• Journal of Korean Society of Forest Science
• /
• v.110 no.4
• /
• pp.610-621
• /
• 2021

The soil creep, primarily caused by earthquakes and torrential rainfall events, has widely occurred across the country. The Korea Forest Service attempted to quantify the soil creep susceptible areas using a discriminant value table to prevent or mitigate casualties and/or property damages in advance. With the advent of advanced computer technologies, machine learning-based classification models have been employed for managing mountainous disasters, such as landslides and debris flows. This study aims to quantify the soil creep susceptibility using several classifiers, namely the k-Nearest Neighbor (k-NN), Naive Bayes (NB), Random Forest (RF), and Support Vector Machine (SVM) models. To develop the classification models, we downscaled 292 data from 4,618 field survey data. About 70% of the selected data were used for training, with the remaining 30% used for model testing. The developed models have the classification accuracy of 0.727 for k-NN, 0.750 for NB, 0.807 for RF, and 0.750 for SVM against test datasets representing 30% of the total data. Furthermore, we estimated Cohen's Kappa index as 0.534, 0.580, 0.673, and 0.585, with AUC values of 0.872, 0.912, 0.943, and 0.834, respectively. The machine learning-based classifications for soil creep susceptibility were RF, NB, SVM, and k-NN in that order. Our findings indicate that the machine learning classifiers can provide valuable information in establishing and implementing natural disaster management plans in mountainous areas.

• Sim-seong Yeon-gu
• /
• v.33 no.1
• /
• pp.61-80
• /
• 2018

We investigated whether the 81 items of the Gray-Wheelwright test correctly measure the concept of Jung's typology and aimed to refine the test. Participants (n=431) completed the Gray-Wheelwright test, and the results were analyzed using factor analysis with the varimax rotation and the maximum likelihood extraction method. A pair of opposing attitudes, introversion/extroversion, or one of the two pairs of opposing functional types, thinking/feeling or intuition/sensation, was labeled to the extracted factor according to the majority type of the items in the factor. The minority items or items not included in any factors were excluded from making a short form of the Gray-Wheelwright test with 45 items. We used intraclass correlation (ICC) coefficient and Cronbach's alpha for the test-retest reliability and internal consistency of the test, respectively. The newly developed short form of the Gray-Wheelwright test measured the Jung's personality types well, which was comparable to the original one while reducing time and effort required for the testing.

• Korean Journal of Clinical Laboratory Science
• /
• v.48 no.2
• /
• pp.54-61
• /
• 2016

Studies on assessment tools for predicting cardiovascular disease risk (CDR), along with the studies to prevent CDR have been consistently reported. The validity of the Framingham risk score (FRS), a commonly known tool, has been verified through the precedent studies. In this study, we examined the differences of FRS according to the application of categories of LDL cholesterol (LDL-C) or Total cholesterol (TC), and attempted to evaluate the agreement of 10 yr CHD risk judgment based on the above-mentioned application. Excluding those diagnosed as cardiovascular diseases, data on subjects (755 men and 775 women) from the 2011 Korean National Health and Nutrition Examination Survey were used. We found differences of FRS and 10 yr CHD risk depending on the application of categories of LDL cholesterol (LDL-C) or Total cholesterol (TC). FRS of TC points were higher than those of LDL-C in both men and women. In classification of low risk (<10%), intermediate risk (10~19%), and high risk (${\geq}20%$), there were disagreements for 106 men and 26 women. Women showed almost perfect agreement from Coefficient of Cohen's Kappa (0.718 in men, and 0.884 in women). In assessment of 10 yr CHD risk, R-squared value from regression including TC was higher than that of LDC-C in both men and women (0.972 vs 0.885). From this result, we can draw a conclusion that correlation coefficients of FRS and CHD risk including TC were higher than those of LDC-C, and women showed a greater degree of agreement than men.

• Journal of Nutrition and Health
• /
• v.52 no.4
• /
• pp.332-341
• /
• 2019

Purpose: Poor nutrition in hospitalized patients is closely linked to an increased risk of infection, which can result in complications affecting mortality, as well as increased length of hospital stay and hospital costs. Therefore, adequate nutritional support is essential to manage the nutritional risk status of patients. Nutritional support needs to be preceded by nutrition screening, in which accuracy is crucial, particularly for the initial screening. To perform initial nutrition screening of hospitalized patients, we used the Catholic Kwandong University (CKU) Nutritional Risk Screening (CKUNRS) tool, originally developed at CKU Hospital. To validate CKUNRS against the Patient-Generated Subjective Global Assessment (PG-SGA) tool, which is considered the gold standard for nutritional risk screening, results from both tools were compared. Methods: Nutritional status was evaluated in 686 adult patients admitted to CKU Hospital from May 1 to July 31, 2018 using both CKUNRS and PG-SGA. Collected data were analyzed, and the results compared, to validate CKUNRS as a nutrition screening tool. Results: The comparison of CKUNRS and PG-SGA revealed that the prevalence of nutritional risk on admission was 15.6% (n = 107) with CKUNRS and 44.6% (n = 306) with PG-SGA. The sensitivity and specificity of CKUNRS to evaluate nutritional risk status were 98.7% (96.8 ~ 99.5) and 33.3% (28.1 ~ 39.0), respectively. Thus, the sensitivity was higher, but the specificity lower compared with PG-SGA. Cohen's kappa coefficient was 0.34, indicating valid agreement between the two tools. Conclusion: This study found concordance between CKUNRS and PG-SGA. However, the prevalence of nutritional risk in hospitalized patients was higher when determined by CKUNRS, compared with that by PG-SGA. Accordingly, CKUNRS needs further modification and improvement in terms of screening criteria to promote more effective nutritional support for patients who have been admitted for inpatient care.

• Journal of Digital Convergence
• /
• v.15 no.3
• /
• pp.77-86
• /
• 2017

This study selected car-related mobile apps for app developers suffering from low revenue and classified car apps assisting users in driving or managing a car. A total of 697 car apps were classified into eight categories. Most apps are in four categories: car news & information (28%), locating service (23%), car rental service (15%), safe/efficient driving service (12%). The remaining categories are buying & selling, driver's communication, maintenance management, and expenses monitoring. Many apps are simple and too similar in their main functions. Only a few apps are designed to be more comprehensive and have functions in two or more categories. For the practicality of the categorization scheme, this study checked the inter-rater reliability in two tests and got 0.886 and 0.828. The result from this study suggests functions that are not implemented yet or need to be combined. Future research will focus on identifying promising car apps or designing multi-functional car apps.

• Korean Journal of Radiology
• /
• v.25 no.2
• /
• pp.146-156
• /
• 2024

Objective: Automated breast ultrasound (ABUS) is a relevant imaging technique for early breast cancer diagnosis and is increasingly being used as a supplementary tool for mammography. This study compared the performance of ABUS and handheld ultrasound (HHUS) in detecting and characterizing the axillary lymph nodes (LNs) in patients with breast cancer. Materials and Methods: We retrospectively reviewed the medical records of women with recently diagnosed early breast cancer (≤ T2) who underwent both ABUS and HHUS examinations for axilla (September 2017-May 2018). ABUS and HHUS findings were compared using pathological outcomes as reference standards. Diagnostic performance in predicting any axillary LN metastasis and heavy nodal-burden metastases (i.e., ≥ 3 LNs) was evaluated. The ABUS-HHUS agreement for visibility and US findings was calculated. Results: The study included 377 women (53.1 ± 11.1 years). Among 385 breast cancers in 377 patients, 101 had axillary LN metastases and 30 had heavy nodal burden metastases. ABUS identified benign-looking or suspicious axillary LNs (average, 1.4 ± 0.8) in 246 axillae (63.9%, 246/385). According to the per-breast analysis, the sensitivity, specificity, positive and negative predictive values, and accuracy of ABUS in predicting axillary LN metastases were 43.6% (44/101), 95.1% (270/284), 75.9% (44/58), 82.6% (270/327), and 81.6% (314/385), respectively. The corresponding results for HHUS were 41.6% (42/101), 95.1% (270/284), 75.0% (42/56), 82.1% (270/329), and 81.0% (312/385), respectively, which were not significantly different from those of ABUS (P ≥ 0.53). The performance results for heavy nodal-burden metastases were 70.0% (21/30), 89.6% (318/355), 36.2% (21/58), 97.3% (318/327), and 88.1% (339/385), respectively, for ABUS and 66.7% (20/30), 89.9% (319/355), 35.7% (20/56), 97.0% (319/329), and 88.1% (339/385), respectively, for HHUS, also not showing significant difference (P ≥ 0.57). The ABUS-HHUS agreement was 95.9% (236/246; Cohen's kappa = 0.883). Conclusion: Although ABUS showed limited sensitivity in diagnosing axillary LN metastasis in early breast cancer, it was still useful as the performance was comparable to that of HHUS.

• Korean Journal of Clinical Laboratory Science
• /
• v.49 no.3
• /
• pp.239-247
• /
• 2017

Among the commonly known tools to diagnose diabetes are fasting plasma glucose (FPG), HbA1c., and OGTT known as gold standard. However, there can be many disagreements on the ways to diagnose diabetes. In this study, we examined the differences of the types of diabetes according to the applicability of FPG and HbA1c. Moreover, we evaluated the concordance of diagnosis. We excluded subjects with missing glucose and HbA1c data, as well as those previously diagnosed with diabetes, and those who fasted less than 8 hours. The data of 4,502 subjects (1,956 men and 2,546 women) from the 2015 KNHNES were analyzed. We divided these patients into three categories which are normal, prediabetes, and diabetes, based on the FPG and HbA1c. In men, the number of subjects with FPG ranging from 100 to 125 mg/dL and HbA1c ${\geq}6.5%$ was 23 out of 664, and the number of subjects with FPG < 126 mg/dL and HbA1c ${\geq}6.5%$ was 39 out of 86 newly diagnosed diabetes patients. The concordance rate was as follows: Normal 80.3%, prediabetes 44.9%, and diabetes 54.7%. The coefficient of Cohen's Kappa was 0.322 in men and 0.362 in women; this suggests that both gender showed a low concordance rate. However, when we divided them into two categories (nondiabetes and diabetes), Kappa was 0.582 in men and 0.637 in women, showing a relatively high concordance rate. While all subjects with FPG ${\geq}126mg/dL$ showed a significantly high HOMA IR, all subjects with FPG < 126 mg/dL showed a significantly high QUICKI. Considering the low concordance rate for the diagnosis of diabetes and characteristic of diagnostic tests, it is necessary to combine the related tests for diagnosing diabetes.

• The Journal of the Korea Contents Association
• /
• v.15 no.11
• /
• pp.313-321
• /
• 2015

To minimize CT examination for Hip FAI diagnosis and operation plan. also, whether the MRI 3D images can replace Hip Clock face image was evaluated when performing Hip FAI MRI by using additional 3D image. This study analyzed Hip MRI and 3D Hip CT images of 31 patients in this hospital. For the purpose of evaluating the images, one orthopedic surgeon and one radiology specialist reconstructed Clock face, at MR and CT modality, by superior 12 o'clock, labrum front 3 o'clock, and the other side 9 o'clock, centering on Hip joint articular transverse ligament 6 o'clock. Afterwards, by the Likert Scale 5 point scale (independent t-test p<0.005), this study evaluated the check-up of A. retinacular vessel, B. head neck junction at 11 o'clock, A. Epiphyseal line, B. Cam lesion at 12 o'clock, and Cam lesion, Posterior Cam lesion at 1,2,3 and 4 o'clock. As for the verification of reliability among observers, this study verified coincidence by Cohen's weighted Kappa verification. As a result of Likert scale for the purpose of qualitative evaluation about the image, 11 o'clock A. retinacular vessel MR average was $3.69{\pm}1.0$ and CT average was $2.8{\pm}0.78$. B. head neck juncton didn't have a difference between two observers (p <0.416). 12 o'clock A. Epiphyseal line MR average was $3.54{\pm}1.00$ and CT average was $4.5{\pm}0.62$(p<0.000). B. Cam lesion didn't have a difference between two observers (p <0.532). 1,2,3,4 Cam lesion and Posterior Cam lesion were not statistically significant (p <0.656, p <0.658). As a result of weighted Kappa verification, 11 o'clock A.retinacular vessel CT K value was 0.663 and the lowest conformity. As a result of coincidence evaluation on respective item, a very high result was drawn, and two observers showed high reliability.

• Korean Journal of Community Nutrition
• /
• v.26 no.4
• /
• pp.280-295
• /
• 2021

Objectives: This study aimed to examine the characteristics of patients according to their nutritional status as assessed by five nutritional screening tools: Patient-Generated Subjective Global Assessment (PG-SGA), NUTRISCORE, Nutritional Risk Index (NRI), Prognostic Nutritional Index (PNI), and Controlling Nutritional Status (CONUT) and to compare the agreement, sensitivity, and specificity of these tools. Methods: A total of 952 gastric cancer patients who underwent gastrectomy and chemotherapy from January 2009 to December 2012 at the Samsung Medical Center were included. We categorized patients into malnourished and normal according to the five nutritional screening tools 1 month after surgery and compared their characteristics. We also calculated the Spearman partial correlation, Cohen's Kappa coefficient, the area under the curve (AUC), sensitivity, and specificity of each pair of screening tools. Results: We observed 86.24% malnutrition based on the PG-SGA and 85.82% based on the NUTRISCORE among gastric cancer patients in our study. When we applied NRI or CONUT, however, the malnutrition levels were less than 30%. Patients with malnutrition as assessed by the PG-SGA, NUTRISCORE, or NRI had lower intakes of energy and protein compared to normal patients. When NRI, PNI, or CONUT were used to identify malnutrition, lower levels of albumin, hemoglobin, total lymphocyte count, total cholesterol, and longer postoperative hospital stays were observed among patients with malnutrition compared to those without malnutrition. We found relatively high agreement between PG-SGA and NUTRISCORE; sensitivity was 90.86% and AUC was 0.78. When we compared NRI and PNI, sensitivity was 99.64% and AUC was 0.97. AUC ranged from 0.50 to 0.67 for comparisons between CONUT and each of the other nutritional screening tools. Conclusions: Our study suggests that PG-SGA and NRI have a relatively high agreement with the NUTRISCORE and PNI, respectively. Further cohort studies are needed to examine whether the nutritional status assessed by PG-SGA, NUTRISCORE, NRI, PNI, and CONUT predicts the gastric cancer prognosis.

• Korean Journal of Community Nutrition
• /
• v.27 no.3
• /
• pp.205-222
• /
• 2022

Objectives: This study examined the characteristics of patients according to nutritional status assessed by five nutritional screening tools: Patient-Generated Subjective Global Assessment (PG-SGA), NUTRISCORE, Nutritional Risk Index (NRI), Prognostic Nutritional Index (PNI), and Controlling Nutritional Status (CONUT) and to compare the agreement, sensitivity, and specificity of these tools. Methods: A total of 952 gastric cancer patients who underwent gastrectomy and chemotherapy from January 2009 to December 2012 were included. The patients were categorized into malnutrition and normal status according to five nutritional screening tools one month after surgery. The Spearman partial correlation, Cohen's Kappa coefficient, the area under the curve (AUC), sensitivity, and specificity of each two screening tools were calculated. Results: Malnutrition was observed in 86.24% of patients based on the PG-SGA and 85.82% based on the NUTRISCORE. When NRI or CONUT were applied, the proportions of malnutrition were < 30%. Patients with malnutrition had lower intakes of energy and protein than normal patients when assessed using the PG-SGA, NUTRISCORE, or NRI. Lower levels of albumin, hemoglobin, total lymphocyte count, and total cholesterol and longer postoperative hospital stays were observed among patients with malnutrition compared to normal patients when NRI, PNI, or CONUT were applied. Relatively high agreement for NUTRISCORE relative to PG-SGA was found; the sensitivity was 90.86%, and the AUC was 0.78. When NRI, PNI, and CONUT were compared, the sensitivities were 23.72% for PNI relative to NRI, 44.53% for CONUT relative to NRI, and 90.91% for CONUT relative to PNI. The AUCs were 0.95 for NRI relative to PNI and 0.91 for CONUT relative to PNI. Conclusions: NUTRISCORE had a high sensitivity compared to PG-SGA, and CONUT had a high sensitivity compared to PNI. NRI had a high specificity compared to PNI. This relatively high sensitivity and specificity resulted in 77.00% agreement between PNI and CONUT and 77.94% agreement between NRI and PNI. Further cohort studies will be needed to determine if the nutritional status assessed by PG-SGA, NUTRISCORE, NRI, PNI, and CONUT predicts the gastric cancer prognosis.