• Journal of Korean Neurosurgical Society
• /
• v.49 no.6
• /
• pp.345-350
• /
• 2011

Objective : It is debatable whether an anterior oblique fracture orientation is really a contraindication to anterior odontoid screw fixation. The purpose of this study was to investigate the feasibility of anterior odontoid screw fixation of type II and rostral shallow type III fracture with an anterior oblique fracture orientation. Methods : The authors evaluated 16 patients with type II and rostral shallow type III odontoid fracture with an anterior oblique fracture orientation. Of these 16 patients, 8 (group 1) were treated by anterior odontoid screw fixation, and 8 (group 2) by a posterior C1-2 arthrodesis. Results : Of the 8 patients in group 1, seven patients achieved solid bone fusion (87.5%), and one experienced screw back-out of the C-2 body two months after anterior screw fixation. All patients treated by posterior C1-C2 fusion in group 2 achieved successful bone fusion. Mean fracture displacements and fracture gaps were not significantly different in two groups. (p=0.075 and 0.782). However, mean fracture orientation angles were $15.3{\pm}3.2$ degrees in group 1, and $28.6{\pm}8.1$ degrees in group 2 (p=0.002), and mean fragment angulations were $3.2{\pm}2.1$ degrees in group 1, and $14.8{\pm}3.7$ degrees in group 2 (p=0.001). Conclusion : Even when the fracture lines of type II and rostral shallow type III fractures are oriented in an anterior oblique direction, anterior odontoid screw fixation can be feasible in carefully selected patients with a relatively small fracture orientation angle and relatively small fragment angulation.

• Korean Journal of Materials Research
• /
• v.26 no.8
• /
• pp.427-429
• /
• 2016

A multi-step deposition process for the gap-filling of submicrometer trenches using dimethyldimethoxysilane (DMDMOS), $(CH_3)_2Si(OCH_3)_2$, and $C_xH_yO_z$ by plasma enhanced chemical vapor deposition (PECVD) is presented. The multi-step process consisted of pre-treatment, deposition, and post-treatment in each deposition step. We obtained low-k films with superior gap-filling properties on the trench patterns without voids or delamination. The newly developed technique for the gap-filling of submicrometer features will have a great impact on inter metal dielectric (IMD) and shallow trench isolation (STI) processes for the next generation of microelectronic devices. Moreover, this bottom up gap-fill mode is expected to be universally for other chemical vapor deposition systems.

• Current Optics and Photonics
• /
• v.8 no.3
• /
• pp.215-224
• /
• 2024

In medium- and long-range optical imaging systems, atmospheric turbulence causes blurring and distortion of images, resulting in loss of image information. An image-restoration method based on an adaptive feature-fusion multi-input-multi-output (MIMO) dense U-shaped network (Unet) is proposed, to restore a single image degraded by atmospheric turbulence. The network's model is based on the MIMO-Unet framework and incorporates patch-embedding shallow-convolution modules. These modules help in extracting shallow features of images and facilitate the processing of the multi-input dense encoding modules that follow. The combination of these modules improves the model's ability to analyze and extract features effectively. An asymmetric feature-fusion module is utilized to combine encoded features at varying scales, facilitating the feature reconstruction of the subsequent multi-output decoding modules for restoration of turbulence-degraded images. Based on experimental results, the adaptive feature-fusion MIMO dense U-shaped network outperforms traditional restoration methods, CMFNet network models, and standard MIMO-Unet network models, in terms of image-quality restoration. It effectively minimizes geometric deformation and blurring of images.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.4
• /
• pp.1234-1257
• /
• 2023

In this paper, we propose a framework for multi-focus image fusion called PATN. In our approach, by aggregating deep features extracted based on the U-type Transformer mechanism and shallow features extracted using the PSA module, we make PATN feed both long-range image texture information and focus on local detail information of the image. Meanwhile, the edge-preserving information value of the fused image is enhanced using a dense residual block containing the Sobel gradient operator, and three loss functions are introduced to retain more source image texture information. PATN is compared with 17 more advanced MFIF methods on three datasets to verify the effectiveness and robustness of PATN.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.6
• /
• pp.1692-1705
• /
• 2024

In this paper, we present a method that integrates a Grammar Transducer as an external language model to enhance the accuracy of the pre-trained Korean End-to-end (E2E) Automatic Speech Recognition (ASR) model. The E2E ASR model utilizes the Connectionist Temporal Classification (CTC) loss function to derive hypothesis sentences from input audio. However, this method reveals a limitation inherent in the CTC approach, as it fails to capture language information from transcript data directly. To overcome this limitation, we propose a fusion approach that combines a clause-level n-gram language model, transformed into a Weighted Finite-State Transducer (WFST), with the E2E ASR model. This approach enhances the model's accuracy and allows for domain adaptation using just additional text data, avoiding the need for further intensive training of the extensive pre-trained ASR model. This is particularly advantageous for Korean, characterized as a low-resource language, which confronts a significant challenge due to limited resources of speech data and available ASR models. Initially, we validate the efficacy of training the n-gram model at the clause-level by contrasting its inference accuracy with that of the E2E ASR model when merged with language models trained on smaller lexical units. We then demonstrate that our approach achieves enhanced domain adaptation accuracy compared to Shallow Fusion, a previously devised method for merging an external language model with an E2E ASR model without necessitating additional training.

• Proceedings of the KWS Conference
• /
• 2007.11a
• /
• pp.141-143
• /
• 2007

Increasing demand of using low weight materials in recent automotive trends has been the challenge to develop a sound welding of aluminum alloys. A heat treatable AA6082-T6 and a non-heat treatable AA5083-0 aluminum alloys were joined in this study. Investigations revealed that about 60 UTS will be reduced due to welding process. Fracture happened in the interface between fusion zone and base metal of top specimen where penetration is shallow. Therefore, lower welding torch angle produced the better strength which allows deeper penetration to the top specimen. PWHT at $560^{\circ}C$ for 2 hours can be used to return the original UTS of the specimens.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2019.06a
• /
• pp.102-105
• /
• 2019

In the classical single-image super-resolution (SISR) reconstruction method using convolutional neural networks, the extracted features are not fully utilized, and the training time is too long. Aiming at the above problems, we proposed an improved SISR method based on a residual network. Our proposed method uses a feature fusion technology based on improved residual blocks. The advantage of this method is the ability to fully and effectively utilize the features extracted from the shallow layers. In addition, we can see that the feature fusion can adaptively preserve the information from current and previous residual blocks and stabilize the training for deeper network. And we use the global residual learning to make network training easier. The experimental results show that the proposed method gets better performance than classic reconstruction methods.

• Journal of the Korean earth science society
• /
• v.37 no.1
• /
• pp.1-10
• /
• 2016

Four entities of the Cenozoic fish fossils were discovered in the Miocene Duho Formation, Pohang, Korea. these fossils were identified as the first Pleuronectiformes in Korea based on the following-the presence of postcleithrum, the elongation of the first proximal pterygiophore of the anal fin ray, almost consistent presence of two proximal pterygiophore of the anal fin rays between the two adjacent hemal spines, the fusion of the first and second hypurals, the fusion of the third and fourth hypurals and the first preural centrum, the presence of well-developed anteriormost plate-like neural spine, the presence of the urohyal like fish-hook and its elongated sciatic part, and the division of the parahypural from the first preural centrum. On the other hand, geological studies about the Duho Formation consistently claimed that shallow-sea creatures were washed away by meteorological events such as a great flood and deposited at the bottom of deep-sea by the turbidity current. However, in Duho Formation, only shallow-sea ones have been discovered thus far. This study reported that Flat fish, deep-sea creature, was discovered in Duho Formation for the first time in Korea.

• Journal of Trauma and Injury
• /
• v.31 no.1
• /
• pp.16-18
• /
• 2018

Delayed esophageal rupture due to blunt injury is not new. However, rupture due to suspected barotrauma is very rare. We describe a case of esophageal rupture in a male 24-year-old patient after diving in shallow waters. The patient was quadriplegic and could not experience the typical chest pain related to rupture and resulting mediastinitis. The rupture was discovered 4 days after emergency decompressive laminectomy and fusion for his cervical spine. The rupture was evidently caused by barotrauma and was discovered four days after admission. He underwent primary closure and pericardial flap as a life-saving procedure.

• The Journal of Korean Society for Radiation Therapy
• /
• v.17 no.2
• /
• pp.95-103
• /
• 2005

Purpose : In the radiosurgery, to obtain CT image to find more accurate tumor position during respiration, and using them, to increase the accuracy of radiation treatment by applying image guided. Materials and Methods : Using the self-made vacuum cushion for the body SRS, CT images were obtained three for each patient during respiration (shallow, inhalation, exhalation). They were transformed to the RTP computer and then were fused. Global GTVs were delineated on the fused images and more appropriated treatment planning was established. Results : We can find the tumor position is moving toward cranio-caudal with max 10 mm margin and volume is transformed. As a result from the comparision of DVH (pre & post radio surgery), we observed about 100% dose to tumor. Conclusion : BSRS was skeptical due to the tumor movement during respiration. More accurate by the combination of the development of immobilization devices and BSRS based on Image Guide, it will be applied to more cases for BSRS.