Search

Abstract To study telomere maintenance mechanism (TMM) activation during malignant transformation, we compared neurofibroma (NF) and malignant peripheral nerve sheath tumor (MPNST) in the same patient with type-1 neurofibromatosis (NF1), a total of 20 NF-MPNST pairs in 20 NF1 patients. These comparisons minimized genetic bias and contrasted only changes associated with malignant transformation, while subtracting changes that developed upon the transformation of normal cells to the benign tumor. TGF-β superfamily genes were found to activate the PAX and SOX transcription factors, leading to TMM activation. BMPER activates PAX6 through BMP2 and PAX7 through BMP4; BMP15 activates SOX14; and INHBC activates PAX9 and SOX14. The activated PAX and SOX genes sequentially establish the core architecture of the RAD52-dependent alternative lengthening of telomeres (ALT). Specifically, PAX7 activates the recombinase (RAD52) and a negative regulator (SLX4IP). PAX6 and SOX14 activate positive regulators (BLM and BRCA2, respectively). PAX9 and SOX14 activate RAD9B and FEN1, which are responsible for the stability of homologous recombination intermediates and increase, together with RAD52, the telomere length. Telomere elongation achieved by the activation of PAX7 and PAX9 is associated with a poor prognosis. We demonstrated that TGF-β superfamily-induced transcriptional activation pathways activated the RAD52-dependent ALT during malignant transformation of MPNSTs.

When fabricating a metal powder filter with a double-layered pore structure, which includes a microporous layer, by coating flake-shaped powder using the Wet Powder Spray (WPS) process, it is crucial to ensure that the coating is applied parallel to the surface of the metal powder filter. Failure to do so can result in a non-uniform pore structure, leading to a rapid decline in particle filtration efficiency. Therefore, it is necessary to improve the alignment of the coated flake-shaped powder on the filter surface. In this study, a metal powder filter with a double-layered pore structure was fabricated by applying a flake-shaped metal powder coating to a tube-shaped SUS316L powder filter using the WPS process. In order to improve the flake-shaped powder orientation of coating layer, the impact of the rolling process was investigated. The characteristics of the pores were analyzed based on the powder size and the rolling time. The microstructure of the fabricated metal powder filter was observed using an optical microscope, surface roughness, variations in thickness, and air permeability were analysed.

Generally, when fabricating porous filters using metal powder, about 1 to 2 wt% of a binder is added to increase the formability of the metal powder. If the binder is not completely removed through a debinding and sintering process, however it can cause defects such as discoloration and the generation of fine particles. In this study, a study was conducted to fabricate a plate-type porous stainless steel powder filter with a different pore structure without the use of a biner. First, the metal powder is charged into a ceramic mold to form a plate-shaped powder filter, and then covered with a ceramic top plate. In order to observe the pore properties according to the pressure, the unpressurized specimen and the specimen pressurized with a pressure of 30 MPa were then sintered at 1050℃ for one hour in a high vacuum atmosphere furnace. The microstructure of the sintered plate-shaped powder filter was observed through an optical microscope and in order to analyze its pore properties as a filter, gas permeability and porosity were measured using a capillary flow porometer and Archimedes’ law.

In this study, we examined the types of social minorities represented in South Korean high school Society and Culture textbooks since the democratization of South Korea and how their portrayals have changed. Throughout the curriculum period, women were the most frequently mentioned social minority groups, whereas sexual minorities were rarely discussed. The manner in which ethnic minorities are described has changed mostly in response to multicultural changes within Korean society, and people with disabilities have usually been mentioned only at an abstract level. Since the democratization of Korea, various social minority groups have developed rights movements and have struggled for their recognition by various parts of society, however, the voices of change have not been faithfully reflected in textbooks. In response to the Korean government’s policy trends of reducing the amount of learning, the content of social minorities has decreased, and social changes have not been fully captured. This finding somewhat differs from the expectations that the inclusion of ‘social minorities’ as a topic of study in the curriculum would improve the representation of the voices of various social minorities.

The implementation of reservoir computing using resistive random-access memory as a physical reservoir has attracted attention due to its low training cost and high energy efficiency during parallel data processing. In this work, a NbOx/Al2O3-based memristor device was fabricated through a sputter and atomic layer deposition process to realize reservoir computing. The proposed device exhibits favorable resistive switching properties (>103 cycle endurance) and demonstrates short-term memory characteristics with current decay. Utilizing the controllability of the resistance state and its variability during cycle repetition, electrical pulses are applied to investigate the synapse-emulating properties of the device. The results showcase the functions of potentiation and depression, the coexistence of short-term and long-term plasticity, excitatory post-synaptic current, and spike-rate dependent plasticity. Building upon the functionalities of an artificial synapse, pulse spikes are categorized into three distinct neural firing patterns (normal, adapt, and boost) to implement 4-bit reservoir computing, enabling a significant distinction between “0” and “1.”

Multilabel remote sensing image classification is a fundamental task that classifies multiple objects and land covers within an image. However, training deep learning models for this task requires a considerable cost of labeling. While several efforts to reduce labeling costs have been made, they often result in decreased label quality and the inclusion of incorrect (i.e., noisy) labels. To tackle this issue, algorithms for training deep learning models robust to multilabel noise have been proposed in the literature. Nonetheless, the efficacy of these algorithms has been evaluated only under instance-independent multilabel noise, where noise is generated regardless of the individual characteristics and features of each remote sensing image. In this article, we introduce generating instance-dependent multilabel noise into multilabel remote sensing image datasets for the first time. We leverage a vision-language model with zero-shot prediction capabilities to compute categorywise prediction scores for each image, based on which we generate multilabel noise in an instance-dependent manner. We demonstrate that the proposed instance-dependent multilabel noise is more feasibly generated with respect to individual images compared to traditional instance-independent multilabel noise. We also demonstrate that a more challenging noise scenario is generated, which leads to a more complex decision boundary and stronger overfitting during deep learning model training. Finally, we re-evaluate existing noise-robust training algorithms under the generated instance-dependent multilabel noise and observe that several algorithms exhibit limited robustness against instance-dependent multilabel noise.

In this paper, a via-guided microstrip line is proposed for minimized microwave circuit designs. The via-guided microstrip line is qualified for a general transmission line requirement of lossless, linear phase, a quasi-TEM mode, and analytical modeling. The equivalent circuit model and element values are provided to compute the characteristic impedance and guided wavelength analysis of the new transmission line. The proposed via-guide microstrip line with a short via-spacing and a distance to microstrip shows a shorter electric wavelength on a thicker and higher dielectric substrate, which is attributed to additional coupling capacitance. The characteristics analysis shows that a via-guided microstrip branch-line hybrid coupler has been successfully designed and implemented with a size reduction of 36.8% when compared to the same circuit on a conventional microstrip line.

Neural Architecture Search (NAS) has proven valuable in many applications such as computer vision. However, its true potential is unveiled when applied to less-explored domains. In this work, we study NAS for addressing problems in diverse fields where we expect to apply deep neural networks in the real world domains. We introduce D2NAS, Differential and Diverse NAS, leveraging techniques such as Differentiable ARchiTecture Search (DARTS) and Diverse-task Architecture SearcH (DASH) for architecture discovery. Our approach outperforms existing models, including Wide ResNet (WRN) and DASH, when evaluated on NAS-Bench-360 tasks, which include 10 numbers of diverse tasks for 1D, 2D, 2D Dense (tightly interconnected data) tasks. Compared to DASH, D2NAS reduces average error rates by 12.2%, while achieving an 85.1% reduction in average parameters (up to 97.3%) and a 91.3% reduction in Floating Point Operations (FLOPs, up to 99.3%). Therefore, D2NAS enables the creation of lightweight architectures that exhibit superior performance across various tasks, extending its applicability beyond computer vision to include mobile applications.

Neural Compact Models (NCMs) have emerged as a crucial tool to meet the stringent demands of Design-Technology Co-Optimization (DTCO) and to overcome the complexities and prolonged development cycles encountered in traditional compact model creation. Despite their efficiency in simulating electronic devices, a significant barrier to the widespread adoption of NCMs in the industry remains: the lack of interpretability. In the semiconductor sector, where inaccuracies or failures can lead to considerable financial consequences, it is critical to ensure that the model’s predictions are both understandable and reliable. This study aims to enhance the interpretability of NCMs used for I-V and C-V characterizations by clarifying the physical significance of latent vectors. A regularization technique is employed to disentangle features within the latent space, and interpolation is used to visualize and elucidate each dimension’s physical impact. Our approach, which offers interpretable insights into the model’s functionality, seeks to encourage broader implementation of NCMs in the industry, thus accelerating advancements in DTCO.

Due to the vast array of NeRF-based techniques, the representation power of Neural Radiance Fields (NeRF) has been quickly rising in recent years. However, it is still difficult to offer fresh perspectives for user-controlled geometry alterations with current techniques; instead, they restrict edits to those that were observed during training or to particular regions chosen by the provider. In this paper, we present NeRFlex, Flexible Neural Radiance Fields with Diffeomorphic Deformation, which lets users process geometry completely unrestricted by using radiance field deformation. Given a condition with a specific viewpoint, a conditional score function is estimated using diffusion time steps. The deformation between the radiance fields before and after an edit is then generated by the score function and the initial radiance field. To guarantee topology preservation, invertibility, and smooth transformation, diffeomorphic constraints are provided to the deformation field. Experiments on various objects demonstrate NeRFlex’s ability to generate flexible deformations and high-quality novel views after geometry edits, which were never observed in the training data. Continuous deformation along the pathway leading to the deformed object is also obtained by diffeomorphism decomposition.

The long-horizon problem has been a persistent challenge in the field of reinforcement learning, leading to the exploration of various solutions. Planning methods have emerged as a prominent approach, generating intermediate plans from the current state to the goal state. However, these methods often rely on the assumption of additional interaction with the environment or the availability of offline data. However, in real-world scenarios, instead of such information, only time-independent random observations of the environment can be provided. To overcome this limitation, we propose the Ricci planner, a novel algorithm capable of generating a plan from the current location to a desired goal using only a limited number of time-independent random samples from the observation space. We drew inspiration from the observation that the most efficient path is one with the minimum length and, based on this, we transformed the problem of finding an efficient path into a shortest path-finding problem. Subsequently, we formulated this as an optimization problem on a path space. However, the length functional in the path space is highly non-convex and multi-modal, which results in numerous suboptima, and makes the problem exceedingly challenging to solve. To address this issue, we employ the Ricci flow to continuously transform the target manifold into a simpler manifold. Initially, we identify the shortest path on the simpler manifold and subsequently convert it to the shortest path on the desired manifold by applying the inverse process of the Ricci flow. We conduct a experimental comparison with graph-based shortest path finding methods. We assessed both the quality of the generated plan itself and its effectiveness when applied to an agent and observed improved results from both perspectives compared to the baseline.

Mixup is a powerful data augmentation strategy that has been shown to improve the generalization and adversarial robustness of machine learning classifiers, particularly in computer vision applications. Despite its simplicity and effectiveness, the impact of Mixup on the fairness of a model has not been thoroughly investigated yet. In this paper, we demonstrate that Mixup can perpetuate or even exacerbate bias presented in the training set. We provide insight to understand the reasons behind this behavior and propose GBMix, a group-balanced Mixup strategy to train fair classifiers. It groups the dataset based on their attributes and balances the Mixup ratio between the groups. Through the reorganization and balance of Mixup among groups, GBMix effectively enhances both average and worst-case accuracy concurrently. We empirically show that GBMix effectively mitigates bias in the training set and reduces the performance gap between groups. This effect is observed across a range of datasets and networks, and GBMix outperforms all the state-of-the-art methods.

Abstract Osteoclasts are the primary target for osteoporosis drug development. Recent animal studies revealed the crucial roles of osteoblasts in regulating osteoclastogenesis and the longer lifespans of osteoclasts than previously thought with fission and recycling. However, existing culture platforms are limited to replicating these newly identified cellular processes. We report a demineralized bone paper (DBP)-based osteoblast culture and osteoclast assay platform that replicates osteoclast fusion, fission, resorption, and apoptosis with high fidelity and analytical power. An osteoid-inspired DBP supports rapid and structural mineral deposition by osteoblasts. Coculture osteoblasts and bone marrow monocytes under biochemical stimulation recapitulate osteoclast differentiation and function. The DBP-based bone model allows longitudinal quantitative fluorescent monitoring of osteoclast responses to bisphosphonate drug, substantiating significantly reducing their number and lifespan. Finally, we demonstrate the feasibility of humanizing the bone model. The DBP-based osteo assay platforms are expected to advance bone remodeling-targeting drug development with improved prediction of clinical outcomes.

LiFePO4 emerges as a viable alternative to cobalt‐containing cathodes, such as Li[Ni1–x–yMnxCoy]O2 and Li[Ni1−x−yCoxAly]O2. As Fe is abundant in nature, LiFePO4 is a low‐cost material. Moreover, stable structure of LiFePO4 imparts long service life and thermal stability. However, the practical implementation of LiFePO4 cathode in energy storage devices is impeded by its low energy density and high ionic/electrical resistance. Herein, the LiFePO4 electrode with high active material loading and low ionic/electrical resistance through the dry process is reported for the first time. The dry process not only enables the uniform distribution of the polymeric binders and conductive additives within the thick electrode but also inhibits the formation of cracks. Furthermore, the bridge‐like connection of polytetrafluoroethylene facilitates the insertion and extraction of Li ions to the LiFePO4 crystal. Hence, the dry‐processed LiFePO4 electrode with high areal capacity (7.8 mAh cm−2) exhibits excellent cycle stability over 300 cycles in full‐cell operation. In addition, it is demonstrated that the estimated energy density of prismatic cell with the dry‐processed LiFePO4 electrode is competitive with state‐of‐the‐art Li[Ni1–x–yMnxCoy]O2‐based battery.

Recent meta-learning algorithms for few-shot learning are based on episodic training where each episode consists of only a few support and query samples to imitate a target few-shot task. However, due to the limited number of categories and few samples in each category, this framework suffers from over-fitting to both a meta-training dataset and the support set of each episode. It also causes a large variance in the accuracy of each episode, which reduces reliability and confidence in model performance. To address this problem, we propose a novel meta-ensemble learning approach based on a recent ensemble method: a multi-input multi-output (MIMO) configuration. Our approach is simply applied to existing meta-learning algorithms. Multiple subnetworks in a single model simultaneously learn multiple episodes and ensemble the predictions, leveraging the model capacity. We show that meta-ensemble learning achieves significant improvement in generalization. It also improves the performance of meta-learning algorithms on few-shot classification benchmarks.

Despite major economic and technological advances, much of the ocean remains unexplored, which has led to the use of remotely operated vehicles (ROVs) and gliders for surveying. ROVs and underwater gliders are essential for ocean data collection. Gliders, which control their own buoyancy, are particularly effective unmanned platforms for long-term observations. The traditional method of recovering the glider on a small boat is a risky operation and depends on the skill of the workers. Therefore, a safer, more efficient, and automated system is needed to recover them. In this study, we propose a lightweight artificial neural network for underwater glider detection that is efficient for learning and inference. In order to have a smaller parameter size and faster inference, a convolutional neural network (CNN) vision encoder in an artificial neural network splits an image of a glider into a number of elongated patches that overlap to better preserve the spatial information of the pixels in the horizontal and vertical directions. Global max-pooling, which computes the maximum over all the spatial locations of an input feature, was used to activate the most salient feature vectors at the end of the encoder. As a result of the inference of the glider detection models on the test dataset, the average precision (AP), which indicates the probability that an object is located within the predicted bounding box, shows that the proposed model achieves AP = 99.7%, while the EfficientDet-D2 model for comparison of detection performance achieves AP = 69.2% at an intersection over union (IOU) threshold of 0.5. Similarly, the proposed model achieves an AP of 78.9% and the EfficientDet-D2 model achieves an AP of 50.5% for an IOU threshold of 0.75. These results show that accurate prediction is possible within a wide range of recall for glider position inference in a real ocean environment.

In this paper, we propose a new coded computation scheme that can alleviate straggler effects in distributed computing. We consider data security and master’s privacy for matrix multiplication tasks. The proposed scheme, called fully private and secure coded matrix multiplication (FPSCMM), ensures data security and master’s privacy on two data matrices for multiplication tasks from colluding workers. We also show that the storage overhead at workers can be reduced by FPSCMM, since it is enough for workers to store the encoded matrices with sub-blocks. Lastly, we compare FPSCMM with the existing master’s privacy-preserving coded matrix multiplication schemes.

We consider information-theoretic privacy in federated submodel learning, where a global server has multiple submodels. Compared to the privacy considered in the conventional federated submodel learning where secure aggregation is adopted for ensuring privacy, information-theoretic privacy provides the stronger protection on submodel selection by the local machine. We propose an achievable scheme that partially adopts the conventional private information retrieval (PIR) scheme that achieves the minimum amount of download. With respect to computation and communication overhead, we compare the achievable scheme with a naïve approach for federated submodel learning with information-theoretic privacy.

Neural architecture search (NAS) methods automatically find optimal neural networks without human assistance. Numerous algorithms for NAS have been studied to find architectures with gradient-based search. Differentiable architecture search (DARTS), one of the key papers of gradient-based search, dramatically reduced search cost, and showed outstanding performance through continuous relaxation and meta-learning based approximation. However, one of the issues with DARTS is that the gradient-based search process is biased due to the nested bi-level optimization structure, and the greedy behavior of the gradient descent. As a result, there is a problem that search spaces are limited to a limited set of architectures. To overcome the bias of the gradient-based search in network architecture search (NAS), we used a dynamic search method. This technique allows gradient-based search to have exploration. In this paper, we present a novel approach, namely, Dynamic-Exploration DARTS (DE-DARTS). For effective exploration, we use dynamic attention networks (DANs) in DE-DARTS, which change model architectures based on input data. As our DANs are activated early in the search, more diverse architectures are considered, depending on input data at the beginning of search. Our algorithm is evaluated in multiple image classification datasets including CIFAR-10, and ImageNet, and shows improved performance.

This paper proposes a coupled-error-based formation control algorithm for the rapid formation completion of multi-robot systems. We consider a multi-robot system with omni-directional robots with swerve-driving mechanisms and a communication system with minimized constraints. This paper introduces a coupled error that links the distance errors with the leading robot and the following robot through a coupling ratio. We propose a controller using the coupled error to achieve the control objectives of this paper. Unlike existing results that only use the information of the preceding robot, this algorithm couples the information of both the preceding robot and one’s follower. Using the proposed error-coupling-based formation control algorithm, multi-robot systems can quickly establish formations for collaboration, allowing tasks to commence swiftly and reducing deformations in formations due to speed variations. With stability analysis and simulation results for the practical application of the proposed algorithm, the approach has been verified to improve the speed of both the completion of the formation and overall system trajectory tracking, balancing trade-offs between them.

Summary: Acoustic tweezers (ATs) are a promising technology that can trap and manipulate microparticles or cells with the focused ultrasound beam without physical contact. Unlike optical tweezers, ATs may be used for in vivo studies because they can manipulate cells through tissues. However, in previous non-invasive microparticle trapping studies, ATs could only trap spherical particles, such as beads. Here, we present a theoretical analysis of how the acoustic beam traps red blood cells (RBCs) with experimental demonstration. The proposed modeling shows that the trapping of a non-spherical, biconcave-shaped RBC could be successfully done by single-beam acoustic tweezers (SBATs). We demonstrate this by trapping RBCs using SBATs in the Rayleigh regime, where the cell size is smaller than the wavelength of the beam. Suggested SBAT is a promising tool for cell transportation and sorting.

Abstract Venadaparib (VEN), a next-generation inhibitor of poly (ADP-ribose) polymerases, is under development for oral use in patients having cancers with deoxyribonucleic acid repair defects. The objective of this study was to develop and validate a sensitive and robust analytical method for quantifying VEN in a small volume of plasma samples from rats and dogs, and to assess the feasibility of the assay for application in pharmacokinetic/toxicokinetic studies. Plasma samples were subjected to deproteination, and an aliquot was injected into an LC–MS/MS system. VEN and imipramine were analyzed in the positive ion mode and quantified by monitoring the transition at m/z 407.2 → 70.0 for VEN and 281.2 → 86.1 for imipramine. The lower and upper limits of the assay were determined to be 1 and 1000 ng/mL, respectively, with acceptable linearity (r 2 > 0.995). Validation parameters, such as accuracy, precision, dilution, recovery, matrix effect, and stability, were within acceptable ranges. This method was adequately applied to the characterization of pharmacokinetics of VEN in rats and dogs at the oral dose of 30 and 0.5 mg/kg, respectively. These findings suggest that the validated assay is applicable to the kinetic studies of VEN with a small volume of plasma samples from the animals.

Network Augmentation (NetAug) is a recent method used to improve the performance of tiny neural networks on large-scale datasets. This method provides additional supervision to tiny models from larger augmented models, mitigating the issue of underfitting. However, the capacity of the augmented models is not fully utilized, resulting in underutilization of resources. In order to fully utilize the capacity of a larger augmented model without exacerbating the underfitting of a tiny model, we propose a new method called Multi-Input Network Augmentation (MINA). MINA converts tiny neural networks into a multi-input configuration, allowing only the augmented model to receive more diverse inputs during training. Additionally, tiny neural network can be converted back into their original single-input configuration after training. Our extensive experiments on large-scale datasets demonstrate that MINA is effective in improving the performance of tiny neural networks. We also demonstrate that MINA is consistently effective in downstream tasks, such as fine-grained image classification tasks and object detection tasks.

As deep learning has been successfully deployed in diverse applications, there is an ever increasing need to explain its decision. To explain decisions, case-based reasoning has proved to be effective in many areas. The prototype-based explanation is a method that provides an explanation of the model’s prediction using the distance between an input and learned prototypes to effectively perform case-based reasoning. However, existing methods are less reliable because distance is not always consistent with human perception. In this study, we construct a latent space which we call an $explanation space$ with distributional embedding and latent space regularization. This explanation space ensures that similar (in terms of human-interpretable features) images share similar latent representations, and therefore provides a reliable explanation for the consistency between distance-based explanation and human perception. The explanation space also provides additional explanation by transition, allowing the user to understand the factors that affect the distance. Throughout extensive experiments including human evaluation, we have shown that the explanation space provides a more human-understandable explanation.

In this study, a flake-shaped metal powder was coated on a tube shaped pre-sintered 316L stainless steel support using a wet powder spraying process to fabricate a double pore structure, and the pore characteristics were analyzed according to coating time and tube rotation speed. The thickness of the coated layer was checked via optical microscopy, and porosity was measured using image analysis software. Air permeability was measured using a capillary flow porometer. As a result of the experiment, the optimal rotation speed of the support tube was established as 200 rpm. When the rotation speed was fixed, the coating thickness and the coating amount of the double pore structure increased as the coating time increased. The porosity of the double pore structure was increased due to the irregular arrangement of the flake-shaped powder. The air permeability of the double pore structure decreased with increasing fine pore layer thickness.

Recently, 3D printing processes have been used to manufacture metal powder filters with manufacturing complex-shape. In this study, metal powder filters of various shapes were manufactured using the metal extrusion additive manufacturing (MEAM) process, which is used to manufacture three-dimensional structures by extruding a filament consisting of a metal powder and a binder. Firstly, filaments were prepared by appropriately mixing SUS316 powder with sizes ranging from 7.5 µm to 50 µm and a binder. These filaments were extruded at temperatures of 100℃ to 160℃ depending on the type of filament being manufactured, to form three types of cylindrical filter. Specimens were sintered in a high vacuum atmosphere furnace at 850℃ to 1050℃ for 1 hour after debinding. The specimens were analyzed for permeability using a capillary flow porometer, porosity was determined by applying Archimedes’ law and microstructure was observed using SEM.

Abstract Hyperspectral endoscopy has shown its potential to improve disease diagnosis in gastrointestinal tracts. Recent approaches in developing hyperspectral endoscopy are mainly focusing on enhancing image speed and quality of spectral information under a clinical environment, but there are many issues in obtaining consistent spectral information due to complicated imaging conditions, including imaging angle, non-uniform illumination, working distance, and low reflected signal. We quantitatively investigated the effect of imaging angle on the distortion of spectral information by exploiting a bifurcated fiber, spectrometer, and tissue-mimicking phantom. Spectral distortion becomes severe as increasing the angle of the imaging fiber or shortening camera exposure time for fast image acquisition. Moreover, spectral ranges from 450 to 550 nm are more susceptible to the angle-dependent spectral distortion than longer spectral ranges. Therefore, imaging angles close to normal and longer target spectral ranges with enough detector exposure time could minimize spectral distortion in hyperspectral endoscopy. These findings will help implement clinical HSI endoscopy for the robust and accurate measurement of spectral information from patients in vivo.

Ethical management is not a choice for companies, but a necessity. This study, therefore, filled a need for further research into ethical management. We looked at the factors involved in a company’s employment of an ethical management system and relationships between factors to determine how the subordinate variable, the ethical management system, affects the performance of an enterprise.In specific, this study identified internal and external factors involved in ethical management systems in Korea’s domestic shipping and port logistics companies. We investigated the impact of these factors on organizational performance variables and each level of the ethical management system.According to empirical findings, only internal factors such as openness and internal integrity, and external factors such as normative and cognitive environments, had positive effects on the use of an ethical management system. On the other hand, transparency, which is an internal factor, and regulatory environment, which is an external factor, had no positive impact.

In this study, five Smart Construction Objects (SCOs) are developed and demonstrated. The usefulness and characteristics of the developed SCOs were evaluated using a tri-axial diagram analysis method. In addition, a smart construction management system associated with the developed SCOs is proposed. The efficacy of the management system is demonstrated by applying it to various types of construction site, including tunnel construction, railway construction, and underground water tank construction. The results of the tri-axial diagram analysis showed that the developed SCOs have improved awareness and autonomy compared to previously available ones. Multiple on-site applications of the developed smart construction management system resulted in a significant reduction in the time and cost required for construction by more than 40% and 70%, respectively, compared to conventional methods. This is attributable to the simplification of construction resource management procedures, JSA (Job Security Analysis), TBM (Tool Box Meeting), PTW (Permit to Work), and nonconformity management.

In contrast to the 4G LTE, one of the key requirements of 5G and beyond networks is ultra-high reliability. Towards this end, we investigate the problem of scheduling with joint power allocation for multi-user multiple-input multiple-output (MU-MIMO) systems by taking the reliability into account. In this study, two novel MU-MIMO scheduling methods are proposed based on the use of bit error rate (BER) as the scheduling metric; however, capacity has been typically considered as an user-selection metric in MU-MIMO systems to increase the sum-throughput. In particular, we also develop a new power allocation strategy that can minimize the BER for a given data transmission rate. The computational complexities of the two aforementioned BER-based scheduling algorithms, as well as the conventional capacity-based algorithm, are analyzed. Our simulation results show that the proposed BER-based scheduling algorithms with the joint power allocation have better performance in terms of BER compared to the conventional MIMO scheduling schemes such as a single-user MIMO (SU-MIMO) and the capacity-based scheduling algorithm. The proposed multi-user scheduling framework can be utilized in various emerging applications demanding higher reliability and availability such as mission-critical services.

In this paper, we consider coded computation for matrix multiplication tasks in distributed computing, which can mitigate the effect of slow workers, called stragglers, using a coding approach. We assume that the straggler computation results can be leveraged at the master by assigning multiple sub-tasks to the workers. In this scenario, a new coded computation scheme is proposed to preserve the data security and privacy from workers, which is called securely straggler-exploiting codes (SSEC). Moreover, the proposed SSEC can efficiently reduce the communication load in distributed computing for assigning the sub-tasks to the workers, by overlapping the encoded matrices in assigning multiple sub-tasks with appropriate polynomial functions. It is also proven that the data security and privacy constraints are satisfied in SSEC in an information-theoretic sense. In conclusion, SSEC shows good performance on the recovery threshold and communication loads and compare them with the existing secure coded computation schemes for matrix multiplication tasks.

In this study, experiments were conducted on 30 subjects by means of electrocardiogram (ECG) and electroencephalogram (EEG) methodologies as well as a money game to examine the effects of stress on creativity in business problem-solving. The study explained the relationship between creativity and human physiological response using the biopsychosocial model of challenge and threat. The subjects were asked to perform a cognitive mapping task. Based on the brain wave theory, we identified the types of brain waves and locations of brain activities that occurred during the creative problem-solving process in a business environment and studied the effects of stress on creativity. The results of the experiments showed significant differences in creativity in business problem-solving depending on whether or not stress was triggered. Differences were found in the time domain (SDNN, RMSSD) and frequency domain (HF, LF/HF ratio) of heart rates, a physiological stress indicator, between the stress group and the no-stress group. A brain wave analysis confirmed that alpha waves increased in the frontal lobe of the brain during creative business problem-solving but decreased when the subjects were under stress, during which beta waves in the brain increased. This study seeks to examine creativity in business problem-solving by studying the effects of stress on human physiological response and cognitive functions in the hope of providing a new and objective interpretation of existing research results.

Background Korea, as one of the fastest-aging countries worldwide, requires an improved healthcare service model for older adults. We evaluated the current healthcare system and developed a service model based on information and communication technologies (ICT) for use in older patients in long-term care facilities (LTCF). Methods We conducted a qualitative literature review, focus group interviews (FGIs), and structured survey to identify the current technology use and status of healthcare systems. We then developed a web-based platform with necessary, high-priority, and usable content for the care of older patients in LTCF. Results We reviewed 60 (23 hypertension, 18 diabetes, and 19 heart failure) articles on information and communication technologies (ICT)-based disease management for clinical effectiveness and improved patient satisfaction. FGIs and structured surveys were used to evaluate the inconvenience in patient and medical information transfer between hospitals and cost and time required for its process. Accordingly, we confirmed the unmet need for an ICT-based service model for management, monitoring, and consultation among older patients and developed the Health-RESPECT (integrated caRE Systems for elderly PatiEnts using iCT), a service platform for older patients residing in LTCF. The medical information exchange system was used to transfer medical information. Health-RESPECT includes an established algorithm for evidence-based comprehensive geriatric assessment and customized management; chronic disease management; management of potentially inappropriate medications; rehabilitation; and consultation and videoconferencing. Conclusion This study identified the current status and unmet needs of healthcare systems for older adults. We developed an ICT-based system to manage older institutionalized patients. However, the Health-RESPECT service model requires further validation.

This paper introduces novel driving methods of the pull-down unit in a gate driver circuit for enhancement- and depletion-mode a-IGZO thin-film transistors (TFTs). The proposed gate driver circuit can achieve uniform output characteristics and effectively reduce the VOUT ripple voltage because the threshold voltage (VTH) of the pull-down units is compensated regardless of the a-IGZO TFT operation characteristics (enhancement mode: positive value of VTH, depletion mode: negative value of VTH). Many groups proposed the VTH compensation method for pull-down TFTs in the gate driver circuit using a diode connection structure. However, the diode connection structure to extract the VTH value cannot be applied in the depletion-mode oxide TFTs because TFT enters the turn-on state even when the VGS value is 0 V. To solve this problem, we adopted the VTH extraction period only once in one frame time. As a result, our circuit can compensate for VTH of the pull-down unit in the enhancement mode and can be normally operated in the depletion mode. Adjunctively, two low signals (VGL1 and VGL2) and QC node were designed to prevent the leakage current path for Q and VOUT nodes. To verify the threshold voltage tolerance for various stress conditions, we demonstrated the reliability of the circuit according to the threshold voltage change of the TFTs. The simulation result shows that all the VOUT waveforms are maintained at +28 V (VGH) under the VTH shift conditions from -7 V to +11 V; further, the rising time and falling time are less than 0.62 μs and 0.96 μs, respectively. Based on a 120Hz ultra-high definition (UHD) graphics (3840×2160) display panel, the proposed circuit has uniform VOUT characteristics compared to previous VTH compensation circuit when ΔVTH changes from -3 V to +11 V. When ΔVTH changes from -4 V to -7 V, there is also no circuit malfunction, even with slight increase in the falling time and power consumption.

In this paper, we consider a power allocation optimization technique for a time-varying fading broadcast channel in energy harvesting communication systems, in which a transmitter with a rechargeable battery transmits messages to receivers using the harvested energy. We first prove that the optimal online power allocation policy for the sum-rate maximization of the transmitter is a monotonically increasing function of harvested energy, remaining battery, and the channel gain of each user. We then construct a lightweight neural network architecture to take advantage of the monotonicity of the optimal policy. This two-step approach, which relies on effective function approximation to provide a mathematical guideline for neural network design, can prevent us from wasting the representational capacity of neural networks. The tailored neural network architectures enable stable learning and eliminate the heuristic network design. For performance assessment, the proposed approach is compared with the closed-form optimal policy for a partially observable Markov problem. Through additional experiments, it is observed that our online solution achieves a performance close to the theoretical upper bound of the performance in a time-varying fading broadcast channel.

In a distributed computing environment, there may exist slow processing workers, which are known as “stragglers”, and they can slow down the whole computing process. In this article, we consider coded computation for matrix multiplication tasks in distributed computing, which can mitigate the effect of stragglers by a coding approach. We propose a new communication-efficient coded computation scheme, namely squeezed polynomial codes, for a straggler-exploiting scenario where multiple sub-tasks are assigned to the workers to partially leverage the computation capability of stragglers. The key idea of squeezed polynomial codes is to overlap the encoded matrices in assigning multiple sub-tasks with appropriate polynomial functions in order to reduce the task-allocation communication load from a master to its workers. We compare squeezed polynomial codes with the existing schemes for distributed matrix multiplication in a communication load perspective. Consequently, we show that squeezed polynomial codes can efficiently reduce the communication load while ensuring the optimal recovery condition at a master to obtain final product.

We recently investigated the design and fabrication of Langevin-type transducers for therapeutic ultrasound. Effect of ultrasonic energy arising from the transducer on biological tissue was examined. In this study, the transducer was set to radiate acoustic energy to mesenchymal stem cells (MSCs) for inducing differentiation into cartilage tissue. The average chondrogenic ratio in area was 20.82% in the control group, for which no external stimulation was given. Shear stress was applied to MSCs as the contrast group, which resulted in 42.66% on average with a 25.92% minimum rate; acoustic pressure from the flat tip causing transient cavitation enhanced chondrogenesis up to 52.96%. For the round tip excited by 20 Vpp, the maximum differentiation value of 69.43% was found, since it delivered relatively high acoustic pressure to MSCs. Hence, the results from this study indicate that ultrasound pressure at the kPa level can enhance MSC chondrogenesis compared to the tens of kHz range by Langevin transducers.

We previously conducted an empirical study on Langevin type transducers in medical use by examining the heat effect on porcine tissue. For maximum acoustic output, the transducer was activated by a continuous sinusoidal wave. In this work, pulsed waves with various duty factors were applied to our transducer model in order to examine their effect on functionality. Acoustic power, electro-acoustic conversion efficiency, acoustic pressure, thermal effect on porcine tissue and bovine muscle, and heat generation in the transducer were investigated under various input conditions. For example, the results of applying a continuous wave of 200 VPP and a pulse wave of 70% duty factor with the same amplitude to the transducer were compared. It was found that continuous waves generated 9.79 W of acoustic power, 6.40% energy efficiency, and 24.84 kPa acoustic pressure. In pulsed excitation, the corresponding values were 9.04 W, 8.44%, and 24.7 kPa, respectively. The maximum temperature increases in bovine muscle are reported to be 83.0 °C and 89.5 °C for each waveform, whereas these values were 102.5 °C and 84.5 °C in fatty porcine tissue. Moreover, the heat generation around the transducer was monitored under continuous and pulsed modes and was found to be 51.3 °C and 50.4 °C. This shows that pulsed excitation gives rise to less thermal influence on the transducer. As a result, it is demonstrated that a transducer triggered by pulsed waves improves the energy efficiency and provides sufficient thermal impact on biological tissues by selecting proper electrical excitation types.

This study proposes a novel autonomous system that is able to retrieve an underwater glider safely and quickly. After a long-period mission, the glider is out of its energy, floats up in the middle of the sea, and needs retrieval. However, most of the recovery processes are conducted manually, which is not effective and even damages the glider or injures the worker. In this study, we propose the use of a remotely operated vehicle (ROV) in a recovery process that consists of three stages: (i) driving the ROV into the near area of the glider, (ii) automatically catching the glider using the ROV’s onboard sensors and capturing mechanisms, and (iii) retrieving the glider to the mother ship or station. Mathematical models were formulated to realize the feasibility of the system. A cascade adaptive sliding mode control-proportional integral controller is designed to cope with the thruster’s characteristics, the dynamics of the ROV and the glider, and oceanic disturbances. Finally, simulation studies were conducted in the system operation scenarios and their results validate the efficiency of the proposed process.

The dominant position of e-commerce is especially being articulated in the retailing industry once again due to several constraints that the world faces in the COVID-19 pandemic era. In this regard, this study explores the significant role of trust transfer (from offline to online) and the moderating effect of consumers’ neurotic traits in the framework of trust-satisfaction-repurchase intention in the e-commerce context based on a survey with 406 Korean e-commerce consumers. Moreover, a prediction-oriented segmentation (POS) technique combined with structural equation models (SEM) was utilized to reveal consumers’ probable hidden heterogeneous characteristics. The outcomes of the global model SEM analysis indicate that offline-online trust transference occurs in e-commerce, and the conveyed trust significantly influences satisfaction and consumers’ repeat purchase intention through satisfaction. Neuroticism also has significant positive effects on trust transfer in the global model. However, results in three subgroups generated by POS show heterogeneous characteristics that considerably differed from the global model test results. The implications from this study will be beneficial to field practitioners in the e-commerce industry in addressing the importance of trust transfer, negative neurotic traits as well as heterogeneous aspects of consumers.

A simple relay selection scheme is proposed for analog network coding using differential modulation (ANC-DM). We propose a relay selection scheme based solely on the received signal power in which their own information is canceled out at each source node. The proposed power-based relay selection scheme is further applied to analog network coding using double-differential modulation (ANC-DDM), which is suitable for channels with carrier offset. Simulation results show that our proposed simple power-based criterion for relay selection in ANC-DM system has little loss compared to the optimal relay selection scheme using ideal BER. Keywords: Analog network coding, Differential modulation, Double-differential modulation, Relay selection

We propose a novel collaborative filtering algorithm based on deep neural networks. We use normalized user-rating vector and normalized item-rating vector as inputs to a neural network. The batch normalization technique is used for each layer to prevent neural networks from overfitting. Experimental results show that the proposed method outperforms conventional collaborative filtering algorithms. Based on the results, its performance is comparable to the well-known Netflix prize winning algorithm by BellKor. The proposed method has another strong advantage that online operation is possible with little extra complexity and performance degradation. Keywords: Collaborative Filtering, Neural Network

Dynamic adaptive streaming over HTTP (DASH) inherently requires many transcoding operations to produce various bit-rate versions. However, transcoding is highly computationally intensive, resulting in significant power consumption in transcoding servers. To limit the amount of energy consumption in transcoding servers, some transcoding tasks can be skipped and executed later, but this poses several questions, such as (1) which versions can be transcoded before an actual streaming session, (2) how to limit total amount of transcoding energy consumption, and (3) how to handle requests to the untranscoded versions. To address this, we first introduce the concept of transcoding gain to express the overall video quality assigned to each transcoded version by considering segment popularity and go on to propose an algorithm that determines when and which versions should be transcoded with the aim of maximizing the overall transcoding gain by limiting energy consumption. To achieve this, it allows lower bit-rate versions to be streamed than are requested, at the cost of minimal video quality degradation by considering the overall video quality. The proposed scheme was implemented on a clustered system where various issues were resolved. The experimental results show that the transcoding selection algorithms can effectively control the amount of power consumption, minimizing the effect of overall video quality degradation. For example, it can save up to 65% energy when video quality degradation is allowed to be 4.3% at maximum compared with the scheme that transcodes all the versions.

A wireless powered communication network (WPCN) is considered, which has a hybrid access point (H-AP) and two users with energy harvesting capabilities. The H-AP transfers energy to the users during downlink (DL) transmission, and the users transmit information to the H-AP during uplink (UL) transmission by utilizing the harvested energy. To enhance user fairness, a user cooperation protocol is employed in which a user close to the H-AP relays the message of the other user by partially using its allocated time and available energy resources. By jointly optimizing the time allocation for DL energy transfer and UL information transmission, we maximize the sum-throughput in the considered network subject to a set of quality of service (QoS) requirements. The optimization problem is solved through a divide-and-conquer approach along with convex optimization techniques, and thus an analytical solution is derived. In particular, the conditions under which the WPCN benefits from the user cooperation in achieving the maximum sum-throughput are fully characterized. Furthermore, an efficient resource allocation algorithm to find the optimal solution under QoS requirements is presented.

In a multi-agent system, multi-job assignment is an optimization problem that seeks to minimize total cost. This can be generalized as a complex problem in which several variations of vehicle routing problems are combined, and as an NP-hard problem. The parameters considered include the number of agents and jobs, the loading capacity, the speed of the agents, and the sequence of consecutive positions of jobs. In this study, a deep neural network (DNN) model was developed to solve the job assignment problem in a constant time regardless of the state of the parameters. To generate a large training dataset for the DNN, the planning domain definition language (PDDL) was used to describe the problem, and the optimal solution that was obtained using the PDDL solver was preprocessed into a sample of the dataset. A DNN was constructed by concatenating the fully-connected layers. The assignment solution obtained via DNN inference increased the average traveling time by up to 13% compared with the ground cost. As compared with the ground cost, which required hundreds of seconds, the DNN execution time was constant at approximately 20 ms regardless of the number of agents and jobs.

In a traditional distributed storage system, a source can be restored perfectly when a certain subset of servers is contacted. The coding is independent of the contents of the source. This paper considers instead a lossy source coding version of this problem where the more servers that are contacted, the higher the quality of the restored source. An example could be video stored on distributed storage. In information theory, this is called the multiple description problem, where the distortion depends on the number of descriptions received. The problem considered in this paper is how to restore the system operation when one of the servers fail and a new server replaces it, that is, repair. The requirement is that the distortions in the restored system should be no more than in the original system. The question is how many extra bits are needed for repair. We find an achievable rate and show that this is optimal in certain cases. One conclusion is that it is necessary to design the multiple description codes with repair in mind; just using an existing multiple description code results in unnecessary high repair rates.

We recently proposed an analytical design method of Langevin transducers for therapeutic ultrasound treatment by conducting parametric study to estimate the effect of compression force on resonance characteristics. In this study, experimental investigations were further performed under various electrical conditions to observe the acoustic power of the fully equipped transducer and to assess its heat-related bioeffect. Thermal index (TI) tests were carried out to examine temperature rise and thermal damage induced by the acoustic energy in fatty porcine tissue. Acoustic power emission, TI values, temperature characteristics, and depth/size of thermal ablation were measured as a function of transducer’s driving voltage. By exciting the transducer with 300 Vpp sinusoidal continuous waveform, for instance, the average power was 23.1 W and its corresponding TI was 4.1, less than the 6 specified by the Food and Drug Administration (FDA) guideline. The maximum temperature and the depth of the affected site were 74.5 °C and 19 mm, respectively. It is shown that thermal ablation is likely to be more affected by steep heat surge for a short duration rather than by slow temperature rise over time. Hence, the results demonstrate the capability of our ultrasonic transducer intended for therapeutic procedures by safely interrogating soft tissue and yet delivering enough energy to thermally stimulate the tissue in depth.

This work was designed to evaluate the interlayer strength of 3D-printed mortar with postinstalled interlayer reinforcement. Two methods of postinstalled interlayer reinforcement were considered according to the amount of overlapping. The first method did not include overlapping of the interlayer reinforcement, while the second method included overlap lengths of 20 and 40 mm. Additionally, two different curing conditions were considered: air-curing conditions and water-curing conditions. The compressive, splitting tensile, and flexural tensile strengths of 3D-printed mortar specimens with different reinforcement methods and curing conditions were investigated under three loading directions. The three loading directions were defined based on the three planes of the printed specimens. The compressive, splitting tensile, and flexural tensile strengths were dependent on the loading directions. In particular, the splitting and flexural tensile strengths decreased considerably when tensile stresses acted on the interlayers of the 3D-printed mortar specimens. However, when longitudinal interlayer reinforcement penetrated the printed layers, the flexural tensile strength or interlayer bonding strength of the printed specimens increased significantly at the interlayers. In addition, mortar specimens reinforced with overlap lengths of 20 and 40 mm were investigated in this study. The flexural tensile strength or interlayer bonding strength of 3D-printed mortar decreased after treatment under air-curing conditions because the interlayers of the printed mortar formed more pores under these conditions and were more vulnerable under loading. Finally, the findings of this study suggested that interlayer reinforcement is a potential method for improving the interlayer bonding strength of 3D-printed mortar.

Purpose: This randomized control trial was designed to examine the effect on the self-managing ability for promoting bone health of mobile type–bone health intervention (mobile type–BHI). Methods: The mobile type–BHI consisted of a mobile application called “Strong bone, Fit body” (SbFb) and group education. A total of 82 college women with low bone mass (Z score < –1) participated. They were assigned randomly to three groups, experimental group I, (n = 28), experimental group II (n = 32), and control group (n = 22). This study ran from June 2014 to January 2015. The outcome variables were bone mineral density, minerals related to bone metabolism (calcium, phosphorus, vitamin D), biochemical markers related to bone remodeling (osteocalcin, C-terminal telopeptide, sclerostin), food intake diary by 24 hours recall, and psychosocial variables related to bone health (knowledge, health belief and self-efficacy). Data were analyzed using SAS program and a computer aided nutritional analysis program. Results: Both the experimental group I, who used mobile type–BHI, and experimental group II, who only received group education, showed outcomes regarding knowledge of the benefits of exercise and calcium as compared with the control group. The two experimental groups also demonstrated results in the serum levels of calcium, vitamin D, and sclerostin compared to those of the control group. Conclusion: Although both experimental groups exhibited positive outcomes in regards to the promotion of bone health, this study did not show an additional effect of the mobile application on self-management ability for the promotion of bone health. Nonetheless, the SbFb application is very meaningful as it is the first application developed with the aim of improving women's bone health.

Although there is a growing trend to use higher strength for concrete and steel in reinforced concrete structures due to the lightness and slenderness of these members together with the simplified arrangement of their reinforcement, there is still the necessity to inspect the reduction of ductility resulting from the gain in strength. Taking into account that this also concerns the design for torsion, this study intends to investigate the regulations related to the torsional minimum reinforcement ratio in view of the minimum ductility requirement with focus on Eurocode 2. To that goal, the relation between the torsional cracking moment and the ductile behavior is discussed for the beam reinforced with the minimum torsional reinforcement ratio to examine the eventual properness of the minimum torsional reinforcement ratio recommended by Eurocode 2. Moreover, a pure torsion test is performed on 18 beams made of 80 MPa concrete reinforced by high-strength bars with rectangular section and various test variables involving the minimum torsional reinforcement ratio, the transverse-to-longitudinal reinforcement ratio, and the total reinforcement ratio. As a result, for the high-strength concrete beams, the minimum torsional reinforcement ratio recommended by Eurocode 2 was insufficient to prevent the sudden loss of strength after the initiation of the torsional cracking. But with regard to the compatibility torsion of statically indeterminate structure, the adoption of the minimum torsional reinforcement ratio recommended by Eurocode 2 might secure enough deformability under displacement-controlled mode to allow the redistribution of the torsional moment.