• Korean Journal of Agricultural and Forest Meteorology
• /
• v.23 no.3
• /
• pp.149-155
• /
• 2021

Early warning systems for weather risk management in the agricultural sector have been developed to predict potential wind damage to crops. These systems take into account the daily maximum wind speed to determine the critical wind speed that causes fruit drops and provide the weather risk information to farmers. In an effort to increase the accuracy of wind risk predictions, an artificial neural network for binary classification was implemented. In the present study, the daily wind speed and other weather data, which were measured at weather stations at sites of interest in Jeollabuk-do and Jeollanam-do as well as Gyeongsangbuk- do and part of Gyeongsangnam- do provinces in 2019, were used for training the neural network. These weather stations include 210 synoptic and automated weather stations operated by the Korean Meteorological Administration (KMA). The wind speed data collected at the same locations between January 1 and December 12, 2020 were used to validate the neural network model. The data collected from December 13, 2020 to February 18, 2021 were used to evaluate the wind risk prediction performance before and after the use of the artificial neural network. The critical wind speed of damage risk was determined to be 11 m/s, which is the wind speed reported to cause fruit drops and damages. Furthermore, the maximum wind speeds were expressed using Weibull distribution probability density function for warning of wind damage. It was found that the accuracy of wind damage risk prediction was improved from 65.36% to 93.62% after re-classification using the artificial neural network. Nevertheless, the error rate also increased from 13.46% to 37.64%, as well. It is likely that the machine learning approach used in the present study would benefit case studies where no prediction by risk warning systems becomes a relatively serious issue.

• Pediatric Infection and Vaccine
• /
• v.28 no.2
• /
• pp.101-109
• /
• 2021

Purpose: Common human coronaviruses (HCoVs) are relatively understudied due to the mild nature of HCoV infection. Given the lack of local epidemiology data on common HCoVs, we aimed to describe clinical and epidemiological characteristics of common HCoVs in children. Methods: Respiratory viral test results from 9,589 respiratory samples from Seoul National University Children's Hospital were analyzed from January 2015 to December 2019. Viral detection was done by the multiplex reverse transcription polymerase chain reaction. Demographics and clinical diagnosis were collected for previously healthy children tested positive for HCoVs. Results: Of the 9,589 samples tested, 1 or more respiratory viruses were detected from 5,017 (52.3%) samples and 463 (4.8%) samples were positive for HCoVs (OC43 2.8%, NL63 1.4%, 229E 0.7%). All 3 types co-circulated during winter months (November to February) with some variation by type. HCoV-OC43 was the most prevalent every winter season. HCoV-NL63 showed alternate peaks in late winter (January to March) and early winter (November to February). HCoV-229E had smaller peaks every other winter. Forty-one percent of HCoV-positive samples were co-detected with additional viruses; human rhinovirus 13.2%, respiratory syncytial virus 13.0%, influenza virus 4.3%. Common clinical diagnosis was upper respiratory tract infection (60.0%) followed by pneumonia (14.8%), croup (8.1%), and bronchiolitis (6.7%). Croup accounted for 17.0% of HCoV-NL63-positive children. Conclusions: This study described clinical and epidemiological characteristics of common HCoVs (OC43, NL63, 229E) in children. Continuing surveillance, perhaps by adding HKU1 in the diagnostic panel can further elucidate the spectrum of common HCoV infections in children.

• Korean Journal of School Psychology
• /
• v.17 no.1
• /
• pp.17-38
• /
• 2020

This study examined the factor structure of the Child Behavior Checklist Dysregulation Profile(CBCL-DP) for school-aged children in Korea identified differences in the level of maladjustment and problematic behaviors between the clinical group which had characteristics of CBCL-DP and the control group which did not. Confirmative factor analysis was performed on three alternative models from the literature to determine which was the most appropriate factor structure for the CBCL-DP. The result showed that the bi-factor model fit the sample data better than both the one and second-factor models. To confirm that the bi-factor model was the most appropriate factor structure, regression paths with relevant variables examined. The showed that CBCL-DP with the bi-factor model was associated with executive function difficulty as reported by parents and with school adjustment and all sub-factors of strength and difficulty as reported by teachers. The results also showed that this model had a different relationship with anxiety/depression, aggressive behavior, and attention problems than the other models. The clinical group was shown to have more executive function difficulty, worse adjustment of school life and to be less likely to engage in desired behaviors than the control group. These results indicate the CBCL-DP is more related to negative outcomes than any other factor, and that the bi-factor model was found to best fit the sample data, consistent with other studies. The early discovery of CBCL-DP can be used to provide interventions for high-risk children who exhibit emotional and behavioral problems, making its detection a significant diagnostic tool. The implications of these result, the limitations of this study, and areas for future research are discussed in this paper.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.25 no.1
• /
• pp.17-27
• /
• 2023

Weeds are one of the factors that reduce crop yield through nutrient and photosynthetic competition. Quantification of weed density are an important part of making accurate decisions for precision weeding. In this study, we tried to quantify the density of weeds in images of maize fields taken by unmanned aerial vehicle (UAV). UAV image data collection took place in maize fields from May 17 to June 4, 2021, when maize was in its early growth stage. UAV images were labeled with pixels from maize and those without and the cropped to be used as the input data of the semantic segmentation network for the maize detection model. We trained a model to separate maize from background using the deep learning segmentation networks DeepLabV3+, U-Net, Linknet, and FPN. All four models showed pixel accuracy of 0.97, and the mIOU score was 0.76 and 0.74 in DeepLabV3+ and U-Net, higher than 0.69 for Linknet and FPN. Weed density was calculated as the difference between the green area classified as ExGR (Excess green-Excess red) and the maize area predicted by the model. Each image evaluated for weed density was recombined to quantify and visualize the distribution and density of weeds in a wide range of maize fields. We propose a method to quantify weed density for accurate weeding by effectively separating weeds, maize, and background from UAV images of maize fields.

• Journal of Chest Surgery
• /
• v.43 no.6
• /
• pp.588-595
• /
• 2010

Background: Base on types of tumor, the types of expressed tumor is diverse and the difference in its expression rate is even more various. Due to such reasons an animal model is absolutely needed for a clinical research of lung cancer. The author attempted oncogenesis by cultivating a cell line of non-small cell carcinoma and then injecting it inside thoracic cavities of nude mice. The author conducted quantitative analyses of HER2/neu tumor gene - an epidermal growth factor receptor (EGFR) related to lung cancer, and TGF-${\beta}_1$, which acts as a resistance to cell growth inhibition and malignant degeneration. In order to investigate achievability of the oncogenesis, histological changes and the expression of cancer gene in case of orthotopic lung cancer is necessary. Material and Method: Among 20 immunity-free male BALB/c, five nude mice were selected as the control group and rest as the experimental group. Their weights ranged from 20 to 25 gm (Orient, Japan). After injection of lung cancer line (SW900 G IV) into the pleural cavity of nude mice, They were raised at aseptic room for 8 weeks. HER2/neu was quantitatively analyzed by separating serum from gathered blood via chemiluminiscent immunoassay (CLIA), and immunosandwitch method was applied to quantitatively analyze TGF-${\beta}_1$. SPSS statistical program (SPSS Version 10.0, USA) was implemented for statistical analysis. Student T test was done, and cases in which p-value is less than 0.05 were considered significant. Result: Even after lung cancer was formed in the normal control group or after intentionally injected lung cancer cell line, no amplification of HER2/neu gene showed reaction. However, the exact quantity of TGF-${\beta}_1$ was $28,490{\pm}8,549pg/mL$, and the quantity in the group injected with lung cancer cell was $42,362{\pm}14,449pg/mL$, meaning 1.48 times highly Significant (p<0.483). It proved that HER2/neu gene TGF-${\beta}_1$ had no meaningful interconnection. Conclusion: TGF-${\beta}_1$ gene expressed approximately 1.48 times amplification in comparison to the control group. The amplification of TGF-${\beta}_1$ meant somatic recuperation inhibition mechanism due to carcinogenesis in nude mice was definitely working. It may be implemented as a quantitative analysis that allows early detection of lung cancer in human body.

• Tuberculosis and Respiratory Diseases
• /
• v.41 no.4
• /
• pp.339-353
• /
• 1994

Background : The p53 gene codes for a DNA-binding nuclear phosphoprotein that appears to inhibit the progression of cells from the G1 to the S phase of the cell cycle. Mutations of the p53 gene are common in a wide variety of human cancers, including lung cancer. In lung cancers, point mutations of the p53 gene have been found in all histological types including approximately 45% of resected NSCLC and even more frequently in SCLC specimens. Mutant forms of the p53 protein have transforming activity and interfere with the cell-cycle regulatory function of the wild-type protein. The majority of p53 gene mutations produce proteins with altered conformation and prolonged half life; these mutant proteins accumulate in the cell nucleus and can be detected by immunohistochemical staining. But protein overexpression has been reported in the absence of mutation. p53 protein overexpression or gene mutation is reported poor prognostic factor in breast cancer, but in lung cancer, its prognostic significance is controversial. Method : We investigated the p53 abnormalities by nucleotide sequencing, polymerase chain reaction-single strand conformation polymorphism(PCR-SSCP), and immunohistochemical staining. We correlated these results with each other and survival in 75 patients with NSCLC resected with curative intent. Overexpression of the p53 protein was studied immunohistochemically in archival paraffin- embedded tumor samples using the D07(Novocastra, U.K.) antibody. Overexpression of p53 protein was defined by the nuclear staining of greater than 25% immunopositive cells in tumors. Detection of p53 gene mutation was done by PCR-SSCP and nucleotide sequencing from the exon 5-9 of p53 gene. Result: 1) Of the 75 patients, 36%(27/75) showed p53 overexpression by immunohistochemical stain. There was no survival difference between positive and negative p53 immunostaining(overall median survival of 26 months, disease free median survival of 13 months in both groups). 2) By PCR-SSCP, 27.6%(16/58) of the patients showed mobility shift. There was no significant difference in survival according to mobility shift(overall median survival of 27 in patients without mobility shift vs 20 months in patients with mobility shift, disease free median survival of 8 months vs 10 months respectively). 3) Nucleotide sequence was analysed from 29 patients, and 34.5%(10/29) had mutant p53 sequence. Patients with the presence of gene mutations showed tendency to shortened survival compared with the patients with no mutation(overall median survival of 22 vs 27 months, disease free median survival of 10 vs 20 months), but there was no statistical significance. 4) The sensitivity and specificity of immunostain based on PCR-SSCP was 67.0%, 74.0%, and that of the PCR-SSCP based on the nucleotide sequencing was 91.8%, 96.2% respectively. The concordance rate between the immunostain and PCR-SSCP was 62.5%, and the rate between the PCR-SSCP and nucleotide sequencing was 95.3%. Conclusion : In terms of detection of p53 gene mutation, PCR-SSCP was superior to immunostaining. p53 gene abnormalities either overexpression or mutation were not a significant prognostic factor in NSCLC patients resected with curative intent. However, patients with the mutated p53 gene showed the trends of early relapse.

• Tuberculosis and Respiratory Diseases
• /
• v.56 no.5
• /
• pp.465-484
• /
• 2004

Background : Insulin-like growth factor (IGF)-binding protein-3 (IGFBP-3) inhibits the proliferation of non-small cell lung cancer (NSCLC) cells by inducing apoptosis. Methods : In this study, we investigated whether hypermethylation of IGFBP-3 promoter play an important role in the loss of IGFBP-3 expression in NSCLC. We also studied the mechanisms that mediate the silencing of IGFBP-3 expression in the cell lines which have hypermethylated IGFBP-3 promoter. Results : The IGFBP-3 promoter has hypermethylation in 7 of 15 (46.7%) NSCLC cell lines and 16 (69.7%) of 23, 7 (77.8%) of 9, 4 (80%) of 5, 4 (66.7 %) of 6, and 6 (100%) of 6 tumor specimens from patients with stage I, II, IIIA, IIIB, and IV NSCLC, respectively. The methylation status correlated with the level of protein and mRNA in NSCLC cell lines. Expression of IGFBP-3 was restored by the demethylating agent 5'-aza-2'-deoxycytidine (5'-aza-dC) in a subset of NSCLC cell lines. The Sp-1/ Sp-3 binding element in the IGFBP-3 promoter, important for promoter activity, was methylated in the NSCLC cell lines which have reduced IGFBP-3 expression and the methylation of this element suppressed the binding of the Sp-1 transcription factor. A ChIP assay showed that the methylation status of the IGFBP-3 promoter influenced the binding of Sp-1, methyl-CpG binding protein-2 (MeCP2), and histone deacetylase (HDAC) to Sp-1/Sp-3 binding element, which were reversed by by 5'-aza-dC. In vitro methylation of the IGFBP-3 promoter containing the Sp-1/Sp-3 binding element significantly reduced promoter activity, which was further suppressed by the overexpression of MeCP2. This reduction in activity was rescued by 5'-aza-dC. Conclusion : These findings indicate that hypermethylation of the IGFBP-3 promoter is one mechanism by which IGFBP-3 expression is silenced and MeCP2, with recruitment of HDAC, may play a role in silencing of IGFBP-3 expression. The frequency of this abnormality is also associated with advanced stages among the patients with NSCLC, suggesting that IGFBP-3 plays an important role in lung carcinogenesis/progression and that the promoter methylation status of IGFBP-3 may be a marker for early molecular detection and/or for monitoring chemoprevention efforts.

• Journal of Preventive Medicine and Public Health
• /
• v.33 no.3
• /
• pp.285-298
• /
• 2000

Objectives : To investigate the positive rate of Helicobacter pylori in patients with dyspepsia; medical compliance and related factors; the eradication rate a year after screening and related factors; the relationship between the eradication of Helicobacter pylori and the improvement of symptoms; and the estimated cost of three alternative approaches to treat Helicobacter pylori in the community. Methods : A total of 510 subjects with dyspeptic symptoms were selected and given the serological test in March 1998. The subjects were all adults over 30 years of age residing in Kyongju city. Results : Of the 510 selected subjects, 375 (73.5%) subjects proved positive for Helicobacter pylori on serological testing. Of these 304 (81.1%) who consented to an endoscopic examination, underwent a Campylobacter-like organism (CLO) test. Of these 304 subjects, 204 (67.1%), who had positive CLO test results, were given the triple therapy - tripotassium dicitrato bismuthate, amoxicillin, and metronidazole. To determine the eradication rate of Helicobacter pylori, 181 (88.1%) out of the 204 subjects who were given the triple therapy completed a follow-up urea breath test one year later. Of these, the Helicobacter pylori of 87(48.1%) subjects was eradicated. Among the 122 subjects who were medication compliant, the Helicobacter pylori eradication rate was 57.4% (70 subjects), while the eradication rates was only 28.8% (17subjects) in the non-compliant group. The Helicobacter pylori eradication was significantly related to compliance (p<0.01), but not to other characteristics and habits. The symptom improvement rate tended to be higher 62.1%), in the Helicobacter pylori eradicated group than in the non-eradicated group (59.6%). Conclusions : When the advantages and disadvantages of each alternative treatment were considered in the light of cost, antibiotic tolerance and the number of patients to be treated, alternative II was favorable in terms of cost. Alternative III was favorable in terms of the number of patients to be treated, antibiotic tolerance and early detection of gastric cancer. Further long-term research analyzing the cost-benefit and cost-effectiveness of each treatment will be needed as supporting material in creating new policies.

• Journal of Chest Surgery
• /
• v.35 no.12
• /
• pp.847-853
• /
• 2002

We previously published the data that proved the safety of retrograde cerebral perfusion for 120 minutes. At this time, we planned to check the neuron-specific enolase and S100-$\beta$ in serum and urine to assess the possibility of early detection of cerebral injury. Material and Method: We used pigs(Landrace species) weighing 35 kg and performed RCP for 120 minutes. After the weaning of cardiopulmonary bypass, we observed the pigs for another 120 minutes. Systemic arterial pressure, central venous pressure, and serum and urine levels of neuron-specific enolose (NSE) and S100$\beta$ protein were checked. Central venous pressure during RCP was maintained in the range of 20 to 25 mmHg. Result: Serum levels of NSE(ng/$m\ell$) were 0.67$\pm$0.18(induction of anesthesia), 0.53$\pm$0.47(soon after CPB), 0.44$\pm$0.27(20min alter CPB), 0.24$\pm$0.09(RCP 20min), 0.37$\pm$0.35(RCP 40min), 0.33$\pm$0.21 (RCP 60min), 0.37$\pm$0.22(RCP 80min), 0.41$\pm$0.23(RCP 100 min), 0.48$\pm$0.26(RCP 120min), 0.42$\pm$0.29(30min after rewarming), 0.35 $\pm$0.32(60min after rewarming, 0.42$\pm$0.37(CPBoff 30min), 0.47$\pm$0.34(CPBOff 60min), 0.47$\pm$0.28(CPBOff 90min), and 0.57$\pm$0.29(CPBOff 120min). There was no statistically significant difference in levels between before and after RCP(ANOVA, p＞0.05). Urine levels of NSE also showed no statistically significant difference in levels between before and after RCP. There was no correlation between urine and serum levels of NSE(Pearson correlation, p＞0.05). Serum levels of S100$\beta$ protein(ng/$m\ell$) during the same time frames were 0.14$\pm$0.08, 0.15$\pm$0.07, 0.22$\pm$0.15, 0.23$\pm$0.07, 0.28$\pm$0.10, 0.40$\pm$0.05, 0.47$\pm$0.03, 0.49$\pm$0.12, 0.43$\pm$0.11, 0.46$\pm$0.15, 0.62$\pm$0.17, 0.77$\pm$0.21, 0.78$\pm$0.23, 0.77$\pm$0.23, and 0.82$\pm$0.33. There was statistically significant difference in levels between before and after RCP(ANOVA, p＜0.05). Urine levels of NSE also showed statistically significant difference in levels between before and after RCP(ANOVA, p＜0.05). There was significant correlation between urine and serum levels of NSE(Pearson correlation, p＜0.05). Conclusion: The author observed the increase in serum and urine levels of S100$\beta$ after 120 minutes of RCP. Significant correlation between serum and urine levels was observed. The results were considered to be the fundamental data that could correlate this study with human-based study.

• Journal of Intelligence and Information Systems
• /
• v.24 no.2
• /
• pp.221-241
• /
• 2018

Deep learning is getting attention recently. The deep learning technique which had been applied in competitions of the International Conference on Image Recognition Technology(ILSVR) and AlphaGo is Convolution Neural Network(CNN). CNN is characterized in that the input image is divided into small sections to recognize the partial features and combine them to recognize as a whole. Deep learning technologies are expected to bring a lot of changes in our lives, but until now, its applications have been limited to image recognition and natural language processing. The use of deep learning techniques for business problems is still an early research stage. If their performance is proved, they can be applied to traditional business problems such as future marketing response prediction, fraud transaction detection, bankruptcy prediction, and so on. So, it is a very meaningful experiment to diagnose the possibility of solving business problems using deep learning technologies based on the case of online shopping companies which have big data, are relatively easy to identify customer behavior and has high utilization values. Especially, in online shopping companies, the competition environment is rapidly changing and becoming more intense. Therefore, analysis of customer behavior for maximizing profit is becoming more and more important for online shopping companies. In this study, we propose 'CNN model of Heterogeneous Information Integration' using CNN as a way to improve the predictive power of customer behavior in online shopping enterprises. In order to propose a model that optimizes the performance, which is a model that learns from the convolution neural network of the multi-layer perceptron structure by combining structured and unstructured information, this model uses 'heterogeneous information integration', 'unstructured information vector conversion', 'multi-layer perceptron design', and evaluate the performance of each architecture, and confirm the proposed model based on the results. In addition, the target variables for predicting customer behavior are defined as six binary classification problems: re-purchaser, churn, frequent shopper, frequent refund shopper, high amount shopper, high discount shopper. In order to verify the usefulness of the proposed model, we conducted experiments using actual data of domestic specific online shopping company. This experiment uses actual transactions, customers, and VOC data of specific online shopping company in Korea. Data extraction criteria are defined for 47,947 customers who registered at least one VOC in January 2011 (1 month). The customer profiles of these customers, as well as a total of 19 months of trading data from September 2010 to March 2012, and VOCs posted for a month are used. The experiment of this study is divided into two stages. In the first step, we evaluate three architectures that affect the performance of the proposed model and select optimal parameters. We evaluate the performance with the proposed model. Experimental results show that the proposed model, which combines both structured and unstructured information, is superior compared to NBC(Naïve Bayes classification), SVM(Support vector machine), and ANN(Artificial neural network). Therefore, it is significant that the use of unstructured information contributes to predict customer behavior, and that CNN can be applied to solve business problems as well as image recognition and natural language processing problems. It can be confirmed through experiments that CNN is more effective in understanding and interpreting the meaning of context in text VOC data. And it is significant that the empirical research based on the actual data of the e-commerce company can extract very meaningful information from the VOC data written in the text format directly by the customer in the prediction of the customer behavior. Finally, through various experiments, it is possible to say that the proposed model provides useful information for the future research related to the parameter selection and its performance.