Search

Mandibular fracture is one of the most frequent injuries in oral and maxillo-facial surgery. Radiologists diagnose mandibular fractures using panoramic radiography and cone-beam computed tomography (CBCT). Panoramic radiography is a conventional imaging modality, which is less complicated than CBCT. This paper proposes the diagnosis method of mandibular fractures in a panoramic radiograph based on a deep learning system without the intervention of radiologists. The deep learning system used has a one-stage detection called you only look once (YOLO). To improve detection accuracy, panoramic radiographs as input images are augmented using gamma modulation, multi-bounding boxes, single-scale luminance adaptation transform, and multi-scale luminance adaptation transform methods. Our results showed better detection performance than the conventional method using YOLO-based deep learning. Hence, it will be helpful for radiologists to double-check the diagnosis of mandibular fractures.

For the image conversion from low dynamic range images to high dynamic range images, we considered the possibility of using the Generative Adversarial Network (GAN) as one of the deep neural network methods. Deep learning requires a lot of data before building a module, but once creation is done, it is convenient to use it. In this paper, we proposed a weighting map for a local luminance based learning to reconstruct locally tone-mapped images.

Purpose The purpose of this study is to investigate the effect of the cross-link position on the rod fracture phenomenon during pedicle subtraction osteotomy (PSO) surgery using finite element model (FEM). Methods A three-dimensional finite element model of a lumbar spine with sagittal imbalance was constructed using computed tomography data of a 65-year-old female patient. After simulating the standard PSO at the L4 level, we constructed four models, specifically a model without a cross-link and three models with a cross-link at three different sites. The peak von Mises stress (PVMS) of the rod around the PSO site was measured after applying physiological loads (flexion, extension, axial rotation, and lateral bending) in each model. Results The measured PVMS outcomes at the PSO site were 135.8, 135.9, 208.9, and 384.7 MPa for model 1, 2, 3, and 4 during flexion, and 180.0, 180.1, 210.1, and 445.7 MPa during extension. These results show that when the cross-link is located at the PSO site, the rod stress at the PSO site increases significantly during flexion and extension. As the cross-link moved away from the PSO site, the effect on the rod stress decreased. When the cross-link was placed two levels away from the PSO site, the rod stress was scarcely affected. Conclusion When the cross-link during PSO surgery was positioned two levels away from the PSO site, the risk of rod fracture did not increase.

This study proposes a method of blending visible and near-infrared (NIR) images to enhance their edge details and local contrast based on the Laplacian pyramid and principal component analysis (PCA). In the proposed method, both the Laplacian pyramid and PCA are implemented to generate a radiance map. Using the PCA algorithm, the soft-mixing method and the mask-skipping filter were applied when the images were fused. The color compensation method uses the ratio between the radiance map fused by the Laplacian pyramid and the PCA algorithm and the luminance channel of the visible image to preserve the chrominance of the visible image. The results show that the proposed method improves edge details and local contrast effectively.

Purpose: Purpose was to explore associations between sleep duration and metabolic syndrome (MS) risks, and to determine factors associated with self-rated health (SRH) of adults with MS compared to other adults. Methods: This is a secondary data analysis based on the Fifth Korea National Health and Nutrition Examination Survey KNHANES V (N=12662). Study instruments included sleep duration, MS risk factors, SRH and health-related quality of life (HRQoL). Results: Mean age of participants was 43.68±12.26 years. Fifty-eight percent were women, and 18.3% were identified as having MS. Systolic blood pressure (SBP), diastolic blood pressure (DBP), and SRH were significantly different according to sleep duration (p

This paper proposes a system model to improve the utilization of smart appliances for smart grid demand response by expanding smart energy profile (SEP) 2.0. The original smart appliance system model was designed assuming that there exists a device with an energy service interface model in the home environment, but this is not common in the real world. Our proposed expansion of SEP 2.0 allows the use of smart appliances even without an energy service interface; thus, it can lead to an increase in the demand response participation of appliances. The effectiveness of the proposed model is verified through a field test conducted with appliance manufacturers and a utility company.

Power line communication (PLC) technology can be a viable solution for the future ubiquitous networks because it provides a cheaper alternative to other wired technology currently being used for communication. However, for the better performance of PLC-based networks, their physical layer and medium access control (MAC) protocols need to be a lot more enhanced. Recently, many physical frame encoding technologies have been developed, but the MAC protocols still need more consideration because of the unique characteristics of power line channels. In this paper, we describe an overview of the Korea Standard power line MAC protocol (called "KS-MAC"). We also report the simulation results performed with the well-known simulation tool of network simulator-2 (i.e., NS-2) and compared with the actual field test data.

For most applications in wireless sensor networks (WSNs), it is often assumed that the deployment of sensor nodes is unmanaged and random, so the density of local node may vary throughout the network. In high density areas, nodes consume more energy due to frequent packet collisions and retransmissions. One of the ways to alleviate this problem is to adjust the transmission power of each sensor node by means of efficient topology control mechanisms. In this paper, we propose an efficient topology control for energy conservation, named "k+ Neigh." In our scheme, each sensor node reduces its transmission power so that it has minimum number of k neighbor nodes. Later, we will show that the preferred value of the k is 2 by simulation. In the performance evaluation, the proposed scheme can make significant energy saving with such a topology structure, while the network connectivity is guaranteed.

In the area of wireless sensor networks, achieving minimum energy consumption is a very important research issue. A number of energy efficient protocols have been proposed, mostly based on a layered design approach, which means that they are focused on designing optimal strategies for “single” layer only. In this paper, we take an alternative approach i.e., a cross-layer design, and present a new MAC protocol named MAC-CROSS. In this new approach, the interactions between MAC and Routing layers are fully exploited to achieve energy efficiency for wireless sensor networks. More precisely, in the proposed MAC-CROSS algorithm, routing information at the network layer is utilized for the MAC layer such that it can maximize a sleep duration of each node. Through implementation on a Mica Mote platform and simulation study using the NS-2, we evaluate a performance of the presented MAC-CROSS and prove its substantial performance gains.