• Journal of the korean academy of Pediatric Dentistry
• /
• 제49권2호
• /
• pp.131-139
• /
• 2022

This study aimed to evaluate the effectiveness of deep convolutional neural networks (CNNs) for diagnosis of interproximal caries in pediatric intraoral radiographs. A total of 500 intraoral radiographic images of first and second primary molars were used for the study. A CNN model (Resnet 50) was applied for the detection of proximal caries. The diagnostic accuracy, sensitivity, specificity, receiver operating characteristic (ROC) curve, and area under ROC curve (AUC) were calculated on the test dataset. The diagnostic accuracy was 0.84, sensitivity was 0.74, and specificity was 0.94. The trained CNN algorithm achieved AUC of 0.86. The diagnostic CNN model for pediatric intraoral radiographs showed good performance with high accuracy. Deep learning can assist dentists in diagnosis of proximal caries lesions in pediatric intraoral radiographs.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• 제26권6_spc호
• /
• pp.651-659
• /
• 2016

Condition monitoring (CM) encounters a large data problem due to sensors that measure vibration data with a continuous, and sometimes, high sampling rate. In this study, compressive sensing approaches for condition monitoring are proposed to demonstrate the efficiency in handling a large amount of data and to improve the damage detection capability of the current condition monitoring process. Compressive sensing is a novel sensing/sampling paradigm that takes much fewer samples compared to traditional sampling methods. For the experiments a built-in rotating system was used and all data were compressively sampled to obtain compressed data. Optimal signal features were then selected without the reconstruction process and were used to detect and classify damage. The experimental results show that the proposed method could improve the data processing speed and the accuracy of condition monitoring of rotating systems.

• Proceedings of the Korea Water Resources Association Conference
• /
• 한국수자원학회 2021년도 학술발표회
• /
• pp.77-77
• /
• 2021

가뭄은 사회·경제적으로 매우 큰 피해를 주는 자연재해이며, 그 시작과 발생 지역을 정확하게 예측하는 데 어려운 문제가 있다. 이에 수문 분야에서는 가뭄에 영향을 미치는 수문·기상인자들을 이용하여 다양한 가뭄지수를 개발하였고 이를 활용하여 가뭄 현상을 모니터링하고 예측 및 전망하는데 다양한 노력을 기울이고 있다. 하지만 가뭄지수들은 실제 가뭄이 어떠한 형태로 발생하는지 파악하기에 많은 한계점을 가지고 있다. 이에 최근 들어 미국과 유럽에서는 실제 농업, 환경, 에너지 등과 같은 다양한 분야에 걸쳐 가뭄 피해로 인해 생기는 가뭄 영향을 보다 체계적이고 상세한 데이터 인벤토리로 구축하고 가뭄지수와의 상관관계, 회귀분석과 같은 연구를 통해 가뭄 영향 예측을 시도하고 있다. 따라서 본 연구에서는 보고서, 데이터베이스, 웹 크롤링(Web-Crawling)을 통한 뉴스 기사 등과 같은 자료를 수집하여 국내 가뭄 영향 인벤토리를 구축하였다. 또한 수문 분야에 널리 사용되고 있는 가뭄지수인 표준 강수 증발산량지수 SPEI(Standardized Precipitation-Evapotranspiration Index)를 기반으로 지역에 따른 가뭄 영향을 예측하기 위해 최근 로지스틱 회귀모형, Random forest, Support vector machine, XGBoost 등의 다양한 머신러닝 기법을 적용하였다. 각 모형의 성능을 Receiver Operating Characteristic(ROC) 곡선을 통해 평가하여 가뭄 영향 예측에 적절한 머신러닝 기법을 제시하였다. 본 연구 결과를 통해 텍스트 기반의 가뭄 영향 자료와 머신러닝 기법을 통한 가뭄 영향 예측 방법론은 가뭄 재난 관리에 유용한 정보를 제공할 수 있다.

• Clinical and Experimental Pediatrics
• /
• 제48권12호
• /
• pp.1362-1369
• /
• 2005

Purpose : This study aimed to investigate the correlation between the plasma level of N-terminal pro-B-type natriuretic peptide(pro-BNP) and several known risk factors influencing outcomes after Fontan operations, and to assess whether pro-BNP levels can be used as predictive risk factors in Fontan operations. Methods : Plasma pro-BNP concentrations were measured in 35 patients with complex cardiac anomalies before catheterization. Cardiac catheterization was performed in all subjects. Mean right atrium pressure, mean pulmonary artery pressure(PAP), and ventricular end-diastolic pressure(EDP) were obtained. Cardiac output and pulmonary vascular resistance were calculated by Fick method. Results : Plasma pro-BNP levels exhibited statistically significant positive correlations with mean PAP(r=0.70, P<0.001), pulmonary vascular resistance(r=0.57, P<0.001), RVEDP(r=0.63, P<0.001), LVEDP(r=0.74, P<0.001), and cardiothoracic ratio(r=0.71, P<0.001). The area under the ROC curve using pro-BNP level to differentiate risk groups in Fontan operations was high : 0.868(95 percent CI, 0.712-1.023, P<0.01). The cutoff value of pro-BNP concentrations for the detection of risk groups in Fontan operations was determined to be 332.4 pg/mL(sensitivity 83.3 percent, specificity 82.7 percent). Conclusion : These data suggest that plasma pro-BNP levels may be used as a predictive risk factor in Fontan operations, and as a guide to determine the mode of therapy during follow-up after Fontan operations.

• The Korean Journal of Nuclear Medicine Technology
• /
• 제12권3호
• /
• pp.241-246
• /
• 2008

Purpose: To study of difference among primary aldosteronism patients and normal groups and essential hypertension patients and to confirm aldosterone/plasma renin activity ratio increase in secondary hypertension group which was diagnosed as primary aldosteronism. Materials and method: 1. Period: from April 2006 to March 2008. 2.Targets: 901 patients who visited seoul national university bundang hospital. 3. Groups: we divided by three groups. (normal group (n=147), essential hypertension (n=709), primary aldosteronism (n=45)) 4. Then calculated aldosterone/plasma renin activity ratio. 5. We used ROC curve to measure sensitivity and specificity. Results: 1. normal groups aldosterone/plasma renin activity ratio: $52.8{\pm}52.46$ essential hypertension patients aldosterone/plasma renin activity ratio: $171.04{\pm}291.56$ primary aldosteronism patients aldosterone/plasma renin activity ratio: $2325{\pm}2200$. 2. Aldosterone/renin ratio was significant in comparing each groups (p<0.001). 3. The sensitivity was 91.1% and the specificity was 92.4% when cut off of aldosterone/renin ratio was 485. Conclusion: It was confirmed that aldosterone/plasma renin activity ratio in primary aldosteronism was higher than normal group. According to this result, we can tell that aldosterone/ plasma renin activity ratio is very useful in diagnosis of primary aldosteronism.

• Nuclear Medicine and Molecular Imaging
• /
• 제41권3호
• /
• pp.201-208
• /
• 2007

Purpose: We investigated whether the diagnostic performance of SPECT scintimammography (SMM) can be improved by adding computer-aided diagnosis (CAD) of ultrasonography (US). Materials and methods: We reviewed breast SPECT SMM images and corresponding US images from 40 patients with breast masses (21 malignant and 19 benign tumors). The quantitative data of SPECT SMM were obtained as the uptake ratio of lesion to contralateral normal breast. The morphologic features of the breast lesions on US were extracted and quantitated using the automated CAD software program. The diagnostic performance of SPECT SMM and CAD of US alone was determined using receiver operating characteristic (ROC) curve analysis. The best discriminating parameter (D-value) combining SPECT SMM and the CAD of US was created. The sensitivity, specificity and accuracy of combined two diagnostic modalities were compared to those of a single one. Results: Both SPECT SMM and CAD of US showed a relatively good diagnostic performance (area under curve = 0.846 and 0.831, respectively). Combining the results of SPECT SMM and CAD of US resulted in improved diagnostic performance (area under curve =0.860), but there was no statistical differerence in sensitivity, specificity and accuracy between the combined method and a single modality. Conclusion: It seems that combining the results of SPECT SMM and CAD of breast US do not significantly improve the diagnostic performance for diagnosis of breast cancer, compared with that of SPECT SMM alone. However, SPECT SMM and CAD of US may complement each other in differential diagnosis of breast cancer.

• Clinical and Experimental Pediatrics
• /
• 제51권11호
• /
• pp.1158-1164
• /
• 2008

Purpose : To investigate the discriminative ability of pediatric index of mortality 2 (PIM2) and pediatric risk of mortality III (PRISM III) in predicting mortality in children admitted into the intensive care unit (ICU). Methods : We retrospectively analyzed variables of PIM2 and PRISM III based on medical records with children cared for in a single hospital ICU from January 2003 to December 2007. Exclusions were children who died within 2 h of admission into ICU or hopeless discharge. We used Students t test and ANOVA for general characteristics and for correlation between survivors and non-survivors for variables of PIM2 and PRISM III. In addition, we performed multiple logistic regression analysis for Hosmer-Lemeshow goodness-of-fit, receiver operating characteristic curve (ROC) for discrimination, and calculated standardized mortality ratio (SMR) for estimation of prediction. Results : We collected 193 medical records but analyzed 190 events because three children died within 2 h of ICU admission. The variables of PIM2 correlated with survival, except for the presence of post-procedure and low risk. In PRISM III, there was a significant correlation for cardiovascular/neurologic signs, arterial blood gas analysis but not for biochemical and hematologic data. Discriminatory performance by ROC showed an area under the curve 0.858 (95% confidence interval; 0.779-0.938) for PIM2, 0.798 (95% CI; 0.686-0.891) for PRISM III, respectively. Further, SMR was calculated approximately as 1 for the 2 systems, and multiple logistic regression analysis showed ${\chi}^2(13)=14.986$, P=0.308 for PIM2, ${\chi}^2(13)=12.899$, P=0.456 for PRISM III in Hosmer-Lemeshow goodness-of-fit. However, PIM2 was significant for PRISM III in the likelihood ratio test (${\chi}^2(4)=55.3$, P<0.01). Conclusion : We identified two acceptable scoring systems (PRISM III, PIM2) for the prediction of mortality in children admitted into the ICU. PIM2 was more accurate and had a better fit than PRISM III on the model tested.

• Korean Journal of Remote Sensing
• /
• 제33권6_2호
• /
• pp.1047-1060
• /
• 2017

In recent years, global warming has been continuing and abnormal weather phenomena are occurring frequently. Especially in the 21st century, the intensity and frequency of hydrological disasters are increasing due to the regional trend of water. Since the damage caused by disasters in urban areas is likely to be extreme, it is necessary to prepare a landslide susceptibility maps to predict and prepare the future damage. Therefore, in this study, we analyzed the landslide vulnerability using the logistic model and assessed the management plan after the landslide through the field survey. The landslide area was extracted from aerial photographs and interpretation of the field survey data at the time of the landslides by local government. Landslide-related factors were extracted topographical maps generated from aerial photographs and forest map. Logistic regression (LR) model has been used to identify areas where landslides are likely to occur in geographic information systems (GIS). A landslide susceptibility map was constructed by applying a LR model to a spatial database constructed through a total of 13 factors affecting landslides. The validation accuracy of 77.79% was derived by using the receiver operating characteristic (ROC) curve for the logistic model. In addition, a field investigation was performed to validate how landslides were managed after the landslide. The results of this study can provide a scientific basis for urban governments for policy recommendations on urban landslide management.

• Korean Journal of Psychosomatic Medicine
• /
• 제26권1호
• /
• pp.9-18
• /
• 2018

Objectives : The aims of this study were to investigate the psychopathology in adolescents with internalizing disorder using the self-report version of Strengths and Difficulties Questionnaire (SDQ-SR) and the Minnesota Multiphasic Personality Inventory for adolescents (MMPI-A), and to explore the complementarity between these two inventories for diagnostic assessment. Methods : Ninety-one patients aged 13-17 were divided into two groups by clinical diagnosis, 44 with internalizing disorder and 47 comparison group with other disorders. The data of SDQ-SR and MMPI-A completed by them were analyzed for the ability to predict internalizing disorder. Results : The logistic regression analysis revealed that diagnostic predictability increased by 2.27 times with every 1 point of SDQ-SR emotional symptom score increment. Comparison of ROC curves for internalizing disorders showed that the SE and SP of SDQ-SR emotional symptom with score over 4 was 88.94 and 78.72, respectively. For A-anx of MMPI-A with score over 56, SE and SP was 77.27 and 74.47, respectively. However, combination of these scales could not enhance the predictability of diagnostic classification more than that of SDQ-SR emotional symptom alone. Conclusions : Emotional symptom scale of SDQ-SR and A-anx, A-aln, A and INTR of MMPI-A should be important subscales for diagnosing the internalizing disorder of adolescents, however, which needs to be examined further with a larger sample size including normal control group.

• Journal of the Korean Association of Geographic Information Studies
• /
• 제22권4호
• /
• pp.1-11
• /
• 2019

With the development of remote sensing sensor technology, it has become possible to acquire satellite images with various spectral information. In particular, since the hyperspectral image is composed of continuous and narrow spectral wavelength, it can be effectively used in various fields such as land cover classification, target detection, and environment monitoring. Change detection techniques using remote sensing data are generally performed through differences of data with same dimensions. Therefore, it has a disadvantage that it is difficult to apply to heterogeneous sensors having different dimensions. In this study, we have developed a change detection method applicable to hyperspectral image and high spat ial resolution satellite image with different dimensions, and confirmed the applicability of the change detection method between heterogeneous images. For the application of the change detection method, the dimension of hyperspectral image was reduced by using correlation analysis and principal component analysis, and the change detection algorithm used CVA. The ROC curve and the AUC were calculated using the reference data for the evaluation of change detection performance. Experimental results show that the change detection performance is higher when using the image generated by adequate dimensionality reduction than the case using the original hyperspectral image.