• Journal of Korean Clinical Nursing Research
• /
• v.26 no.1
• /
• pp.117-130
• /
• 2020

Purpose: This study aimed to develop a bedside nursing shift report protocol and evaluate the effect of the protocol in a tertiary hospital in South Korea. Methods: The bedside nursing handoff protocol with patient engagement was developed based on the literature review and the validation of an expert group. The effect of the protocol on clinical implication was tested in three medical-surgical units in a tertiary hospital. Outcomes were assessed by patient perception, nurse perception, and reporting time. Data collected from June to August in 2018 and analyzed with descriptive statistics and One-way ANOVA using SPSS version 25.0. Results: The bedside nursing shift report protocol with patient engagement consisted of two steps: nurse to nurse report and bedside report with patients. Nurse's perception with patient engagement was significantly increased after applying protocol (F=17.85, p<.001). Patient's perception was significantly improved in the areas of discharge plan (F=7.86, p<.001), health information privacy (F=4.46, p=.012) and identify attending nurse (F=3.19, p=.042). There were no differences in reporting time between the bedside nursing shift report and a traditional shift report (F=0.61, p=.054). Conclusion: Patient perception was significantly increased, while nurse perception was not different after applying this protocol. For the change in the perception of nurses, education may be preceded to improve nurses' competence for the bedside shift report. Furthermore, the support in enough nurse staffing should be needed for encouraging the bedside shift report. The bedside shift report may enhance patient engagement. Therefore it may improve patient safety and health outcome in clinics.

• Journal of Advanced Navigation Technology
• /
• v.25 no.5
• /
• pp.426-432
• /
• 2021

Continuous wave (CW) Doppler radar has the advantage of being able to solve the privacy problem unlike camera and obtains signals in a non-contact manner. Therefore, this paper proposes a human activity recognition (HAR) system using CW Doppler radar, and presents the hardware design and implementation results for acceleration. CW Doppler radar measures signals for continuous operation of human. In order to obtain a single motion spectrogram from continuous signals, an algorithm for counting the number of movements is proposed. In addition, in order to minimize the computational complexity and memory usage, binarized neural network (BNN) was used to classify human motions, and the accuracy of 94% was shown. To accelerate the complex operations of BNN, the FPGA-based BNN accelerator was designed and implemented. The proposed HAR system was implemented using 7,673 logics, 12,105 registers, 10,211 combinational ALUTs, and 18.7 Kb of block memory. As a result of performance evaluation, the operation speed was improved by 99.97% compared to the software implementation.

• Journal of Convergence for Information Technology
• /
• v.9 no.3
• /
• pp.140-145
• /
• 2019

With the development of IT technology, many changes are taking place in the health service environment over the past. However, even if medical technology is converged with IT technology, the problem of medical costs and management of health services are still one of the things that needs to be addressed. In this paper, we propose a model for hospitals that have established the IoT system to efficiently analyze and manage the personal information of users who receive medical services. The proposed model aims to efficiently check and manage users' medical information through an in-house IoT system. The proposed model can be used in a variety of heterogeneous cloud environments, and users' medical information can be managed efficiently and quickly without additional human and physical resources. In particular, because users' medical information collected in the proposed model is stored on servers through the IoT gateway, medical staff can analyze users' medical information accurately regardless of time and place. As a result of performance evaluation, the proposed model achieved 19.6% improvement in the efficiency of health care services for occupational health care staff over traditional medical system models that did not use the IoT system, and 22.1% improvement in post-health care for users who received medical services. In addition, the burden on medical staff was 17.6 percent lower on average than the existing medical system models.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.18 no.1
• /
• pp.115-137
• /
• 2008

As stock market gets bigger, use of HTS(Home Trading System) is getting increased in stock exchange. HTS provides lots of functions such as inquiry about stock quotations, investment counsel and so on. Thus, despite the fact that the functions fur convenience and usefulness are developed and used, security functions for privacy and trade safety are insufficient. In this paper, we analyze the security system of HTS service through the key-logging and sniffing and suggest that many private information is unintentionally exposed. We also find out a vulnerable point of the system, and show the advisable criteria of secure HTS.

• The Journal of the Korea Contents Association
• /
• v.22 no.7
• /
• pp.81-88
• /
• 2022

This study is to investigate the perception of domestic appraisers about the possibility of using artificial intelligence (AI) and related risks from the use of AI in the appraisal industry. We conducted a mobile survey of evaluators from February 10 to 18, 2022. We collected survey data from 193 respondents. Frequency analysis and multiple response analysis were performed for basic analysis. When AI is used in the appraisal industry, factor analysis was used to analyze various types of risks. Although appraisers have a positive perception of AI introduction in the appraisal industry, they considered collateral, consulting, and taxation, mainly in areas where AI is likely to be used and replaced, mainly negative effects related to job losses and job replacement. They were more aware of the alternative risks caused by AI in the field of human labor. I was very aware of responsibilities, privacy and security, and the risk of technical errors. However, fairness, transparency, and reliability risks were generally perceived as low risk issues. Existing studies have mainly studied analysis methods that apply AI to mass evaluation models, but this study focused on the use and risk of AI. Understanding industry experts' perceptions of AI utilization will help minimize potential risks when AI is introduced on a large scale.

• Korea Journal of Hospital Management
• /
• v.27 no.2
• /
• pp.16-33
• /
• 2022

Purposes: Prompt evaluation of routes and contact tracing are very important for epidemiological investigations of coronavirus disease 2019 (COVID-19). To ensure better adoption of contact tracing apps, it is necessary to understand users' expectations, preferences, and concerns. This study aimed to identify main reasons why people use the apps, appropriate services, and basis for voluntary app services that can improve app participation rates and data sharing. Methodology/Approach: This study conducted an online survey from November 11 to December 6, 2020, and received a total of 1,048 survey responses. This study analyzed the questionnaire survey findings of 883 respondents in areas with many confirmed cases of COVID-19. This study used a multiple regression analysis. Findings: Respondents who had experience of using related apps showed a high intention to use contact-tracing apps. Participants wished for the contact tracking apps to be provided by the government or public health centers (74%) and preferred free apps (93.88%). The factors affecting the participants' intention to use these apps were their preventive value, performance expectancy, perceived risk, facilitative ability, and effort expectancy. The results highlighted the need to ensure voluntary participation to address participants' concerns regarding privacy protection and personal information exposure. Practical Implications: The results can be used to accurately identify user needs and appropriate services and thereby improve the development of contact tracking apps. The findings provide the basis for voluntary app that can enhance app participation rates and data sharing. The results will also serve as the basis for developing trusted apps that can facilitate epidemiological investigations.

• Smart Media Journal
• /
• v.12 no.2
• /
• pp.56-65
• /
• 2023

In broadcasts such as news and coverage programs, voice is modulated to protect the identity of the informant. Adjusting the pitch is commonly used voice modulation method, which allows easy voice restoration to the original voice by adjusting the pitch. Therefore, since broadcast voice modulation methods cannot properly protect the identity of the speaker and are vulnerable to security, a new voice modulation method is needed to replace them. In this paper, using the Lightweight speech de-identification model as the evaluation target model, we compare speech de-identification performance with broadcast voice modulation method using pitch modulation. Among the six modulation methods in the Lightweight speech de-identification model, we experimented on the de-identification performance of Korean speech as a human test and EER(Equal Error Rate) test compared with broadcast voice modulation using three modulation methods: McAdams, Resampling, and Vocal Tract Length Normalization(VTLN). Experimental results show VTLN modulation methods performed higher de-identification performance in both human tests and EER tests. As a result, the modulation methods of the Lightweight model for Korean speech has sufficient de-identification performance and will be able to replace the security-weak broadcast voice modulation.

• Smart Media Journal
• /
• v.13 no.4
• /
• pp.33-48
• /
• 2024

With the advent of 5G, characterized by Enhanced Mobile Broadband (eMBB), Ultra-Reliable Low Latency Communications (URLLC), and Massive Machine Type Communications (mMTC), efficient network management and service provision are becoming increasingly critical. This paper proposes a novel approach to address key challenges of 5G networks, namely ultra-high speed, ultra-low latency, and ultra-reliability, while dynamically optimizing network slicing and resource allocation using machine learning (ML) and deep learning (DL) techniques. The proposed methodology utilizes prediction models for network traffic and resource allocation, and employs Federated Learning (FL) techniques to simultaneously optimize network bandwidth, latency, and enhance privacy and security. Specifically, this paper extensively covers the implementation methods of various algorithms and models such as Random Forest and LSTM, thereby presenting methodologies for the automation and intelligence of 5G network operations. Finally, the performance enhancement effects achievable by applying ML and DL to 5G networks are validated through performance evaluation and analysis, and solutions for network slicing and resource management optimization are proposed for various industrial applications.

• Fire Science and Engineering
• /
• v.34 no.2
• /
• pp.156-165
• /
• 2020

In this paper, the application of evacuation delay time (Cognition time + initiation time) and examine the degree of opening of fire doors in households for evaluating evacuation safety and suggest a realistic alternative. In order to proceed with this study, first of all, the preliminary investigation on evacuation safety evacuation of residential-type buildings (Apartment, urban living houses, etc.) among the performance-oriented design targets of Gwangju Metropolitan City, which was implemented until June 2018. Then, for the two representative types that are commonly used among the previously surveyed buildings, evacuation delay time is applied to W1, W2, and respectively simulating the opening of the doors is applied to th full open, 1/4 open, the leakage gap and evacuation safety evaluation was performed. As a result of evaluating evacuation safety was found that it is difficult to secure evacuation safety regardless of evacuation delay time W1 and W2 when the fire door is fully open and 1/4 open, Only when the leakage gap is applied evacuation safety was ensured even if evacuation delay time W2 was applied. Therefore, when a residential building is subject to performance-oriented design, evaluating the application of W2 rather than W1 is considered for evacuation delay time to reflect concern about privacy infringement due to CCTV installation, etc. In order to secure the Smoke blocking performance of the fire door and to improve the performance-oriented design, I would like to propose to consider the method of applying a leak gap to the degree of opening of the fire door. Through this, it is expected that the performance-oriented design will be a step further by performing evacuation safety evaluation with more realistic data.

• Journal of Intelligence and Information Systems
• /
• v.25 no.1
• /
• pp.163-177
• /
• 2019

As smartphones are getting widely used, human activity recognition (HAR) tasks for recognizing personal activities of smartphone users with multimodal data have been actively studied recently. The research area is expanding from the recognition of the simple body movement of an individual user to the recognition of low-level behavior and high-level behavior. However, HAR tasks for recognizing interaction behavior with other people, such as whether the user is accompanying or communicating with someone else, have gotten less attention so far. And previous research for recognizing interaction behavior has usually depended on audio, Bluetooth, and Wi-Fi sensors, which are vulnerable to privacy issues and require much time to collect enough data. Whereas physical sensors including accelerometer, magnetic field and gyroscope sensors are less vulnerable to privacy issues and can collect a large amount of data within a short time. In this paper, a method for detecting accompanying status based on deep learning model by only using multimodal physical sensor data, such as an accelerometer, magnetic field and gyroscope, was proposed. The accompanying status was defined as a redefinition of a part of the user interaction behavior, including whether the user is accompanying with an acquaintance at a close distance and the user is actively communicating with the acquaintance. A framework based on convolutional neural networks (CNN) and long short-term memory (LSTM) recurrent networks for classifying accompanying and conversation was proposed. First, a data preprocessing method which consists of time synchronization of multimodal data from different physical sensors, data normalization and sequence data generation was introduced. We applied the nearest interpolation to synchronize the time of collected data from different sensors. Normalization was performed for each x, y, z axis value of the sensor data, and the sequence data was generated according to the sliding window method. Then, the sequence data became the input for CNN, where feature maps representing local dependencies of the original sequence are extracted. The CNN consisted of 3 convolutional layers and did not have a pooling layer to maintain the temporal information of the sequence data. Next, LSTM recurrent networks received the feature maps, learned long-term dependencies from them and extracted features. The LSTM recurrent networks consisted of two layers, each with 128 cells. Finally, the extracted features were used for classification by softmax classifier. The loss function of the model was cross entropy function and the weights of the model were randomly initialized on a normal distribution with an average of 0 and a standard deviation of 0.1. The model was trained using adaptive moment estimation (ADAM) optimization algorithm and the mini batch size was set to 128. We applied dropout to input values of the LSTM recurrent networks to prevent overfitting. The initial learning rate was set to 0.001, and it decreased exponentially by 0.99 at the end of each epoch training. An Android smartphone application was developed and released to collect data. We collected smartphone data for a total of 18 subjects. Using the data, the model classified accompanying and conversation by 98.74% and 98.83% accuracy each. Both the F1 score and accuracy of the model were higher than the F1 score and accuracy of the majority vote classifier, support vector machine, and deep recurrent neural network. In the future research, we will focus on more rigorous multimodal sensor data synchronization methods that minimize the time stamp differences. In addition, we will further study transfer learning method that enables transfer of trained models tailored to the training data to the evaluation data that follows a different distribution. It is expected that a model capable of exhibiting robust recognition performance against changes in data that is not considered in the model learning stage will be obtained.