• Journal of Digital Convergence
• /
• v.19 no.6
• /
• pp.251-258
• /
• 2021

In this paper, we proposed an encoder-decoder model utilizing residual learning to improve the accuracy of the U-Net-based semantic segmentation method. U-Net is a deep learning-based semantic segmentation method and is mainly used in applications such as autonomous vehicles and medical image analysis. The conventional U-Net occurs loss in feature compression process due to the shallow structure of the encoder. The loss of features causes a lack of context information necessary for classifying objects and has a problem of reducing segmentation accuracy. To improve this, The proposed method efficiently extracted context information through an encoder using residual learning, which is effective in preventing feature loss and gradient vanishing problems in the conventional U-Net. Furthermore, we reduced down-sampling operations in the encoder to reduce the loss of spatial information included in the feature maps. The proposed method showed an improved segmentation result of about 12% compared to the conventional U-Net in the Cityscapes dataset experiment.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.3
• /
• pp.19-28
• /
• 2021

Esophagogastroduodenoscopy is a method commonly used for early diagnosis of upper gastrointestinal lesions. However, 10-20 percent of the gastric lesions are reported to be missed, due to human error. And countries including the US, the UK, and Japan, the World Endoscopy Organization (WEO) suggested guidelines about essential gastrointestinal parts to take pictures of so that all gastric lesions are observed. In this paper, we propose deep learning techniques for classification of anatomical sites, aiming for the system that informs practitioners whether they successfully did the gastroscopy without blind spots. The proposed model uses pre-processing modules and data augmentation techniques suitable for gastroscopy images. Not only does the experiment result with a maximum F1 score of 99.6%, but it also shows a error rate of less than 4% based on the actual data. Given the performance results, we found the model to be explainable with the potential to be utilized in the clinical area.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.5
• /
• pp.534-544
• /
• 2022

Medical ultrasound system has been widely used to visualize the lesion for diagnostics in most medical service site including hospitals and clinics thanks to its advantages such as real time operation, ease of use, safety. Among many signal processing blocks of the system, one of the most important part that governs the image quality is the beamformer, and technologies for this part has been continuously developed in long time. The synthetic aperture imaging method, that is one of the major technologies of beamforming, was introduced to maximize utilizing the information delivered from the patient's body through the probe, and contributed to breakthrough of the image quality since it was introduced in around 1990's, and evolved continuously in decades. This paper reviews and surveys the process of development of this technology and expects future evolution.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.9 no.1
• /
• pp.51-60
• /
• 2014

In this study, ICT medical service provider's level of knowledge fire fighting safety and methods on coping with fires in the regions of Gwangju and Jeonam Province of Korea were investigated to determine the elements affecting such levels and provide basic information on the manuals for educating how to cope with the fire fighting safety in medical facilities. The data were analyzed using SPSS Win 14.0. The scores of level of knowledge fire fighting safety of ICT medical service provider's were 7.06(10 point scale), and the scores of level of recognizing how to cope with fire fighting safety were 6.61(11 point scale). level of recognizing how to cope with fire fighting safety were significantly different according to gender(t=4.12, p<.001), age(${\chi}^2$=17.24, p<.001), length of career(${\chi}^2$=22.76, p<.001), experience with fire fighting safety education(t=6.10, p<.001), level of subjective knowledge on fire fighting safety(${\chi}^2$=53.83, p<.001). In order to enhance the level of understanding of fire fighting safety and methods of coping by the ICT medical service providers it is found that: self-directed learning through avoiding the education just conveying knowledge by lecture tailored learning for individuals fire fighting education focused on experiencing actual work by developing various contents emphasizing cooperative learning deploying patients by classification systems using simulations and a study on the implementation of digital anti-fire monitoring system with multipoint communication protocol, a design and development of the smoke detection system using infra-red laser for fire detection in the wide space, video based fire detection algorithm using gaussian mixture mode developing an education manual for coping with fire fighting safety through multi learning approach at the medical facilities are required.

• The Korean Journal of Nuclear Medicine Technology
• /
• v.13 no.3
• /
• pp.110-122
• /
• 2009

Purpose: Nowadays, A medical institution assesment could get more interest about a quality of medical services from many hospitals that developed the active activities for improving medical services. Also, there is an other additional issue which is the patients risk management. Uijeongbu ST. Mary's hospital Nuclear Medicine department has been changed many work process after PET-CT introduction and renovation of its place since 2008. Therefore, modified structure and the way of existing work process have contained risk factors. The purpose of this study would be the appropriate risk management process while imaging examination process, the removal risk factors and improved activities through the analysis risk factors. Materials and Methods: Nuclear Medicine department new process should analysis through many-sided, Firstly, make and trained risk management manual after then apply an actual work. Result analysis showed the number of risk accident occurrence that comparing the last year and after the improved activities. Secondly, producing risk management public relations film has been showed an applicable patient after then the customer service measurement checked for a hundred patient by questionnaire. Lastly, Risk factors were eliminated through the facilities participation improving activities which could change for the better risk factors. Results: The number of safety accident occurrence(medication error, fall and collision) were checked as zero after the improving activities both PET-CT and gamma camera examination. The results of questionnaire showed as follows; 74% marked as understanding of the test process and 81% checked "satisfaction" after the public relations film showing. The question "Did you consider about the risk factors?", both PET-CT and gamma camera checked as 94% and 89% respectively. Customer risk management could be accomplished effectively through the improving activities at the nuclear medicine department. Conclusions: The study would be an opportunity that spread risk factors were systematically showed and analyzied. Also, It showed the possibility of the minimized safety accident and its feedback, if application of the response manuel that could be a standard of radiology technician's work method to react safety accident. It was the more effective that visual material could be easy to approach as a methodology of risk factors. As far as I have concerned that It could help the safety and convenience through continuous and detailed activities that offer to patients.

• Journal of the Korean Society for information Management
• /
• v.23 no.3 s.61
• /
• pp.5-21
• /
• 2006

In recent years, theories of image and sound analysis have been proposed to work with text retrieval systems and have progressed quickly with the rapid progress in data processing speeds. This study proposes a common representation format for multimedia documents (CRFMD) composed of both images and text to form a single data structure. It also shows that image classification of a given test set is dramatically improved when text features are encoded together with image features. CRFMD might be applicable to other areas of multimedia document retrieval and processing, such as medical image retrieval, World Wide Web searching, and museum collection retrieval.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.7 no.8
• /
• pp.1792-1801
• /
• 2003

In these days, image mosaic is getting interest in the field of advertisement, tourism, game, medical imaging, and so on with the development of internet technology and the performance of personal computers. The main problem of mage mosaic is searching corresponding points correctly in the overlapped area between images. However, previous methods requires a lot of CPU times and data processing for finding corresponding points. And they need repeated recording with a revolution of 360 degree around objects or background. This paper presents a new image mosaic method which generates a panorama image from a video sequence recorded by a general video camera. Our method finds the corresponding points between two successive images by using a new direction oriented 3­step block matching methods. Experimental results show that the suggested method is more efficient than the methods based on existing block matching algorithm, such as full search and K­step search algorithm.

• The Korean Society of Law and Medicine
• /
• v.22 no.2
• /
• pp.111-138
• /
• 2021

'Unlicensed medical practice by a non-medical practitioner' often represented by surrogate surgery or so-called 'ghost surgery,' causes irreparable damage to life or body, and therefore calls for very strict and effective controls. The 'bill on installment of CCTVs in an operating room' to prevent unlicensed surrogate surgery has been discussed for a long time, but due to numerous issues and heated confrontations, it has been pending in the National Assembly. Nevertheless, it is expected that the bill will be discussed again in earnest in the National Assembly because surrogate surgery and factory-type cosmetic surgery, which has been performed mainly in the field of cosmetic surgery, has also been occurring in the field of therapeutic surgery. In general, an operating room is considered as being locked or closed, as well as disallowing implicit complicity among insiders. Hence, if the insiders conspire to commit or cover up an illegal act, or if a surgeon performs rapid cosmetic surgery and then leaves the recipient (or medical institution) so as to perform more operations for profit - even if it is legitimate practice - it may result in serious consequences in terms of the recovery of a patient. In this case, installation of CCTVs can be of great help in identifying an illegal act and assessing any occurrence of negligence. On the other hand, while the fundamental purpose of therapeutic surgery is to restore a patient's life or body - that is, lifesaving - installation of CCTVs may base the relationship between a surgeon and a patient on distrust and surveillance, so it may increase the number of requests for CCTV footage or lead to more disputes, as well as placing a burden on the surgeon when best results are not achieved for a patient. As a result, the surgeon may choose non-invasive treatment contrary to conscience instead of risky but necessary surgery, or he/she may have significant difficulty in determining the timing of surgery, which may limit the provision of effective surgical medical care. Then, in terms of the relationship between a surgeon and patient, and in the long run, there could be significant disadvantages for the public and patients if CCTV footage is allowed. In this paper, we review domestic and overseas cases and issues regarding installation of CCTVs in an operating room, and present various viewpoints and suggestions to promote legislation with minimized legal problems and side effects, thereby contributing to protection of the lives and health of the public, patients, and recipients of surgery.

• The KIPS Transactions:PartB
• /
• v.14B no.6
• /
• pp.431-436
• /
• 2007

Magnetic Resonance Imaging(MRI) has chemical shift phenomenon between fat and water, and the phenomenon has influence on structure enclosed by fat. Strong signals emitted from fat often generate false artefact, which reflects the importance of fat suppression techniques. There have been a number of researches on fat suppression techniques, but using fat suppression method alone in MRI can cause difficultproblems in diagnosis. This paper aims to study a fat suppression method by Chemical Shift Selective saturation(CHESS). This research describes the theoretical background and the experiment on water and fat phantom with MR instruments. In the experiment, CHESS pulse was designed by utilising Matlap program, and the pulse diagram was generated for the Pre-saturation process. The experiment using water and fat phantom was applied to C-spine, L-spine and Breast, and produced successful fat suppression results. This experiment has proved that the CHESSpulse fat suppression is a very helpful technique in diagnosing medical imaging. This method is a robust and useful technique for both clinical and basic investigators..(Experiment with Chungnam national university hospital G.E 1.5T MR)

• The Journal of Korean Institute of Next Generation Computing
• /
• v.14 no.5
• /
• pp.31-39
• /
• 2018

Recently, there has been active studies to provide a user-friendly interface in a virtual reality environment by recognizing user hand gestures based on deep learning. However, most studies use separate sensors to obtain hand information or go through pre-process for efficient learning. It also fails to take into account changes in the external environment, such as changes in lighting or some of its hands being obscured. This paper proposes a hand gesture recognition method based on deep learning that is strong in external environments without the need for pre-process of RGB images obtained from general webcam. In this paper we improve the VGGNet and the GoogLeNet structures and compared the performance of each structure. The VGGNet and the GoogLeNet structures presented in this paper showed a recognition rate of 93.88% and 93.75%, respectively, based on data containing dim, partially obscured, or partially out-of-sight hand images. In terms of memory and speed, the GoogLeNet used about 3 times less memory than the VGGNet, and its processing speed was 10 times better. The results of this paper can be processed in real-time and used as a hand gesture interface in various areas such as games, education, and medical services in a virtual reality environment.