Background: Automated classification systems using Artificial Intelligence (AI) and Machine Learning (ML) can enhance accuracy and efficiency in diagnosing pet skin diseases within veterinary medicine. Objectives: This study created a system that classifies pet skin diseases by evaluating multiple ML models to determine which method is most effective. Design: Comparative experimental study. Methods: Pet skin disease images were obtained from AIHub. Models, including Multi-Layer Perceptron (MLP), Boosted Stacking Ensemble (BSE), H2O AutoML, Random Forest, and Tree-based Pipeline Optimization Tool (TPOT), were trained and their accuracy assessed. Results: The TPOT achieved the highest accuracy (94.50 percent), due to automated pipeline optimization and ensemble learning. H2O AutoML also performed well at 94.25 percent, illustrating the effectiveness of automated selection for intricate imaging tasks. Other models scored lower. Conclusion: These findings highlight the potential of AI-driven solutions for faster and more precise pet skin disease diagnoses. Future investigations should incorporate broader disease varieties, multimodal data, and clinical validations to solidify the practicality of these approaches in veterinary medicine.

PURPOSES : For autonomous vehicles, abnormal situations, such as sudden changes in driving speed and sudden stops, may occur when they leave the operational design domain. This may adversely affect the overall traffic flow by affecting not only autonomous vehicles but also the driving environment of manual vehicles. Therefore, to minimize the traffic problems and adverse effects that may occur in mixed traffic situations involving manual and autonomous vehicles, an autonomous vehicle driving support system based on traffic operation optimization is required. The main purpose of this study was to build a big-data-classification system by specifying data classification to support the self-driving of Lv.4 autonomous vehicles and matching it with spatio-temporal data. METHODS : The research methodology is explained through a review of related literature, and a traffic management index and big-dataclassification system were built. After collecting and mapping the ITS history traffic information data of an actual Living Lab city, the data were classified using the traffic management indexing method. An AI-based model was used to automatically classify traffic management indices for real-time driving support of Lv.4 autonomous vehicles. RESULTS : By evaluating the AI-based model performance using the test data from the Living Lab city, it was confirmed that the data indexing accuracy was more than 98% for the KNN, Random Forest, LightGBM, and CatBoost algorithms, but not for Logistics Regression. The data were severely unbalanced, and it was necessary to classify very low probability nonconformities; therefore, precision is also important. All four algorithms showed similarly good performances in terms of accuracy. CONCLUSIONS : This paper presents a method for efficient data classification by developing a traffic management index to easily fuse and analyze traffic data collected from various institutions and big data collected from autonomous vehicles. Additionally, EdgeRSU is presented to support the driving of Lv.4 autonomous vehicles in mixed autonomous and manual vehicles traffic situations. Finally, a database was established by classifying data automatically indexed through AI-based models to quickly collect and use data in real-time in large quantities.

교통안전시설물의 관리는 도로교통의 안전과 직결되는 문제이다. 운전자는 신호등, 표지판, 노면표시 등을 통해 운전에 필요한 정보 를 얻는다. 노후된 표지판과 노면표시는 운전자에게 잘못된 정보를 제공할 수 있으므로 주기적인 시설물의 관리가 필요하다. 본 연구 는 딥 러닝 기술을 활용해 운전자 시각의 영상 자료에서 교통안전표지를 자동으로 탐지하고자 하며, 교통안전표지의 공통된 색상 특 징을 기반으로 클래스를 그룹으로 묶어 데이터셋을 구축하는 방법을 제안한다. 객체탐지의 성능지표로 널리 활용되는 mAP를 사용해 클래스 묶음 여부에 따른 탐지 성능을 비교한 결과, 색상 기반 클래스 묶음을 적용한 모델의 탐지 성능이 비교군에 비해 약 36% 상승 함을 확인하였다.

운전 시뮬레이션을 통해 3-수준 자율주행 중 차량 전방에 장애물이 출현하는 상황에서 서로 다른 연령대의 운전자 들이 보이는 제어권 전환 반응시간과 상황인식, 그리고 차량통제 수행에서의 차이를 장애물 회피 이전(before the obstacle avoidance: BOA)과 이후(after the obstacle avoidance: AOA) 구간으로 구분하여 분석하였다. 본 연구의 결 과를 요약하면 다음과 같다. 첫째, 실험참가자들의 상황인식은 AOA 구간에 비해 BOA 구간에서, 그리고 청년운전자 집단에 비해 고령운전자 집단에서 더 낮았는데, 이러한 경향은 AOA 구간에 비해 BOA 구간에서 더 뚜렷하였다. 둘째, 제어권 인수 시간은 청년운전자 집단에 비해 고령운전자 집단에서 유의하게 더 느렸다. 셋째, 네 가지 차량통 제 측정치 모두에서 BOA 구간보다는 AOA 구간에서, 그리고 청년운전자 집단보다는 고령운전자 집단에서 더 저하 된 수행이 관찰되었으나 차량통제 수행에서의 연령집단간 차이는 BOA 구간보다는 AOA 구간에서 더 컸다. 이러한 결과는 자율주행 중 제어권을 인수받아 수동으로 운전하여 장애물을 회피하는 상황에서 운전자의 상황인식과 차량 통제는 BOA 구간과 AOA 구간에 따라 달라질 수 있음을 시사한다.

This study evaluates the analytical performance of a newly developed miniaturized disposable U-tube for an automated blood viscometer and compares it to conventional viscometers. Whole blood viscosity (WBV), essential for circulatory function, exhibits non-Newtonian behavior, posing challenges for measurement at low shear rates. The blood viscometer, based on a scanning capillary tube method, used disposable U-tubes to measure viscosities across a shear rate range of 1s⁻¹ to 1,000s⁻¹. Precision evaluation showed stable coefficients of variation (CV) across different viscosity levels. Repeatability assessment indicated consistent CV values, demonstrating the reliability of the device. The agreement with the LV-III Brookfield viscometer and MCR 92 Rheometer was analyzed using Bland-Altman plots, which revealed minor systematic biases and consistent differences across the measurement range. Correlation analysis using Passing-Bablok regression showed high correlation coefficients (R > 0.96) with regression slopes close to 1. The newly developed miniaturized disposable U-tube exhibits excellent precision, reliable repeatability, and high correlation with established methods, enhancing laboratory productivity and offering potential for clinical applications. Further studies with human blood samples are recommended to confirm its clinical applicability.

PURPOSES : This study aims to calculate the estimation of travel time saving benefits from smart expressway construction by considering the willingness to pay for automated vehicles. METHODS : In this study, data were collected from 809 individual drivers through a stated preference survey. A multinomial logit model was constructed to analyze the choice behavior between arterial roads, expressways, and smart expressways. Through this, the values of time and benefits were estimated. RESULTS : The value of time was calculated at 19,379 won per vehicle per hour for arterial roads and expressways and 23,061 won per vehicle per hour for smart expressways. Applying these values to the Jungbu Naeryuk expressway, we evaluated the demand change and benefits resulting from the improvement to the smart expressways. The results show that the traffic volume on the Jungbu Naeryuk expressway is expected to increase by 4.7% to 20.7% depending on the changes in capacity. CONCLUSIONS : The travel time saving benefits are estimated as positive, resulting from the construction of smart expressways. The benefits resulting from the construction of new smart expressways are expected to be enhanced due to the anticipation of more significant time-saving effects.

Due to climate change and the rise in international transportation, there is an emerging potential for outbreaks of mosquito-borne diseases such as malaria, dengue, and chikungunya. Consequently, the rapid detection of vector mosquito species, including those in the Aedes, Anopheles, and Culex genera, is crucial for effective vector control. Currently, mosquito population monitoring is manually conducted by experts, consuming significant time and labor, especially during peak seasons where it can take at least seven days. To address this challenge, we introduce an automated mosquito monitoring system designed for wild environments. Our method is threefold: It includes an imaging trap device for the automatic collection of mosquito data, the training of deep-learning models for mosquito identification, and an integrated management system to oversee multiple trap devices situated in various locations. Using the well-known Faster-RCNN detector with a ResNet50 backbone, we’ve achieved mAP (@IoU=0.50) of up to 81.63% in detecting Aedes albopictus, Anopheles spp., and Culex pipiens. As we continue our research, our goal is to gather more data from diverse regions. This not only aims to improve our model’s ability to detect different species but also to enhance environmental monitoring capabilities by incorporating gas sensors.

구조물의 동적 해석 자동화는 구조 통합 시스템에서 중요한 역할을 한다. 해석 결과에 따른 신속한 대피 또는 경고 조치가 신속하게 이루어지도록 해석 모듈은 짧은 실시간에 해석 결과를 출력해야 한다. 구조 해석법으로 세계적으로 가장 많이 사용되는 방법은 유한 요소법이다. 유한요소법이 널리 사용되는 이유 중 하나는 사용의 편리다. 그러나 사용자가 유한요소망을 입력해야 하는데 요소망의 요소 수는 계상량과 정비례하고 요소망의 적절성은 에러와 연관된다. 본 연구는 시간 영역 동적 해석에서 전 단계 해석 결과를 사용하 여 계산된 대표 변형률 값으로 오차를 평가하고, 요소 세분화는 절점 이동인 r-법과 요소 분할인 h-법의 조합으로 효율적으로 계산하 는 적응적 요소망 형성 전략을 제시한다. 적용한 캔틸레버보와 간단한 프레임 예제를 통하여 적절한 요소망 형성, 정확성, 그리고 연 산 효율성을 검증하였다. 이 방법의 간단함이 지진 하중, 풍하중, 비선형 해석 등에 의한 복잡한 구조 동적 해석에도 효율적으로 사용 될 수 있는 것을 보여 준다.

본 연구에서는 운전 시뮬레이션을 사용하여 자율주행 환경을 구현한 후 3-수준 자율주행 조건에서 자율주행 차량 (automated vehicle: AV)으로부터 운전자에게 전달되는 제어권 인수 요구(takeover request: TOR) 정보의 양상(시각, 청각 및 시각+청각) 및 도로 형태(직선도로와 곡선도로)에 따라 운전자의 제어권 인수 시간(takeover time: TOT) 및 정신적 작업부하(제어권 인수 이후에 운전자들이 경험한 주관적 작업부하와 심장박동수에서의 변화)가 어떻게 차별 화되는지 분석하였다. 본 연구의 결과를 요약하면 다음과 같다. 먼저, AV로부터 TOR이 제시된 이후 실험참가자들 이 보인 TOT에 대한 분석 결과, TOR 정보양상의 측면에서는 시각 정보가 가장 빠른 TOT를 이끌어 낸 반면 청각 정보 조건에서 가장 느렸고, 도로 형태 측면에서는 직선도로 조건에 비해 곡선도로 조건에서의 TOT가 유의하게 더 느렸으며, 특히 청각 정보 조건에서 도로 형태에 따른 TOT에서의 차이가 가장 컸다. 둘째, 정신적 작업부하에 대한 분석 결과, TOR 정보가 시각 혹은 시각+청각적으로 제시된 조건에 비해 청각적으로 제시된 조건에서 주관적 작업부 하 측정치와 심장박동수 변화 크기 모두 전반적으로 더 낮았고 특히, 심장박동수 변화의 경우 이러한 경향은 곡선도 로 조건에서만 관찰되었다. 이러한 결과는 TOR 정보의 양상과 도로 형태에 따라 운전자의 TOT와 정신적 작업부하 수준이 달라질 수 있고, 특히 TOT가 빠를수록 정신적 작업부하 수준은 상대적으로 더 높아질 수 있음을 시사한다.

PURPOSES : This study is to initiated to estimate the impact of mixed traffic flow on expressway section according to the market penetration rate(MPR) of automated vehicles(AVs) using a enhanced intelligent driver model(EIDM). METHODS : To this end, microscopic traffic simulation and EIDM were used to implement mixed traffic flow on basic expressway section and simulation network was calibrated to understand the change of impact in mixed traffic flow due to the MPR of AVs. Additionally, MOEs of mobility aspects such as average speed and travel time were extracted and analyzed. RESULTS : The result of the impact of mobility MOEs by MPR and level of service indicated that 100% MPR of AVs normally affect positive impact on expressway at all level of service. However, it was analyzed that improvements in the level of service from LOS A to C are minimal until the MPR of AVs reaches 75% or higher. CONCLUSIONS : This research shows that impact of MPR of AVs using EIDM of mixed traffic flow on basic expressway. Increasing MPR of AVs affects positive impact on expressway at all level of services. However, MPR from 25% to 75% of AVs in LOS A to C shows minimal impacts. Therefore, to maximize the effectiveness of AVs, appropriate traffic operation and management strategies are necessary.

PURPOSES : This study is to develop an comprehensive validation methodology for autonomous mobility-on-demand system with level 4 automated driving system. METHODS : The proposed method includes the quantitative techniques for validating both automated driving system and center system using each optimal indicators. In addition, a novel method for validating the whole system applying multi-criteria decision methodology is suggested. RESULTS : The relative weights for the vehicle system was higher than the center systems. Moreover, the relative weights of failure rate for validating the vehicle system was the highest, in addition to, a relative weight for accuracy of dynamic routing algorithm within center system was the highest. CONCLUSIONS : The proposed methodology will be applicable to validate the autonomous mobility on demand system quantitatively considering the relative weights for each systems.

PURPOSES : This study aims to develop a congestion mitigation strategy at lane drop bottleneck with low Connected and Automated Vehicle (CAV) penetration. METHODS : The proposed strategy is designed to assign a role of a moving bottleneck to CAVs to reduce low-speed lane changes at bottleneck locations, which are the main cause of bottleneck capacity drop. Through this, it aims to induce proactive upstream lane changes for Human-Driven Vehicles (HDVs,). Therefore, this study includes the control algorithm for CAVs, and the evaluation of the strategy assumes penetration rates of 5% and 10% in a Microsimulation VISSIM environment. The assessment is conducted by comparing the capacity drop and total travel time. Additionally, a sensitivity test for the parameter of the CAV control algorithm, reduced speed, is performed to find the optimal parameter. RESULTS : In this study, three scenarios, a) Base, b) CAV with no control, and c) CAV with control, are designed to evaluate the effects of the CAV control strategy. Analysis of segment density and lane change distribution reveals that the control strategy effectively prevented vehicle congestion due to the bottleneck effect. Additionally, the analysis of capacity changes before and after the bottleneck and total travel time shows the effectiveness of the control strategy. The sensitivity test on CAV control speed emphasized the importance of selecting an appropriate speed for maintaining efficient traffic flow. Lastly, as the CAV penetration rate increased, the control strategy exhibited greater effectiveness in mitigating capacity drop. CONCLUSIONS : The proposed strategy is intended for use at low CAV penetration rates and is expected to provide assistance in mitigating congestion at bottlenecks during the early stages of CAV commercialization. Furthermore, since the role of CAV in the strategy can be performed by CVs or even HDVs, it can be applied not only immediately but also in the near future.

The radioactive cesium, released from the normal operation or the accidental operation of nuclear facilities, should be regularly monitored for environmental regulatory compliance. The 135Cs/137Cs isotopic ratios, potentially useful for long-term tracking Cs transport in seawater, can be used as a tool of understanding how radionuclides are transported from different nuclear production source terms and distributed in the ocean. The ultra-high sensitive mass spectrometers (TIMS, SF-ICP-MS and TQ-ICP-MS) have been used to measure the 135Cs/137Cs isotopic ratios. However, the radiochemical separation of Cs from the seawater matrix is essential for the analysis of Cs using the mass spectrometers. An automated radiochemical procedure for the separation of Cs in seawater was developed for the analysis of 135Cs/137Cs isotopic ratios using a sequential column chromatography with AMPPAN and AG50Wx8 cation exchange resins. National Instrument’s LabVIEW is a graphical programming language and a powerful tool for the instrument control. A virtual instrument system for the automated separation of cesium isotopes was developed by the state machine of the fundamental design patterns in LabVIEW. In this study, the conceptual designs of an automated separation system of cesium isotopes, its virtual instrument system based on the LabVIEW state machine architectures and an automated radiochemical procedure were described for the purification of cesium isotopes at trace levels found in seawater discharged from the various nuclear facilities.

As part of strengthening pyro safety measures, the Korea Atomic Energy Research Institute is developing LIBS (Laser-Induced Breakdown Spectroscopy) application technology to analyze molten salt components in electrolytic recovery device in real time. LIBS performs qualitative and quantitative analysis by analyzing the spectrum of energy emitted by atomizing and ionizing elements on the surface of a salt sample with a high-focused laser. Since salt easily corrodes metal, it must be managed in an environment with a dew point of -40°C or lower. In this study, we designed and manufactured a device that places a rod-type sampling stick on a mounting base, automatically moves it to the optimal measurement position for LIBS, and retrieves the sample. Its characteristics are as follows. First, LIBS is stationary and does not move. Second, the sample stick is placed on a mounting base and can rotate 360 degrees. Third, according to the command, the sample stick automatically moves to the optimal measurement position of LIBS with three degrees of freedom (X, Y, Z). Fourth, the salt attached to the sampling stick is recovered for chemical analysis by driving the gripper mounted at the bottom of the Z axis, Z axis, and rotation axis (R). The X, Y, and Z movement distances of this device are each 100 mm, rotation is 360 degrees, grip stroke is 50 mm, and position accuracy is ±20 m. Once the performance test of the automated salt sample analysis device is completed, it will be installed in a dry room with a dew point of - 40°C or lower. Samples will be collected remotely in connection with the electrolytic recovery device and gantry robot built in the dry room. We plan to conduct experiments to seat the sample stick. Ultimately, we plan to conduct comprehensive experiments in conjunction with LIBS.

Studies on automatic scoring systems in writing assessments have also evaluated the relationship between human and machine scores for the reliability of automated essay scoring systems. This study investigated the magnitudes of indices for inter-rater agreement and discrepancy, especially regarding human and machine scoring, in writing assessment. The mean of the overall population correlation between automated and human scoring in essay writing was .78. The overall common d effect size was 0.001. Results from this meta-analysis indicated a strong relationship with no discrepancies between automated and human scoring. Both the I2 and Q values suggested that the population correlation values studied seemed to be heterogeneous, in contrast to homogenous d effect sizes. Therefore, it is necessary to investigate the sources of the between-studies variations for r correlations. Practical implications for ways of reporting results of automatic-scoring systems research and limitations of the study are also discussed.