• Smart Structures and Systems
• /
• v.30 no.6
• /
• pp.673-686
• /
• 2022

Deep learning algorithms for Structural Health Monitoring (SHM) have been extracting the interest of researchers and engineers. These algorithms commonly used loss functions and evaluation indices like the mean square error (MSE) which were not originally designed for SHM problems. An updated loss function which was specifically constructed for deep-learning-based structural damage detection problems has been proposed in this study. By tuning the coefficients of the loss function, the weights for damage localization and quantification can be adapted to the real situation and the deep learning network can avoid unnecessary iterations on damage localization and focus on the damage severity identification. To prove efficiency of the proposed method, structural damage detection using convolutional neural networks (CNNs) was conducted on a truss bridge model. Results showed that the validation curve with the updated loss function converged faster than the traditional MSE. Data augmentation was conducted to improve the anti-noise ability of the proposed method. For reducing the training time, the normalized modal strain energy change (NMSEC) was extracted, and the principal component analysis (PCA) was adopted for dimension reduction. The results showed that the training time was reduced by 90% and the damage identification accuracy could also have a slight increase. Furthermore, the effect of different modes and elements on the training dataset was also analyzed. The proposed method could greatly improve the performance for structural damage detection on both the training time and detection accuracy.

• Journal of the Korean Society of Systems Engineering
• /
• v.18 no.2
• /
• pp.94-107
• /
• 2022

Accidents prevention and mitigation is the highest priority of nuclear power plant (NPP) operation, particularly in the aftermath of the Fukushima Daiichi accident, which has reignited public anxieties and skepticism regarding nuclear energy usage. To deal with accident scenarios more effectively, operators must have ample and precise information about key safety parameters as well as their future trajectories. This work investigates the potential of machine learning in forecasting NPP response in real-time to provide an additional validation method and help reduce human error, especially in accident situations where operators are under a lot of stress. First, a base-case SGTR simulation is carried out by the best-estimate code RELAP5/MOD3.4 to confirm the validity of the model against results reported in the APR1400 Design Control Document (DCD). Then, uncertainty quantification is performed by coupling RELAP5/MOD3.4 and the statistical tool DAKOTA to generate a large enough dataset for the construction and training of neural-based machine learning (ML) models, namely LSTM, GRU, and hybrid CNN-LSTM. Finally, the accuracy and reliability of these models in forecasting system response are tested by their performance on fresh data. To facilitate and oversee the process of developing the ML models, a Systems Engineering (SE) methodology is used to ensure that the work is consistently in line with the originating mission statement and that the findings obtained at each subsequent phase are valid.

• Advances in nano research
• /
• v.12 no.2
• /
• pp.185-195
• /
• 2022

Deep learning is another field of artificial intelligence (AI) utilized for computer aided diagnosis (CAD) and image processing in scientific research. Considering numerous mechanical repetitive tasks, reading image slices need time and improper with geographical limits, so the counting of image information is hard due to its strong subjectivity that raise the error ratio in misdiagnosis. Regarding the highest mortality rate of Lung cancer, there is a need for biopsy for determining its class for additional treatment. Deep learning has recently given strong tools in diagnose of lung cancer and making therapeutic regimen. However, identifying the pathological lung cancer's class by CT images in beginning phase because of the absence of powerful AI models and public training data set is difficult. Convolutional Neural Network (CNN) was proposed with its essential function in recognizing the pathological CT images. 472 patients subjected to staging FDG-PET/CT were selected in 2 months prior to surgery or biopsy. CNN was developed and showed the accuracy of 87%, 69%, and 69% in training, validation, and test sets, respectively, for T1-T2 and T3-T4 lung cancer classification. Subsequently, CNN (or deep learning) could improve the CT images' data set, indicating that the application of classifiers is adequate to accomplish better exactness in distinguishing pathological CT images that performs better than few deep learning models, such as ResNet-34, Alex Net, and Dense Net with or without Soft max weights.

• Steel and Composite Structures
• /
• v.41 no.5
• /
• pp.681-695
• /
• 2021

Many existing transmission or communication towers designed several decades ago have undergone nonreversible performance degradation, making it hardly meet the additional requirements from upgrades in wind load design codes and extra services of electricity and communication. Therefore, a new-type non-destructive reinforcement method was proposed to reduce the on-site operation of drilling and welding for improving the quality and efficiency of reinforcement. Six built-up steel angle members were tested under compression to examine the reinforcement performance. Subsequently, the cyclic loading test was conducted on a pair of steel angle tower sub-structures to investigate the reinforcement effect, and a simplified prediction method was finally established for calculating the buckling bearing capacity of those new-type retrofitted built-up steel angles. The results indicates that: no apparent difference exists in the initial stiffness for the built-up specimens compared to the unreinforced steel angles; retrofitting the steel angles by single-bolt clamps can guarantee a relatively reasonable reinforcement effect and is suggested for the reduced additional weight and higher construction efficiency; for the substructure test, the latticed substructure retrofitted by the proposed reinforcement method significantly improves the lateral stiffness, the non-deformability and energy dissipation capacity; moreover, an apparent pinching behavior exists in the hysteretic loops, and there is no obvious yield plateau in the skeleton curves; finally, the accuracy validation result indicates that the proposed theoretical model achieves a reasonable agreement with the test results. Accordingly, this study can provide valuable references for the design and application of the non-destructive upgrading project of steel angle towers.

• Journal of Digital Convergence
• /
• v.19 no.11
• /
• pp.415-424
• /
• 2021

The purpose of the current study is to to examine the psychometric properties of the Subjective Agingwell Scale (SAS). Three hundred and forty two elders completed the SAS and the scales assessing subjective well-being, optimism, perceived control and adhering to healthy behavior. The 11-item SAS that displays good internal reliability and good fit to the three-factor model consisting of cognitive satisfaction, positive affect, and spiritual fullness. Correlational analyses with measures of subjective agingwell, subjective well-being, and optimism provide evidence for construct validity. Moreover, the results from hierarchical regression analyses show criterion-related validity of the SAS. This scale could be used in the field to measure and promote subjective agingwell.

• Journal of Engineering Education Research
• /
• v.25 no.4
• /
• pp.21-34
• /
• 2022

This study is conducted with the aim of identify the factors constituting learning competencies for engineering college students, and developing and validating the scale to measure them. To this end, literature and prior research were reviewed and focus group interview was conducted with high-achieving learners of K University in the capital region of Korea. According to previous research, 3 learning competency groups, 12 learning competencies and 41 sub-competencies were derived. Delphi survey was carried out twice, 28 sub-competencies were derived among the 41 sub-competencies through this process. 166 initial items were developed through literature review and FGI. Then, 130 items were confirmed by verifying content validity in the second Delphi survey. Based on this, pilot test were performed with 110 students in K university, and an interview was conducted with 50 students who participated in the pilot test. Reflecting the pilot test results, 1 sub-competency and 22 items were deleted. After the confirmed pilot test results, the main test were performed with all current students in K University. According to the main test result, the validity of the scale and the model fit was verified for the response data of 823 students, and the scale consisting of a total of 105 items was confirmed. The final learning competencies scale included three competency groups and 10 learning competencies. The scale developed in this study can be used as an independent scale for each competency group as needed. It is expected that this scale can be contributed to support the development their learning competencies for academic success of engineering college students, who are future learners.

• Journal of Hospice and Palliative Care
• /
• v.25 no.3
• /
• pp.110-120
• /
• 2022

Purpose: The purpose of this study was to confirm the factor structure of the McGill Quality of Life Questionnaire-Revised (MQOL-R) in the context of Korean culture and to verify its reliability and validity. Methods: The participants comprised terminal cancer patients aged 25 or older, and data from 164 participants were analyzed. The study was conducted in the following order: translation, expert review, reverse translation, preliminary investigation and interviews, and completion of the final version. Confirmatory factor analysis was applied to evaluate the validity of the instrument, and the Beck Depression Inventory, Korean version (K-BDI) was applied to confirm the criterion validity of the MQOL-R Korean version. The Cronbach's alpha coefficient, representing internal consistency, was measured to evaluate reliability. Results: Cronbach's alpha for all 14 questions was 0.862. The model fit indices for confirmatory factor analysis were within the acceptance criteria. The factor loadings of all four factors were over 0.50, and convergent validity and discriminant validity were confirmed. Regarding criterion validity, a negative correlation was found between the four factors of MQOL-R Korean version and the K-BDI. Conclusion: The MQOL-R Korean version, the reliability and validity of which were verified in this study, is a 15-item tool consisting of 14 items dealing with four physical, psychological, existential, and social factors and a single item evaluating the overall quality of life. The MQOL-R Korean version is an instrument that can more concisely and effectively measure the quality of life of patients with life-threatening diseases.

• BMB Reports
• /
• v.55 no.6
• /
• pp.281-286
• /
• 2022

Hepatocellular carcinoma is a major health burden, and though various treatments through much research are available, difficulties in early diagnosis and drug resistance to chemotherapy-based treatments render several ineffective. Cancer stem cell model has been used to explain formation of heterogeneous cell population within tumor mass, which is one of the underlying causes of high recurrence rate and acquired chemoresistance, highlighting the importance of CSC identification and understanding the molecular mechanisms of CSC drivers. Extracellular CSC-markers such as CD133, CD90 and EpCAM have been used successfully in CSC isolation, but studies have indicated that increasingly complex combinations are required for accurate identification. Pseudogene-derived long non-coding RNAs are useful candidates as intracellular CSC markers - factors that regulate pluripotency and self-renewal - given their cancer-specific expression and versatile regulation across several levels. Here, we present the use of microarray data to identify stemness-associated factors in liver cancer, and selection of sole pseudogene-derived lncRNA ZNF204P for experimental validation. ZNF204P knockdown impairs cell proliferation and migration/invasion. As the cytosolic ZNF204P shares miRNA binding sites with OCT4 and SOX2, well-known drivers of pluripotency and self-renewal, we propose that ZNF204P promotes tumorigenesis through the miRNA-145-5p/OCT4, SOX2 axis.

• Imaging Science in Dentistry
• /
• v.52 no.4
• /
• pp.399-408
• /
• 2022

Purpose: The purpose of this study was to compare volume measurements obtained using 2 image software packages on Digital Imaging and Communications in Medicine (DICOM) images acquired from 1 multidetector computed tomography and 5 cone-beam computed tomography devices, using different protocols for physical volume measurements. Materials and Methods: Four pieces of bovine leg were prepared. Marrow was removed from 3 pieces, leaving cortical bone exposed. The resulting space of 1 piece was filled with water, another was filled with propylene glycol, and the third was left unfilled. The marrow in the fourth sample was left fully intact. Volume measurements were obtained after importing DICOM images into the Dolphin Imaging 11.95 and ITK-SNAP software programs. Data were analyzed using 3-way analysis of variance with a generalized linear model to determine the effects of voxel size, software, and content on percentage mean volume differences between tomographic protocols. A significance level of 0.05 was used. Results: The intraclass correlation coefficients for intraobserver and interobserver reliability were, respectively, 0.915 and 0.764 for the Dolphin software and 0.894 and 0.766 for the ITK-SNAP software. Three sources of statistically significant variation were identified: the interaction between software and content (P=0.001), the main effect of content (P=0.014), and the main effect of software (P=0.001). Voxel size was not associated with statistically significant differences in volume measurements. Conclusion: Both content and software influenced the accuracy of volume measurements, especially when the content had gray values similar to those of the adjacent tissues.

• Wind and Structures
• /
• v.36 no.4
• /
• pp.221-236
• /
• 2023

The lateral component of turbulence and the vortices shed in the wake of a structure result in introducing dynamic wind load in the acrosswind direction and the resulting level of motion is typically larger than the corresponding alongwind motion for a dynamically sensitive structure. The underlying source mechanisms of the acrosswind load may be classified into motion-induced, buffeting, and Strouhal components. This study proposes a frequency domain framework to decompose the overall load into these components based on output-only measurements from wind tunnel experiments or full-scale measurements. First, the total acrosswind load is identified based on measured acceleration response by solving the inverse problem using the Kalman filter technique. The decomposition of the combined load is then performed by modeling each load component in terms of a Bayesian filtering scheme. More specifically, the decomposition and the estimation of the model parameters are accomplished using the unscented Kalman filter in the frequency domain. An aeroelastic wind tunnel experiment involving a tall circular cylinder was carried out for the validation of the proposed framework. The contribution of each load component to the acrosswind response is assessed by re-analyzing the system with the decomposed components. Through comparison of the measured and the re-analyzed response, it is demonstrated that the proposed framework effectively decomposes the total acrosswind load into components and sheds light on the overall underlying mechanism of the acrosswind load and attendant structural response. The delineation of these load components and their subsequent modeling and control may become increasingly important as tall slender buildings of the prismatic cross-section that are highly sensitive to the acrosswind load effects are increasingly being built in major metropolises.