Bum Joo Shin

The Development of Integrated Sensor System for Measuring Simultaneously ECG, PPG and PPW

The pulse transit time(PTT), which is determined by measuring the electrocardiogram(ECG) and pulse wave, gives comprehensive information about the cardiovascular system. However, a little movement of body and/or inaccurate pressure applied to skin during the measurement of pulse wave leads to acquire incorrect results. To overcome such problem, we developed an integrated sensor system which makes it possible to measure ECG, pressure pulse wave(PPW) and photoplethysmograph(PPG) at the same time. Futhermore, we implemented a new metal electrode which enables to continuously measure ECG. We verified that both integrated sensor system and new electrode provide useful effect.

Intravenous Infiltration Risk by Catheter Dwell Time Among Hospitalized Children

Purpose: This study was aimed to examine the cumulative risk for infiltration over IV catheter dwell time by general or catheterization‐specific characteristics of pediatric patients with IV therapy. Design and methods: This secondary data analysis was done with the data of 1596 children who received peripheral IV therapy at least once during their hospital stay between August 1st and October 30th, 2011 and in June, 2013 in an academic medical center, Yangsan, Republic of Korea. The survival functions of infiltration were determined by using the Kaplan–Meier analysis. Result: The cumulative risk for infiltration had rapidly increased from 1.5% after 24 hours of catheter dwell time to 17.3% after 96 hours. The survival functions were significantly different in the medical than in the surgical department (p = .005), lower extremities than upper ones (p = .001), and use of 10% dextrose (p = .001), ampicillin/sulbactam (p < .001), vancomycin (p = .024), high‐concentration electrolytes (p = .001), and phenytoin (p < .001). Conclusion: When catheter dwell times are similar, the cumulative risk for infiltration was higher in cases wherein the patient had a risk factor. The cumulative risk for infiltration has rapidly increased after 24 hours in patients who have 10% dextrose, high‐concentration electrolytes, and phenytoin. Practice implications: The results suggest that nurses are required to assess the IV site every hour after 24 hours of catheter dwell time for the infusion of irritants for a safer practice of IV therapy. However, this monitoring time may be modified by the age of child, previous IV complications, and/or hemodynamic issues which may impact IV integrity. HighlightsCumulative risk for infiltration has increased by 1.8–2 times per 24 hours between 48 and 120 hours after the IV catheter insertion.Cumulative risk for infiltration was approximately 10% after 72 hours, 17.3% after 96 hours, and then 29.3% after 108 hours.When catheter dwell times are same, the cumulative risk for infiltration was higher in medical patients, patients who had IV catheter in their lower extremities, patients who had 10% dextrose, ampicillin/sulbactam, vancomycin, high‐concentration electrolytes, and phenytoin.

Real-time face tracking system using adaptive face detector and Kalman filter

In this paper, we propose a real-time face tracking system using adaptive face detector and the Kalman filter. Basically, the features used for face detection are five types of simple Haar-like features. To only extract the more significant features from these features, we employ principal component analysis (PCA). The extracted features are used for a learning vector of the support vector machine (SVM), which classifies the faces and non-faces. The face detector locates faces from the face candidates separated from the background by using real-time updated skin color information. We trace the moving faces with the Kalman filter, which uses the static information of the detected faces and the dynamic information of changes between previous and current frames. In this experiment, the proposed system showed an average tracking rate of 97.3% and a frame rate of 23.5 frames per s, which can be adapted into a real-time tracking system.

Performance hotspot based CUDA acceleration

With the introduction of many-core GPUs, there is widespread interest in using GPUs to accelerate non-graphics applications such as bioinformatics, energy, finance and several research areas. Even though the GPUs provide highly parallel processing capability, performance improvement is not always achievable due to multiple reasons. One of them will be the application mapping to the hardware acceleration module. In this paper, we investigate performance hotspot functions of an application and used it for application mapping for CUDA acceleration. Based on our experiments with three non-graphic applications, hotspot function based CUDA acceleration shows 15%-40% performance improvement on GPGPU with minimal efforts.

Support of Automatic Calibration for Computerized Puretone Audiometer Using Linux Real-time Spectrum Analyzer

Compared to conventional pure-tone audiometer(PTA), the computerized one has several advantages such as relatively lower price, better flexibility, and easy operation. In addition, the computerized PTA can be easily extended to support automatic calibration. In this study, the support of automatic calibration in the computerized PTA has been addressed using the Linux realtime spectrum analyzer and network. In terms of accuracy, the new automatic calibration machine was compared to the existing manually operated one. According to the comparison, normal operation and high accuracy were observed within the scope of the international standard.

Auto ABLB Audiometry System Supporting One-to-many Model

ABLB (alternate binaural loudness balance) test is one of the medical assessments to diagnose detailed lesion of sensory-neural hearing loss based on a recruitment phenomenon. However, current ABLB audiometry takes an operational model, so called face-to-face model, in which model one audiometrist can assess only one subject at a time. As a result, this face-to-face model leads to expensive audiometrist`s labor cost and lengthy wait when there exist many subjects. As a solution, this paper suggests an ABLB audiometry system supporting one-to-many model in which model an audiometrist enables to assess several subjects concurrently. By providing such capabilities as real-time transfer of assessment result, video monitoring of subject and video chat, this solution can provide same effect as face-to-face model but overcome weakness of the existing face-to-face model.

Development of Laser Ruler in Rigid Laryngoscope

Objectives The objective of this study was to develop a new device that provides a simple, noninvasive method of measuring accurate lesion size while using an endoscope. Methods We developed a rigid laryngoscope with a built-in laser-ruler using a one-light emitting diode and an acrylic plate. The invention incorporates a built-in laser diode that projects an auto-parallel beam into the optical path of the rigid laryngoscope to form two spots in the field of view. Results While the interspot distance remains consistent despite changes in focal plane, magnification, or viewing angle of the laryngoscope, projection to an uneven surface introduces certain variations in the shape, and size of the spots, and the distance between the two spots. Conclusion The device enables a laryngologist to easily measure the distance between landmarks, as well as the change in real size, and the progressive change of vocal fold lesions in an outpatient setting.

Human motion modeling using multivision

In this paper, we propose a gesture modeling system based on computer vision in order to recognize a gesture naturally without any trouble between a system and a user using real-time 3D modeling information on multiple objects. It recognizes a gesture after 3D modeling and analyzing the information pertaining to the users body shape in stereo views for human movement. In the 3D-modeling step, 2D information is extracted from each view by using an adaptive color difference detector. Potential objects such as faces, hands, and feet are labeled by using the information from 2D detection. We identify reliable objects by comparing the similarities of the potential objects that are obtained from both the views. We acquire information on 2D tracking from the selected objects by using the Kalman filter and reconstruct it as a 3D gesture. A joint of each part of a body is generated in the combined objects. We experimented on ambiguities using occlusion, clutter, and irregular 3D gestures to analyze the efficiency of the proposed system. In this experiment, the proposed gesture modeling system showed a good detection and a processing time of 30 frames per second, which can be used in a real-time.

Modeling of places based on feature distribution

In this paper, a place model based on a feature distribution is proposed for place recognition. In many previous proposed methods, places are modeled as images or a set of extracted features. In those methods, a database of images or feature sets should be built. The cost of search time will grow exponentially when the database goes large. The proposed feature distribution method uses global information of each place and the search space grows linearly according to the number of places. In the experiments, we evaluate the performance using different number of frames and features for the recognition each time. Additionally, we have shown that the proposed method is applicable to many real-time applications such as robot navigation, wearable computing systems, and so on.

Computer interface to use eye and mouse movement

In this paper, an interface using eye and mouse for the handicapped people is proposed. The eye regions are localized using neural network (NN)-based texture classifier that discriminates the facial region into eye class and non-eye class, and then are tracked using mean-shift procedure. The mouse region is detected based on edge information and then tracked using template matching. To assess the validity of the proposed system, it was applied to the interface system and was tested on a group of 25 users. The results show that our system has the accuracy of 99%.