug, idebenone. Pseudoternary phase diagrams were used evaluate self-microemulsification existence area, and release rate idebenone investigated. The mixtures consisting Labrafac hydro or Labrafil 2609 (HLB values > 4) with surfactant (Labrasol containing 80% Transcutol) cosurfactant (Plurol oleique WL 1173) found be...
Abstract A deep-neural-network (DNN) was successfully used to predict clinically-acceptable dose distributions from organ contours for intensity-modulated radiotherapy (IMRT). To provide the next step in DNN-based plan automation, we propose a DNN that directly generates beam fluence maps and volumetric distributions, without inverse planning. We collected 240 prostate IMRT plans train using distributions. After training done, made 45 synthetic (SPs) generated fluence-maps compared them with...
Abstract This work aims to investigate the clinical feasibility of deep learning-based synthetic CT images for cervix cancer, comparing them MR calculating attenuation (MRCAT). Patient cohort with 50 pairs T2-weighted and from cervical cancer patients was split into 40 training 10 testing phases. We conducted deformable image registration Nyul intensity normalization maximize similarity between as a preprocessing step. The processed were plugged learning model, generative adversarial...
Predicting radiation dermatitis (RD), a common radiotherapy toxicity, is essential for clinical decision-making regarding toxicity management. This prospective study aimed to develop and validate machine-learning model predict the occurrence of grade ≥ 2 RD using thermal imaging in early stages head neck cancer. Thermal images skin surfaces were acquired weekly during radiotherapy. A total 202 used calculate difference map temperature analyze extract features. Changes features treatment...
Background/Aim: Log data from radiation treatment machines can play a crucial role in quality assurance by enabling the recalculation of delivered dose distribution and identification deviations delivery. This article proposes novel method for recalculating carbon-ion therapy using log data. Materials Methods: The proposed approach leverages existing functionality commercial planning systems, thus eliminating need specialized in-house software calculation evaluation. performed tests entail...
Our results suggest that musical training alters the functional anatomy of rapid spectrotemporal processing, resulting in improved behavioral performance along with a more efficient network primarily involving traditional language regions. This finding may have important implications for improving language/reading skills, especially children struggling dyslexia.
Recently, several efforts have been made to develop the deep learning (DL) algorithms for automatic detection and segmentation of brain metastases (BM). In this study, we developed an advanced DL model BM segmentation, especially small-volume BM. From institutional cancer registry, contrast-enhanced magnetic resonance images 65 patients 603 were collected train evaluate our model. Of patients, 12 with 58 assigned test-set performance evaluation. Ground-truth was one radiation oncologist...
Heart toxicity, such as major acute coronary events (ACE), following breast radiation therapy (RT) is of utmost concern. Thus, many studies have been investigating the effect mean heart dose (MHD) and received in sub-structures on toxicity. Most focused thresholds its sub-structures, while few adopted computational methods deep neural networks (DNN) radiomics. This work aims to construct a feature-driven predictive model for ACE after RT.A recently proposed two-step that extracts number...
This study aimed to develop a new approach predict radiation dermatitis (RD) by using the skin dose distribution in actual area of RD occurrence determine predictive grade.
Fiber reinforced polymer (FRP) is one of the best retrofitting materials for blast resistance strengthening due to a number advantages, such as excellent strength self-weight ratio, large fatigue capacity, etc. Presently, many different types FRP models are used in analysis, but they randomly without fundamental scientific reasoning, thereby creating erroneous analytical results. Therefore, this study carried out assess and compare comprehensive finite element model that can properly...
We report technical approaches for enhancing brightness of OLED microdisplays based on Si backplanes (OLEDoS) by changing in white device architectures and incorporating the microcavity technology. In addition, effective strategies further increasing performances OLEDoS panels are suggested.
We aimed to investigate the deliverability of dynamic conformal arc therapy (DCAT) by gantry wobble owing intrinsic inter-segment break Elekta linear accelerator (LINAC) and its adverse influence on dose patient. The DCAT was evaluated according plan parameters, which affect rotation speed resultant positional inaccuracies; number control points rates investigated using treatment machine log files dosimetry devices, respectively. A non-negligible degradation in due observed both devices....
In this paper, we present offline-to-online knowledge distillation (OOKD) for video instance segmentation (VIS), which transfers a wealth of from an offline model to online consistent prediction. Unlike previous methods that have adopted either or model, our single takes advantage both models by distilling knowledge. To transfer correctly, propose query filtering and association (QFA), filters irrelevant queries exact instances. Our KD with QFA increases the robustness feature matching...
In breast cancer radiation therapy, minimizing radiation-related risks and toxicity is vital for improving life expectancy. Tailoring radiotherapy techniques treatment positions can reduce doses to normal organs mitigate treatment-related toxicity. This study entailed a dosimetric comparison of six different external beam whole-breast irradiation in both supine prone positions. We selected fourteen patients, generating plans per patient. assessed target coverage at risk (OAR) evaluate the...
Adjuvant radiation therapy improves the overall survival and loco-regional control in patients with breast cancer. However, radiation-induced heart disease, which occurs after treatment from incidental exposure to cardiac organ, is an emerging challenge. This study aimed generate synthetic contrast-enhanced computed tomography (SCECT) non-contrast CT (NCT) using deep learning (DL) investigate its role contouring substructures. We also determine applicability for a retrospective on...
Abstract We assessed the effects of soluble molecules (supernatants) produced by pro‐ (Th1) and anti‐ (Th2) inflammatory T‐cell lines on capacity adult human CNS‐derived microglia to express or produce selected cell surface that regulate immune reactivity impact tissue protection/repair within CNS. Treatment with supernatants from allo‐antigen myelin basic protein‐specific Th1 augmented expression MHC class II, CD80, CD86, CD40, CD54, enhanced functional antigen‐presenting in a mixed...
A new air interface scheme based on adaptive resource allocation method is proposed for downlink in cellular OFDM systems. The featured as a novel technique high spectral efficiency (i.e. frequency reuse factor approaching one) and power efficiency. In order to reduce the co-channel interference (CCI) improve capacity cell, cell partitioned into three 120/spl deg/ sectors adjacent from different cells are composed virtual which centrally controlled. Assuming knowledge of instantaneous...