• Proceedings of the Korea Information Processing Society Conference
• /
• 2024.05a
• /
• pp.621-622
• /
• 2024

본 논문은 Exemplar Transformer를 활용하여 단일 프로세서 환경에서 동작하는 시각적 객체 추적(Visual Object Tracking) 모델인 ETTrack(Exemplar Transformer Track)을 리눅스(우분투 20.04) 운영체제를 사용하는 임베디드 시스템 라떼판다 알파(LattePanda Alpha)에 탑재하였다. 데스크톱 프로세서인 Intel i7-8700와 라떼판다 알파의 Intel m3-8100y에서의 객체 추적 성능과 속도를 AUC(Area Under the ROC Curve)와 FPS를 통해서 비교평가하였다. 평가 결과 기존 트랜스포머 기반 추적 모델(TransT)과 유사한 성능, 3 FPS 빠른 추적 속도를 나타내는 것을 구현 증명하였다.

• Communications for Statistical Applications and Methods
• /
• v.31 no.4
• /
• pp.365-375
• /
• 2024

Diverse methods to evaluate the prediction model of a time to event have been proposed in the context of right censored data where all subjects are subject to be susceptible. A time-dependent AUC (area under curve) measures the predictive ability of a marker based on case group and control one which are varying over time. When a substantial portion of subjects are event-free, a population consists of a susceptible group and a cured one. An uncertain curability of censored subjects makes it difficult to define both case group and control one. In this paper, our goal is to propose a time-dependent AUC for a cure rate model when a censoring distribution is related with covariates. A class of inverse probability of censoring weighted (IPCW) AUC estimators is proposed to adjust the possible sampling bias. We evaluate the finite sample performance of the suggested methods with diverse simulation schemes and the application to the melanoma dataset is presented to compare with other methods.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.21 no.5
• /
• pp.842-849
• /
• 1997

The aim of this study is an investigation of the interaction of two micro hole defects affecting fatigue crack initation life and propagation behavior. The locatio of two micro hole defects was considered as an angle of alignment and the distance between the centers of two micro hole defects. The fatigue cracking behavior is experimented under bending. When micro defects are located close to each other, the fatigue crack initiation lives are varied with their relative locations. In the experiments, the area of local plastic strain strongly played a role in the fatigue crack initiation lives. Therefore we introduce a parameter which contains the plastic deformation area at stress concentrations and propose a fatigue crack initiation life prediction curve. In addition, the directions and propagation rates of fatigue cracks initiated at two micro hole defects are studied experimentally.

• Steel and Composite Structures
• /
• v.33 no.3
• /
• pp.375-388
• /
• 2019

This paper aims to investigate the axial load behavior and stability strength of square tubed steel-reinforced concrete (TSRC) columns. Unlike concrete filled steel tubular (CFST) column, the outer steel tube of a TSRC column is mainly used to provide confinement to the core concrete. Ten specimens were tested under axial compression, and the main test variables included length-to-width ratio (L/B) of the specimens, width-to-thickness ratio (B/t) of the steel tubes, and with or without stud shear connectors on the steel sections. The failure mode, ultimate strength and load-tube stress response of each specimen were summarized and analyzed. The test results indicated that the axial load carried by square tube due to friction and bond of the interface increased with the increase of L/B ratio, while the confinement effect of tube was just the opposite. Parametric studies were performed through ABAQUS based on the test results, and the feasibility of current design codes has also been examined. Finally, a method for calculating the ultimate strength of this composite column was proposed, in which the slenderness effect on the tube confinement was considered.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.175-175
• /
• 2023

Detection and monitoring of wildfires are essential for limiting their harmful effects on ecosystems, human lives, and property. In this research, we propose a novel method running in the Google Earth Engine platform for identifying and characterizing burnt regions using a hybrid of Sentinel-1 (C-band synthetic aperture radar) and Sentinel-2 (multispectral photography) images. The 2022 Uljin wildfire, the severest event in South Korean history, is the primary area of our investigation. Given its documented success in remote sensing and land cover categorization applications, we select the Random Forest (RF) method as our primary classifier. Next, we evaluate the performance of our model using multiple accuracy measures, including overall accuracy (OA), Kappa coefficient, and area under the curve (AUC). The proposed method shows the accuracy and resilience of wildfire identification compared to traditional methods that depend on survey data. These results have significant implications for the development of efficient and dependable wildfire monitoring systems and add to our knowledge of how machine learning and remote sensing-based approaches may be combined to improve environmental monitoring and management applications.

• Journal of the Korean Association of Oral and Maxillofacial Surgeons
• /
• v.49 no.3
• /
• pp.135-141
• /
• 2023

Objectives: This study aimed to develop and validate machine learning (ML) models using H₂O-AutoML, an automated ML program, for predicting medication-related osteonecrosis of the jaw (MRONJ) in patients with osteoporosis undergoing tooth extraction or implantation. Patients and Methods: We conducted a retrospective chart review of 340 patients who visited Dankook University Dental Hospital between January 2019 and June 2022 who met the following inclusion criteria: female, age ≥55 years, osteoporosis treated with antiresorptive therapy, and recent dental extraction or implantation. We considered medication administration and duration, demographics, and systemic factors (age and medical history). Local factors, such as surgical method, number of operated teeth, and operation area, were also included. Six algorithms were used to generate the MRONJ prediction model. Results: Gradient boosting demonstrated the best diagnostic accuracy, with an area under the receiver operating characteristic curve (AUC) of 0.8283. Validation with the test dataset yielded a stable AUC of 0.7526. Variable importance analysis identified duration of medication as the most important variable, followed by age, number of teeth operated, and operation site. Conclusion: ML models can help predict MRONJ occurrence in patients with osteoporosis undergoing tooth extraction or implantation based on questionnaire data acquired at the first visit.

• Korean Circulation Journal
• /
• v.52 no.5
• /
• pp.368-378
• /
• 2022

Background and Objectives: Recurrence rates after radiofrequency catheter ablation (RFCA) in atrial fibrillation (AF) patients are not low especially in non-paroxysmal AF. The diameter of left atrium (LA) has been widely used to predict the recurrence after RFCA for decades. However, LA diameter represents structural remodeling of LA and does not reflect electrical remodeling. We aimed to determine the predictive value of electrical remodeling of LA which is represented by the amount of low voltage zone (LVZ). Methods: We performed a retrospective cohort analysis of AF patients who underwent de novo RFCA in a single-center. Results: A total of 3,120 AF patients with de novo RFCA were analyzed. Among these patients, 537 patients underwent an electroanatomic mapping with bipolar voltage measurement of LA. The diameter of LA and flow velocity of LA appendage (LAA) differed significantly according to quartile group of LVZ area and percentage: patients with high LVZ had large LA diameter and low LAA flow velocity (p<0.001). Freedom from late recurrence (LR) was significantly lower in patients with high LVZ area and percentage (p<0.001). The diameter and surface area of LA had area under curve (AUC) of 0.592 and 0.593, respectively (p=0.002 for both). The predictive value of LVZ area (AUC, 0.676) and percentage (AUC, 0.671) were both superior compared with LA diameter (p=0.011 and 0.027 for each comparison). Conclusions: In conclusion, LVZ can predict freedom from LR after RFCA in AF patients. Predictive value was higher in parameters reflecting electrical rather than structural remodeling of LA.

• Composites Research
• /
• v.15 no.2
• /
• pp.1-10
• /
• 2002

The strength of glass/epoxy fabric joints under pin-loading is estimated based on the characteristics length method and experiment. To investigate the effect of finite element idealization for the contact between pin and laminate, three modeling cases are analyzed; assuming the cosine load distribution around the contact area, constraining the radial displacement at the hole boundary, and using the contact element. To study the effect of failure criteria, Tsai-Wu and Yamada-Sun methods are applied on the characteristic curve. The results of the nonlinear analysis using the contact element showed good agrements with experimental data in both laminates made of uni-directional prepreg tapes and fabrics. In terms of failure criteria, Tsai-Wu method showed better agreement with experimental results than the one by Yamada-Sun laminate.

• Biomolecules & Therapeutics
• /
• v.13 no.2
• /
• pp.90-94
• /
• 2005

Prazosin hydrochloride is an antihypertensive drug with selective ${\alpha}_1$-adrenoreceptor blocking effects. A simple high performance liquid chromatographic method has been developed and validated for the quantitative determination of prazosin in human plasma. A reversed-phase C18 column was used for the separation of prazosin and terazosin (internal standard) with a mobile phase composed of water, acetonitrile and triethylamine(75:25:0.1, V/V;pH5.0) at a flow rate of 1.5 ml/min. the fluorescence detector was set at excitation and emissionwavelengths of 250 and 370 nm, respectively. Intra-and inter-day precision and accuracy were acceptable for all quality control samples including the lower limit of quantification of 0.5 ng/ml. Good recovery (>80%) was seen in plasma. Prazosin was stable in human plasma under various storage conditions. This method was used successfully for a pharmacokinetic study in plasma after oral administration of a single 2-mg dose as prazosin base to 16 healthy volunteers. The maximum plasma concentration of prazosin was 23.1 ${\pm}$ 16.5 ng/ml at 2.1 h, and the mean area under the curve and elimination half-life were calculated to be 108.4 ${\pm}$ 74.2 ng ${\cdot}$hr/ml and 2.5 ${\pm}$ 0.6 h, respectively.

• Journal of Biomedical Engineering Research
• /
• v.31 no.3
• /
• pp.177-186
• /
• 2010

In 3D ultrasound color Doppler imaging (CDI), 8-16 pulse transmissions (ensembles) per each scanline are used for effective clutter rejection and flow estimation, but it yields a low volume acquisition rate. In this paper, we have evaluated three flow estimation methods: autoregression (AR), eigendecomposition (ED), and autocorrelation combined with adaptive clutter rejection (AC-ACR) for a small ensemble size (E=4). The performance of AR, ED and AC-ACR methods was compared using 2D and 3D in vivo data acquired under different clutter conditions (common carotid artery, kidney and liver). To evaluate the effectiveness of three methods, receiver operating characteristic (ROC) curves were generated. For 2D kidney in vivo data, the AC-ACR method outperforms the AR and ED methods in terms of the area under the ROC curve (AUC) (0.852 vs. 0.793 and 0.813, respectively). Similarly, the AC-ACR method shows higher AUC values for 2D liver in vivo data compared to the AR and ED methods (0.855 vs. 0.807 and 0.823, respectively). For the common carotid artery data, the AR provides higher AUC values, but it suffers from biased estimates. For 3D in vivo data acquired from a kidney transplant patient, the AC-ACR with E=4 provides an AUC value of 0.799. These in vivo experiment results indicate that the AC-ACR method can provide more robust flow estimates compared to the AR and ED methods with a small ensemble size.