• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.63 no.7
• /
• pp.945-949
• /
• 2014

In cardiac magnetic resonance imaging (CMRI), heart and respiratory motions are one of main obstacles in obtaining diagnostic quality of images. To synchronize CMRI to the physiological motions, ECG and respiratory gatings are commonly used. In this paper multi-biological signal (ECG, respiratory, and SPO2) based smart trigger system is proposed. By using multi-biological signal, the proposed system is robust to the induced noise such as eddy current when gradient pulsing is continuously applied during the examination. Digital conversion of the multi-biological signal makes the system flexible in implementing smart and intelligent algorithm to detect cardiac and respiratory motion and to reject arrhythmia of the heart. The digital data is used for real-time trigger, as well as signal display, and data storage which may be used for retrospective signal processing.

• Korean Journal of Artificial Intelligence
• /
• v.5 no.1
• /
• pp.29-37
• /
• 2017

The range of problems that can be handled by the activation of big data and the development of hardware has been rapidly expanded and machine learning such as deep learning has become a very versatile technology. In this paper, mnist data set is used as experimental data, and the Cross Entropy function is used as a loss model for evaluating the efficiency of machine learning, and the value of the loss function in the steepest descent method is We applied the Gradient Descent Optimize algorithm to minimize and updated weight and bias via backpropagation. In this way we analyze optimal reliability value corresponding to the number of exercises and optimal reliability value without overfitting. And comparing the overfitting time according to the number of data changes based on the number of training times, when the training frequency was 1110 times, we obtained the result of 92%, which is the optimal reliability value without overfitting.

• Journal of computational fluids engineering
• /
• v.7 no.2
• /
• pp.8-16
• /
• 2002

A numerical study on the use of the momentum interpolation method for flows with a large body force is presented. The inherent problems of the momentum interpolation method are discussed first. The origins of problems of the momentum interpolation methods are the validity of linear assumptions employed for the evaluation of the cell-face velocities, the enforcement of mass conservation for the cell-centered velocities and the specification of pressure and pressure correction at the boundary. Numerical experiments are performed for a typical flow involving a large body force. The numerical results are compared with those by the staggered grid method. The fact that the momentum interpolation method may result in physically unrealistic solutions is demonstrated. Numerical experiments changing the numerical grid have shown that a simple way of removing the physically unrealistic solution is a proper grid refinement where there is a large pressure gradient. An effective way of specifying the pressure and pressure correction at the boundary by a local mass conservation near the boundary is proposed, and it is shown that this method can effectively remove the inherent problem of the specification of pressure and pressure correction at the boundary when one uses the momentum interpolation method.

• Journal of Electrical Engineering and Technology
• /
• v.10 no.4
• /
• pp.1899-1909
• /
• 2015

Glaucoma is one of the most common causes of blindness which is caused by increase of fluid pressure in the eye which damages the optic nerve and eventually causing vision loss. An automated technique to diagnose glaucoma disease can reduce the physicians’ effort in screening of Glaucoma in a person through the fundal retinal images. In this paper, optimal hyper analytic wavelet transform for Glaucoma detection technique from fundal retinal images is proposed. The optimal coefficients for transformation process are found out using the hybrid GSO-Cuckoo search algorithm. This technique consists of pre-processing module, optimal transformation module, feature extraction module and classification module. The implementation is carried out with MATLAB and the evaluation metrics employed are accuracy, sensitivity and specificity. Comparative analysis is carried out by comparing the hybrid GSO with the conventional GSO. The results reported in our paper show that the proposed technique has performed well and has achieved good evaluation metric values. Two 10- fold cross validated test runs are performed, yielding an average fitness of 91.13% and 96.2% accuracy with CGD-BPN (Conjugate Gradient Descent- Back Propagation Network) and Support Vector Machines (SVM) respectively. The techniques also gives high sensitivity and specificity values. The attained high evaluation metric values show the efficiency of detecting Glaucoma by the proposed technique.

• Journal of Mechanical Science and Technology
• /
• v.14 no.6
• /
• pp.605-621
• /
• 2000

A mobile manipulator-a serial connection of a mobile platform and a task robot-is redundant by itself. Using its redundant freedom, a mobile manipulator can move in various modes, i. e., can perform dexterous tasks. In this paper, to improve task execution efficiency utilizing redundancy, optimal configurations of the mobile manipulator are maintained while it is moving to a new task point. Assuming that a task robot can perform the new task by itself, a desired configuration for the task robot can be pre-determined. Therefore, a cost function for optimality can be defined as a combination of the square errors of the desired and actual configurations of the mobile platform and of the task robot. In the combination of the two square errors, a newly defined mobility of a mobile platform is utilized as a weighting index. With the aid of the gradient method, the cost function is minimized, so the tasle that the mobile manipulator performs is optimized. The proposed algorithm is experimentally verified and discussed with a mobile manipulator, PURL-II.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.16 no.12 s.117
• /
• pp.1192-1200
• /
• 2006

Viscoelastic damping materials are widely used to reduce noise and vibration because of its low cost and easy implementation, for examples, on the body structure of passenger cars, air planes, electric appliances and ships. To design the damped structures, the material property such as elastic modulus and loss factor is essential information. The four-parameter fractional derivative model well describes the dynamic characteristics of the viscoelastic damping materials with respect to both frequency and temperature. However, the identification procedure of the four-parameter is very time-consuming one. In this study a new identification procedure of the four-parameters is proposed by using an FE model and a gradient-based numerical search algorithm. The identification procedure goes two sequential steps to make measured frequency response functions(FRF) coincident with simulated FRFs: the first one is a peak alignment step and the second one is an amplitude adjustment step. A numerical example shows that the proposed method is useful in identifying the viscoelastic material parameters of fractional derivative model.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.11 no.3
• /
• pp.135-140
• /
• 1999

To simulate the wave transformation by an energy dissipation region, a numerical model is suggested by discretizing the elliptic mild-slope equation. Generalized conjugate gradient method is used as solution algorithm to apply parabolic approximation to open boundary condition. To demonstrate the applicabil-ity of the numerical procedure suggested, the wave scattering by a circular damping region is examined. The feature of reflection in front of the damping region is captured clearly by the numerical solution. The effect of the size of dissipation coefficient is examined for a rectangular damping region. The recovery of wave height by diffraction occurs very slowly with distance behind the damping region.

• Journal of Biomedical Engineering Research
• /
• v.28 no.1
• /
• pp.132-138
• /
• 2007

MR(magnetic resonance) image of moving organ such as heart shows serious distortion of MR image due to motion itself. To eliminate motion artifacts, MRI(magnetic resonance imaging) scan sequences requires a trigger pulse like ECG(electro-cardiography) R-wave. ECG-gating using cardiac cycle synchronizes the MRI sequence acquisition to the R-wave in order to eliminate image motion artifacts. In this paper, we designed ECG/PPG(photo-plethysmography) gating system which is for eliminating motion artifacts due to moving organ. This system uses nonmagnetic carbon electrodes, lead wire and shield case for minimizing RF(radio-frequency) pulse and gradient effect. Also, we developed a ECG circuit for preventing saturation by magnetic field and a finger plethysmography sensor using optic fiber. And then, gating pulse is generated by adaptive filtering based on NLMS(normalized least mean square) algorithm. To evaluate the developed system, we measured and compared MR imaging of heart and neck with and without ECG/PPG gating system. As a result, we could get a clean image to be used in clinically. In conclusion, the designed ECG/PPG gating system could be useful method when we get MR imaging of moving organ like a heart.

• Structural Engineering and Mechanics
• /
• v.32 no.1
• /
• pp.95-126
• /
• 2009

The present contribution addresses the parallelization of advanced simulation methods for structural reliability analysis, which have recently been developed for large-scale structures with a high number of uncertain parameters. In particular, the Line Sampling method and the Subset Simulation method are considered. The proposed parallel algorithms exploit the parallelism associated with the possibility to simultaneously perform independent FE analyses. For the Line Sampling method a parallelization scheme is proposed both for the actual sampling process, and for the statistical gradient estimation method used to identify the so-called important direction of the Line Sampling scheme. Two parallelization strategies are investigated for the Subset Simulation method: the first one consists in the embarrassingly parallel advancement of distinct Markov chains; in this case the speedup is bounded by the number of chains advanced simultaneously. The second parallel Subset Simulation algorithm utilizes the concept of speculative computing. Speedup measurements in context with the FE model of a multistory building (24,000 DOFs) show the reduction of the wall-clock time to a very viable amount (<10 minutes for Line Sampling and ${\approx}$ 1 hour for Subset Simulation). The measurements, conducted on clusters of multi-core nodes, also indicate a strong sensitivity of the parallel performance to the load level of the nodes, in terms of the number of simultaneously used cores. This performance degradation is related to memory bottlenecks during the modal analysis required during each FE analysis.

• Structural Monitoring and Maintenance
• /
• v.5 no.1
• /
• pp.51-65
• /
• 2018

A number of sensing techniques have been implemented for detecting defects in civil infrastructures instead of onsite human inspections in structural health monitoring. However, the issue of faults in sensors has not received much attention. This issue may lead to incorrect interpretation of data and false alarms. To overcome these challenges, this article presents a deep learning-based method with a new architecture of Stateful Long Short Term Memory Neural Networks (S-LSTM NN) for detecting sensor fault without going into details of the fault features. As LSTMs are capable of learning data features automatically, and the proposed method works without an accurate mathematical model. The detection of four types of sensor faults are studied in this paper. Non-stationary acceleration responses of a three-span continuous bridge when under operational conditions are studied. A deep network model is applied to the measured bridge data with estimation to detect the sensor fault. Another set of sensor output data is used to supervise the network parameters and backpropagation algorithm to fine tune the parameters to establish a deep self-coding network model. The response residuals between the true value and the predicted value of the deep S-LSTM network was statistically analyzed to determine the fault threshold of sensor. Experimental study with a cable-stayed bridge further indicated that the proposed method is robust in the detection of the sensor fault.