• Journal of the Korean Society of Industry Convergence
• /
• v.27 no.4_2
• /
• pp.1001-1007
• /
• 2024

Recent deep learning algorithms for object detection in real-time play a crucial role in various applications such as autonomous driving, traffic monitoring, health care, and water quality monitoring. The size of small objects, in particular, significantly impacts the accuracy of detection models. However, data containing small objects can lead to underfitting issues in models. Therefore, this study developed a deep learning model capable of quickly detecting small objects to provide more accurate predictions. The RE-SOD (Residual block based Small Object Detector) developed in this research enhances the detection performance for small objects by using RGB separation preprocessing and residual blocks. The model achieved an accuracy of 1.0 in image classification and an mAP50-95 score of 0.944 in object detection. The performance of this model was validated by comparing it with real-time detection models such as YOLOv5, YOLOv7, and YOLOv8.

• Safety and Health at Work
• /
• v.10 no.4
• /
• pp.461-469
• /
• 2019

Background: Collisions involving workers and mobile machines continue to be a major concern in underground coal mines. Over the last 30 years, these collisions have resulted in numerous injuries and fatalities. Recently, the Mine Safety and Health Administration (MSHA) proposed a rule that would require mines to equip mobile machines with proximity detection systems (PDSs) (systems designed for automated collision avoidance). Even though this regulation has not been enacted, some mines have installed PDSs on their scoops and hauling machines. However, early implementation of PDSs has introduced a variety of safety concerns. Past findings show that workers' trust can affect technology integration and influence unsafe use of automated technologies. Methods: Using a mixed-methods approach, the present study explores the effect that factors such as mine of employment, age, experience, and system type have on workers' trust in PDSs for mobile machines. The study also explores how workers are trained on PDSs and how this training influences trust. Results: The study resulted in three major findings. First, the mine of employment had a significant influence on workers' trust in mobile PDSs. Second, hands-on and classroom training was the most common types of training. Finally, over 70% of workers are trained on the system by the mine compared with 36% trained by the system manufacturer. Conclusion: The influence of workers' mine of employment on trust in PDSs may indicate that practitioners and researchers may need to give the organizational and physical characteristics of each mine careful consideration to ensure safe integration of automated systems.

• Korean Journal of Veterinary Service
• /
• v.36 no.2
• /
• pp.79-86
• /
• 2013

Mycobacterium (M.) bovis, a member of the M. tuberculosis complex (MTC), is a re-emerging, zoonotic agent of bovine tuberculosis whose prevalence probably depends on variations in direct exposure to cattle and ingestion of raw milk. Accurate species differentiation of M. bovis and M. tuberculosis is needed to distinguish between human and zoonotic tuberculosis. This study successfully developed a loop-mediated isothermal amplification (LAMP) assay for rapid detection and differentiation of M. bovis and M. tuberculosis, however showed negative reactions in eight non-tuberculous mycobacteria (NTM) samples and ten other bacterial species. Sensitivity of this assay for detection of genomic M. bovis DNA was 10 $fg/{\mu}l$. And this assay successfully detected M. bovis in bovine clinical specimens. In conclusion, the LAMP assay is a simple and powerful tool for rapid detection of M. bovis in both pure bacterial culture and in clinical samples.

• Journal of Dairy Science and Biotechnology
• /
• v.39 no.1
• /
• pp.9-19
• /
• 2021

Enterobacter sakazakii (Cronobacter spp.) causes meningitis, necrotizing enterocolitis, sepsis, and bacteremia in neonates and children and has a high mortality rate. For rapid E. sakazakii detection, various differential and selective media containing α-glucosidase substrates, such as 5-bromo-4-chloro-3-indolyl-α-D-glucopyranoside (BCIG) or 4-methylumbelliferyl-α-D-glucoside (α-MUG), have been developed as only E. sakazakii exhibits α-glucosidase activity in the genus Enterobacter. However, Escherichia vulneris (family: Enterobacteriaceae) can also utilize α-glucosidase substrates, thereby resulting in false positives. Various iron sources are known to promote the growth of gram-negative bacteria. This study aimed to develop a selective medium containing α-glucosidase substrates for E. sakazakii detection that would eliminate false positives, such as those of E. vulneris, and to determine the role of iron source in the medium. Three previously developed (TPD) media, i.e., Oxoid, OK, and VRBG, and the medium developed in this study, i.e., NGTE, were evaluated using 58 E. sakazakii and 5 non-E. sakazakii strains. Fifty-four E. sakazakii strains appeared as fluorescent or chromogenic colonies on all four media that were assessed. Two strains showed colonies on NGTE medium and not on TPD media. In contrast, the remaining two strains showed colonies on TPD media and not on NGTE medium. None of the non-E. sakazakii strains showed fluorescent or chromogenic colonies on any of the evaluated media except E. vulneris, which showed colonies on TPD media and not on NGTE medium. This study demonstrated that the newly developed NGTE medium was not only equally efficient in promoting the growth of bacterial colonies when compared with the currently available media but also eliminated false positives, such as E. vulneris.

• Asian Pacific Journal of Cancer Prevention
• /
• v.14 no.11
• /
• pp.6241-6243
• /
• 2013

Objective: To explore the predictive value of tumor markers, including cancer antigen 72-4 (CA72-4), cancer antigen 15-3 (CA15-3) and cancer antigen 125 (CA125), in single or combined detection, for the diagnosis of ovarian cancer. Methods: 120 patients diagnosed with ovarian cancer from August 2011 to March 2013 and 80 patients diagnosed with benign ovarian tumors were enrolled in this test, along with 50 health examination women randomly selected from the database as controls. Serum levels of CA72-4, CA15-3 and CA125 in this study were determined by electrochemiluminescence (ECL). Results: Serum levels of CA72-4, CA15-3 and CA125 in ovarian cancer were higher than those in healthy group and benign group (P<0.01).The sensitivity of combined detection of those three tumor markers for diagnosis of ovarian cancer was obviously higher than with single detection with each marker (P<0.01). Conclusions: CA72-4, CA15-3 and CA125 could be a good combination in the diagnosis of ovarian cancer. Patients whose tumor markers continue to increase should be highly suspected of malignancy.

• Smart Structures and Systems
• /
• v.5 no.4
• /
• pp.483-494
• /
• 2009

Damage detection has been proven to be a challenging task in structural health monitoring (SHM) due to the fact that damage cannot be measured. The difficulty associated with damage detection is related to electing a feature that is sensitive to damage occurrence and evolution. This difficulty increases as the damage size decreases limiting the ability to detect damage occurrence at the micron and submicron length scale. Damage detection at this length scale is of interest for sensitive structures such as aircrafts and nuclear facilities. In this paper a new photonic sensor based on photonic crystal (PhC) technology that can be synthesized at the nanoscale is introduced. PhCs are synthetic materials that are capable of controlling light propagation by creating a photonic bandgap where light is forbidden to propagate. The interesting feature of PhC is that its photonic signature is strongly tied to its microstructure periodicity. This study demonstrates that when a PhC sensor adhered to polymer substrate experiences micron or submicron damage, it will experience changes in its microstructural periodicity thereby creating a photonic signature that can be related to damage severity. This concept is validated here using a three-dimensional integrated numerical simulation.

• Clinical Endoscopy
• /
• v.56 no.4
• /
• pp.460-469
• /
• 2023

Background/Aims: Diagnosis of Helicobacter pylori-naive gastric cancer (HPNGC) is becoming increasingly important. This study aimed to explore the quality indicators for HPNGC detection. Methods: We conducted a cross-sectional, nationwide, web-based survey of gastrointestinal endoscopists in Japan. In addition to questions about the number of HPNGC cases detected in a year and basic information, the questionnaire also consisted of 28 questions: (1) 18 about HPNGC awareness, (2) six about diagnostic proactiveness, and (3) four about interest in HPNGC. Results: Valid responses were obtained from 712 endoscopists. The Japan Gastroenterological Endoscopy Society-certified endoscopists had a significantly higher HPNGC detection rate than the nonspecialists (0.42% vs. 0.32%, respectively; p=0.008). The results of the multiple regression analysis showed that Japan Gastroenterological Endoscopy Society certification and high awareness and interest scores were independent predictors of the HPNGC detection rate (p=0.012, p<0.001, p=0.024, respectively). Principal component analysis showed that the endoscopists who attended conferences for collecting information on HPNGC had a higher level of awareness. Conclusions: To improve the detection of HPNGC, it is necessary to increase the awareness of the disease. It is hoped that relevant societies will play an important role in endoscopists' education.

• Journal of Wind Energy
• /
• v.9 no.4
• /
• pp.16-23
• /
• 2018

A damage detection method for the tripod support structure of offshore wind turbines is presented for structural health monitoring. A finite element model of a prototype tripod support structure is established and the modal properties are calculated. The degree and location of the damage are estimated based on the neural network technique using the changes of natural frequencies and mode shape due to the damage. The stress distribution occurring in the support structure is obtained by a dynamic analysis for the wind turbine system to select the output data of the neural network. The natural frequencies and mode shapes for 36 possible damage scenarios were used for the input data of the learned neural network for damage assessment. The estimated damages agreed reasonably well with the accurate ones. The presented method could be effectively applied for damage detection and structural health monitoring of various types of support structures of offshore wind turbines.

• Journal of Ginseng Research
• /
• v.32 no.4
• /
• pp.390-396
• /
• 2008

For evaluating the quality of ginseng, simple and fast analysis methods are needed to determine the ginsenoside content of the ginseng products. The aim of this study was therefore to optimize conditions for fast analysis of the ginsenosides, the active ingredients in extracts of Korean red ginseng. When tandem HPLC mass spectrometry (HPLC-MS/MS) was used, four forms of ginsenoside, Rb1, Rb2, Rc, and Re, were readily separated in seven minutes using a gradient mobile phase (acetonitrile and water containing acetic acid). This is the shortest separation time reported among the studies of major ginsenoside analysis. When gradient HPLC with UV detection was used, the detection limit was high, but separation of these four ginsenosides required 25 minutes using acetonitrile and water containing formic acid as a mobile phase. HPLC-MS/MS was able to separate ginsenoside Rg1 easily regardless of the mobile phase condition, but the HPLC-UV could not separate Rg1 because acetonitrile concentration in the mobile phase had to be maintained below 20%. Ginsenoside peaks were clearer and had more sensitive detection limits when Korean red ginseng extract was analyzed by the HPLC-MS/MS, but the UV detection was useful for chromatographic fingerprinting of all four major ginsenosides of the extract: Rb1, Rb2, Rc, and Re. Extracts were found to contain 2.17 mg, 1.51 mg, 1.29 mg, and 0.46 mg of ginsenoside Rb1, Rb2, Rc, Re, respectively, per gram weight. The ratios of each ginsenoside in the extracts were 1.0 : 0.7 : 0.6 : 0.2, respectively. Taken together, the results indicate that HPLC-MS/MS spectrometry could be the most useful method for rapid analysis of even small amounts of major ginsenosides, while HPLC with UV detection could also be used for rapid analysis of major ginsenosides and for quality control of ginseng products.

• Smart Structures and Systems
• /
• v.10 no.4_5
• /
• pp.443-458
• /
• 2012

The structural health monitoring (SHM) benchmark study on optimal sensor placement problem for the instrumented Canton Tower has been launched. It follows the success of the modal identification and model updating for the Canton Tower in the previous benchmark study, and focuses on the optimal placement of vibration sensors (accelerometers) in the interest of bettering the SHM system. In this paper, the sensor placement problem for the Canton Tower and the benchmark model for this study are first detailed. Then an information entropy based sensor placement method with the purpose of damage detection is proposed and applied to the benchmark problem. The procedure that will be implemented for structural damage detection using the data obtained from the optimal sensor placement strategy is introduced and the information on structural damage is specified. The information entropy based method is applied to measure the uncertainties throughout the damage detection process with the use of the obtained data. Accordingly, a multi-objective optimal problem in terms of sensor placement is formulated. The optimal solution is determined as the one that provides equally most informative data for all objectives, and thus the data obtained is most informative for structural damage detection. To validate the effectiveness of the optimally determined sensor placement, damage detection is performed on different damage scenarios of the benchmark model using the noise-free and noise-corrupted measured information, respectively. The results show that in comparison with the existing in-service sensor deployment on the structure, the optimally determined one is capable of further enhancing the capability of damage detection.