Mechanistic modelling for African swine fever transmission in the Republic of Korea

김으뜸 대한수의학회 2023 Journal of Veterinary Science Vol.24 No.2

Under the current African swine fever (ASF) epidemic situation, a science-based ASF-control strategy is required. An ASF transmission mechanistic model can be used to understand the disease transmission dynamics among susceptible epidemiological units and evaluate the effectiveness of an ASF-control strategy by simulating disease spread results with different control options. The force of infection, which is the probability that a susceptible epidemiological unit becomes infected, could be estimated by applying an ASF transmission mechanistic model. The government needs to plan an ASF-control strategy based on an ASF transmission mechanistic model.

Spatial pattern of avian influenza (AI) risk using data from routine AI surveillance in the Republic of Korea, 2014–2015

김으뜸,박선일 한국예방수의학회 2019 예방수의학회지 Vol.43 No.3

Epidemiological research to investigate the spatial characteristics of poultry farms confirmed with avian influenza (AI) infection can help increase the efficacy of AI surveillance as well as AI control strategies. The spatial characteristics of poultry farms confirmed with AI infection can provide insights on effective AI-surveillance and AI-control strategies to policymakers by providing a visualization of the geographical pattern of AI distribution. The goal of the current study was to investigate the spatial characteristics of the risk of a farm being AI-positive by using data from routine AI-surveillance performed during the period 2014–2015. To achieve this goal, we applied a spatial model because it improves the estimation of the relative risk by taking into account spatial dependence between epidemiological units. The results revealed there was a lack of dependency between districts in the risk of a farm being AI-positive. The estimates for the spatial autocorrelation coefficient in the spatial model for chicken farms were 0.006 in 2014 (p = 0.9496) and -0.064 in 2015 (p = 0.6052) and for duck farms were -0.066 in 2014 (p = 0.4380) and 0.047 in 2015. Likewise, Moran’s I statistic estimates for chicken farms were 0.0243 in 2014 (p = 0.3183) and -0.0174 in 2015 (p = 0.5657) and for duck farms were -0.0342 in 2014 (p = 0.6678) and -0.0230 in 2015.

Use of Likelihood Ratios in Evidence-based Clinical Decision Making

김으뜸,박선일 한국임상수의학회 2008 한국임상수의학회지 Vol.25 No.3

During the clinical decision making practitioners are often faced with performing diagnostic tests to solveterms such as sensitivity, specificity, and positive (PPV) and negative predictive value (NPV). Although well known,clinicians are frequently unclear about the concept and application of these terms in everyday evidence-based clinicaldecision making. Sensitivity and specificity, which are intrinsic properties of diagnostic tests, sumarizes thecharacteristics of the test over a population. The PPV and NPV are greatly dependent on the population prevalenceof disease, and thus they do not transferable to different patients or clinical settings. Besides, considering the factthat clinicians more often interested in knowing the extent to which a test result could confirm or exclude of a condition(LR) using the information contained in sensitivity and specificity are becoming increasingly popular for reportingthe usefulness of diagnostic tests because this term provide an indication of posttest probability as a function of thepretest probability. In this article, clinical applications of LR are illustrated with some practical examples. Discussionis also included of the inherent limitations regarding diagnostic test characteristics.

The contribution of farm vehicle movements for a highly pathogenic avian influenza epidemic in 2014 in the Republic of Korea

김으뜸,박선일 한국예방수의학회 2019 예방수의학회지 Vol.43 No.4

The goal of the current study was to explore the relationship between vehicle movement frequency and a disease outbreak by using the example of the highly pathogenic avian influenza (HPAI) outbreak in 2014 in the Republic of Korea. To explore the relationship between the HPAI outbreak status of Korean provinces and vehicle movements, both an ordinary least square model (OLS) and a maximum entropy model (MaxEnt) were built. The HPAI outbreak status of each province was used as a dependent variable. The number of poultry farm vehicle movements within the province (within variable), the number of poultry farm vehicle movements from one province to another province (outbound variable), the number of poultry farm vehicle movements from other provinces to one province (inbound variable), and the number of poultry farms in each province were included in the models as independent variables. Results of the OLS model were as follows: the estimated coefficient of the log-transformed within variable was -0.30, that of the log-transformed outbound variable was 0.71, that of the log-transformed inbound variable was -0.30, and that of the number of poultry farms was 0.07; however, only the number of poultry farms per province was statistically significant. Results of the MaxEnt model were as follows: the median relative contribution of the log-transformed outbound variable was 52.0 (range: 12.2–83.9), that of the log-transformed inbound variable was 34.4 (range: 8.8–83.4), that of the log-transformed within variable was 3.7 (range: 1.8–7.3), and that of the number of poultry farms per province was 0.7 (range: 0.0–11.7). The area under the receiver operating characteristics curve was 0.683. The results of current study should be helpful for planning a national HPAI surveillance program to locate surveillance resources with the consideration of risk level of provinces.

Global and local models of poultry farm vehicle movement contributions to a 2014 highly pathogenic avian influenza epidemic in the Republic of Korea

김으뜸,박선일 한국예방수의학회 2019 예방수의학회지 Vol.43 No.4

The goal of the current study was to estimate the contribution of poultry farm vehicle movement frequency to the 2014 highly pathogenic avian influenza (HPAI) epidemic using both global and local regression models. On one hand, the global model did not consider the hypothesis that a relationship between predictors and the outcome variable might vary across the country (spatially homogeneous), while on the other hand, the local model considered that there was spatial heterogeneity within the country. The HPAI outbreak status in each province was used as a dependent variable and the number of poultry farm vehicle movements within each province (within variable), the number of poultry farm vehicle movement from one province to another province (outbound variable), the number of poultry farm vehicle movements from other provinces to one province (inbound variable), and the number of poultry farms in each province were included in the model as independent variables. The results of a global model were as follows: estimated coefficient of the log-transformed within variable was 0.73, that of the log-transformed outbound variable was 2.04, that of the log-transformed inbound variable was 0.74, and that of the number of poultry farms was 1.08. Only the number of poultry farms was a statistically significant variable (p-value < 0.001). The AIC score of the global model was 1397.5. The results of the local model were as follows: estimated median coefficient of the log-transformed within variable was 0.75, that of the log-transformed outbound variable was 2.54, that of the log-transformed inbound variable was 0.60, and that of the number of poultry farms was 0.07. The local model’s AIC score was 1382.2. The results of our study indicate that a local model would provide a better understanding of the relationship between HPAI outbreak status and poultry farm vehicle movements than that provided by a global model.

Species distribution modeling for wild boar (Sus scropa) in the Republic of Korea using MODIS data

김으뜸,박선일 한국예방수의학회 2020 예방수의학회지 Vol.44 No.2

The distribution of wild boar (Sus scropa) in the Republic of Korea was forecasted using environmental factors. A species distribution model was applied with the standard normalized difference vegetation index (NDVI), enhanced vegetation index (EVI), solar zenith angle (SUNZ), daytime land surface temperature (dTemp), and nighttime land surface temperature (nTemp). Understanding wild boar distribution is important for controlling African swine fever (ASF) because the disease could be endemic in wild boar or spread from wild boars to domestic pigs. Among the five predictors, the NDVI was the most influencing factor for the wild boar distribution. The relative contributions of the predictors were 67.4 for NDVI, 16.9 for dTemp, 10.5 for SUNZ, 4.4 for EVI, and 0.8 for nTemp. The area size under the receiver-operating curve of the receiver-operating characteristics for the current model was 0.62, but the real wild boar observation data overlapped with the predicted high-density wild boar distribution area. The wild boar distribution density was relatively higher in Gangwon-do, Gyeongsangbuk-do, Gyeongsangnam-do, and Jeollanam-do. Given the ASF epidemics, contact between ASF-infected animals and ASF-susceptible animals in high-density wild boar distribution areas should be prevented by long-range fencing or active surveillance.

온도 센싱을 위한 크롬 박막 패브리-페로 간섭계 설계

김으뜸,전수진,김선호,김도현,이재상,정미,정태경,최영완 한국정보통신설비학회 2018 한국정보통신설비학회 학술대회 Vol.2018 No.08

다층모형을 이용한 국내 양돈농가의 돼지생식기호흡기증후군 위험요인 분석

김으뜸,이경기,김성희,박선일 한국임상수의학회 2017 한국임상수의학회지 Vol.34 No.2

The goal of this study was to investigate risk factors associated with porcine reproductive and respiratory syndrome (PRRS) in pig farms in the Republic of Korea using logistic regression and a multilevel model. A crosssectional study was applied to 305 pig farms with a questionnaire-based interview by veterinarians between March 2014 and February 2015. The questionnaire comprised eight categories: proximity to neighbors, disinfection, visitors, vehicles, insecticides, wild animals, gilts, and feeding. In total, 61 questions in eight categories related to pig farm biosecurity were investigated. Farms were classified as PRRS stable or unstable based on the results of an antibody test and PCR. For univariate analysis, keeping production records with computers (OR = 0.283, 95% CI = 0.056 − 1.425), accredited farm with no use of antibiotics (OR = 0.412, 95% CI = 0.134 − 1.269), reviewing health record of semen prior to purchasing (OR = 0.492, 95% CI = 0.152 − 1.589), complete isolation of runt pigs (OR = 0.264, 95% CI = 0.084 − 0.829), compulsory registering for visitors (OR = 0.424, 95% CI = 0.111 − 1.612), keeping records of insecticide history (OR = 0.406, 95% CI = 0.089 − 1.846), routine on-farm monitoring by veterinarians (OR = 0.314, 95% CI = 0.069 − 1.423), and use of on-farm checklist for biosecurity monitoring (OR = 0.313, 95% CI = 0.063 − 1.553) were found to decrease the probability of PRRS infection. Multivariate and multilevel analysis revealed only two factors, complete isolation of runt pigs (OR = 0.165, 95% CI = 0.045 − 0.602 and OR = 0.208, 95% CI = 0.055 − 0.782) and compulsory registering for visitors (OR = 0.106, 95% CI = 0.017 − 0.655 and OR = 0.119, 95% CI = 0.017 − 0.809) were found to decrease the probability of PRRS infection. The intracluster correlation coefficient of a province for multilevel model was 0.05. The results of this study might facilitate biosecurity measures for individual farms to reduce the probability of PRRS infection.

Poultry farm vehicle movements as risk factors for the 2014/15 highly pathogenic avian influenza epidemics in the Republic of Korea

김으뜸,박선일 한국예방수의학회 2020 예방수의학회지 Vol.44 No.2

Since the first HPAI epidemics in 2003, there has been little epidemiological research on the association between HPAI epidemics and vehicle movements around poultry farms. This study examined the relationship between vehicle movements around poultry farms and the 2014/15 HPAI epidemics in the Republic of Korea using two methods: a boosted regression trees (BRT) model and logistic regression of a generalized linear model (GLM). The BRT model considers the non-linearity association between the frequency of vehicle movements around poultry farms and the HPAI outbreak status per province using a machine learning technique. In contrast, a GLM assesses the relationship based on the traditional frequentist method. Among the three types of vehicle movements (outbound, inbound, and within), only the outbound was found to be a risk factor of the 2014/15 HPAI epidemics according to both the BRT model and multivariate logistic regression of GLM. In the BRT model results, the median relative contribution of the log-transformed outbound variable was 53.68 (range: 39.99 – 67.58) in the 2014 epidemics and 49.79 (range: 33.90 – 56.38) in the 2015 epidemics. In the GLM results, the odds ratio of the log-transformed outbound variable was 1.22 for the 2014 HPAI epidemics (p < 0.001) and 2.48 for the 2015 HPAI epidemics (p < 0.001), respectively. The results indicated that intensive disinfection measures on outbound movement were needed to reduce the risk of HPAI spread. The current BRT models are suitable for risk analysis because the median area under the receiver operating characteristic curve was 0.83 (range: 0.74 – 0.91) and 0.85 (range: 0.73 – 0.87) for the 2014 and 2015 epidemics models, respectively. The Akaike information criterion scores for the multivariate logistic regression of GLM were 150.27 and 78.21 for the 2014 and 2015 epidemics models, respectively. These scores were relatively lower than those from the univariate logistic regression of GLM.

Tabu search algorithm을 이용한 optical fiber alignment 시간 단축 방법

김으뜸,김선호,박준희,전수진,김지훈,정태경,최영완 한국정보통신설비학회 2017 한국정보통신설비학회 학술대회 Vol.2017 No.08