• Journal of the Korean Society of Safety
• /
• v.36 no.3
• /
• pp.31-39
• /
• 2021

In Korea, more than half of work-related fatalities have occurred on construction sites. To reduce such occupational accidents, safety inspection by government agencies is essential in construction sites that present a high risk of serious accidents. To address this issue, this study developed risk prediction models of serious accidents in construction sites using five machine learning methods: support vector machine, random forest, XGBoost, LightGBM, and AutoML. To this end, 15 proactive information (e.g., number of stories and period of construction) that are usually available prior to construction were considered and two over-sampling techniques (SMOTE and ADASYN) were used to address the problem of class-imbalanced data. The results showed that all machine learning methods achieved 0.876~0.941 in the F1-score with the adoption of over-sampling techniques. LightGBM with ADASYN yielded the best prediction performance in both the F1-score (0.941) and the area under the ROC curve (0.941). The prediction models revealed four major features: number of stories, period of construction, excavation depth, and height. The prediction models developed in this study can be useful both for government agencies in prioritizing construction sites for safety inspection and for construction companies in establishing pre-construction preventive measures.

• Proceedings of the National Institute of Ecology of the Republic of Korea
• /
• v.2 no.1
• /
• pp.1-14
• /
• 2021

The study has been carried out with an objective to prepare Siberian roe deer habitat potential maps in South Korea based on three geographic information system-based models including frequency ratio (FR) as a bivariate statistical approach as well as convolutional neural network (CNN) and long short-term memory (LSTM) as machine learning algorithms. According to field observations, 741 locations were reported as roe deer's habitat preferences. The dataset were divided with a proportion of 70:30 for constructing models and validation purposes. Through FR model, a total of 10 influential factors were opted for the modelling process, namely altitude, valley depth, slope height, topographic position index (TPI), topographic wetness index (TWI), normalized difference water index, drainage density, road density, radar intensity, and morphological feature. The results of variable importance analysis determined that TPI, TWI, altitude and valley depth have higher impact on predicting. Furthermore, the area under the receiver operating characteristic (ROC) curve was applied to assess the prediction accuracies of three models. The results showed that all the models almost have similar performances, but LSTM model had relatively higher prediction ability in comparison to FR and CNN models with the accuracy of 76% and 73% during the training and validation process. The obtained map of LSTM model was categorized into five classes of potentiality including very low, low, moderate, high and very high with proportions of 19.70%, 19.81%, 19.31%, 19.86%, and 21.31%, respectively. The resultant potential maps may be valuable to monitor and preserve the Siberian roe deer habitats.

• Imaging Science in Dentistry
• /
• v.51 no.1
• /
• pp.55-62
• /
• 2021

Purpose: This study evaluated the influence of a metal artifact reduction (MAR) tool in a cone-beam computed tomography (CBCT) device on the diagnosis of vertical root fractures (VRFs) in teeth with different root filling materials. Materials and Methods: Forty-five extracted human premolars were classified into three subgroups; 1) no filling; 2) gutta-percha; and 3) metallic post. CBCT images were acquired using an Orthopantomograph 300 unit with and without a MAR tool. Subsequently, the same teeth were fractured, and new CBCT scans were obtained with and without MAR. Two oral radiologists evaluated the images regarding the presence or absence of VRF. Receiver operating characteristic (ROC) curves and diagnostic tests were performed. Results: The overall area under the curve values were 0.695 for CBCT with MAR and 0.789 for CBCT without MAR. The MAR tool negatively influenced the overall diagnosis of VRFs in all tested subgroups, with lower accuracy (0.45-0.72), sensitivity (0.6-0.67), and specificity (0.23-0.8) than were found for the images without MAR. In the latter group, the accuracy, sensitivity, and specificity values were 0.68-0.77, 0.67-083, and 0.53-087, respectively. However, no significant difference was found between images with and without MAR for the no filling and gutta-percha subgroups (P>0.05). In the metallic post subgroup, CBCT showed a significant difference according to MAR use (P<0.05). Conclusion: The OP 300 MAR tool negatively influenced the detection of VRFs in teeth with no root canal filling, gutta-percha, or metallic posts. Teeth with metallic posts suffered the most from the negative impact of MAR.

• Food Science of Animal Resources
• /
• v.41 no.2
• /
• pp.312-323
• /
• 2021

The purpose of this study was to use 1H nuclear magnetic resonance (1H NMR) to quantify taste-active and bioactive compounds in chicken breasts and thighs from Korean native chicken (KNC) [newly developed KNCs (KNC-A, -C, and -D) and commercial KNC-H] and white-semi broiler (WSB) used in Samgye. Further, each breed was differentiated using multivariate analyses, including a machine learning algorithm designed to use metabolic information from each type of chicken obtained using 1H-13C heteronuclear single quantum coherence (2D NMR). Breast meat from KNC-D chickens were superior to those of conventional KNC-H and WSB chickens in terms of both taste-active and bioactive compounds. In the multivariate analysis, meat portions (breast and thigh) and chicken breeds (KNCs and WSB) could be clearly distinguished based on the outcomes of the principal component analysis and partial least square-discriminant analysis (R2=0.945; Q2=0.901). Based on this, we determined the receiver operating characteristic (ROC) curve for each of these components. AUC analysis identified 10 features which could be consistently applied to distinguish between all KNCs and WSB chickens in both breast (0.988) and thigh (1.000) meat without error. Here, both 1H NMR and 2D NMR could successfully quantify various target metabolites which could be used to distinguish between different chicken breeds based on their metabolic profile.

• Journal of the Korea Convergence Society
• /
• v.13 no.3
• /
• pp.33-41
• /
• 2022

In this article, we proposed to predict natural gas (NG) leakage levels through feature selection based on a factor analysis (FA) of the integrating the Korean Meteorological Agency data and natural gas leakage data for considering complex factors. The paper has been divided into three modules. First, we filled missing data based on the linear interpolation method on the integrated data set, and selected essential features using FA with OrdinalEncoder (OE)-based normalization. The dataset is labeled by K-means clustering. The final module uses four algorithms, K-nearest neighbors (KNN), decision tree (DT), random forest (RF), Naive Bayes (NB), to predict gas leakage levels. The proposed method is evaluated by the accuracy, area under the ROC curve (AUC), and mean standard error (MSE). The test results indicate that the OrdinalEncoder-Factor analysis (OE-F)-based classification method has improved successfully. Moreover, OE-F-based KNN (OE-F-KNN) showed the best performance by giving 95.20% accuracy, an AUC of 96.13%, and an MSE of 0.031.

• Journal of Biomedical Engineering Research
• /
• v.44 no.1
• /
• pp.25-32
• /
• 2023

Recognizing the size and location of prostate cancer is critical for prostate cancer diagnosis, treatment, and predicting prognosis. This paper proposes a model to classify the tumor region and normal tissue with cross-sectional visual images of prostatectomy tissue. We used specimen images of 44 prostate cancer patients who received prostatectomy at Gachon University Gil Hospital. A total of 289 prostate slice images consist of 200 slices including tumor region and 89 slices not including tumor region. Images were divided based on the presence or absence of tumor, and a total of 93 features from each slice image were extracted using Radiomics: 18 first order, 24 GLCM, 16 GLRLM, 16 GLSZM, 5 NGTDM, and 14 GLDM. We compared feature selection techniques such as LASSO, ANOVA, SFS, Ridge and RF, LR, SVM classifiers for the model's high performances. We evaluated the model's performance with AUC of the ROC curve. The results showed that the combination of feature selection techniques LASSO, Ridge, and classifier RF could be best with an AUC of 0.99±0.005.

• Investigative Magnetic Resonance Imaging
• /
• v.25 no.4
• /
• pp.323-331
• /
• 2021

Purpose: To investigate the diagnostic criteria of T1-weighted imaging (T1W) and time-of-flight (TOF) imaging for detecting intraplaque hemorrhage (IPH) of a vertebrobasilar artery (VBA) compared with simultaneous non-contrast angiography and intraplaque hemorrhage (SNAP) imaging. Materials and Methods: Eighty-seven patients with VBA atherosclerosis who underwent high resolution MR imaging for evaluation of VBA plaque were reviewed. The presence and location of VBA plaque and IPH on SNAP were determined. The signal intensity (SI) of the VBA plaque on T1W and TOF imaging was manually measured and the SI ratio against adjacent muscles was calculated. The receiver-operating characteristic (ROC) curve was used to compare the diagnostic accuracy for detecting VBA IPH. Results: Of 87 patients, 67 had IPH and 20 had no IPH on SNAP. The SI ratio between VBA IPH and temporalis muscle on T1W was significantly higher than that in the no-IPH group (235.9 ± 16.8 vs. 120.0 ± 5.1, P < 0.001). The SI ratio between IPH and temporalis muscle on TOF was also significantly higher than that in the no-IPH group (236.8 ± 13.3 vs. 112.8 ± 7.4, P < 0.001). Diagnostic efficacies of SI ratios on TOF and TIW were excellent (AUC: 0.976 on TOF and 0.964 on T1W; cutoff value: 136.7% for TOF imaging and 135.1% for T1W imaging). Conclusion: Compared with SNAP, cutoff levels of the SI ratio between VBA plaque and temporalis muscle on T1W and TOF imaging for detecting IPH were approximately 1.35 times.

• Journal of Veterinary Clinics
• /
• v.40 no.1
• /
• pp.31-37
• /
• 2023

Vertebral left atrial size (VLAS) is an important indicator to predict myxomatous mitral valve degeneration (MMVD) in dogs. When the caudal margin of cardiac silhouette and the dorsal margin of caudal vena cava (CdVC) could not be seen exactly, another way to evaluate VLAS is needed. The objective of this study was to assess whether a new modified VLAS (m-VLAS) could be used as an indicator to predict MMVD in 57 small breed dogs with MMVD. The m-VLAS was also used to classify American College of Veterinary Internal Medicine staging groups and left heart enlargement confirmed with echocardiograph (EchoLHE) groups. The m-VLAS was measured as the distance from the ventral aspect of the carina to the dorsal aspect of the intersection of the cardiac silhouette and the farthest LA caudal margin, not the CdVC, followed by drawing the same line beginning at the cranial edge of T4. Based on VLAS values and m-VLAS values measured for dogs with MMVD, correlations between these values and left heart enlargement groups were then evaluated. There were significant differences in both the VLAS and the m-VLAS between EchoLHE groups. The AUC of the ROC curve of the m-VLAS to detect EchoLHE was higher than that of the VLAS. The optimal cutoff value for the m-VLAS was >2.7, which had a higher specificity (86.84%) than the VLAS specificity (71.05%). This study reveals that a new m-VLAS is a more specific indicator than the VLAS for predicting left side heart enlargement in small breed dogs. Therefore, the m-VLAS can be used as a clinically useful radiographic measurement alternative to or better than the VLAS.

• Journal of Korean Critical Care Nursing
• /
• v.16 no.3
• /
• pp.24-33
• /
• 2023

Purpose : This study aims to establish an optimal cut-off score on the Braden scale for the assessment of pressure injury to detect pressure injury risks among inpatients in a South Korean tertiary hospital. Methods : This retrospective study used electronic medical records, from January to December 2022. A total of 654 patients were included in the study. Of these, 218 inpatients with pressure injuries and 436 without pressure injuries were classified and analyzed using 1:2 Propensity Score Matching (PSM), and the generalized estimating equation was performed using SPSS Version 26 and the R Machlt package program. Results : The cut-off value on the Braden scale for distinguishing pressure injury was 17 points, and the AUC (area under the ROC curve) was 0.531 (0.484-0.579). The sensitivity was 56.6% (45.5-67.7%) and the specificity was 69.7% (66.0-73.4%). With 17 points, the Braden scale cut-off distinguished those who had pressure injuries from those who did not at the time of admission (p < .03). In the pressure injury group, the Braden score on the day of the pressure injury was 14, with significant results in all subcategories except the moisture category. Conclusion : Our findings revealed that a cut-off value of 17 was optimal for predicting the risk of pressure injuries among tertiary hospital inpatients. Future studies should evaluate the optimal cut-off values in different clinical environments. Additionally, it is necessary to conduct multicenter large sample studies to verify the effectiveness of a 17 value in PI risk assessments.

• Atmosphere
• /
• v.33 no.5
• /
• pp.531-548
• /
• 2023

We developed the Aviation Convective Index (ACI) for predicting deep convective area using the operational global Numerical Weather Prediction model of the Korea Meteorological Administration. Seasonally optimized ACI (ACI_SnOpt) was developed to consider seasonal variabilities on deep convections in Korea. Yearly optimized ACI (ACI_YrOpt) in Part 1 showed that seasonally averaged values of Area Under the ROC Curve (AUC) and True Skill Statistics (TSS) were decreased by 0.420% and 5.797%, respectively, due to the significant degradation in winter season. In Part 2, we developed new membership function (MF) and weight combination of input variables in the ACI algorithm, which were optimized in each season. Finally, the seasonally optimized ACI (ACI_SnOpt) showed better performance skills with the significant improvements in AUC and TSS by 0.983% and 25.641% respectively, compared with those from the ACI_YrOpt. To confirm the improvements in new algorithm, we also conducted two case studies in winter and spring with observed Convectively-Induced Turbulence (CIT) events from the aircraft data. In these cases, the ACI_SnOpt predicted a better spatial distribution and intensity of deep convection. Enhancements in the forecast fields from the ACI_YrOpt to ACI_SnOpt in the selected cases explained well the changes in overall performance skills of the probability of detection for both "yes" and "no" occurrences of deep convection during 1-yr period of the data. These results imply that the ACI forecast should be optimized seasonally to take into account the variabilities in the background conditions for deep convections in Korea.