• Journal of radiological science and technology
• /
• v.44 no.6
• /
• pp.655-661
• /
• 2021

Positron emission tomography (PET) images is affected by acquisition time, short acquisition times results in low gamma counts leading to degradation of image quality by statistical noise. Noise2Void(N2V) is self supervised denoising model that is convolutional neural network (CNN) based deep learning. The purpose of this study is to evaluate denoising performance of N2V for PET image with a short acquisition time. The phantom was scanned as a list mode for 10 min using Biograph mCT40 of PET/CT (Siemens Healthcare, Erlangen, Germany). We compared PET images using NEMA image-quality phantom for standard acquisition time (10 min), short acquisition time (2min) and simulated PET image (S2 min). To evaluate performance of N2V, the peak signal to noise ratio (PSNR), normalized root mean square error (NRMSE), structural similarity index (SSIM) and radio-activity recovery coefficient (RC) were used. The PSNR, NRMSE and SSIM for 2 min and S2 min PET images compared to 10min PET image were 30.983, 33.936, 9.954, 7.609 and 0.916, 0.934 respectively. The RC for spheres with S2 min PET image also met European Association of Nuclear Medicine Research Ltd. (EARL) FDG PET accreditation program. We confirmed generated S2 min PET image from N2V deep learning showed improvement results compared to 2 min PET image and The PET images on visual analysis were also comparable between 10 min and S2 min PET images. In conclusion, noisy PET image by means of short acquisition time using N2V denoising network model can be improved image quality without underestimation of radioactivity.

• Nutrition Research and Practice
• /
• v.16 no.2
• /
• pp.272-283
• /
• 2022

BACKGROUND/OBJECTIVES: Most child and adolescent food literacy measurement tools focus on nutrition and food safety. However, the importance of aspects related to the food system such as food distribution and food waste and their effects on environmental sustainability is growing. We therefore developed and validated a two-dimensional tool for children (8-12 years old) and adolescents (13-18 years old) that can comprehensively measure food literacy. The association of food literacy with diet quality and self-reported health was assessed. SUBJECTS/METHODS: First, we developed a food literacy conceptual framework that contains food system and literacy dimensions through a literature review, focus group interviews, and expert review. After a face validity study, we conducted the main survey (n = 200) to validate the questionnaire. Construct validity and reliability were assessed using exploratory factor analysis (EFA), confirmatory factor analysis (CFA), and Cronbach's alpha. RESULTS: As a result of the Delphi study, content validity was confirmed for the remaining 30 items after two items were excluded (content validity ratio = 0.86). Eleven items were excluded from the EFA results, while the CFA results indicated appropriate fit indices for the proposed model (comparative fit index = 0.904, root mean square error of approximation = 0.068). The final food literacy questionnaire consisted of 19 questions and comprised 5 factors: production, distribution, selection, preparation and cooking, and intake. Food literacy was positively associated with diet quality, as assessed by the Nutrition Quotient score, in both children and adolescents and with self-reported health in adolescents.

• Journal of Environmental Science International
• /
• v.30 no.12
• /
• pp.1053-1065
• /
• 2021

In this study, the prediction technology of Hydrological Quantitative Precipitation Forecast (HQPF) was improved by optimizing the weather predictors used as input data for machine learning. Results comparison was conducted using bias and Root Mean Square Error (RMSE), which are predictive accuracy verification indicators, based on the heavy rain case on August 21, 2021. By comparing the rainfall simulated using the improved HQPF and the observed accumulated rainfall, it was revealed that all HQPFs (conventional HQPF and improved HQPF 1 and HQPF 2) showed a decrease in rainfall as the lead time increased for the entire grid region. Hence, the difference from the observed rainfall increased. In the accumulated rainfall evaluation due to the reduction of input factors, compared to the existing HQPF, improved HQPF 1 and 2 predicted a larger accumulated rainfall. Furthermore, HQPF 2 used the lowest number of input factors and simulated more accumulated rainfall than that projected by conventional HQPF and HQPF 1. By improving the performance of conventional machine learning despite using lesser variables, the preprocessing period and model execution time can be reduced, thereby contributing to model optimization. As an additional advanced method of HQPF 1 and 2 mentioned above, a simulated analysis of the Local ENsemble prediction System (LENS) ensemble member and low pressure, one of the observed meteorological factors, was analyzed. Based on the results of this study, if we select for the positively performing ensemble members based on the heavy rain characteristics of Korea or apply additional weights differently for each ensemble member, the prediction accuracy is expected to increase.

• Journal of radiological science and technology
• /
• v.45 no.2
• /
• pp.103-109
• /
• 2022

To maintain improved image quality in mammography, the quality control process is performed using the ACR (American college of radiology) phantom. In addition, many studied were performed by fabricating the customized breast phantom to provide more information in mammography. Thus, the purpose of this study was to evaluate the image quality by designing the modified ACR phantoms. The five modified acrlylic ACR phantoms were designed by considering insert position and phantom thickness. The phantoms were consisted of 4.5, 3.0, and 1.5 cm in terms of phantom thickness, and 3.0, 2.0, and 0.5 cm in terms of insert position, respectively. The acquired images were evaluated by PSNR (peak signal to noise ratio), RMSE (root mean square error), CC (correlation coefficient), CNR (contrast to noise ratio), and COV (coefficient of variation). Based on the similarity analysis, the result is suitable between conventional and new designed phantoms. In addition, the CNR and COV results in terms of insert position showed that image quality for 0.5 cm was 2.3 and 27.4% improved compared with 2 and 3 cm, respectively. According to phantom thickness results, the CNR result for 1.5 cm and COV result for 4.5 cm were 50.1 and 62.7% improved compared with that those conditions. In conclusion, we confirmed that the image quality depends on the breast size and thickness through modified ACR phantom study.

• Journal of ILASS-Korea
• /
• v.27 no.2
• /
• pp.57-65
• /
• 2022

The purpose of this study is to use machine learning to build a model capable of predicting the flash boiling spray characteristics. In this study, the flash boiling spray was visualized using Shadowgraph visualization technology, and then the spray image was processed with MATLAB to obtain quantitative data of spray characteristics. The experimental conditions were used as input, and the spray characteristics were used as output to train the machine learning model. For the machine learning model, the XGB (extreme gradient boosting) algorithm was used. Finally, the performance of machine learning model was evaluated using R² and RMSE (root mean square error). In order to have enough data to train the machine learning model, this study used 12 injectors with different design parameters, and set various fuel temperatures and ambient pressures, resulting in about 12,000 data. By comparing the performance of the model with different amounts of training data, it was found that the number of training data must reach at least 7,000 before the model can show optimal performance. The model showed different prediction performances for different spray characteristics. Compared with the upstream spray angle and the downstream spray angle, the model had the best prediction performance for the spray tip penetration. In addition, the prediction performance of the model showed a relatively poor trend in the initial stage of injection and the final stage of injection. The model performance is expired to be further enhanced by optimizing the hyper-parameters input into the model.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.24 no.2
• /
• pp.124-132
• /
• 2022

This study was aimed to visualize the spatial distribution of PM₁₀ data measured at non-uniformly distributed observation sites in South Korea. Different spatial interpolation methods were applied to irregularly distributed PM₁₀ observation data from January, 2019, when the concentration was the highest and in July, 2019, when the concentration was the lowest. Four interpolation methods with different parameters were used: Inverse Distance Weighted (IDW), Ordinary Kriging (OK), radial base function, and scattered interpolation. Six cases were cross-validated and the normalized root-mean-square error for each case was compared. The results showed that IDW using smoothing-related factors was the most appropriate method, while the OK method was least appropriate. Our results are expected to help users select the proper spatial interpolation method for PM₁₀ data analysis with comparative reliability and effectiveness.

• Journal of Environmental Science International
• /
• v.31 no.1
• /
• pp.61-76
• /
• 2022

In this study, surface particulate matter (PM_2.5) concentrations were calculated based on empirical equations using measurements of ceilometer backscatter intensities and meteorological variables taken over 19 months. To quantify the importance of meteorological conditions on the calculations of surface PM_2.5 concentrations, eight different meteorological conditions were considered. For each meteorological condition, the optimal upper limit height for an integration of ceilometer backscatter intensity and coefficients for the empirical equations were determined using cross-validation processes with and without considering meteorological variables. The results showed that the optimal upper limit heights and coefficients depended heavily on the meteorological conditions, which, in turn, exhibited extensive impacts on the estimated surface PM_2.5 concentrations. A comparison with the measurements of surface PM_2.5 concentrations showed that the calculated surface PM_2.5 concentrations exhibited better results (i.e., higher correlation coefficient and lower root mean square error) when considering meteorological variables for all eight meteorological conditions. Furthermore, applying optimal upper limit heights for different weather conditions revealed better results compared with a constant upper limit height (e.g., 150 m) that was used in previous studies. The impacts of vertical distributions of ceilometer backscatter intensities on the calculations of surface PM_2.5 concentrations were also examined.

• Nutrition Research and Practice
• /
• v.16 no.6
• /
• pp.765-774
• /
• 2022

BACKGROUND/OBJECTİVES: It is important to determine Dysfunctional eating behaviors such as dietary restraint and overeating tendencies in order to provide weight management and acquire the right habits in children. The purpose of this study was to test the reliability and validity of Dutch Eating Behaviour Questionnaire Children (DEBQ-C) with Turkish preadolescent children. MATERIALS/METHODS: This research included 440 preadolescents (9.3 ± 6.9 years and 235 girls, 205 boys). The instrument is divided into three subscales, each with 20 items. Emotional eating, restrained eating, and external eating are the three subscales. Confirmatory factor analysis (CFA) was used to assess the construct validity of the Turkish version of the DEBQ-C, and Cronbach α values were computed to evaluate the subscale reliabilities. There were 20 observable variables and three latent variables in the hypothesized model. RESULTS: Fit indices for the hypothesized model were good (×2/degree of freedom = 1.96; root mean square error of approximation = 0.05; comparative fit index = 0.95; goodness of fit index = 0.93). These findings revealed that the Turkish version of the DEBQ-C has a factor structure that was identical to the three-factor structure of the original scale. The Turkish version of the DEBQ-C subscales has internal consistency coefficients ranging from 0.72 (external eating) to 0.86. (emotional eating). CONCLUSIONS: The DEBQ-C Turkish version is a viable and reliable tool for measuring overeating tendencies in Turkish preadolescents, according to the findings.

• Korean Journal of Agricultural Science
• /
• v.49 no.3
• /
• pp.483-493
• /
• 2022

It is important to improve the efficiency of plant breeding and crop yield to fulfill increasing food demands. In plant phenotyping studies, the capability to correlate morphological traits such as plant height, stem diameter, leaf length, leaf width, leaf angle and size of panicle of the plants has an important role. However, manual phenotyping of plants is prone to human errors and is labor intensive and time-consuming. Hence, it is important to develop techniques that measure plant phenotypic traits accurately and rapidly. The aim of this study was to determine the feasibility of point cloud data based on a 3D light detection and ranging (LiDAR) system for plant phenotyping. The obtained results were then verified through manually acquired data from the sorghum samples. This study measured the plant height, plant crown diameter and the panicle height and diameter. The R² of each trait was 0.83, 0.94, 0.90, and 0.90, and the root mean square error (RMSE) was 6.8 cm, 1.82 cm, 5.7 mm, and 7.8 mm, respectively. The results showed good correlation between the point cloud data and manually acquired data for plant phenotyping. The results indicate that the 3D LiDAR system has potential to measure the phenotypes of sorghum in a rapid and accurate way.

• Geomechanics and Engineering
• /
• v.28 no.6
• /
• pp.599-611
• /
• 2022

Tendon reinforced cemented soil is applied extensively in foundation stabilisation and improvement, especially in areas with soft clay. To solve the deterioration problem led by steel corrosion, the glass fiber-reinforced polymer (GFRP) tendon is introduced to substitute the traditional steel tendon. The interface bond strength between the cemented soil matrix and GFRP tendon demonstrates the outstanding mechanical property of this composite. However, the lack of research between the influence factors and bond strength hinders the application. To evaluate these factors, back propagation neural network (BPNN) is applied to predict the relationship between them and bond strength. Since adjusting BPNN parameters is time-consuming and laborious, the particle swarm optimisation (PSO) algorithm is proposed. This study evaluated the influence of water content, cement content, curing time, and slip distance on the bond performance of GFRP tendon-reinforced cemented soils (GTRCS). The results showed that the ultimate and residual bond strengths were both in positive proportion to cement content and negative to water content. The sample cured for 28 days with 30% water content and 50% cement content had the largest ultimate strength (3879.40 kPa). The PSO-BPNN model was tuned with 3 neurons in the input layer, 10 in the hidden layer, and 1 in the output layer. It showed outstanding performance on a large database comprising 405 testing results. Its higher correlation coefficient (0.908) and lower root-mean-square error (239.11 kPa) were obtained compared to multiple linear regression (MLR) and logistic regression (LR). In addition, a sensitivity analysis was applied to acquire the ranking of the input variables. The results illustrated that the cement content performed the strongest influence on bond strength, followed by the water content and slip displacement.