• Journal of Intelligence and Information Systems
• /
• v.29 no.1
• /
• pp.121-142
• /
• 2023

As the number and weight of imported food are steadily increasing, safety management of imported food to prevent food safety accidents is becoming more important. The Ministry of Food and Drug Safety conducts on-site inspections of foreign food facilities before customs clearance as well as import inspection at the customs clearance stage. However, a data-based safety management plan for imported food is needed due to time, cost, and limited resources. In this study, we tried to increase the efficiency of the on-site inspection by preparing a machine learning prediction model that pre-selects the companies that are expected to fail before the on-site inspection. Basic information of 303,272 foreign food facilities and processing businesses collected in the Integrated Food Safety Information Network and 1,689 cases of on-site inspection information data collected from 2019 to April 2022 were collected. After preprocessing the data of foreign food facilities, only the data subject to on-site inspection were extracted using the foreign food facility_code. As a result, it consisted of a total of 1,689 data and 103 variables. For 103 variables, variables that were '0' were removed based on the Theil-U index, and after reducing by applying Multiple Correspondence Analysis, 49 characteristic variables were finally derived. We build eight different models and perform hyperparameter tuning through 5-fold cross validation. Then, the performance of the generated models are evaluated. The research purpose of selecting companies subject to on-site inspection is to maximize the recall, which is the probability of judging nonconforming companies as nonconforming. As a result of applying various algorithms of machine learning, the Random Forest model with the highest Recall_macro, AUROC, Average PR, F1-score, and Balanced Accuracy was evaluated as the best model. Finally, we apply Kernal SHAP (SHapley Additive exPlanations) to present the selection reason for nonconforming facilities of individual instances, and discuss applicability to the on-site inspection facility selection system. Based on the results of this study, it is expected that it will contribute to the efficient operation of limited resources such as manpower and budget by establishing an imported food management system through a data-based scientific risk management model.

• Korean Journal of Remote Sensing
• /
• v.35 no.6_3
• /
• pp.1173-1185
• /
• 2019

We, for the first time, retrieved sulfur dioxide (SO₂) vertical column density (VCD) in industrial and volcanic areas from TROPOspheric Monitoring Instrument (TROPOMI) using the Principle component analysis(PCA) algorithm. Furthermore, SO₂ VCDs retrieved by the PCA algorithm from TROPOMI raw data were compared with those retrieved by the Differential Optical Absorption Spectroscopy (DOAS) algorithm (TROPOMI Level 2 SO₂ product). In East Asia, where large amounts of SO₂ are released to the surface due to anthropogenic source such as fossil fuels, the mean value of SO₂ VCD retrieved by the PCA (DOAS) algorithm was shown to be 0.05 DU (-0.02 DU). The correlation between SO₂ VCD retrieved by the PCA algorithm and those retrieved by the DOAS algorithm were shown to be low (slope = 0.64; correlation coefficient (R) = 0.51) for cloudy condition. However, with cloud fraction of less than 0.5, the slope and correlation coefficient between the two outputs were increased to 0.68 and 0.61, respectively. It means that the SO₂ retrieval sensitivity to surface is reduced when the cloud fraction is high in both algorithms. Furthermore, the correlation between volcanic SO₂ VCD retrieved by the PCA algorithm and those retrieved by the DOAS algorithm is shown to be high (R = 0.90) for cloudy condition. This good agreement between both data sets for volcanic SO₂ is thought to be due to the higher accuracy of the satellite-based SO₂ VCD retrieval for SO₂ which is mainly distributed in the upper troposphere or lower stratosphere in volcanic region.

• Journal of Korea Water Resources Association
• /
• v.49 no.7
• /
• pp.625-634
• /
• 2016

Three leading agencies under different ministries - Korea Meteorological Administration (KMA) in the ministry of Environment, Han river control office in the Ministry of Land, Infrastructure and Transport (MOLIT) and Weather Group of ROK Air Force in the Ministry of National Defense (MND) - have been operated radars in the purpose of observing weather, hydrology and military operational weather in Korea. Eight S-band dual-pol. radars have been newly installed or replaced by these ministries over different places by 2015. However each ministry has different aims of operating radars, observation strategies, data processing algorithms, etc. Due to the differences, there is a wide level of accuracy on observed radar data as well as the composite images made of the cross governmental radar measurement. Gaining fairly high level of accuracy on radar data obtained by different agencies has been shared as a great concern by the ministries. Thus, "an agreement of harmonizing weather and hydrological radar products" was made by the three ministries in 2010. Particularly, this is very important to produce better rainfall estimation using the cross governmental radar measurement. Weather Radar Center(WRC) in KMA has been developed an empirical method using measurements observed by Yongin testbed radar. This study is aiming to examine the efficiency of the empirical method to improve the accuracies of radar rainfalls estimated from cross governmental dual-pol. radar measurements. As a result, the radar rainfalls of three radars (Baengnyeongdo, Biseulsan, and, Sobaeksan Radar) were shown improvement in accuracy (1-NE) up to 70% using data from May to October in 2015. Also, the range of the accuracies in radar rainfall estimation, which were from 30% to 60% before adjusting polarimetric variables, were decreased from 65% to 70% after adjusting polarimetric variables.

• Journal of the Korean Society of Radiology
• /
• v.16 no.1
• /
• pp.1-6
• /
• 2022

Deep learning is a collection of algorithms that enable learning by summarizing the key contents of large amounts of data; it is being developed to diagnose lesions in the medical imaging field. To evaluate the accuracy of the cerebral hemorrhage diagnosis, we used a convolutional neural network (CNN) to derive the diagnostic accuracy of cerebral parenchyma computed tomography (CT) images and the cerebral parenchyma CT images of areas where cerebral hemorrhages are suspected of having occurred. We compared the accuracy of CNN with different numbers of hidden layers and discovered that CNN with more hidden layers resulted in higher accuracy. The analysis results of the derived CT images used in this study to determine the presence of cerebral hemorrhages are expected to be used as foundation data in studies related to the application of artificial intelligence in the medical imaging industry.

• Journal of radiological science and technology
• /
• v.40 no.1
• /
• pp.79-85
• /
• 2017

The purpose of this study was to measure contrast to noise ratio (CNR) and signal to noise ratio (SNR) according to change of reconstruction from region of interest (ROI) in breast positron emission tomography-computed tomography (PET-CT), and to analyze the CNR and SNR statically. We examined images of breast PET-CT of 100 patients in a University-affiliated hospital, Seoul, Korea. Each patient's image of breast PET-CT were calculated by using ImageJ. Differences of CNR and SNR among four reconstruction algorithms were tested by SPSS Statistics21 ANOVA test for there was statistical significance (p<0.05). We have analysis socio-demographical variables, CNR and SNR according to reconstruction images, 95% confidence according to CNR and SNR of reconstruction and difference in a mean of CNR and SNR. SNR results, with the quality of distributions in the order of PSF_TOF, Iterative and Iterative-TOF, FBP-TOF. CNR, with the quality of distributions in the order of PSF_TOF, Iterative and Iterative-TOF, FBP-TOF. CNR and SNR of PET-CT reconstruction methods of the breast would be useful to evaluate breast diseases.

• Steel and Composite Structures
• /
• v.34 no.5
• /
• pp.743-767
• /
• 2020

Due to the impressive flexural performance, enhanced compressive strength and more constrained crack propagation, Fibre-reinforced concrete (FRC) have been widely employed in the construction application. Majority of experimental studies have focused on the seismic behavior of FRC columns. Based on the valid experimental data obtained from the previous studies, the current study has evaluated the seismic response and compressive strength of FRC rectangular columns while following hybrid metaheuristic techniques. Due to the non-linearity of seismic data, Adaptive neuro-fuzzy inference system (ANFIS) has been incorporated with metaheuristic algorithms. 317 different datasets from FRC column tests has been applied as one database in order to determine the most influential factor on the ultimate strengths of FRC rectangular columns subjected to the simulated seismic loading. ANFIS has been used with the incorporation of Particle Swarm Optimization (PSO) and Genetic algorithm (GA). For the analysis of the attained results, Extreme learning machine (ELM) as an authentic prediction method has been concurrently used. The variable selection procedure is to choose the most dominant parameters affecting the ultimate strengths of FRC rectangular columns subjected to simulated seismic loading. Accordingly, the results have shown that ANFIS-PSO has successfully predicted the seismic lateral load with R² = 0.857 and 0.902 for the test and train phase, respectively, nominated as the lateral load prediction estimator. On the other hand, in case of compressive strength prediction, ELM is to predict the compressive strength with R² = 0.657 and 0.862 for test and train phase, respectively. The results have shown that the seismic lateral force trend is more predictable than the compressive strength of FRC rectangular columns, in which the best results belong to the lateral force prediction. Compressive strength prediction has illustrated a significant deviation above 40 Mpa which could be related to the considerable non-linearity and possible empirical shortcomings. Finally, employing ANFIS-GA and ANFIS-PSO techniques to evaluate the seismic response of FRC are a promising reliable approach to be replaced for high cost and time-consuming experimental tests.

• Asian Pacific Journal of Cancer Prevention
• /
• v.15 no.10
• /
• pp.4265-4269
• /
• 2014

Purpose: This study aimed at summarizing epidemiological evidence of the association between gestational diabetes mellitus (GDM) and subsequent risk of cancer. Materials and Methods: We searched Medline, Embase, Cancer Lit and CINAHL for epidemiological studies published by February 1, 2014 examining the risk of cancer in patients with history of GDM using highly inclusive algorithms. Information about first author, year of publication, country of study, study design, cancer sites, sample sizes, attained age of subjects and methods used for determining GDM status were extracted by two researchers and Stata version 11.0 was used to perform the meta-analysis and estimate the pooled effects. Results: A total of 9 articles documented 5 cohort and 4 case-control studies containing 10,630 cancer cases and 14,608 women with a history of GDM were included in this review. Taken together, the pooled odds ratio (OR) between GDM and breast cancer risk was 1.01 (0.87-1.17); yet the same pooled ORs of case-control and cohort studies were 0.87 (0.71-1.06) and 1.25 (1.00-1.56) respectively. There are indications that GDM is strongly associated with higher risk of pancreatic cancer (HR=8.68) and hematologic malignancies (HR=4.53), but no relationships were detected between GDM and other types of cancer. Conclusions: Although GDM increases the risk of certain types of cancer, these results should be interpreted with caution becuase of some methodological flaws. The issue merits added investigation and coordinated efforts between researchers, antenatal clinics and cancer treatment and registration agencies to help attain better understanding.

• Korean Journal of Radiology
• /
• v.22 no.9
• /
• pp.1514-1524
• /
• 2021

Objective: To develop a radiomics risk score based on dynamic contrast-enhanced (DCE) MRI for prognosis prediction in patients with glioblastoma. Materials and Methods: One hundred and fifty patients (92 male [61.3%]; mean age ± standard deviation, 60.5 ± 13.5 years) with glioblastoma who underwent preoperative MRI were enrolled in the study. Six hundred and forty-two radiomic features were extracted from volume transfer constant (K^trans), fractional volume of vascular plasma space (V_p), and fractional volume of extravascular extracellular space (V_e) maps of DCE MRI, wherein the regions of interest were based on both T1-weighted contrast-enhancing areas and non-enhancing T2 hyperintense areas. Using feature selection algorithms, salient radiomic features were selected from the 642 features. Next, a radiomics risk score was developed using a weighted combination of the selected features in the discovery set (n = 105); the risk score was validated in the validation set (n = 45) by investigating the difference in prognosis between the "radiomics risk score" groups. Finally, multivariable Cox regression analysis for progression-free survival was performed using the radiomics risk score and clinical variables as covariates. Results: 16 radiomic features obtained from non-enhancing T2 hyperintense areas were selected among the 642 features identified. The radiomics risk score was used to stratify high- and low-risk groups in both the discovery and validation sets (both p < 0.001 by the log-rank test). The radiomics risk score and presence of isocitrate dehydrogenase (IDH) mutation showed independent associations with progression-free survival in opposite directions (hazard ratio, 3.56; p = 0.004 and hazard ratio, 0.34; p = 0.022, respectively). Conclusion: We developed and validated the "radiomics risk score" from the features of DCE MRI based on non-enhancing T2 hyperintense areas for risk stratification of patients with glioblastoma. It was associated with progression-free survival independently of IDH mutation status.

• Journal of Korean Academy of Oral and Maxillofacial Radiology
• /
• v.28 no.2
• /
• pp.387-413
• /
• 1998

Image resampling is of particular interest in digital radiology. When resampling an image to a new set of coordinate, there appears blocking artifacts and image changes. To enhance image quality, interpolation algorithms have been used. Resampling is used to increase the number of points in an image to improve its appearance for display. The process of interpolation is fitting a continuous function to the discrete points in the digital image. The purpose of this study was to determine the effects of the seven interpolation functions when image resampling in digital periapical images. The images were obtained by Digora, CDR and scanning of Ektaspeed plus periapical radiograms on the dry skull and human subject. The subjects were exposed to intraoral X-ray machine at 60kVp and 70 kVp with exposure time varying between 0.01 and 0.50 second. To determine which interpolation method would provide the better image, seven functions were compared; (1) nearest neighbor (2) linear (3) non-linear (4) facet model (5) cubic convolution (6) cubic spline (7) gray segment expansion. And resampled images were compared in terms of SNR(Signal to Noise Ratio) and MTF(Modulation Transfer Function) coefficient value. The obtained results were as follows ; 1. The highest SNR value(75.96dB) was obtained with cubic convolution method and the lowest SNR value(72.44dB) was obtained with facet model method among seven interpolation methods. 2. There were significant differences of SNR values among CDR, Digora and film scan(P<0.05). 3. There were significant differences of SNR values between 60kVp and 70kVp in seven interpolation methods. There were significant differences of SNR values between facet model method and those of the other methods at 60kVp(P<0.05), but there were not significant differences of SNR values among seven interpolation methods at 70kVp(P>0.05). 4. There were significant differences of MTF coefficient values between linear interpolation method and the other six interpolation methods (P< 0.05). 5. The speed of computation time was the fastest with nearest -neighbor method and the slowest with non-linear method. 6. The better image was obtained with cubic convolution, cubic spline and gray segment method in ROC analysis. 7. The better sharpness of edge was obtained with gray segment expansion method among seven interpolation methods.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.1062-1062
• /
• 2001

The concept of “precision agriculture” or “site-specific farming” is usually confined to the fields of soil science, crop science and agronomy. However, because plants grow in soil, animals eat plants, and humans eat animal products, it could be argued (perhaps with some poetic licence) that the fields of feed quality, animal nutrition and animal production should also be considered in this context. NIR spectroscopy has proved over the last 20 years that it can provide a firm foundation for quality measurement across all of these fields, and with the continuing developments in instrumentation, computer capacity and software, is now a major cog in the wheel of precision agriculture. There have been a few giant leaps and a lot of small steps in the impact of NIR on the animal world. These have not been confined to the amazing advances in hardware and software, although would not have occurred without them. Rapid testing of forages, grains and mixed feeds by NIR for nutritional value to livestock is now commonplace in commercial laboratories world-wide. This would never have been possible without the pioneering work done by the USDA NIR Forage Research Network in the 1980's, following the landmark paper of Norris et al. in 1976. The advent of calibration transfer between instruments, algorithms which utilize huge databases for calibration and prediction, and the ability to directly scan whole grains and fresh forages can also be considered as major steps, if not leaps. More adventurous NIR applications have emerged in animal nutrition, with emphasis on estimating the functional properties of feeds, such as in vivo digestibility, voluntary intake, protein degradability and in vitro assays to simulate starch digestion. The potential to monitor the diets of grazing animals by using faecal NIR spectra is also now being realized. NIR measurements on animal carcasses and even live animals have also been attempted, with varying degrees of success, The use of discriminant analysis in these fields is proving a useful tool. The latest giant leap is likely to be the advent of relatively low-cost, portable and ultra-fast diode array NIR instruments, which can be used “on-site” and also be fitted to forage or grain harvesters. The fodder and livestock industries are no longer satisfied with what we once thought was revolutionary: a 2-3 day laboratory turnaround for fred quality testing. This means that the instrument needs to be taken to the samples rather than vice versa. Considerable research is underway in this area, but the challenge of calibration transfer and maintenance of instrument networks of this type remains. The animal world is currently facing its biggest challenges ever; animal welfare, alleged effects of animal products on human health, environmental and economic issues are difficult enough, but the current calamities of BSE and foot and mouth disease are “the last straw” NIR will not of course solve all these problems, but is already proving useful in some of these areas and will continue to do so.