• 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.

• Proceedings of the IEEK Conference
• /
• 2003.07c
• /
• pp.2911-2914
• /
• 2003

Conventionally, biological manipulations have been performed manually with long training and pretty low success rates. To overcome this problem, a novel biological manipulation system has been developed to manipulate biological cells without any interference of a human operator, In this paper, we demonstrate a development of tole-autonomous Cell Manipulation System (CMS) using an image processing at a remote site. The CMS consists of two manipulators, a plane stage, and an optical microscope. We developed deformable template-model-matching algorithm for micro objects and pattern matching algorithm of end effect for these manipulators in order to control manipulators and the stage. Through manipulation of biological cells using these algorithms, the performance of the CMS is verified experimentally.

• 한국전산유체공학회:학술대회논문집
• /
• 2009.04a
• /
• pp.39-42
• /
• 2009

The aerodynamic characteristics of aircraft winglet with structural deformation was investigated using the static FSI(Fluid-Structure Interaction) system. The system, comprised of CAD, CFD, CSD, VSI, and grid regeneration modules, was constructed. In the process VSI, grid regeneration, and integration modules were developed to combine CSD and CFD modules. As a test model, KC-135A, the double winglet suggested by Whitcomb, was selected and its aerodynamic characteristics for the rigid and deformable models was calculated by applying the static FSI system. As a result, the lift and drag coefficients of test models were reduced to 11% and 1.3%, respectively.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.04c
• /
• pp.343-345
• /
• 2003

본 논문에서는 색상 정보와 변형 모델을 이용한 얼굴 영역 및 얼굴의 특징 영역의 자동 검출 방법을 제시한다. 영상으로부터 획득할 수 있는 정보 중 가장 빠르고 쉽게 얻을 수 있는 정보가 색상 정보이며, 색상정보는 사물을 판단함에 있어서 가장 효율적이면서 컴퓨터의 계산량을 줄일 수 있다는 장점을 갖고 있기 때문에 얼굴 영역 검출 방법으로 많이 이용되고 있다. 본 연구에서는 얼굴영역 및 얼굴 특성 추출함에 있어 컬러모델 사용 시 외부 조명의 영향을 줄여주는 조명 보정 방법을 제시하고, 조명 보정에 의해 평활화된 YCbCr 색상모델에 적용하여 각 성분 특성을 고려한 얼굴영역 및 얼굴의 특성 영역에 해당하는 후보 영역을 검출하는 방법을 제시한다. 검출된 얼굴후보 영역 및 특성 영역은 가변 모델인 동적 윤곽선 모델의 초기값으로 자동 적용되어 윤곽선 모델 적용시 문제점가운데 하나인 초기값 설정문제를 해결함과 동시에 얼굴 및 얼굴 특징 정보의 정확한 윤곽선을 추출하는데 사용된다. 실험 결과 제시된 방법을 적용한 결과 빠르고 효과적으로 얼굴 및 특성 영역을 검출 할 수 있음을 입증 할 수 있었다.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.10 no.4
• /
• pp.57-63
• /
• 2011

The aim of this study is to analyze turning process using commercial FEM simulation code. Various simulation models of orthogonal cutting process for 3 layers of metallic material have been simulated and analyzed. The workpiece material used for the orthogonal plane-strain metal cutting simulation consists of three layers, which are Allow Tool Steel, Aluminum and Stainless Steel. The finite element model is composed of a deformable workpiece and a rigid tool. The tool penetrates through the workpiece at a constant speed and constant feed rate. As an analytical result, detailed cutting temperature, strain, pressure, residual stress for both a tool and each layer of workpiece were obtained during the turning process. It has been closely observed that the chip flow curve deforms continuously.

• Structural Engineering and Mechanics
• /
• v.51 no.1
• /
• pp.163-180
• /
• 2014

In the displacement based finite element analysis of composite beams that consist of two Euler-Bernoulli beams juxtaposed with a deformable shear connection, the coupling of the displacement fields may cause oscillations in the interlayer slip field and reduction in optimal convergence rate, known as slip-locking. In this study, the B-bar procedure is proposed to alleviate the locking effects. It is also shown that by changing the primary dependent variables in the mathematical model, to be able to interpolate the interlayer slip field directly, oscillations in the slip field can be completely eliminated. Examples are presented to illustrate the performance and the numerical characteristics of the proposed methods.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.1524-1529
• /
• 2005

Particle filter is a kind of conditional density propagation model. Its similar characteristics to both selection and mutation operator of evolutionary strategy (ES) due to its Bayesian inference rule structure, shows better performance than any other tracking algorithms. When a new object is entering the region of interest, particle filter sets which have been swarming around the existing objects have to move and track the new one instantaneously. Moreover, there is another problem that it could not track multiple objects well if they were moving away from each other after having been overlapped. To resolve reinitialization problem, we use competitive-AVQ algorithm of neural network. And we regard interfarme difference (IFD) of background images as potential field and give priority to the particles according to this IFD to track multiple objects independently. In this paper, we showed that the possibility of real-time object tracking as intelligent interfaces by simulating the deformable contour particle filters.

• 제어로봇시스템학회:학술대회논문집
• /
• 2002.10a
• /
• pp.57.4-57
• /
• 2002

$\textbullet$ Image segmentation is an essential technique of image analysis. In spite of the traditional issues in contour initialization and boundary concavities, active contour models(snakes) are popular and known as successful methods for segmentation. $\textbullet$ We could find in experiment that snake using Gaussian External Force is fast in time but low in accuracy and snake using Gradient Vector Flow by Chenyang Xu and Jerry L. Prince is high in accuracy but slow in time. $\textbullet$ In this paper, we presented a new active contour model, GGF snake, for segmentation of endoscopic image. Proposed GGF snake made up for the defects of the traditional snakes in contour initialization and boundary...

• Steel and Composite Structures
• /
• v.20 no.5
• /
• pp.963-981
• /
• 2016

A nonlocal trigonometric shear deformation beam theory based on neutral surface position is developed for bending, buckling, and vibration of functionally graded (FG) nanobeams using the nonlocal differential constitutive relations of Eringen. The present model is capable of capturing both small scale effect and transverse shear deformation effects of FG nanobeams, and does not require shear correction factors. The material properties of the FG nanobeam are assumed to vary in the thickness direction. The equations of motion are derived by employing Hamilton's principle, and the physical neutral surface concept. Analytical solutions are presented for a simply supported FG nanobeam, and the obtained results compare well with those predicted by the nonlocal Timoshenko beam theory.

• Structural Engineering and Mechanics
• /
• v.31 no.3
• /
• pp.315-331
• /
• 2009

Mechanical connections are recognized as extremely important elements in the aspect of strength and structural safety. However, classical structural model does not consider the connection stiffness properties, and are based on models with pinned or rigid joints only. In fact, mechanical connections are deformable and behave not linearly, affecting the whole structure and inducing nonlinear behavior as well. The quantification of this effect, however, depends on the description of the working of the connectors and the wood response under embedment. The theoretical modeling of wood structures with semi-rigid connections involves not only the structural analysis, but also the modeling of both single and grouped moment resisting connectors and the study of the wood properties under embedment. The proposal of this paper is to approach these aspects, and to quantitatively study the influence of the moment resistant connection in wooden framed structures. Comparisons between rigid and semi-rigid connections and between linear and nonlinear analysis lead to quantitative results.