• Progress in Medical Physics
• /
• v.32 no.4
• /
• pp.172-178
• /
• 2021

Purpose: This study aimed to develop a deep learning architecture combining two task models to generate synthetic computed tomography (sCT) images from low-tesla magnetic resonance (MR) images to improve metallic marker visibility. Methods: Twenty-three patients with cervical cancer treated with intracavitary radiotherapy (ICR) were retrospectively enrolled, and images were acquired using both a computed tomography (CT) scanner and a low-tesla MR machine. The CT images were aligned to the corresponding MR images using a deformable registration, and the metallic dummy source markers were delineated using threshold-based segmentation followed by manual modification. The deformed CT (dCT), MR, and segmentation mask pairs were used for training and testing. The sCT generation model has a cascaded three-dimensional (3D) U-Net-based architecture that converts MR images to CT images and segments the metallic marker. The performance of the model was evaluated with intensity-based comparison metrics. Results: The proposed model with segmentation loss outperformed the 3D U-Net in terms of errors between the sCT and dCT. The structural similarity score difference was not significant. Conclusions: Our study shows the two-task-based deep learning models for generating the sCT images using low-tesla MR images for 3D ICR. This approach will be useful to the MR-only workflow in high-dose-rate brachytherapy.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.24 no.2
• /
• pp.127-136
• /
• 2016

The camera based object detection systems should satisfy the recognition performance as well as real-time constraints. Particularly, in safety-critical systems such as Autonomous Emergency Braking (AEB), the real-time constraints significantly affects the system performance. Recently, multi-core processors and system-on-chip technologies are widely used to accelerate the object detection algorithm by distributing computational loads. However, due to the advanced hardware, the complexity of system architecture is increased even though additional hardwares improve the real-time performance. The increased complexity also cause difficulty in migration of existing algorithms and development of new algorithms. In this paper, to improve real-time performance and design complexity, a task scheduling strategy is proposed for visual object tracking systems. The real-time performance of the vision algorithm is increased by applying pipelining to task scheduling in a multi-core processor. Finally, the proposed task scheduling algorithm is applied to crosswalk detection and tracking system to prove the effectiveness of the proposed strategy.

• Journal of the Korean Society of Physical Medicine
• /
• v.12 no.3
• /
• pp.79-84
• /
• 2017

PURPOSE: The purpose of this study is to examine how task-oriented training focused on lower extremity strengthening can affect mobility function and postural stability. METHODS: The study's subjects included 10 children with cerebral palsy: 7 girls and 3 boys between the ages of 4 and 9 whose Gross Motor Functional Classification System (GMFCS) level was I or II. Their functional mobility was gauged using the Gross Motor Function Measurement (GMFM), and their postural stability was evaluated using a force platform. Participants received task-oriented training focused on lower extremity strengthening for 5 weeks. The study used a paired t-test to investigate the difference in mobility function and postural stability of children with cerebral palsy before and after the lower extremity strengthening exercise. RESULTS: The GMFM dimensions D (standing) (p<.02) and E (walking) (p<.001) improved significantly between the pre-test and post-test. A significant increase in the posturographic center of pressure (CoP) shift and surface area of the CoP were found overall between the pre-test and post-test (p<.001). CONCLUSION: The present study provides evidence that an 8-week task-oriented training focused on strengthening the lower extremities is an effective and feasible strategy for improving the mobility function and postural stability of children with cerebral palsy.

• Journal of The Korean Society of Integrative Medicine
• /
• v.1 no.3
• /
• pp.19-28
• /
• 2013

Purpose : The purpose of this study was to find out the impact of action-observation training and task-oriented training on activities of daily living performance of stroke patients. Method : 30 stroke patients hospitalized in D hospital located in Busan and treated were randomly allocated to Action-Observation Training Group and Task-Oriented Training Group in fifteens. To compare activities of daily living performance before and after therapy intervention, Korea-modified Barthel index (K-MBI) was carried out. Result : In both groups, activities of daily living performance of stroke patients before and after therapy intervention showed statistically significant differences (p<.05) and activities of daily living performance between two groups after therapy intervention showed statistically significant differences. Conclusion : It was found that action-observation training and task-oriented training improved activities of daily living performance of stroke patients. It is considered that the application of action-observation training and task-oriented training to clinical occupational therapy will show a positive effect on the improvement of activities of daily living performance.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.6
• /
• pp.1667-1677
• /
• 1999

An effective allocation of requested number of processors to newly incoming tasks in two-dimensional (2D) mesh interconnection networks is very important for achieving the desired high performance and resource utilization. It also needs to guarantee the complete recognition of the free submeshes based on contiguous and available processors with minimum overhead. An efficient task allocation algorithm for 2D meshes is presented in this paper. By employing a new approach for searching the one-dimensional array, the proposed algorithm can find the available submesh without the scanning of the entire 2D array unlike earlier designs. As a result, the new algorithm can significantly reduce the task allocation time. Comprehensive computer simulation shows that the average allocation time and waiting delay are much smaller than earlier irrespective of the size of meshes. The hardware overhead is comparable to other algorithms.

• The Journal of Small Business Innovation
• /
• v.20 no.3
• /
• pp.31-48
• /
• 2017

Teams play an indispensable role in helping an organization achieve its goal. Since constant interaction among team members is the cornerstone of team competitiveness, it is necessary for the team members to commit themselves to shared goals-team commitment, a type of positive, emotional attitude of team members. Constant interactions among team members, however, inevitably breed a byproduct called conflict. Though intra-team conflict has both positive and negative effects on team performance, little research has been done to clarify the roles that relationship, task and process conflicts play in team commitment. This research is an empirical investigation of the relationship among the three types of conflicts - relationship, task and process conflicts - based on data collected from 232 R&D teams of 13 Korean companies. The analysis suggests that, while relationship conflict has a negative impact on team commitment, task and process conflicts have no significant impact on team commitment. As for the relationships among those three types of conflict, process conflict has positively influenced task, and relationship conflicts and task conflict also has had a positive impact on relationship conflict. Based on these results, this study has presented its implications and directions for future research.

• Proceedings of the KIEE Conference
• /
• 2000.07d
• /
• pp.2846-2848
• /
• 2000

In this paper we propose a path planning method for an autonomous mobile robot based on genetic algorithms(GAs). There have been a number of methods proposed for the path planning of a mobile robot. However, few algorithms have been developed for an on-line path planning of a mobile robot with the uncertain information of a task environment. Therefore, we propose a path planning algorithms based on GAs that has ability of creating path planning without the perfect information of a task environment and an ability of planning the efficient path by on-line process. Then we show a possibility of the practical use with the results of simulations and experiments.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10b
• /
• pp.130-134
• /
• 1992

The ability of a robot system to comply to an environment via the control of tool-environment interaction force is of vital for the successful task accomplishment in many robot application. This paper presents the implementation of external force control for two dimensional contour following task using a commercial robot system. Force accommodation is used since a constraint imposed in our work is not to modify the commercial robot system. A linear, decoupled model of two dimensional contour following system in the discrete time domain is derived first. Then the experimental verification of linear control is obtained using a PUMA 560 manipulator with standard Unimation controller, Astek FS6-120A six axis wrist force sensor attached externally to the arm and LSI-11173 microcomputer. Experimentally obtained data shows that the RMS contact force error is 0.8246 N when following the straight edge and 2.3768 N when following 40 mm radius curved contour.

• Journal of the Korean Dietetic Association
• /
• v.23 no.2
• /
• pp.155-179
• /
• 2017

The purpose of this study was to compare the task elements categorized into required and additional man power's implemental between present and future-oriented duties of school nutrition teachers. The survey consisted of five duties, 27 tasks, 93 task elements, and 270 work details in the task elements of school nutrition teachers. A pilot-test was first conducted on nutrition teachers to confirm the survey contents, and then a main survey was performed on 240 school nutrition teachers, using a self-administrated online method, from July 16 to September 5, 2016. To compare present and future-oriented tasks, frequency analyses were conducted. Work details in the task elements were categorized into 'required' and 'additional man power's implemental', depending on school nutrition teachers' responses, based on a 50% cut-off percentage. The results showed that 13 work details among 60 work details (21.7%) in the 'Duty C. Safety and hygiene management of school foodservice', and 15 work details out of 106 work details (14.2%) in 'Duty B. Foodservice management practices' were identified as additional man power's implemental in future oriented duties. As to 'Duty A. Nutrition management', only three work details among 55 work details (5.5%) were identified as additional man power's implemental. On the other hand, all work details in 'Duty D. Nutrition diet education and counseling and 'Duty E. Reinforce professionalism' were identified as "required" as school nutrition teachers' duties. These findings imply that school nutrition teachers perceive nutrition management and education as their primary duties to the fulfill school foodservice' mission of promoting students' health and fostering students' dietary behaviors. The study offers practical and governmental implications, which can encourage school nutrition teachers to perform their primary duties.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.38C no.11
• /
• pp.1000-1009
• /
• 2013

This study investigated changes in behavioral performance before and after watching a multi-view 3D content by using auditory stimuli based on the selective attention theory in order to quantitatively evaluate 3D visual fatigue. Twenty-one undergraduates were asked to report on their current visual and physical condition both in the pre- and post-experiment. A selective attention task was conducted before and after mobile 3D viewing to compare the changes in performance. After performing a Wilcoxon's matched-pairs signed-ranks test on the subjective ratings of 3D visual fatigue, participants were categorized into two groups, unfatigued and fatigued group with a definite criterion. For the unfatigued group, no significant fatigue effects were found in behavioral response times and accuracies to specific auditory targets. In sharply contrast to the unfatigued group, the fatigued group showed significantly delayed response times and less response accuracies. However, no significant changes in accuracies for a working memory task were observed in both groups.