• Title/Summary/Keyword: smoothed kernel

Search Result 30, Processing Time 0.032 seconds

Training Sample and Feature Selection Methods for Pseudo Sample Neural Networks (의사 샘플 신경망에서 학습 샘플 및 특징 선택 기법)

  • Heo, Gyeongyong;Park, Choong-Shik;Lee, Chang-Woo
    • Journal of the Korea Society of Computer and Information
    • /
    • v.18 no.4
    • /
    • pp.19-26
    • /
    • 2013
  • Pseudo sample neural network (PSNN) is a variant of traditional neural network using pseudo samples to mitigate the local-optima-convergence problem when the size of training samples is small. PSNN can take advantage of the smoothed solution space through the use of pseudo samples. PSNN has a focus on the quantity problem in training, whereas, methods stressing the quality of training samples is presented in this paper to improve further the performance of PSNN. It is evident that typical samples and highly correlated features help in training. In this paper, therefore, kernel density estimation is used to select typical samples and correlation factor is introduced to select features, which can improve the performance of PSNN. Debris flow data set is used to demonstrate the usefulness of the proposed methods.

Development of WMLS-based Particle Simulation Method for Solving Free-Surface Flow (자유표면 유동해석을 위한 WMLS 기반 입자법 기술 개발)

  • Nam, Jung-Woo;Park, Jong-Chun;Park, Ji-In;Hwang, Sung-Chul;Heo, Jae-Kyung;Jeong, Se-Min
    • Journal of Ocean Engineering and Technology
    • /
    • v.28 no.2
    • /
    • pp.93-101
    • /
    • 2014
  • In general, particle simulation methods such as the MPS(Moving Particle Simulation) or SPH(Smoothed Particle Hydrodynamics) methods have some serious drawbacks for pressure solutions. The pressure field shows spurious high fluctuations both temporally and spatially. It is well known that pressure fluctuation primarily occurs because of the numerical approximation of the partial differential operators. The MPS and SPH methods employ a pre-defined kernel function in the approximation of the gradient and Laplacian operators. Because this kernel function is constructed artificially, an accurate solution cannot be guaranteed, especially when the distribution of particles is irregular. In this paper, we propose a particle simulation method based on the moving least-square technique for solving the partial differential operators using a Taylor-series expansion. The developed method was applied to the hydro-static pressure and dam-broken problems to validate it.

Gaussian Filtering Effects on Brain Tissue-masked Susceptibility Weighted Images to Optimize Voxel-based Analysis (화소 분석의 최적화를 위해 자화감수성 영상에 나타난 뇌조직의 가우시안 필터 효과 연구)

  • Hwang, Eo-Jin;Kim, Min-Ji;Jahng, Geon-Ho
    • Investigative Magnetic Resonance Imaging
    • /
    • v.17 no.4
    • /
    • pp.275-285
    • /
    • 2013
  • Purpose : The objective of this study was to investigate effects of different smoothing kernel sizes on brain tissue-masked susceptibility-weighted images (SWI) obtained from normal elderly subjects using voxel-based analyses. Materials and Methods: Twenty healthy human volunteers (mean $age{\pm}SD$ = $67.8{\pm}6.09$ years, 14 females and 6 males) were studied after informed consent. A fully first-order flow-compensated three-dimensional (3D) gradient-echo sequence ran to obtain axial magnitude and phase images to generate SWI data. In addition, sagittal 3D T1-weighted images were acquired with the magnetization-prepared rapid acquisition of gradient-echo sequence for brain tissue segmentation and imaging registration. Both paramagnetically (PSWI) and diamagnetically (NSWI) phase-masked SWI data were obtained with masking out non-brain tissues. Finally, both tissue-masked PSWI and NSWI data were smoothed using different smoothing kernel sizes that were isotropic 0, 2, 4, and 8 mm Gaussian kernels. The voxel-based comparisons were performed using a paired t-test between PSWI and NSWI for each smoothing kernel size. Results: The significance of comparisons increased with increasing smoothing kernel sizes. Signals from NSWI were greater than those from PSWI. The smoothing kernel size of four was optimal to use voxel-based comparisons. The bilaterally different areas were found on multiple brain regions. Conclusion: The paramagnetic (positive) phase mask led to reduce signals from high susceptibility areas. To minimize partial volume effects and contributions of large vessels, the voxel-based analysis on SWI with masked non-brain components should be utilized.

The Study on the Comparative Analysis of the Aquaculture Production Efficiency Regarding Methods and Species (양식업의 양식방법별 어종별 생산효율성 비교분석에 관한 연구)

  • Park, Cheol-Hyung
    • The Journal of Fisheries Business Administration
    • /
    • v.43 no.2
    • /
    • pp.79-94
    • /
    • 2012
  • The purpose of this study is to investigate the production efficiencies of the Korean aquaculture fishery with respect to species and methods using a Data Envelopment Analysis. The study extracted the 8 fishes in each of the sea cage culture, aquarium basin, and enclosed aquaculture for the analytical purposes. First, the study estimated the technical, pure technical, and scale efficiencies of the total of 24 aquaculture fishes based on the traditional DEA under the assumptions of both CRS and VRS. 2 fishes were identified as the efficient DMUs under the CCR-model, and 6 fishes under the BCC-model. Second, we tested to see if there was any difference in production efficiencies regarding those three different methods of aquaculture. we could not find any evidence of the differences in efficiency using a rank sum test based on the traditional DEA. However, we could do find that the pure technical efficiency in the sea cage culture was lower than others at 1% level of significance and the pure technical efficiency in enclosed aquaculture was also lower than others at 5% level of significance using Bilateral-DEA, which could explicitly consider the heterogeneity in the 3 production methods of aquaculture. Finally, the study obtained the 95% confidence intervals of the efficiency scores for the 24 fishes under our study using the smoothed bootstraping method in the process of the re-sampling in cooperation with both a kernel density estimation and a reflection method. At the same time, we could estimate the bias-corrected efficiency scores while the traditionally estimated efficiency scores suffered from the biases in the process of solving a linear programming with the deterministic nature of a production frontier. And hence, we could distinguish the differences in production efficiencies of the 8 fishes with respect to those 3 methods of aquaculture.

A Voxel-Based Morphometry of Gray Matter Reduction in Patients with Dementia of the Alzheimer's Type (화소 기반 형태분석 방법을 이용한 알츠하이머 치매환자의 회백질 용적감소의 정량적 분석)

  • Lim, Hyun-Kook;Choi, Eun-Hyung;Lee, Chang-Uk
    • Korean Journal of Biological Psychiatry
    • /
    • v.15 no.2
    • /
    • pp.118-125
    • /
    • 2008
  • Objectives : The purpose of this study was to find brain regions in which gray matter volume was reduced and to show the capability of voxel-based morphpmetry(VBM) analysis for lateralizing clinically significant brain regions in dementia of Alzheimer's type patients compared to healthy group. Methods : MR T1-weighted images of the 20 dementia of Alzheimer's type patients were compared with those of the 20 normal controls. Images were transformed to standard MNI space. In order to observe gray matter volume change. Gray matter was smoothed with a Gaussian kernel. After these preprocessing, statistical analysis was performed using statistical parametric mapping software(SPM2). Results : Gray matter volume was significantly reduced in the bilateral parahippocampal gyri, Lt. anterior cingulate gyrus, Lt. posterior cingulate gyrus, bilateral superior temporal gyri Lt. middle temporal gyrus, Lt. superior, bilateral middle, Rt. anterior frontal gyri and Rt. precuneus in dementia of Alzheimer's type patient group. Conclusions : These VBM results confirm previous findings of temporal lobe and limbic lobe atrophic changes in dementia of Alzheimer's type, and suggest that these abnormalities may be confined to specific sites within that lobe, rather than showing a widespread distribution.

  • PDF

Software development for the visualization of brain fiber tract by using 24-bit color coding in diffusion tensor image

  • Oh, Jung-Su;Song, In-Chan;Ik hwan Cho;Kim, Jong-Hyo;Chang, Kee-Hyun;Park, Kwang-Suk
    • Proceedings of the KSMRM Conference
    • /
    • 2002.11a
    • /
    • pp.133-133
    • /
    • 2002
  • Purpose: The purpose of paper is to implement software to visualize brain fiber tract using a 24-bit color coding scheme and to test its feasibility. Materials and Methods: MR imaging was performed on GE 1.5 T Signa scanner. For diffusion tensor image, we used a single shot spin-echo EPI sequence with 7 non-colinear pulsed-field gradient directions: (x, y, z):(1,1,0),(-1,1,0),(1,0,1),(-1,0,1),(0,1,1),(0,1,-1) and without diffusion gradient. B-factor was 500 sec/$\textrm{mm}^2$. Acquisition parameters are as follows: TUTE=10000ms/99ms, FOV=240mm, matrix=128${\times}$128, slice thickness/gap=6mm/0mm, total slice number=30. Subjects consisted of 10 normal young volunteers (age:21∼26 yrs, 5 men, 5 women). All DTI images were smoothed with Gaussian kernel with the FWHM of 2 pixels. Color coding schemes for visualization of directional information was as follows. HSV(Hue, Saturation, Value) color system is appropriate for assigning RGB(Red, Green, and Blue) value for every different directions because of its volumetric directional expression. Each of HSV are assigned due to (r,$\theta$,${\Phi}$) in spherical coordinate. HSV calculated by this way can be transformed into RGB color system by general HSV to RGB conversion formula. Symmetry schemes: It is natural to code the antipodal direction to be same color(antipodal symmetry). So even with no symmetry scheme, the antipodal symmetry must be included. With no symmetry scheme, we can assign every different colors for every different orientation.(H =${\Phi}$, S=2$\theta$/$\pi$, V=λw, where λw is anisotropy). But that may assign very discontinuous color even between adjacent yokels. On the other hand, Full symmetry or absolute value scheme includes symmetry for 180$^{\circ}$ rotation about xy-plane of color coordinate (rotational symmetry) and for both hemisphere (mirror symmetry). In absolute value scheme, each of RGB value can be expressed as follows. R=λw|Vx|, G=λw|Vy|, B=λw|Vz|, where (Vx, Vy, Vz) is eigenvector corresponding to the largest eigenvalue of diffusion tensor. With applying full symmetry or absolute value scheme, we can get more continuous color coding at the expense of coding same color for symmetric direction. For better visualization of fiber tract directions, Gamma and brightness correction had done. All of these implementations were done on the IDL 5.4 platform.

  • PDF

Improvement of Endoscopic Image using De-Interlacing Technique (De-Interlace 기법을 이용한 내시경 영상의 화질 개선)

  • 신동익;조민수;허수진
    • Journal of Biomedical Engineering Research
    • /
    • v.19 no.5
    • /
    • pp.469-476
    • /
    • 1998
  • In the case of acquisition and displaying medical Images such as ultrasonography and endoscopy on VGA monitor of PC system, image degradation of tear-drop appears through scan conversion. In this study, we compare several methods which can solve this degradation and implement the hardware system that resolves this problem in real-time with PC. It is possible to represent high quality image display and real-time processing and acquisition with specific de-interlacing device and PCI bridge on our hardware system. Image quality is improved remarkably on our hardware system. It is implemented as PC-based system, so acquiring, saving images and describing text comment on those images and PACS networking can be easily implemented.metabolism. All images were spatially normalized to MNI standard PET template and smoothed with 16mm FWHM Gaussian kernel using SPM96. Mean count in cerebral region was normalized. The VOls for 34 cerebral regions were previously defined on the standard template and 17 different counts of mirrored regions to hemispheric midline were extracted from spatially normalized images. A three-layer feed-forward error back-propagation neural network classifier with 7 input nodes and 3 output nodes was used. The network was trained to interpret metabolic patterns and produce identical diagnoses with those of expert viewers. The performance of the neural network was optimized by testing with 5~40 nodes in hidden layer. Randomly selected 40 images from each group were used to train the network and the remainders were used to test the learned network. The optimized neural network gave a maximum agreement rate of 80.3% with expert viewers. It used 20 hidden nodes and was trained for 1508 epochs. Also, neural network gave agreement rates of 75~80% with 10 or 30 nodes in hidden layer. We conclude that artificial neural network performed as well as human experts and could be potentially useful as clinical decision support tool for the localization of epileptogenic zones.

  • PDF

Suspension of Sediment over Swash Zone (Swash대역에서의 해빈표사 부유거동에 관한 연구)

  • Cho, Yong Jun;Kim, Kwon Soo;Ryu, Ha Sang
    • KSCE Journal of Civil and Environmental Engineering Research
    • /
    • v.28 no.1B
    • /
    • pp.95-109
    • /
    • 2008
  • We numerically analyzed the nonlinear shoaling, a plunging breaker and its accompanying energetic suspension of sediment at a bed, and a redistribution of suspended sediments by a down rush of preceding waves and the following plunger using SPH with a Gaussian kernel function, Lagrangian Dynamic Smagorinsky model (LDS), Van Rijn's pick up function. In that process, we came to the conclusion that the conventional model for the tractive force at a bottom like a quadratic law can not accurately describe the rapidly accelerating flow over a swash zone, and propose new methodology to accurately estimate the bottom tractive force. Using newly proposed wave model in this study, we can successfully duplicate severely deformed water surface profile, free falling water particles, a queuing splash after the landing of water particles on the free surface and a wave finger due to the structured vortex on a rear side of wave crest (Narayanaswamy and Dalrymple, 2002), a circulation of suspended sediments over a swash zone, net transfer of sediments clouds suspended over a swash zone toward the offshore, which so far have been regarded very difficult features to mimic in the computational fluid mechanics.

Automatic Interpretation of Epileptogenic Zones in F-18-FDG Brain PET using Artificial Neural Network (인공신경회로망을 이용한 F-18-FDG 뇌 PET의 간질원인병소 자동해석)

  • 이재성;김석기;이명철;박광석;이동수
    • Journal of Biomedical Engineering Research
    • /
    • v.19 no.5
    • /
    • pp.455-468
    • /
    • 1998
  • For the objective interpretation of cerebral metabolic patterns in epilepsy patients, we developed computer-aided classifier using artificial neural network. We studied interictal brain FDG PET scans of 257 epilepsy patients who were diagnosed as normal(n=64), L TLE (n=112), or R TLE (n=81) by visual interpretation. Automatically segmented volume of interest (VOI) was used to reliably extract the features representing patterns of cerebral metabolism. All images were spatially normalized to MNI standard PET template and smoothed with 16mm FWHM Gaussian kernel using SPM96. Mean count in cerebral region was normalized. The VOls for 34 cerebral regions were previously defined on the standard template and 17 different counts of mirrored regions to hemispheric midline were extracted from spatially normalized images. A three-layer feed-forward error back-propagation neural network classifier with 7 input nodes and 3 output nodes was used. The network was trained to interpret metabolic patterns and produce identical diagnoses with those of expert viewers. The performance of the neural network was optimized by testing with 5~40 nodes in hidden layer. Randomly selected 40 images from each group were used to train the network and the remainders were used to test the learned network. The optimized neural network gave a maximum agreement rate of 80.3% with expert viewers. It used 20 hidden nodes and was trained for 1508 epochs. Also, neural network gave agreement rates of 75~80% with 10 or 30 nodes in hidden layer. We conclude that artificial neural network performed as well as human experts and could be potentially useful as clinical decision support tool for the localization of epileptogenic zones.

  • PDF

Voxel-based Morphometry (VBM) Based Assessment of Gray Matter Loss in Medial Temporal Lobe Epilepsy: Comparison with FDG PET (화소기반 형태분석 방법을 이용한 내측측두엽 간질환자의 회백질 부피/농도 감소평가; FDG PET과의 비교)

  • Kang, Hye-Jin;Lee, Ho-Young;Lee, Jae-Sung;Kang, Eun-Joo;Lee, Sang-Gun;Chang, Kee-Hyun;Lee, Dong-Soo
    • The Korean Journal of Nuclear Medicine
    • /
    • v.38 no.1
    • /
    • pp.30-40
    • /
    • 2004
  • Purpose: The aims of this study were to find brain regions in which gray matter volume was reduced and to show the capability of voxel-based morphometry (VBM) analysis for lateralizing epileptogenic zones in medial temporal lobe epilepsy (mTLE). The findings were compared with fluorodeoxyglucose positron omission tomography (FDG PET). Materials and Methods: MR T1-weighted images of 12 left mTLE and 11 right mTLE patients were compared with those of 37 normal controls. Images were transformed to standard MNI space and averaged in order to create study-specific brain template. Each image was normalized to this local template and brain tissues were segmented. Modulation VBM analysis was performed in order to observe gray matter volume change. Gray matter was smoothed with a Gaussian kernel. After these preprocessing, statistical analysis was peformed using statistical parametric mapping software (SPM99). FDG PET images were compared with those of 22 normal controls using SPM. Results: Gray matter volume was significantly reduced in the left amygdala and hippocampus in left mTLE. In addition, volume of cerebellum, anterior cingulate, and fusiform gyrus in both sides and left insula was reduced. In right mTLE, volume was reduced significantly in right hippocampus. In contrast, FDG uptake was decreased in broad areas of left or right temporal lobes in left TLE and right TLE, respectively. Conclusions: Gray matter loss was found in the ipsilateral hippocampus by modulation VBM analysis in medial temporal lobe epilepsy. This VBM analysis might be useful in lateralizing the epileptogenic zones in medial temporal lobe epilepsy, while SPM analysis of FDG PET disclosed hypometabolic epileptogenic zones.