• Journal of the Korean Academy of Clinical Electrophysiology
• /
• v.10 no.1
• /
• pp.23-28
• /
• 2012

Purpose : The objectives of this study were to examine changes in gait parameters related to the COP and gait velocity resulting from the use of ankle foot orthosis, and to provide information for the prescription of ankle foot orthosis and gait rehabilitation training. Methods : We sampled a control group of 14 normal children of the same age as the spastic diplegia (spastic diplegia used ankle foot orthosis) and measured their COP at barefoot gait and their gait parameters at three gait velocities. Change in the COP according to the use of ankle foot orthosis measured one-way ANOVA and Gait parameters according to gait velocity used repeated measures ANOVA. Results : As a result of analyzing change in the COP it became close to that of normal children, and changes in gait parameters such as step width, stride length and stride time also became close to those of normal children. Conclusion : In conclusion, when spastic diplegia used ankle foot orthosis, their balancing ability was improved as a result of decreased change in the COP, and their changes in gait parameters also became close to those of the control group. These results show that the use of ankle foot orthosis improves spastic diplegia gait functions.

• The Bulletin of The Korean Astronomical Society
• /
• v.46 no.2
• /
• pp.46.3-47
• /
• 2021

Recently a multilayer spectral inversion (MLSI) model has been proposed to infer the physical parameters of plasmas in the solar chromosphere. The inversion solves a three-layer radiative transfer model using the strong absorption line profiles, H alpha and Ca II 8542 Å, taken by the Fast Imaging Solar Spectrograph (FISS). The model successfully provides the physical plasma parameters, such as source functions, Doppler velocities, and Doppler widths in the layers of the photosphere to the chromosphere. However, it is quite expensive to apply the MLSI to a huge number of line profiles. For example, the calculating time is an hour to several hours depending on the size of the scan raster. We apply deep neural network (DNN) to the inversion code to reduce the cost of calculating the physical parameters. We train the models using pairs of absorption line profiles from FISS and their 13 physical parameters (source functions, Doppler velocities, Doppler widths in the chromosphere, and the pre-determined parameters for the photosphere) calculated from the spectral inversion code for 49 scan rasters (~2,000,000 dataset) including quiet and active regions. We use fully connected dense layers for training the model. In addition, we utilize a skip connection to avoid a problem of vanishing gradients. We evaluate the model by comparing the pairs of absorption line profiles and their inverted physical parameters from other quiet and active regions. Our result shows that the deep learning model successfully reproduces physical parameter maps of a scan raster observation per second within 15% of mean absolute percentage error and the mean squared error of 0.3 to 0.003 depending on the parameters. Taking this advantage of high performance of the deep learning model, we plan to provide the physical parameter maps from the FISS observations to understand the chromospheric plasma conditions in various solar features.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.3
• /
• pp.1100-1118
• /
• 2021

In a wide range of ML applications, the training data contains privacy-sensitive information that should be kept secure. Training the ML systems by privacy-sensitive data makes the ML model inherent to the data. As the structure of the model has been fine-tuned by training data, the model can be abused for accessing the data by the estimation in a reverse process called model inversion attack (MIA). Although, MIA has been applied to shallow neural network models of recognizers in literature and its threat in privacy violation has been approved, in the case of a deep learning (DL) model, its efficiency was under question. It was due to the complexity of a DL model structure, big number of DL model parameters, the huge size of training data, big number of registered users to a DL model and thereof big number of class labels. This research work first analyses the possibility of MIA on a deep learning model of a recognition system, namely a face recognizer. Second, despite the conventional MIA under the white box scenario of having partial access to the users' non-sensitive information in addition to the model structure, the MIA is implemented on a deep face recognition system by just having the model structure and parameters but not any user information. In this aspect, it is under a semi-white box scenario or in other words a gray-box scenario. The experimental results in targeting five registered users of a CNN-based face recognition system approve the possibility of regeneration of users' face images even for a deep model by MIA under a gray box scenario. Although, for some images the evaluation recognition score is low and the generated images are not easily recognizable, but for some other images the score is high and facial features of the targeted identities are observable. The objective and subjective evaluations demonstrate that privacy cyber-attack by MIA on a deep recognition system not only is feasible but also is a serious threat with increasing alert state in the future as there is considerable potential for integration more advanced ML techniques to MIA.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.2
• /
• pp.671-686
• /
• 2020

Due to the long training time and high training cost of traditional surgical training methods, the emerging virtual surgical training method has gradually replaced it as the mainstream. However, the virtual surgical system suffers from poor authenticity and high computational cost problems. For overcoming the deficiency of these problems, we propose an optimized model for the local compression deformation of soft tissue. This model uses a simulated annealing algorithm to optimize the parameters of the soft tissue model to improve the authenticity of the simulation. Meanwhile, although the soft tissue deformation is divided into local deformation region and non-deformation region, our proposed model only needs to calculate and update the deformation region, which can improve the simulation real-time performance. Besides, we define a compensation strategy for the "superelastic" effect which often occurs with the mass-spring model. To verify the validity of the model, we carry out a compression simulation experiment of abdomen and human foot and compare it with other models. The experimental results indicate the proposed model is realistic and effective in soft tissue compression simulation, and it outperforms other models in accuracy and real-time performance.

• The Journal of the Korea Contents Association
• /
• v.11 no.4
• /
• pp.207-214
• /
• 2011

The purpose of this study was to investigate effect of feedback methods and ambulatory assistive aids on accuracy of partial weight bearing in healthy adults. Twenty subjects were recruited and trained to 30% weight bearing of body weight (BW) using 3-point gait crutches and wheeled walker pattern. Dynamic feedback group (n=11) was received dynamic postresponse feedback and static group (n=9) received static feedback. Force plate was used to measure %BW and GAITRite used to measure gait parameters in immediately and after 3 days of training. Immediately after training, there was not significantly at 30%BW target load in dynamic group with crutch gait (p>.05). There were significantly differences in %BW according to feedback methods and ambulatory aids (p<.05). There was not significant difference between immediately and after 3 days of training (p>0.05). There were significantly differences in gait velocity and stance ratio between crutches and wheeled walker gait (p<.05). Thus, it was suggested that if possible, use crutches, training for partial weight bearing via dynamic feedback is necessary.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.2464-2467
• /
• 2005

The purpose of this study was to explore the effectives of a virtual bicycle system in improving the ability of equilibrium sense of normal healthy adults. Experiments were performed to find the factors related to the training of equilibrium sense. The subjects consisted of young and elderly people and the group of young people was compared against the group of elderly people. We investigated three different running modes of virtual bicycle system with two successive sets in total. W measured the parameters related to the running time, the velocity, the weight movement, the degree of the deviation from the road, and the location of the center of pressure (COP). The results showed that the running capability of the elderly became much better after repeated training. In addition, it was found out that the ability to control postural balance and the capability of equilibrium sensory were improved with the presentation of the visual feedback information of the distribution of weight. We also found that the running time and the running velocity reduced when there was no visual feedback information. From the results, our newly developed bicycle system seems to be effective in the diagnosis of equilibrium sense as well as in the improvement of the sense of sight, and vestibular function of the elderly in the field of rehabilitation training.

• Journal of the Korean Society for Precision Engineering
• /
• v.24 no.10
• /
• pp.137-145
• /
• 2007

The purpose of this study is developing a virtual bicycle system for improving the ability of postural balance control for adults in various age groups. The system consists of an exercise bicycle that allows tilt in accordance with the postural balance of the subject in the system, a visual display that shows virtual road, and a visual feedback system. The rider of the system tries to maintain balance on the bicycle with a visual feedback of a virtual road while the pedaling speed, the heading direction, and various weight distribution information are updated to the subject as visual feedbacks in the display. A series of experiments were performed with various subjects to find the factors related to postural balance control in the system. The related parameters obtained were weight shift, magnitude of the deviation from the center of the virtual road, and variables related to the movement of the center of pressure. The results found that the ability to control postural balance in the system improved with the presentation of visual feedback information of the distribution of weight. It was also found that the general performance of the subject on balance in the system improved after ten days long training. The results show that the newly developed system can be used for the diagnosis of postural balance as well as for the stimulation of various senses such as vision and somatic sense in the field of rehabilitation training.

• Asian Pacific Journal of Cancer Prevention
• /
• v.16 no.9
• /
• pp.3875-3879
• /
• 2015

Background: The neutrophil/lymphocyte ratio (NLR) is a simple index of systemic inflammatory response, and has been shown to be a prognostic indicator in some types of cancer. Inflammation has been implicated in the initiation and progression of thyroid cancer. The aim of this study was to examine the relationship of NLR with papillary thyroid cancer (PTC) and different benign thyroid pathologies like multinodular goiter (MNG) and lymphocytic thyroiditis (LT). Materials and Methods: We retrospectively evaluated the neutrophil, lymphocyte counts and NLR calculated from these parameters of 232 patients with histologically confirmed as multinodular goiter (group MNG) (n=70), lymphocytic thyroiditis (group LT) (n=97), LT with PTC (group LT-PTC) (n=25) and PTC (group PTC) (n=40). The optimal cut-off value for NLR was determined. Results: NLR level was significantly higher in groups LT-PTC and PTC as compared to groups MNG and LT (p<0.05). NLR of LT subgroups according to TSH levels were not different (p>0.05). When we grouped the patients as benign and malignant according to PTC presence, the optimum NLR cut-off point obtained from ROC analysis was 1.91 (sensitivity 89.0% and specificity 54.5%). Conclusions: Since NLR was significantly elevated in group LT-PTC and group PTC, NLR value may give an opinion as a potential marker in differentiation of benign and malign thyroid disorders. For this purpose a cut-off value of 1.91 for NLR may be accepted.

• The Journal of the Acoustical Society of Korea
• /
• v.24 no.1
• /
• pp.58-68
• /
• 2005

In this paper, we propose a scheme to improve discriminative property in the feature compensation method for robust speech recognition under noisy environments. The estimation of noisy speech model used in existing feature compensation methods do not guarantee the computation of posterior probabilities which discriminate reliably among the Gaussian components. Estimation of Posterior probabilities is a crucial step in determining the discriminative factor of the Gaussian models, which in turn determines the intelligibility of the restored speech signals. The proposed scheme employs minimum classification error (MCE) training for estimating the parameters of the noisy speech model. For applying the MCE training, we propose to identify and determine the 'competing components' that are expected to affect the discriminative ability. The proposed method is applied to feature compensation based on parallel combined mixture model (PCMM). The performance is examined over Aurora 2.0 database and over the speech recorded inside a car during real driving conditions. The experimental results show improved recognition performance in both simulated environments and real-life conditions. The result verifies the effectiveness of the proposed scheme for increasing the performance of robust speech recognition systems.

• Journal of Nutrition and Health
• /
• v.35 no.8
• /
• pp.825-833
• /
• 2002

The effects of dietary supplementation of Eleutherococcus senticosus, taurine and carnitine on maximal endurance exercise performance along with other related parameters were evaluated in rats that underwent aerobic exercise training for 6 weeks. Thirty-two male rats (4 weeks old) were randomly divided into 4 groups, and fed experimental diets and/or aerobic exercise trained according to the protocol: SC (sedentary control group), EC (exercise-trained control group), EE (exercise-trained Eleutherococcus senticosus-supplemented group), and EETC (exercise-trained Eleutherococcus senticosus, taurine and carnitine-supplemented group). The food efficiency ratio of EC rats was significantly lower than the value for SC rats (p < 0.01). Exercise-trained control animals (92 $\pm$ 8.8 min) could run significantly longer until exhausted on the treadmill than sedentary control rats (11 $\pm$ 0.8 min) (p < 0.001). Animals fed an Eleutherococcus senticosus-supplemented diet, and an Eleuthherococcus sonticosus, taurine and carnitine- supplemented diet while undergoing aerobic exercise training for 6 weeks exhibited, respectively, 8 and 5 minutes longer running performance until exhausted than the rats fed the control diet. The gastrocnemius muscle glycogen concentration of the rats, measured at 48 hours post maximal exercise performance test, was 43% higher in EC rats than the value for SC rats (p < 0.05), but was not different among EC, EE, and EETC rats. The mitochondrial citrate synthase activity of the soleus muscle was significantly higher in EC rats compared to the value for SC rats (p < 0.01), and showed a tendency to increase, without statistical significance, in EE or EETC rats compared to the value for EC rats. These results indicate that aerobic exercise training for 6 weeks significantly improved maximal exercise performance, muscle glycogen content along with citrate synthase activity, which are important in the energy metabolism of muscle under aerobic exercise. Dietary supplementation of Eleutherococcus senticosus in rats while undergoing aerobic exercise training improved maximal endurance exercise performance without significantly affecting muscle glycogen content and enzyme activities involved in energy metabolism during exercise. Taurine and carnitine supplementation failed to show an additive effect on maximal endurance exercise performance when consumed along with Eleutherococcus senticosus.