- Advanced Sensor and Energy Harvesting Materials
- Video Surveillance and Tracking Methods
- Conducting polymers and applications
- Architecture and Computational Design
- Organ Transplantation Techniques and Outcomes
- Turtle Biology and Conservation
- Data Management and Algorithms
- Electrochemical Analysis and Applications
- Human Pose and Action Recognition
- Advanced Electron Microscopy Techniques and Applications
- Data Mining Algorithms and Applications
- Robotic Path Planning Algorithms
- Tactile and Sensory Interactions
- Advanced Neural Network Applications
- Airway Management and Intubation Techniques
- Liver Disease and Transplantation
- Animal Behavior and Reproduction
- Intensive Care Unit Cognitive Disorders
- Autonomous Vehicle Technology and Safety
- Anomaly Detection Techniques and Applications
- Glycogen Storage Diseases and Myoclonus
- Music Therapy and Health
- Dermatology and Skin Diseases
- Nanoporous metals and alloys
- Hand Gesture Recognition Systems
Soonchunhyang University Hospital
2024
Massachusetts Institute of Technology
2021-2022
IIT@MIT
2021
California Institute of Technology
2010-2020
Korea Advanced Institute of Science and Technology
2013-2019
Catholic University of Korea
2008-2018
Seoul St. Mary's Hospital
2013-2018
Institute for Infocomm Research
2014-2017
Sungshin Women's University
2016
The University of Tokyo
2006
Tactile sensors that can mechanically decouple, and therefore differentiate, various tactile inputs are highly important to properly mimic the sensing capabilities of human skin. Herein, we present an all-solution processable pressure insensitive strain sensor utilizes difference in structural change upon application tensile strain. Under strain, microcracks occur within multiwalled carbon nanotube (MWCNT) network, inducing a large resistance with gauge factor ∼56 at 70% On other hand, under...
Abstract Sensor‐to‐sensor variability and high hysteresis of composite‐based piezoresistive pressure sensors are two critical issues that need to be solved enable their practical applicability. In this work, a sensor composed an elastomer template with uniformly sized arranged pores, chemically grafted conductive polymer film on the surface pores is presented. Compared randomly which had coefficient variation (CV) in relative resistance change 69.65%, our exhibit much higher uniformity CV...
This letter proposes a robust vehicle localization method based on prior point cloud in urban area. The high resolution collected six months ago is provided from Singapore Land Authority around One-north area Singapore, because the data are outdated there many changed aspects of environment such as redrawn road markings, construction areas, and changing tree shapes. In response, this paper novel fusion algorithm particle filter using vertical intensity information for localization. Whereas...
Vehicle following is one of the fundamental functions an autonomous driving system. Detection and recognition tail light signal important to prevent vehicle from rear-end collisions or accidents. Although sensors like acoustic sonar commercialized Advanced Driving Assistance System (ADAS) products such as mobileye could be used for collision warning, a cost-effective approach expected. In this paper, we have developed novel two-stage detect vehicles recognize brake lights single image in...
Jellyfish blooms have caused great damage to the fishery industry. In efforts solve this problem, various systems remove jellyfish been proposed. This letter presents preliminary results of applying an image-based distribution recognition algorithm increase efficiency existing removal system. By using a convolutional neural network and dedicated image processing techniques, experimental show reasonable performance for real-world application.
Abstract The combination of imaging with electrochemical quantification in liquid cell transmission electron microscopy (TEM) provides opportunities for visualizing material processes good spatial and temporal resolution a way that is inaccessible bench‐top experiments. electrode used TEM determines the reliability consistency measurements also influences when on electrode. Here, arising from use 2D materials electrochemistry are explored. Through modeling, it demonstrated graphene...
The medical treatment available for patients with essential tremor (ET) is often inadequate. Furthermore, the efficacy of treatments currently ET cranial nerve areas less satisfactory than that involving upper extremities. This pilot study was performed to evaluate whether zonisamide (ZNS) effective in isolated head tremor.All subjects were randomly treated either ZNS or propranolol. After a washout period, switched alternative drug.ZNS found be more No severe adverse effects reported...
Abstract A highly sensitive bending sensor composed of patterned Pt lines, integrated with energy harvesting capability, is reported. The sensitivity the increases as width lines decreases, owing to increase in crack density decreasing line width. Furthermore, increasing cycles, but saturates at around 1000 cycles. Such a behavior corresponds and eventual saturation microstructured polydimethylsiloxane layer placed on top serve triboelectric layer, where human skin are utilized electrodes....
Magnesium has protective effects in ischaemia-reperfusion injury, and is involved immunomodulation. We investigated the of magnesium pretreatment on secretion T helper (Th) cytokines severity post-reperfusion syndrome (PRS) patients undergoing living donor liver transplantation (LDLT). forty were allocated to two groups 20 (magnesium saline groups). Blood samples for cytokine analysis collected before infusion study solution at end anhepatic phase (time point 1), as well five min 30 after...
In article number 1901744, Steve Park and co-workers depict the fabrication of a highly uniform low hysteresis porous pressure sensor. Using microfluidic emulsion, water droplets are generated assembled in oil solution containing elastomer precursor. Upon annealing, with pore size is formed. The then grafted conductive polymer.
Top-k high utility itemset mining refers to the discovery of top-k patterns using a user-specified value k by considering items in transactional database. Since existing algorithms are based on pattern-growth method, they search two steps. Therefore, generation many candidates and additional database scan for calculating exact utilities unavoidable. In this paper, we propose new algorithm, TKUL-Miner, mine itemsets efficiently. It utilizes utility-list structure which stores necessary...
Prolonged mechanical ventilation after liver transplantation has been associated with deleterious clinical outcomes, so early tracheal extubation posttransplant is now increasing. However, there no universal criterion for predicting in living-donor (LDLT). We investigated specific predictors of LDLT.Perioperative data adult patients undergoing LDLT were reviewed. "Early" was defined as the operating room or intensive care unit (ICU) within 1 h posttransplant, and we divided into (EX) non-EX...
Prisoner suicide has been one of the leading causes death in correctional facilities. In particular, since hanging is most common method, various systems have proposed to detect and prevent attempts. This letter presents preliminary results applying 3-D image recognition techniques a partial suspension attempt. frequently occurring method prison settings, but difficult recognize using conventional methods. By camera random forest, experimental show that detection algorithm can attempts with...
Acute liver failure (ALF) is a rapidly progressing and fatal disease for which transplantation (LT) the only treatment. Posttransplant mechanical ventilation tends to be more prolonged in patients with ALF than other LT patients. The present study examined clinical effects of posttransplant (PMV), identified risk factors PMV following ALF.We reviewed data undergoing between January 2005 June 2011. After grouping according administration (≥ 24 h), donor recipient perioperative variables were...
The preoperative evaluation of a patient for anesthesia increases the practitioner's awareness patient's medical condition and becomes foundation perioperative management plans. Furthermore, visits interviews should also provide information about itself instructions with respect to period [1]. following case report details importance interview by describing undiscovered butyrylcholinesterase (BCHE) deficiency.
Recently, several studies have been carried out on the direct control of behavior in insects and other lower animals order to apply these behaviors performance specialized tasks an attempt find more efficient means carrying than artificial intelligence agents. While most current methods cause involuntary by electronically stimulating corresponding brain area or muscle, we show that, turtles, it is also possible certain types behavior, such as movement trajectory, evoking appropriate...
We present a novel approach to reduce the processing time required derive estimation uncertainty map in deep learning-based optical flow determination methods. Without aware reasoning, model, especially when it is used for mission critical fields such as robotics and aerospace, can cause catastrophic failures. Although several approaches ones based on Bayesian neural networks have been proposed handle this issue, they are computationally expensive. Thus, speed up time, our applies generative...
This paper deals with the problem of steering a group mobile robots along reference path while maintaining desired geometric formation. To solve this problem, overall formation is decomposed into numerous patterns composed pairs robots, and state defined. A control algorithm for proposed based on Nash equilibrium strategies incorporating receding horizon (RHC), also known as model predictive (MPC). Each robot calculates input over finite prediction transmits to its neighbor. Considering...
This paper presents a novel approach at the intersection of machine learning and number theory, focusing on classification prime non-prime numbers. At core our research is development highly sparse encoding method, integrated with conventional neural network architectures. combination has shown promising results, achieving recall over 99\% in identifying numbers 79\% for from an inherently imbalanced sequential series integers, while exhibiting rapid model convergence before completion...
Titanium (Ti) is an adhesion and contact metal commonly used in nanoelectronics two-dimensional (2D) materials research. However, when Ti deposited on graphene (Gr), we obtain dramatically different film morphology depending the experimental conditions. Through a combination of transmission electron microscopy, Raman spectroscopy, ab initio density functional theory calculations, show that most critical parameters are number Gr layers, nature support, deposition temperature. Particularly...