• Computers and Concrete
• /
• v.32 no.3
• /
• pp.327-337
• /
• 2023

Concrete is the most widely used building material, with various types including high- and ultra-high-strength, reinforced, normal, and lightweight concretes. However, accurately predicting concrete properties is challenging due to the geotechnical design code's requirement for specific characteristics. To overcome this issue, researchers have turned to new technologies like machine learning to develop proper methodologies for concrete specification. In this study, we propose a highly accurate deep learning-based predictive model to investigate the compressive strength (UCS) of lightweight concrete with natural aggregates (pumice). Our model was implemented on a database containing 249 experimental records and revealed that water, cement, water-cement ratio, fine-coarse aggregate, aggregate substitution rate, fine aggregate replacement, and superplasticizer are the most influential covariates on UCS. To validate our model, we trained and tested it on random subsets of the database, and its performance was evaluated using a confusion matrix and receiver operating characteristic (ROC) overall accuracy. The proposed model was compared with widely known machine learning methods such as MLP, SVM, and DT classifiers to assess its capability. In addition, the model was tested on 25 laboratory UCS tests to evaluate its predictability. Our findings showed that the proposed model achieved the highest accuracy (accuracy=0.97, precision=0.97) and the lowest error rate with a high learning rate (R2=0.914), as confirmed by ROC (AUC=0.971), which is higher than other classifiers. Therefore, the proposed method demonstrates a high level of performance and capability for UCS predictions.

• Korean Journal of Radiology
• /
• v.23 no.8
• /
• pp.821-827
• /
• 2022

Objective: To determine the anatomical characteristics of the petrous ridge and trigeminal nerve in trigeminal neuralgia (TN) without neurovascular compression (NVC). Materials and Methods: From May 2017 to March 2021, 66 patients (49 female and 17 male; mean age ± standard deviation [SD], 56.8 ± 13.3 years) with TN without NVC and 57 controls (46 female and 11 male; 52.0 ± 15.6 years) were enrolled. The angle of the petrous ridge (APR) and angle of the trigeminal nerve (ATN) were measured using magnetic resonance imaging with a high-resolution three-dimensional T2 sequence. Data on the symptomatic side were compared with those on the asymptomatic side in patients and with the mean measurements of the bilateral sides in controls. Receiver operating characteristic (ROC) analysis was conducted to evaluate the performance of APR and ATN in distinguishing TN patients from controls. Results: In TN patients without NVC, the mean ± standard deviation (SD) of APR on the symptomatic side (98.40° ± 19.75°) was significantly smaller than that of the asymptomatic side (105.59° ± 22.45°, p = 0.019) and controls (108.44° ± 15.98°, p = 0.003). The mean ATN ± SD on the symptomatic side (144.41° ± 8.92°) was significantly smaller than that of the asymptomatic side (149.67° ± 8.09°, p = 0.003) and controls (150.45° ± 8.48°, p = 0.001). The area under the ROC curve for distinguishing TN patients from controls was 0.673 (95% confidence interval [CI]: 0.579-0.758) for APR and 0.700 (CI: 0.607-0.782) for ATN. The sensitivity and specificity using the diagnostic cutoff yielding the highest Youden index were 81.8% (54/66) and 49.1% (28/57), respectively, for APR (with a cutoff score of 94.30°) and 65.2% (43/66) and 66.7% (38/57), respectively, for ATN (cutoff score, 148.25°). Conclusion: In patients with TN without NVC, APR and ATN were smaller than those in controls, which may explain the potential cause of TN and provide additional information for diagnosis.

• Korean Journal of Radiology
• /
• v.22 no.4
• /
• pp.612-623
• /
• 2021

Objective: To evaluate the diagnostic performance of a deep learning algorithm for the automated detection of developmental dysplasia of the hip (DDH) on anteroposterior (AP) radiographs. Materials and Methods: Of 2601 hip AP radiographs, 5076 cropped unilateral hip joint images were used to construct a dataset that was further divided into training (80%), validation (10%), or test sets (10%). Three radiologists were asked to label the hip images as normal or DDH. To investigate the diagnostic performance of the deep learning algorithm, we calculated the receiver operating characteristics (ROC), precision-recall curve (PRC) plots, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) and compared them with the performance of radiologists with different levels of experience. Results: The area under the ROC plot generated by the deep learning algorithm and radiologists was 0.988 and 0.988-0.919, respectively. The area under the PRC plot generated by the deep learning algorithm and radiologists was 0.973 and 0.618-0.958, respectively. The sensitivity, specificity, PPV, and NPV of the proposed deep learning algorithm were 98.0, 98.1, 84.5, and 99.8%, respectively. There was no significant difference in the diagnosis of DDH by the algorithm and the radiologist with experience in pediatric radiology (p = 0.180). However, the proposed model showed higher sensitivity, specificity, and PPV, compared to the radiologist without experience in pediatric radiology (p < 0.001). Conclusion: The proposed deep learning algorithm provided an accurate diagnosis of DDH on hip radiographs, which was comparable to the diagnosis by an experienced radiologist.

• Journal of Korea Multimedia Society
• /
• v.10 no.12
• /
• pp.1559-1565
• /
• 2007

The many target-detection methods that use forward-looking infrared (FUR) images can deal with large targets measuring $70{\times}40$ pixels, utilizing their shape features. However, detection small targets is difficult because they are more obscure and there are many target-like objects. Therefore, few studies have examined how to detect small targets consisting of fewer than $30{\times}10$ pixels. This paper presents a small target detection method using clutter rejection with stochastic hypothesis testing for FLIR imagery. The proposed algorithm consists of two stages; detection and clutter rejection. In the detection stage, the mean of the input FLIR image is first removed and then the image is segmented using Otsu's method. A closing operation is also applied during the detection stage in order to merge any single targets detected separately. Then, the residual of the clutters is eliminated using statistical hypothesis testing based on the t-test. Several FLIR images are used to prove the performance of the proposed algorithm. The experimental results show that the proposed algorithm accurately detects small targets (Jess than $30{\times}10$ pixels) with a low false alarm rate compared to the center-surround difference method using the receiver operating characteristics (ROC) curve.

• The Journal of Asian Finance, Economics and Business
• /
• v.7 no.7
• /
• pp.85-93
• /
• 2020

In peer-to-peer (P2P) loan markets, as most lenders are unskilled and inexperienced ordinary individuals, it is important to know the characteristics of borrowers that significantly impact their repayment performance. This study investigates the effects and importance of borrowers' past repayment performance track record within the platform to identify its predictive power. To this end, I analyze the detailed loan repayment data from two leading P2P lending platforms in Korea using a Cox proportional hazard, multiple linear regression, and logit models. Furthermore, the predictive power of the factors proxied by borrowers' track records are evaluated through the receiver operating characteristic (ROC) curves. As a result, it is found that the borrowers' past track record within the platform have the most important impact on the repayment performance of their current loans. In addition, this study also reveals that the borrowers' track record is much more predictive of their repayment performance than any other factor. The findings of this study emphasize that individual lenders must take into account the quality of borrowers' past transaction history when making a funding decision, and that platform operators should actively share the borrowers' past records within the markets with lenders.

• Journal of muscle and joint health
• /
• v.21 no.1
• /
• pp.75-84
• /
• 2014

Purpose: The purpose of this study was to test the validity and reliability of the Korean version of the Geriatric Anxiety Inventory (K-GAI). Methods: Two hundreds and thirty six elderly were participated to test K-GAI. Goldberg's short screening scale for anxiety was tested for criterion validity. Receiver operating characteristics (ROC) analysis was used for measuring sensitivity and specificity. Results: The obtained internal consistency was 0.88. There were significant associations between test and retest results. K-GAI scores was significantly associated with Goldberg's short screening scale for anxiety (r=.694, p<.001). We found that a score of seven and greater was optimal for a criterion of anxiety among elderly Koreans. At this cut point, sensitivity was 78.9% and specificity was 73.1%. Conclusion: The K-GAI displayed good psychometric properties. This tool would be useful for early detection of anxiety among elderly Koreans with various situations including cognitive disorder, low education, or physical disability.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2016.05a
• /
• pp.90-90
• /
• 2016

본 연구에서는 증발산, 토양수분, 태양복사에너지, 식생 활동 등과 같은 수문기상인자들을 활용하여 새로운 가뭄 지수(Energy-based Water Deficit Index(EWDI)를 개발하였고 이는 Moderate Resolution Imaging Spectroradiometer(MODIS)에서 제공되는 산출물들을 활용하였다. EWDI는 물의 순환과 탄소 순환을 동시에 고려하여 기존에 활용되는 다른 가뭄지수들보다 다양한 측면에서 가뭄을 분석할 수 있는 장점을 가지고 있으며 산정된 EWDI는 증발산 기반의 가뭄지수인 Stand-alone MODIS-based Evaporative Stress Index(stMOD_ESI)와 함께 시공간적인 변동성을 비교하여 전 세계적으로 가뭄 피해가 심각한 지역인 몽골, 호주, 한반도 지역에 대해 2000년에서 2010년까지 적용성을 파악하였다. 또한, 본 연구에서는 각 지수들 간의 상관관계를 파악하고 수문기상 인자들과 가뭄 현상 사이에 관계성을 파악하기 위해 Receiver Operating Characteristics(ROC) 분석을 수행하였다. 위에서 언급한 여러 분석 결과를 토대로, EWDI와 stMOD_ESI는 기존에 많이 쓰였던 가뭄 지수인 표준강수지수(Standardized Precipitation Index, SPI)에 비해 가뭄 상태를 더욱 잘 파악할 수 있는 것으로 나타났으며 EWDI와 stMOD_ESI가 광역적인 범위에서의 적용성이 높음을 파악하였다. 본 연구를 통해 수문기상학 및 수자원 분야에서의 인공위성을 활용한 가뭄 분석 연구의 기틀이 마련되길 기대해 볼 수 있다.

• Journal of Veterinary Science
• /
• v.19 no.6
• /
• pp.840-845
• /
• 2018

The purpose of this study was to propose a standard for differentiation between normal dogs and patients with hyperadrenocorticism (HAC) by measuring skin thickness via ultrasonography in small breed dogs. Significant changes in skin thickness of patients treated with prednisolone (PDS) or patients with HAC treated with trilostane were evaluated. Skin thickness was retrospectively measured on three abdominal digital images obtained from small breed dogs weighing < 15 kg that underwent abdominal ultrasonography. Mean skin thickness of normal dogs was $1.03{\pm}0.25mm$ (mean ${\pm}$ SD). Both the HAC and PDS groups showed significantly thinner skin than that in the normal group. Seven of the 10 HAC patients treated with trilostane had increased skin thickness. The area under the curve value of 0.807 was based on the receiver operating characteristics (ROC) curve for differentiating normal dogs from HAC patients. Sensitivity was 76% and specificity was 73% when skin thickness was less than the 0.83 mm cutoff value. In conclusion, measurement of skin thickness in small breed dogs by using ultrasonography is likely to provide clinical information useful in differentiating HAC patients from normal dogs. However, exposure to PDS, trilostane, and other conditions may have a significant effect on skin thickness.

• Economic and Environmental Geology
• /
• v.56 no.1
• /
• pp.65-73
• /
• 2023

As one of the most vulnerable countries to floods, there should be an increased necessity for accurate and reliable flood forecasting in Indonesia. Therefore, a new prediction model using a machine learning algorithm is proposed to provide daily flood prediction in Indonesia. Data crawling was conducted to obtain daily rainfall, streamflow, land cover, and flood data from 2008 to 2021. The model was built using a Random Forest (RF) algorithm for classification to predict future floods by inputting three days of rainfall rate, forest ratio, and stream flow. The accuracy, specificity, precision, recall, and F1-score on the test dataset using the RF algorithm are approximately 94.93%, 68.24%, 94.34%, 99.97%, and 97.08%, respectively. Moreover, the AUC (Area Under the Curve) of the ROC (Receiver Operating Characteristics) curve results in 71%. The objective of this research is providing a model that predicts flood events accurately in Indonesian regions 3 months prior the day of flood. As a trial, we used the month of June 2022 and the model predicted the flood events accurately. The result of prediction is then published to the website as a warning system as a form of flood mitigation.

• Journal of Korea Water Resources Association
• /
• v.53 no.4
• /
• pp.293-301
• /
• 2020

This study determined appropriate threshold level (cumulative period and percentage) of precipitation for drought management in dam basin. The 5 dam basins were selected, the daily dam storage level and daily precipitation data were collected. MAP (Mean Areal Precipitation was calculated by using Thiessen polygon method, and MAP were converted to accumulated values for 6 cumulative periods (30-, 60-, 90-, 180-, 270-, and 360-day). The correlation coefficient and ratio of variation coefficient between storage level and MAP for 6 cumulative periods were used to determine the appropriate cumulative period. Correlation of cumulative precipitation below 90-day was low, and that of 270-day was high. Correlation was high when the past precipitation during the flood period was included within the cumulative period. The ratio of variation coefficient was higher for the shorter cumulative period and lower for the longer in all dam, and that of 270-day precipitation was closed to 1.0 in every month. ROC (Receiver Operating Characteristics) analysis with TLWSA (Threshold Line of Water Supply Adjustment) was used to determine the percentage of precipitation shortages. It is showed that the percentage of 270-day cumulative precipitation on Boryung dam and other 4-dam were less than 90% and 80% as threshold level respectively, when the storage was below the attention level. The relationship between storage and percentage of dam outflow and precipitation were analyzed to evaluate the impact of artificial dam operations on drought analysis, and the magnitude of dam outflow caused uncertainty in the analysis between precipitation and storage data. It is concluded that threshold level should be considered for dam drought analysis using based on precipitation.