• Journal of information and communication convergence engineering
• /
• v.20 no.3
• /
• pp.219-225
• /
• 2022

In this study, we propose training a support vector machine (SVM) model on top of deep networks for detecting Covid-19 from chest X-ray images. We started by gathering a real chest X-ray image dataset, including positive Covid-19, normal cases, and other lung diseases not caused by Covid-19. Instead of training deep networks from scratch, we fine-tuned recent pre-trained deep network models, such as DenseNet121, MobileNet v2, Inception v3, Xception, ResNet50, VGG16, and VGG19, to classify chest X-ray images into one of three classes (Covid-19, normal, and other lung). We propose training an SVM model on top of deep networks to perform a nonlinear combination of deep network outputs, improving classification over any single deep network. The empirical test results on the real chest X-ray image dataset show that deep network models, with an exception of ResNet50 with 82.44%, provide an accuracy of at least 92% on the test set. The proposed SVM on top of the deep network achieved the highest accuracy of 96.16%.

• Journal of Korean Public Health Nursing
• /
• v.23 no.2
• /
• pp.153-161
• /
• 2009

Purpose: This study was done to evaluate the differences between an experimental group and a control group of nursing students for their knowledge of CPR and their practical ability after undergoing CPR training. Method: This experiment was done with nursing students, who are divided into the experimental group (20 students) and the control group (23 students) in Chunbuk C city. The data were analyzed using the SPSS PC+ 12.0 program for the Chi-square tests, t-tests and Wilcoxon rank sum tests. Results: The 1st hypothesis, that the CPR knowledge scores for the experimental group will be higher than the scores for the control group (t=-3.934, p=<.001), was supported. On the other hand, the control group showed a conspicuous and meaningful improvement (t=-3.932, p=<.001). The 2st hypothesis, that the practical ability scores for the experimental group will be higher than the scores for the control group (t=-3.926, p=<.001), was supported. Conclusion: The CPR training in combination with theory and clinical placement is seen as a means to effectively develop the knowledge and practical ability of CPR.

• Journal of Animal Science and Technology
• /
• v.66 no.1
• /
• pp.167-177
• /
• 2024

Iris segmentation is an initial step for identifying the biometrics of animals when establishing a traceability system for livestock. In this study, we propose a deep learning framework for pixel-wise segmentation of bovine iris with a minimized use of annotation labels utilizing the BovineAAEyes80 public dataset. The proposed image segmentation framework encompasses data collection, data preparation, data augmentation selection, training of 15 deep neural network (DNN) models with varying encoder backbones and segmentation decoder DNNs, and evaluation of the models using multiple metrics and graphical segmentation results. This framework aims to provide comprehensive and in-depth information on each model's training and testing outcomes to optimize bovine iris segmentation performance. In the experiment, U-Net with a VGG16 backbone was identified as the optimal combination of encoder and decoder models for the dataset, achieving an accuracy and dice coefficient score of 99.50% and 98.35%, respectively. Notably, the selected model accurately segmented even corrupted images without proper annotation data. This study contributes to the advancement of iris segmentation and the establishment of a reliable DNN training framework.

• Structural Monitoring and Maintenance
• /
• v.11 no.1
• /
• pp.1-18
• /
• 2024

Self-compacting concrete is widely used around the globe today due to its special and unique properties. This study examines the effect of natural and crushed gravel combinations in different percentages in short-and long-term properties of concrete. The best utilized sand had a fineness modulus of 2.7. In the mentioned mix designs, silica fume was used with 0 and 7% of the weight of the cement. In order to check the properties of fresh and hardened concrete, 9 and 5 test types were performed, respectively. The carried out tests were slump flow, V-funnel, J-ring, L-box, U-box and column segregation for fresh concrete, and compressive, tensile and flexural strengths for hardened concrete. A mix with only 100% natural gravel was considered as the control mix. According to the results, the control mix design and the one containing 100% crushed gravel with silica fume were the best in fresh and hardened concrete tests, respectively. Finally, using the optimization method, a mix design with 25% natural gravel, 75% crushed gravel and silica fume was introduced as the best mix in terms of the results of both fresh and hardened concrete tests.

• Nuclear Engineering and Technology
• /
• v.56 no.6
• /
• pp.2266-2273
• /
• 2024

A machine learning platform is proposed for the diagnosis of a severe accident progression in a nuclear power plant. To predict the key parameters for accident management including lost signals, a long short term memory (LSTM) network is proposed, where multiple accident scenarios are used for training. Training and test data were produced by MELCOR simulation of the Fukushima Daiichi Nuclear Power Plant (FDNPP) accident at unit 3. Feature variables were selected among plant parameters, where the importance ranking was determined by a recursive feature elimination technique using RandomForestRegressor. To answer the question of whether a reduced order ML model could predict the complex transient response, we performed a systematic sensitivity study for the choices of target variables, the combination of training and test data, the number of feature variables, and the number of neurons to evaluate the performance of the proposed ML platform. The number of sensitivity cases was chosen to guarantee a 95 % tolerance limit with a 95 % confidence level based on Wilks' formula to quantify the uncertainty of predictions. The results of investigations indicate that the proposed ML platform consistently predicts the target variable. The median and mean predictions were close to the true value.

• International Journal of Industrial Entomology and Biomaterials
• /
• v.26 no.2
• /
• pp.74-80
• /
• 2013

Three new Multi ${\times}$ Bi combinations M.Con.4 ${\times}$ (SK6 ${\times}$ SK7), M.Con.4 ${\times}$ NB4D2 and Nistari ${\times}$ (SK6 ${\times}$ SK7) were studied in five seasons i.e. during the period of $18^{th}$ June - $12^{th}$ July, $30^{th}$ August - $23^{rd}$ September, $3^{rd}$ November - $2^{nd}$ December, $29^{th}$ January- $2^{nd}$ March and $30^{th}$ March-$25^{th}$ April with one control Nistari ${\times}$ NB4D2 considering nine reeling characteristics along with shell percent and cocoon yield per 100 disease free layings (DFLs) to establish the seasonal effect on all the parameters. Results reveal that all the characters performed well in November-December. Highest yield per 100 DFLs was recorded during November-December in all combinations showing >60 kg yield. Higher correlation of yield/100 DFLs with filament length, non-breakable filament length, denier, raw silk percent, reelability percent and recovery percent was recorded in all combinations of Multi ${\times}$ Bi considered in this study. Variability observed among four combinations indicates the effect of different temperature and humidity of different seasons on expression of different traits. The results indicate that there is significant relationship of raw silk percent with reelability percent, recovery percent and evenness. Reelability percent showed higher correlation with recovery percent. It was observed that neatness was positively correlated with evenness. The overall performance of the newly evolved combinations with regard to productivity and reeling characteristics is discussed emphasizing their utilization at commercial level.

• Nutrition Research and Practice
• /
• v.5 no.4
• /
• pp.294-300
• /
• 2011

Takju is a Korean alcoholic beverage made from rice, and is brewed with the yeast Saccharomyces cerevisiae. This study was conducted to evaluate the effects of exercise training and moderate Takju consumption on learning ability in 6-week old Sprague-Dawley male rats. The rats were treated with exercise and alcohol for 4 weeks in six separate groups as follows: non-exercised control (CC), exercised control (EC), non-exercised consuming ethanol (CA), exercised consuming ethanol (EA), non-exercised consuming Takju (CT), and exercised consuming Takju (ET). An AIN-93M diet was provided ad libitum. Exercise training was performed at a speed of 10 m/min for 15 minutes per day. Ethanol and Takju were administered daily for 6-7 hours to achieve an intake of about 10 ml after 12 hours of deprivation, and, thereafter, the animals were allowed free access to deionized water. A Y-shaped water maze was used from the third week to understand the effects of exercise and alcohol consumption on learning and memory. After sacrifice, brain acetylcholinesterase (AChE) activity was analyzed. Total caloric intake and body weight changes during the experiment were not significantly different among the groups. AChE activity was not significantly different among the groups. The number of errors for position reversal training in the maze was significantly smaller in the EA group than that in the CA and ET groups, and latency times were shorter in the EA group than those in the CC, EC, CT, and ET groups. The latency difference from the first to the fifth day was shortest in the ET group. The exercised groups showed more errors and latency than those of the non-exercised groups on the first day, but the data became equivalent from the second day. The results indicate that moderate exercise can increase memory and learning and that the combination of exercise and Takju ingestion may enhance learning ability.

• Journal of Society of Preventive Korean Medicine
• /
• v.20 no.3
• /
• pp.83-91
• /
• 2016

Objectives : The Levator Ani Muscles Training(LAMT) is one of the well known very effective prevention and treatment method of some specific diseases in the anal organs, digestion system, urinary organs and genital organs, which means LAMT could be considered as an independent exercise. Methods : Chinese databases were searched and found one good source named CNKI(China National Knowledge Infrastructure) where there were downloaded 154 articles, out of which were finally selected 82 used for the introduction of the historical background, health-effectiveness, working mechanism, training methods, etc. Results : LAMT as an independent exercise, is considered very effective for the prevention of diseases in the anal organs, e.g. hemorrhoids, anal fissure, rectal prolapse; urinary organs, e.g. urinary incontinence, frequent micturition; digestion system, e.g. visceroptosis, diarrhea, chronic enteritis; genital organs, e.g. prostatitis, impotence, premature ejaculation, feminine vaginal relaxation, sexual indifference, etc. It is also told LAMT is effective for coronary heart disease, hyperyension and varicose veins of lower extremity to a certain degree. LAMT is also very similar to the Kegel exercise in couple of aspects. They are equally effective for the urinary incontinence and impotence. Conclusions : LAMT as an independent exercise has developed and become more and more health-effective, especially through the combination with the Qigong exercise, and thus many training methods have been introduced. There are not many scientific research and development with very limited accomplishments even in China. It is strongly required to strengthen the research and development activities so that LAMT will become one of the very effective natural healing soonest possible.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.15 no.6
• /
• pp.3759-3765
• /
• 2014

This study examined the effects of action observations on the balance of patients with chronic stroke. The study period was between July 8, 2013 and August 10 2013. The participants were divided randomly into either the landscape picture observation group (n=20) and the action observation group (n=20). Each group observed the landscape pictures and video-recorded action-related balance for 5 minutes, respectively. Both groups had a balance training session for 10 minutes after the picture and video observations. The balance training sessions were repeated twice a day, three times a week for 6 weeks. As a result, significant improvement was noted in the action observation group than the landscape picture observation group for static balance and dynamic balance (p<.05). Consequently, the combination of action observation and balance training is believed to help improve the balance recovery for patients with chronic stroke.

• The Journal of the Korea Contents Association
• /
• v.20 no.9
• /
• pp.520-530
• /
• 2020

The purpose of this study was to investigate the effects of occupation-based bilateral upper extremity training and transcranial direct current stimulation on upper limb function in stroke patients. The study group was divided into 13 experimental groups with occupation-based bilateral upper extremity training and transcranial direct current stimulation, and 13 controls with only occupation-based bilateral upper extremity training. A total of 4 weeks, 50 minutes, 5 times a week conducted, the patients were tested with Canadian Occupational Performance Measure(COPM), Accelerometer, Fugle-Meyer Assessment(FMA), and Motor Activity Log(MAL). As a result of the study, the experimental group and the control group showed significant improvement in both occupation satisfaction and performance, usage of the affected side and the tendon side, recovery of upper limb function, and quality of movement, In particular, the experimental group showed a significant difference in the amount of the affected side than the control group. Therefore, it was found that the combination of occupation-based bilateral upper extremity training and transcranial direct current stimulation had a positive effect on the recovery of upper limb function in stroke patients.