• Clinical Pain
• /
• 제18권2호
• /
• pp.92-96
• /
• 2019

Torticollis is an abnormal, asymmetric head or neck position which usually caused by imbalance of paracervical muscles. The traumatic torticollis can be caused by following events; atlantoaxial rotatory subluxation, atlantoaxial dislocation, cervical vertebral fractures, and injury to the cervical musculature. Especially, acute traumatic atlantoaxial rotatory subluxation usually presents limitation of cervical range of motion without pain or neurologic deficit. We report a case of a 58 year-old man who developed the acute atlantoaxial rotatory subluxation right after the chiropractic therapy, which induced the limitation of cervical range of motion to 52.5% of normal range. The magnetic resonance image revealed the facture of the odontoid process and the partial injury in transverse ligaments of the atlas. He underwent intramuscular botulinum toxin injection and 10 days of continuous cervical traction 15 hours a day using a 5 kg weight. The range of the cervical motion restored up to 90.2% of normal range.

• 산업경영시스템학회지
• /
• 제46권4호
• /
• pp.312-320
• /
• 2023

Recently, there has been an increasing attempt to replace defect detection inspections in the manufacturing industry using deep learning techniques. However, obtaining substantial high-quality labeled data to enhance the performance of deep learning models entails economic and temporal constraints. As a solution for this problem, semi-supervised learning, using a limited amount of labeled data, has been gaining traction. This study assesses the effectiveness of semi-supervised learning in the defect detection process of manufacturing using the MixMatch algorithm. The MixMatch algorithm incorporates three dominant paradigms in the semi-supervised field: Consistency regularization, Entropy minimization, and Generic regularization. The performance of semi-supervised learning based on the MixMatch algorithm was compared with that of supervised learning using defect image data from the metal casting process. For the experiments, the ratio of labeled data was adjusted to 5%, 10%, 25%, and 50% of the total data. At a labeled data ratio of 5%, semi-supervised learning achieved a classification accuracy of 90.19%, outperforming supervised learning by approximately 22%p. At a 10% ratio, it surpassed supervised learning by around 8%p, achieving a 92.89% accuracy. These results demonstrate that semi-supervised learning can achieve significant outcomes even with a very limited amount of labeled data, suggesting its invaluable application in real-world research and industrial settings where labeled data is limited.

• 드라이브 ㆍ 컨트롤
• /
• 제21권1호
• /
• pp.38-45
• /
• 2024

The power delivery is crucial to designing agricultural machinery. Therefore, the tractor-mounted potato harvester was used in this study to conduct the field experiment and analyze the power delivery for each step. This study was focused on an analysis of power delivery from the engine to the hydraulic components for the tractor-mounted harvester during potato harvesting. Finally, the simulation model of a self-propelled potato harvester was developed and validated using the experimental dataset of the tractor-mounted potato harvester. The power delivery analysis showed that approximately 90.22% of the engine power was used as traction power to drive the tractor-mounted harvester, and only 5.10% of the engine power was used for the entire hydraulic system of the tractor and operated the harvester. The statistical analysis of the simulation and experimental results showed that the coefficient of determinations (R²) ranged from 0.80 to 0.96, which indicates that the simulation model was performed with an accuracy of over 80%. The regression models were correlated linearly with the simulation and experimental results. Therefore, we believe that this study could contribute to the design methodology and performance test procedure of agricultural machinery. This basic study would be helpful in the design of a self-propelled potato harvester.

• 지식경영연구
• /
• 제24권4호
• /
• pp.153-176
• /
• 2023

최근 생성형 AI, 특히 ChatGPT와 같은 대화형 인공지능이 고객 서비스를 자동화하는 기술적 대안으로 주목받고 있다. 그러나 고객 서비스 자동화에 있어 현재의 생성형 AI 기술이 기존 인간 관리자를 효과적으로 대체할 수 있는지, 조건이나 환경에 따라 어떤 상황에서는 유리하고 다른 상황에서는 불리한지에 대한 연구는 아직 충분히 이루어지지 않은 상태이다. 이러한 배경에서 본 연구는 "고객 서비스 활동과 관련하여 생성형 AI가 인간 관리자를 대체할 수 있는가?"라는 질문에 답하기 위해, 음식 배달 플랫폼의 고객 온라인 리뷰에 대한 실험과 설문조사를 수행하였다. 또한 고객의 온라인 리뷰가 긍정적일 때와 부정적일 때에 따라 차이가 있는지 정교화 가능성 모델의 관점을 적용하여 가설을 도출하고 해당 가설이 지지되는지를 분석을 통해 확인하였다. 분석 결과, 긍정적인 리뷰에 대해서는 생성형 AI가 인간 관리자를 효과적으로 대체할 수 있지만, 부정적인 리뷰에 대해서는 완벽한 대체가 어려워 인간 관리자의 개입이 더 바람직한 것으로 확인되었다. 이러한 본 연구의 결과는 생성형 AI를 이용하여 고객 서비스 자동화하고자 하는 기업들에게 유의미한 실무적인 통찰을 제공해 줄 수 있을 것이다.

• 농업과학연구
• /
• 제50권4호
• /
• pp.773-784
• /
• 2023

In this study, we developed a dynamic model and steering controller model for an autonomous tractor and evaluated their performance. The traction force was measured using a 6-component load cell, and the rotational speed of the wheels was monitored using proximity sensors installed on the axles. Torque sensors were employed to measure the axle torque. The PI (proportional integral) controller's coefficients were determined using the trial-error method. The coefficient of the P varied in the range of 0.1 - 0.5 and the I coefficient was determined in 3 increments of 0.01, 0.05, and 0.1. To validate the simulation model, we conducted RMS (root mean square) comparisons between the measured data of axle torque and the simulation results. The performance of the steering controller model was evaluated by analyzing the damping ratio calculated with the first and second overshoots. The average front and rear axle torque ranged from 3.29 - 3.44 and 6.98 - 7.41 kNm, respectively. The average rotational speed of the wheel ranged from 29.21 - 30.55 rpm at the front, and from 21.46 - 21.63 rpm at the rear. The steering controller model exhibited the most stable control performance when the coefficients of P and I were set at 0.5 and 0.01, respectively. The RMS analysis of the axle torque results indicated that the left and right wheel errors were approximately 1.52% and 2.61% (at front) and 7.45% and 7.28% (at rear), respectively.

• 한국인터넷방송통신학회논문지
• /
• 제24권2호
• /
• pp.177-184
• /
• 2024

데이터가 빠른 속도로 증가하고 있는 가운데 가능한 최고의 정확도를 달성하기 위해 모든 종류의 복잡한 앙상블 및 딥 러닝 알고리즘이 사용되고 있다. 그렇지만, 이러한 모델이 알 수 없는 데이터를 예측/분류/인식/추적하는 방법과 관련하여 예측, 분류, 인식, 추적이 항상 신뢰할 수 있는 것은 아니다. 데이터 부족, 불균형 데이터 세트, 편향된 데이터 세트 등과 같은 다양한 이유가 학습 모델에 의해 포착되는 결정에 영향을 미칠 수 있다. 이와 관련하여 현재 모델의 설명 가능성에 관한 연구가 관심을 끌고 있다. 현재 설명 가능성 기법과 관련하여 LIME과 SHAP가 보편적으로 사용되고 있지만, 출력 결과들은 다소 상이한 측면을 나타내고 있다. 이에 본 연구에서는 LIME과 SHAP을 결합하는 방식을 소개하고, 데모와 관련해서 IEEE CIS 데이터 세트에서 거래를 사기로 분류할 때 LightGBM 및 Keras 모델이 내린 결정에 대한 설명 가능성을 분석한다.

• Structural Engineering and Mechanics
• /
• 제89권3호
• /
• pp.225-238
• /
• 2024

In this paper, a quasi-3D hyperbolic shear deformation theory for the bending responses of a functionally graded (FG) porous micro-beam is based on a modified couple stress theory requiring only one material length scale parameter that can capture the size influence. The model proposed accounts for both shear and normal deformation effects through an illustrative variation of all displacements across the thickness and satisfies the zero traction boundary conditions on the top and bottom surfaces of the micro-beam. The effective material properties of the functionally graded micro-beam are assumed to vary in the thickness direction and are estimated using the homogenization method of power law distribution, which is modified to approximate the porous material properties with even and uneven distributions of porosity phases. The equilibrium equations are obtained using the virtual work principle and solved using Navier's technique. The validity of the derived formulation is established by comparing it with the ones available in the literature. Numerical examples are presented to investigate the influences of the power law index, material length scale parameter, beam thickness, and shear and normal deformation effects on the mechanical characteristics of the FG micro-beam. The results demonstrate that the inclusion of the size effects increases the microbeams stiffness, which consequently leads to a reduction in deflections. In contrast, the shear and normal deformation effects are just the opposite.

• Structural Engineering and Mechanics
• /
• 제90권3호
• /
• pp.233-252
• /
• 2024

This research explores a new finite element model for the free vibration analysis of bi-directional functionally graded (BDFG) beams. The model is based on an efficient higher-order shear deformation beam theory that incorporates a trigonometric warping function for both transverse shear deformation and stress to guarantee traction-free boundary conditions without the necessity of shear correction factors. The proposed two-node beam element has three degrees of freedom per node, and the inter-element continuity is retained using both C1 and C0 continuities for kinematics variables. In addition, the mechanical properties of the (BDFG) beam vary gradually and smoothly in both the in-plane and out-of-plane beam's directions according to an exponential power-law distribution. The highly elevated performance of the developed model is shown by comparing it to conceptual frameworks and solution procedures. Detailed numerical investigations are also conducted to examine the impact of boundary conditions, the bi-directional gradient indices, and the slenderness ratio on the free vibration response of BDFG beams. The suggested finite element beam model is an excellent potential tool for the design and the mechanical behavior estimation of BDFG structures.

• Structural Engineering and Mechanics
• /
• 제90권3호
• /
• pp.253-261
• /
• 2024

This research explores a new finite element model for the free vibration analysis of bi-directional functionally graded (BDFG) beams. The model is based on an efficient higher-order shear deformation beam theory that incorporates a trigonometric warping function for both transverse shear deformation and stress to guarantee traction-free boundary conditions without the necessity of shear correction factors. The proposed two-node beam element has three degrees of freedom per node, and the inter-element continuity is retained using both C1 and C0 continuities for kinematics variables. In addition, the mechanical properties of the (BDFG) beam vary gradually and smoothly in both the in-plane and out-of-plane beam's directions according to an exponential power-law distribution. The highly elevated performance of the developed model is shown by comparing it to conceptual frameworks and solution procedures. Detailed numerical investigations are also conducted to examine the impact of boundary conditions, the bi-directional gradient indices, and the slenderness ratio on the free vibration response of BDFG beams. The suggested finite element beam model is an excellent potential tool for the design and the mechanical behavior estimation of BDFG structures.

• 한국동물생명공학회지
• /
• 제39권1호
• /
• pp.62-65
• /
• 2024

Dystocia, a challenging condition in obstetrics, can arise from various causes, including fetal monsters with structural abnormalities. This case report presents a unique case of dystocia due to a fetal monster known as Perosomus Elumbis in a beetal breed goat from Pakistan. The 4-years-old pregnant doe presented with prolonged straining and failure to deliver the fetus after 8 hours of labor. Upon examination, the cervix was dilated, and only the forelimbs of the fetus were visible in the birth canal. The subsequent delivery involved the application of manual traction by using a dystocia kit, and the removal of edematous fluid from the legs. The monster fetus exhibited absence of hair growth, along with the absence of thoracic vertebrae. Two other fetuses were present, with one found dead and the other alive. Posttreatment involved fluid therapy, antibiotics, and supportive care for the doe. This case report sheds light on the occurrence of Perosomus Elumbis fetal monsters and their impact on dystocia in goat breeding. Understanding the underlying causes and implementing appropriate management strategies are crucial for successful outcomes in similar cases.