• International conference on construction engineering and project management
• /
• 2022.06a
• /
• pp.877-885
• /
• 2022

Vision-based activity recognition has been widely attempted at construction sites to estimate productivity and enhance workers' health and safety. Previous studies have focused on extracting an individual worker's postural information from sequential image frames for activity recognition. However, various trades of workers perform different tasks with similar postural patterns, which degrades the performance of activity recognition based on postural information. To this end, this research exploited a concept of human-object interaction, the interaction between a worker and their surrounding objects, considering the fact that trade workers interact with a specific object (e.g., working tools or construction materials) relevant to their trades. This research developed an approach to understand the context from sequential image frames based on four features: posture, object, spatial features, and temporal feature. Both posture and object features were used to analyze the interaction between the worker and the target object, and the other two features were used to detect movements from the entire region of image frames in both temporal and spatial domains. The developed approach used convolutional neural networks (CNN) for feature extractors and activity classifiers and long short-term memory (LSTM) was also used as an activity classifier. The developed approach provided an average accuracy of 85.96% for classifying 12 target construction tasks performed by two trades of workers, which was higher than two benchmark models. This experimental result indicated that integrating a concept of the human-object interaction offers great benefits in activity recognition when various trade workers coexist in a scene.

• Journal of Dental Anesthesia and Pain Medicine
• /
• v.23 no.5
• /
• pp.281-285
• /
• 2023

Background: Dentists bend needles prior to certain injections; however, there are concerns regarding needle fracture, lumen occlusion, and sharps handling. A previous study found that a 30-gauge needle fractures after four to nine 90° bends. This fatigue life study evaluated how many 90° bends a 30-gauge dental needle will sustain before fracture when bent using a needle guide. Methods: Two operators at Element Materials Technology, an independent testing, inspection, and certification company tested 48 30-gauge needles. After applying the needle guide, the operators bent the needle to a 90° angle and expressed the anesthetic from the tip. The needle was then bent back to a 0° angle, and the functionality was tested again. This process was repeated until the anesthetic failed to pass through the end of the needle due to fracture or obstruction. Each operator tested 24 needles (12 needles from each lot), and the number of sustained bends before the needle fracture was recorded. Results: The average number of sustained bends before needle failure was 40.33 (95% confidence interval = 37.41-43.26), with a minimum of 20, median of 40, and a maximum of 54. In each trial, the lumen remained patent until the needle fractured. The difference between the operators was statistically significant (P < 0.001). No significant differences in performance between needle lots were observed (P = 0.504). Conclusion: Our results suggest that using a needle guide increases the number of sustained bends before needle fracture (P < 0.000001) than those reported in previous studies. Future studies should further evaluate the use of needle guides with other needle types across a variety of operators. Furthermore, additional opportunities lie in exploring workplace safety considerations and clinical applications of anesthetic delivery using a bent needle.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_1
• /
• pp.621-635
• /
• 2023

The assessment of structural condition is a crucial process for evaluating its usability and determining the diagnostic cycle. The currently employed manpower-based methods suffer from issues related to safety, efficiency, and objectivity. To address these concerns, research based on deep learning using images is being conducted. However, acquiring structural damage data is challenging, making it difficult to construct a substantial amount of training data, thus limiting the effectiveness of deep learning-based condition assessment. In this study, we propose a foundation model-based 2-step structural damage analysis to overcome the lack of training data in image-based structural condition assessments. We subdivided the elements of structural condition assessment into instantiation and quantification. In the quantification step, we applied a foundation model for image segmentation. Our method demonstrated a 10%-point increase in mean intersection over union compared to conventional image segmentation techniques, with a notable 40%-point improvement in the case of rebar exposure. We anticipate that our proposed approach will enhance performance in domains where acquiring training data is challenging.

• Journal of Internet of Things and Convergence
• /
• v.9 no.6
• /
• pp.37-44
• /
• 2023

As a primary technology of Industry 4.0, human-robot collaboration (HRC) requires additional measures to ensure worker safety. Previous studies on avoiding collisions between collaborative robots and workers mainly detect collisions based on sensors and cameras attached to the robot. This method requires complex algorithms to continuously track robots, people, and objects and has the disadvantage of not being able to respond quickly to changes in the work environment. The present study was conducted to implement a web-based platform that manages collaborative robots by recognizing the emotions of workers - specifically their perception of danger - in the collaborative process. To this end, we developed a web-based application that collects and stores emotion-related brain waves via a wearable device; a deep-learning model that extracts and classifies the characteristics of neutral, positive, and negative emotions; and an Internet-of-things (IoT) interface program that controls motor operation according to classified emotions. We conducted a comparative analysis of our system's performance using a public open dataset and a dataset collected through actual measurement, achieving validation accuracies of 96.8% and 70.7%, respectively.

• Journal of the Korean Society of Industry Convergence
• /
• v.26 no.5
• /
• pp.875-884
• /
• 2023

Hydrogen infiltration into metals has been reported to induce alterations in their mechanical properties under load. In this study, we conducted CTOD (Crack Tip Opening Displacement) tests on steel specimens designed for use in liquid hydrogen storage systems. Electrochemical hydrogen charging was performed using both FCC series austenitic stainless steel and BCC series structural steel specimens, while CTOD testing was carried out using a 500kN-class material testing machine. Results indicate a notable divergence in behavior: SS400 test samples exhibited a higher susceptibility to failure compared to austenitic stainless steel counterparts, whereas SUS 316L test samples displayed minimal changes in displacement and maximum load due to hydrogen charging. However, SEM (Scanning Electron Microscopy) analysis results presented challenges in clearly explaining the mechanical degradation phenomenon in the tested materials. This study's resultant database holds significant promise for enhancing the safety design of liquid hydrogen storage systems, providing invaluable insights into the performance of various steel alloys under the influence of hydrogen embrittlement.

• Journal of Ginseng Research
• /
• v.47 no.6
• /
• pp.735-742
• /
• 2023

Background: Gintonin is a new material of ginseng that acts through the ginseng-derived lysophosphatidic acid (LPA) receptor ligand. The gintonin-enriched fraction (GEF) inhibits amyloid plaque accumulation in the cortex and hippocampus, improves cognitive dysfunction by increasing acetylcholine levels, and promoted hippocampal neurogenesis in an animal model of Alzheimer's disease. We evaluated the effect of the GEF on the cognitive performance of subjects with subjective memory impairment (SMI). Methods: In this eight-week, randomized, assessor- and participant-blinded, placebo-controlled study, participants with SMI were assigned to three groups receiving placebo, GEF 300 mg/day or GEF 600 mg/day. The Korean versions of the Alzheimer's Disease Assessment Scale (K-ADAS), Mini-Mental State Examination (K-MMSE), and Stroop color-word test (K-SCWT) were also evaluated along with the safety profiles. Results: One hundred thirty-six participants completed the study. After eight weeks, we analyzed intergroup differences in primary or secondary outcome score changes. When we compared the GEF group with the placebo group, we observed significant improvements in the K-ADAS and K-SCWT scores. The GEF group did not show a significant improvement in K-MMSE and BDI scores compared to the placebo group. No adverse events were observed in the gintonin and placebo groups for eight weeks. Conclusion: The GEF is safe and effective in improving subjective cognitive impairment related to both the K-ADAS and K-SCWT in this study. However, further large-scale and randomized controlled studies are warranted to secure other cognitive function tests besides the K-ADAS and K-SCWT, and to confirm the findings of the current study.

• Computers and Concrete
• /
• v.32 no.6
• /
• pp.577-594
• /
• 2023

Engineered cementitious composites with calcined clay limestone cement (LC³-ECC) as a kind of green, low-carbon and high toughness concrete, has recently received significant investigation. However, the complicated relationship between potential influential factors and LC³-ECC compressive strength makes the prediction of LC³-ECC compressive strength difficult. Regarding this, the machine learning-based prediction models for the compressive strength of LC³-ECC concrete is firstly proposed and developed. Models combine three novel meta-heuristic algorithms (golden jackal optimization algorithm, butterfly optimization algorithm and whale optimization algorithm) with support vector regression (SVR) to improve the accuracy of prediction. A new dataset about LC³-ECC compressive strength was integrated based on 156 data from previous studies and used to develop the SVR-based models. Thirteen potential factors affecting the compressive strength of LC³-ECC were comprehensively considered in the model. The results show all hybrid SVR prediction models can reach the Coefficient of determination (R²) above 0.95 for the testing set and 0.97 for the training set. Radar and Taylor plots also show better overall prediction performance of the hybrid SVR models than several traditional machine learning techniques, which confirms the superiority of the three proposed methods. The successful development of this predictive model can provide scientific guidance for LC³-ECC materials and further apply to such low-carbon, sustainable cement-based materials.

• Journal of Korean Physical Therapy Science
• /
• v.30 no.3
• /
• pp.49-58
• /
• 2023

Background: This study aims to examine the useful- ness of the portable spirometer "The Spirokit" as a clinical diagnostic device through technology introduction, precision test, and correction. Design: Technical note Methods: "The Spirokit" was developed using a propeller-type flow rate and flow rate measurement method using infrared and light detection sensors. The level of agreement between the Pulmonary Waveform Generator and the measured values was checked to determine the precision of "The Spirokit", and the correction equation was included using the Pulmonary Waveform Generator software to correct the error range. The analysis was requested using the ATS 24/26 waveform recognized by the Ministry of Food and Drug Safety and the American Thoracic Society for the values of Forced Voluntary Capacity (FVC), Forced Expiratory Volume in 1second (FEV1), and Peak Expiratory Flow (PEF), which are used as major indicators for pulmonary function tests. All tests were repeated five times to derive an average value, and FVC and FEV1 presented accuracy and PEF presented accuracy as the result values. Results: FVC and FEV1 of 'The Spirokit' developed in this study showed accuracy within ± 3% of the error level in the ATS 24 waveform. The PEF value of 'The Spirokit' showed accuracy within the error level ± 12% of the ATS 26 waveform. Conclusion: Through the results of this study, the precision of 'The Spirokit' as a clinical diagnosis device was identified, and it was confirmed that it can be used as a portable pulmonary function test that can replace a spirometer.

• Korean Medical Education Review
• /
• v.25 no.3
• /
• pp.212-228
• /
• 2023

Systems thinking, a linking domain of health systems science (HSS), is an approach that investigates specific problems from a holistic perspective. It supports improving patients' health, fulfilling their health needs, and anticipating issues that threaten patient safety within the healthcare system. It also helps solve problems through critical thinking and ref lection. This study aimed to develop an curriculum on systems thinking, explore the effectiveness of the course, and investigate the applicability of HSS education at individual universities. In this study, the ADDIE (Analysis, Design, Development, Implementation, and Evaluation) model was utilized to design, develop, implement, and evaluate an elective course on systems thinking. In the design process, learning outcomes and goals were developed, and educational content, teaching-learning methods, and student evaluation methods were linked. In the development process, class materials and evaluation materials were prepared. In the implementation process, the course was implemented, and the evaluation process analyzed the results of learning performance and curriculum assessments. The evaluation found the following results. First, the students in the study realized the importance of systems thinking and experienced the need for systems thinking through non-medical and medical situations. Second, the students were very satisfied with the learning activities in the course (mean=4.84), and the results of the self-competence evaluation, conducted before and after the course, also showed a significant improvement. This study confirmed the effectiveness of the elective course, and its results can serve as a reference for developing an HSS curriculum.

• Journal of Digital Policy
• /
• v.2 no.4
• /
• pp.33-39
• /
• 2023

With a continuous occurrence of right-turn traffic accidents at intersections, there is an increasing demand for measures to address these incidents. In response, a technology has been developed to detect the presence of pedestrians through object detection in CCTV footage at right-turn areas and display warning messages on the screen to alert drivers. The YOLO (You Only Look Once) model, a type of object detection model, was employed to assess the performance of object detection. An algorithm was also devised to address misidentification issues and generate warning messages when pedestrians are detected. The accuracy of recognizing pedestrians or objects and outputting warning messages was measured at approximately 82%, suggesting a potential contribution to preventing right-turn accidents