• The Journal of the Acoustical Society of Korea
• /
• v.42 no.6
• /
• pp.544-551
• /
• 2023

Speech enhancement used to improve the perceptual quality and intelligibility of noise speech has been studied as a method using a complex-valued spectrum that can improve both magnitude and phase in a method using a magnitude spectrum. In this paper, a study was conducted on how to apply attention mechanism to complex-valued spectrum-based speech enhancement systems to further improve the intelligibility and quality of noise speech. The attention is performed based on additive attention and allows the attention weight to be calculated in consideration of the complex-valued spectrum. In addition, the global average pooling was used to consider the importance of the feature map. Complex-valued spectrum-based speech enhancement was performed based on the Deep Complex U-Net (DCUNET) model, and additive attention was conducted based on the proposed method in the Attention U-Net model. The results of the experiments on noise speech in a living room environment showed that the proposed method is improved performance over the baseline model according to evaluation metrics such as Source to Distortion Ratio (SDR), Perceptual Evaluation of Speech Quality (PESQ), and Short Time Object Intelligence (STOI), and consistently improved performance across various background noise environments and low Signal-to-Noise Ratio (SNR) conditions. Through this, the proposed speech enhancement system demonstrated its effectiveness in improving the intelligibility and quality of noisy speech.

• Journal of Korea Society of Industrial Information Systems
• /
• v.28 no.5
• /
• pp.1-14
• /
• 2023

Soybeans are one of the world's top five staple crops and a major source of plant-based protein. Due to their susceptibility to climate change, which can significantly impact grain production, the National Agricultural Science Institute is conducting research on crop phenotypes through growth analysis of various soybean varieties. While the process of capturing growth progression photos of soybeans is automated, the verification, recording, and analysis of growth stages are currently done manually. In this paper, we designed and trained a YOLOv5s model to detect soybean leaf objects from image data of soybean plants and a Convolution Neural Network (CNN) model to judgement the unfolding status of the detected soybean leaves. We combined these two models and implemented an algorithm that distinguishes layers based on the coordinates of detected soybean leaves. As a result, we developed a program that takes time-series data of soybeans as input and performs growth analysis. The program can accurately determine the growth stages of soybeans up to the second or third compound leaves.

• Korean Journal of Radiology
• /
• v.22 no.11
• /
• pp.1909-1917
• /
• 2021

Objective: Muscle quantity and quality can be measured with an automated system on CT. However, the effects of contrast phases on the muscle measurements have not been established, which we aimed to investigate in this study. Materials and Methods: Muscle quantity was measured according to the skeletal muscle area (SMA) measured by a convolutional neural network-based automated system at the L3 level in 89 subjects undergoing multiphasic abdominal CT comprising unenhanced phase, arterial phase, portal venous phase (PVP), or delayed phase imaging. Muscle quality was analyzed using the mean muscle density and the muscle quality map, which comprises normal and low-attenuation muscle areas (NAMA and LAMA, respectively) based on the muscle attenuation threshold. The SMA, mean muscle density, NAMA, and LAMA were compared between PVP and other phases using paired t tests. Bland-Altman analysis was used to evaluate the inter-phase variability between PVP and other phases. Based on the cutoffs for low muscle quantity and quality, the counts of individuals who scored lower than the cutoff values were compared between PVP and other phases. Results: All indices showed significant differences between PVP and other phases (p < 0.001 for all). The SMA, mean muscle density, and NAMA increased during the later phases, whereas LAMA decreased during the later phases. Bland-Altman analysis showed that the mean differences between PVP and other phases ranged -2.1 to 0.3 cm² for SMA, -12.0 to 2.6 cm² for NAMA, and -2.2 to 9.9 cm² for LAMA.The number of patients who were categorized as low muscle quantity did not significant differ between PVP and other phases (p ≥ 0.5), whereas the number of patients with low muscle quality significantly differed (p ≤ 0.002). Conclusion: SMA was less affected by the contrast phases. However, the muscle quality measurements changed with the contrast phases to greater extents and would require a standardization of the contrast phase for reliable measurement.

• Korean Journal of Radiology
• /
• v.22 no.4
• /
• pp.624-633
• /
• 2021

Objective: To evaluate the reliability of CT measurements of muscle quantity and quality using variable CT parameters. Materials and Methods: A phantom, simulating the L2-4 vertebral levels, was used for this study. CT images were repeatedly acquired with modulation of tube voltage, tube current, slice thickness, and the image reconstruction algorithm. Reference standard muscle compartments were obtained from the reference maps of the phantom. Cross-sectional area based on the Hounsfield unit (HU) thresholds of muscle and its components, and the mean density of the reference standard muscle compartment, were used to measure the muscle quantity and quality using different CT protocols. Signal-to-noise ratios (SNRs) were calculated in the images acquired with different settings. Results: The skeletal muscle area (threshold, -29 to 150 HU) was constant, regardless of the protocol, occupying at least 91.7% of the reference standard muscle compartment. Conversely, normal attenuation muscle area (30-150 HU) was not constant in the different protocols, varying between 59.7% and 81.7% of the reference standard muscle compartment. The mean density was lower than the target density stated by the manufacturer (45 HU) in all cases (range, 39.0-44.9 HU). The SNR decreased with low tube voltage, low tube current, and in sections with thin slices, whereas it increased when the iterative reconstruction algorithm was used. Conclusion: Measurement of muscle quantity using HU threshold was reliable, regardless of the CT protocol used. Conversely, the measurement of muscle quality using the mean density and narrow HU thresholds were inconsistent and inaccurate across different CT protocols. Therefore, further studies are warranted in future to determine the optimal CT protocols for reliable measurements of muscle quality.

• Korean Journal of Radiology
• /
• v.22 no.7
• /
• pp.1213-1224
• /
• 2021

Objective: To develop a machine learning (ML) pipeline based on radiomics to predict Coronavirus Disease 2019 (COVID-19) severity and the future deterioration to critical illness using CT and clinical variables. Materials and Methods: Clinical data were collected from 981 patients from a multi-institutional international cohort with real-time polymerase chain reaction-confirmed COVID-19. Radiomics features were extracted from chest CT of the patients. The data of the cohort were randomly divided into training, validation, and test sets using a 7:1:2 ratio. A ML pipeline consisting of a model to predict severity and time-to-event model to predict progression to critical illness were trained on radiomics features and clinical variables. The receiver operating characteristic area under the curve (ROC-AUC), concordance index (C-index), and time-dependent ROC-AUC were calculated to determine model performance, which was compared with consensus CT severity scores obtained by visual interpretation by radiologists. Results: Among 981 patients with confirmed COVID-19, 274 patients developed critical illness. Radiomics features and clinical variables resulted in the best performance for the prediction of disease severity with a highest test ROC-AUC of 0.76 compared with 0.70 (0.76 vs. 0.70, p = 0.023) for visual CT severity score and clinical variables. The progression prediction model achieved a test C-index of 0.868 when it was based on the combination of CT radiomics and clinical variables compared with 0.767 when based on CT radiomics features alone (p < 0.001), 0.847 when based on clinical variables alone (p = 0.110), and 0.860 when based on the combination of visual CT severity scores and clinical variables (p = 0.549). Furthermore, the model based on the combination of CT radiomics and clinical variables achieved time-dependent ROC-AUCs of 0.897, 0.933, and 0.927 for the prediction of progression risks at 3, 5 and 7 days, respectively. Conclusion: CT radiomics features combined with clinical variables were predictive of COVID-19 severity and progression to critical illness with fairly high accuracy.

• Journal of Internet of Things and Convergence
• /
• v.10 no.2
• /
• pp.137-148
• /
• 2024

The purpose of this study is to determine how the use of retailtech technology affects consumers' purchase intention. Furthermore, this study aims to investigate the mediating effects of technology usefulness and ease of use on this influence relationship and whether experiential marketing moderates consumers' purchase intention. The survey was conducted from August 1, 2023 to September 30, 2023, and a total of 257 people participated in the study. For statistical analysis, hierarchical regression analysis, three-stage mediation regression analysis, and hierarchical three-stage controlled regression analysis were conducted to test the hypothesis. The results of the study are as follows. First, it was confirmed that big data-AI utilization, mobile-SNS utilization, live commerce utilization, and IoT utilization affect purchase intention in retail technology utilization. Second, technology usefulness has a mediating effect on IoT utilization, mobile-SNS utilization, and big data-AI utilization. Third, perceived ease of use of technology mediated the effects of IoT utilization, mobile-SNS utilization, live-commerce utilization, and big data-AI utilization. Fourth, escapist experience has a moderating effect on mobile SNS utilization and live commerce utilization. Fifth, esthetic experience has a moderating effect on mobile-SNS utilization and big data-AI utilization. Through this study, we hope that the domestic distribution industry will contribute to national competitiveness by securing the competitive advantage of companies by utilizing new technologies in entering the global market.

• Journal of Ginseng Research
• /
• v.48 no.1
• /
• pp.77-88
• /
• 2024

Background: Lung inflammation occurs in many lung diseases, but has limited effective therapeutics. Ginseng and its derivatives have anti-inflammatory effects, but their unstable physicochemical and metabolic properties hinder their application in the treatment. Panaxadiol (PD) is a stable saponin among ginsenosides. Inhalation administration may solve these issues, and the specific mechanism of action needs to be studied. Methods: A mouse model of lung inflammation induced by lipopolysaccharide (LPS), an in vitro macrophage inflammation model, and a coculture model of epithelial cells and macrophages were used to study the effects and mechanisms of inhalation delivery of PD. Pathology and molecular assessments were used to evaluate efficacy. Transcriptome sequencing was used to screen the mechanism and target. Finally, the efficacy and mechanism were verified in a human BALF cell model. Results: Inhaled PD reduced LPS-induced lung inflammation in mice in a dose-dependent manner, including inflammatory cell infiltration, lung tissue pathology, and inflammatory factor expression. Meanwhile, the dose of inhalation was much lower than that of intragastric administration under the same therapeutic effect, which may be related to its higher bioavailability and superior pharmacokinetic parameters. Using transcriptome analysis and verification by a coculture model of macrophage and epithelial cells, we found that PD may act by inhibiting TNFA/TNFAR and IL7/IL7R signaling to reduce macrophage inflammatory factor-induced epithelial apoptosis and promote proliferation. Conclusion: PD inhalation alleviates lung inflammation and pathology by inhibiting TNFA/TNFAR and IL7/IL7R signaling between macrophages and epithelial cells. PD may be a novel drug for the clinical treatment of lung inflammation.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.17 no.1
• /
• pp.17-24
• /
• 2024

In this paper, we propose a highly secure technique to hide confidential data in image pixels using a quadruple encryption techniques. In the proposed technique, the boundary surface where the image outline exists and the flat surface with little change in pixel values are investigated. At the boundary of the image, in order to preserve the characteristics of the boundary, one bit of confidential data that has been multiply encrypted is spatially encrypted again in the LSB of the pixel located at the boundary to hide the confidential data. At the boundary of an image, in order to preserve the characteristics of the boundary, one bit of confidential data that is multiplely encrypted is hidden in the LSB of the pixel located at the boundary by spatially encrypting it. In pixels that are not on the border of the image but on a flat surface with little change in pixel value, 2-bit confidential data that is multiply encrypted is hidden in the lower 2 bits of the pixel using location-based encryption and spatial encryption techniques. When applying the proposed technique to hide confidential data, the image quality of the stego-image is up to 49.64dB, and the amount of confidential data hidden increases by up to 92.2% compared to the existing LSB method. Without an encryption key, the encrypted confidential data hidden in the stego-image cannot be extracted, and even if extracted, it cannot be decrypted, so the security of the confidential data hidden in the stego-image is maintained very strongly. The proposed technique can be effectively used to hide copyright information in general commercial images such as webtoons that do not require the use of reversible data hiding techniques.

• Science of Emotion and Sensibility
• /
• v.26 no.4
• /
• pp.103-114
• /
• 2023

Augmented reality (AR) and virtual reality (VR) differ fundamentally, with AR overlaying computer-generated information onto the real world in a nonimmersive way. Despite extensive research on cybersickness in VR, its occurrence in AR has received less attention (Vovk et al., 2018). This study examines cybersickness and discomfort associated with AR usage, focusing on the impact of content intensity and exposure time. Participants viewed 30-minute racing simulation game clips through AR equipment, varying in racing speed to alter content intensity. Cybersickness was assessed subjectively using the Simulator sickness questionnaire (SSQ; Kennedy et al., 1993). Findings revealed a progressive increase in cybersickness with longer exposure, persisting even after removing the AR equipment. Contrarily, content intensity did not significantly influence cybersickness levels. Analysis of the SSQ subscales revealed higher oculomotor (O) scores compared to nausea (N) and disorientation (D), suggesting that discomfort primarily stemmed from oculomotor strain. The study highlights distinct differences in user experience between AR and VR, specifically in subjective responses.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.19 no.1
• /
• pp.137-142
• /
• 2024

Recently, deep learning technology has become those methods as de facto standards in the area of medical data representation. But, deep learning inherently requires a large amount of training data, which poses a challenge for its direct application in the medical field where acquiring large-scale data is not straightforward. Additionally, brain signal modalities also suffer from these problems owing to the high variability. Research has focused on designing deep neural network structures capable of effectively extracting spectro-spatio-temporal characteristics of brain signals, or employing self-supervised learning methods to pre-learn the neurophysiological features of brain signals. This paper analyzes methodologies used to handle small-scale data in emerging fields such as brain-computer interfaces and brain signal-based state prediction, presenting future directions for these technologies. At first, this paper examines deep neural network structures for representing brain signals, then analyzes self-supervised learning methodologies aimed at efficiently learning the characteristics of brain signals. Finally, the paper discusses key insights and future directions for deep learning-based brain signal analysis.