• Proceedings of the Korean Vacuum Society Conference
• /
• 2012.02a
• /
• pp.239-240
• /
• 2012

With advancements in semiconductor device technologies, manufacturing processes are getting more complex and it became more difficult to maintain tighter process control. As the number of processing step increased for fabricating complex chip structure, potential fault inducing factors are prevail and their allowable margins are continuously reduced. Therefore, one of the key to success in semiconductor manufacturing is highly accurate and fast fault detection and classification at each stage to reduce any undesired variation and identify the cause of the fault. Sensors in the equipment are used to monitor the state of the process. The idea is that whenever there is a fault in the process, it appears as some variation in the output from any of the sensors monitoring the process. These sensors may refer to information about pressure, RF power or gas flow and etc. in the equipment. By relating the data from these sensors to the process condition, any abnormality in the process can be identified, but it still holds some degree of certainty. Our hypothesis in this research is to capture the features of equipment condition data from healthy process library. We can use the health data as a reference for upcoming processes and this is made possible by mathematically modeling of the acquired data. In this work we demonstrate the use of recurrent neural network (RNN) has been used. RNN is a dynamic neural network that makes the output as a function of previous inputs. In our case we have etch equipment tool set data, consisting of 22 parameters and 9 runs. This data was first synchronized using the Dynamic Time Warping (DTW) algorithm. The synchronized data from the sensors in the form of time series is then provided to RNN which trains and restructures itself according to the input and then predicts a value, one step ahead in time, which depends on the past values of data. Eight runs of process data were used to train the network, while in order to check the performance of the network, one run was used as a test input. Next, a mean squared error based probability generating function was used to assign probability of fault in each parameter by comparing the predicted and actual values of the data. In the future we will make use of the Bayesian Networks to classify the detected faults. Bayesian Networks use directed acyclic graphs that relate different parameters through their conditional dependencies in order to find inference among them. The relationships between parameters from the data will be used to generate the structure of Bayesian Network and then posterior probability of different faults will be calculated using inference algorithms.

• Journal of Navigation and Port Research
• /
• v.46 no.4
• /
• pp.367-374
• /
• 2022

The purpose of this study was to propose a deep learning algorithm that applies to the fault diagnosis of fuel pumps and purifiers of autonomous ships. A deep learning algorithm reflecting the time dependence of the measured signal was configured, and the failure pattern was trained using the vibration signal, measured in the equipment's regular operation and failure state. Considering the sequential time-dependence of deterioration implied in the vibration signal, this study adopts Conv1D with sliding window computation for fault detection. The time dependence was also reflected, by transferring the measured signal from two-dimensional to three-dimensional. Additionally, the optimal values of the hyper-parameters of the Conv1D model were determined, using the grid search technique. Finally, the results show that the proposed data preprocessing method as well as the Conv1D model, can reflect the sequential dependency between the fault and its effect on the measured signal, and appropriately perform anomaly as well as failure detection, of the equipment chosen for application.

• Journal of Preventive Medicine and Public Health
• /
• v.21 no.1 s.23
• /
• pp.89-98
• /
• 1988

While there have been not a few reports on the seroepidemiological characteristics of hepatitis B virus (HBV) infection in Korea, most of them, however, have had several limitations; operational definition of HBV infection, validity of detection methods of HBV serologic markers, size of the study population, and confirmation of the vaccination history against HBV, etc. In order to avoid such limitations, authors randomly selected 1,495 healthy adults among the 217,511 insured (target population) of Korean Medical Insurance Corporation, living in seoul, and tested HBV serologic markers by RIA method and conducted direct interview to them. Although HBV serologic markers (HBsAg, anti-HBs and anti-HBc) of all the subjects were tested, 392(26.2%) of interview failure cases and 361 vaccinee were excluded from the actual population. Finally, the serologic markers tested of 742 nonvaccinee (study population) only were analysed for the seroepidemiologic observation of the natural infection of HBV. The seroepidemiological characteristics of HBV infection in Korea were as follows ; 1. Point prevalence of HBs antigenemia was 11.7(9.1{\sim}14.3)% in male, which was slightly higher than that of female, 9.5($3.7{\sim}15.3$)%. This level was one of the highest among those of Asian-Pacific countries. Decreasing tendency of HBsAg prevalence alter the age of 50 was observed, which seems to be due to selective attrition of HBV chronic carriers among the healthy adults and/or to the limited-lasting duration of the HBs antigenemia, in part. 2. Point prevalence of anti-HBc(78.8% in male,50.9% in female) was higher than that of anti-HBs(65.2% in male,46.6% in female), respectively. And both of them were higher in male than in female. Increasing tendency of the prevalence of both antibodies was observed by age, which seems to be largely due to recurrent infection in adults and to some cumulative effect, in part, of their relatively longer-lasting duration. 3. The level of HBV infection defined by positive for at least one of the 3 serologic markers of HBV by RIA method was 84.7($81.8{\sim}87.6$)% in male and 61.2($51.9{\sim}70.5$)% in female, which was also one of the highest among those of Asian-Pacific countries. The proportion of susceptible population to HBV infection among healthy adults was 15.3% in male and 38.8% in female. 4. The relative frequency of current or past infection and chronic carrier among HBV infected person was estimated. The currently or past infected was estimated 75.7% in male and 71.8% in female, and chronic carrier state, 13.8% in male and 14.1% in female. The analysis of the geometric mean of the antibody titer in anti-HBs positive sera indicated also to be compatible with the above findings, suggesting that active, even though inapparent, infection of HBV occur so frequently among healthy adults in Korea.

• Journal of agricultural medicine and community health
• /
• v.16 no.1
• /
• pp.61-69
• /
• 1991

Serodiagnosis of Clonorchis sinensis infections will probably be a first choice tool for screening of clonorchiasis in a future because of increasing difficulties in collection and examination of stools. The sensitive test such as ELISA can he used effectively. However there are some limitations in serological diagnosis for the detection of serum antibody. One of the major problems is the non-specificity of the antigens which produce cross reaction with other helminthic infection sera. To solve this problem. many investigators have tried to purify the antigens used. In this study, we determined the antigenic profile of the crude saline extract antigen of C. sinensis at early developmental stage based on SDS-PAGE and immunoblotting techniques for the purpose of understanding the nature of C. sinensis worm antigen The following results were obtained : 1) The SDS-PAGE showed many protein hands ranging from 10Kd to 91Kd relative molecular weight. Among them, 66, 46, 40, 33, 27, 24, 16, 14 and 10Kd bands were observed as a principle bands. The protein components of C. sinensis changed chronologically during their early developmental period. 44Kd band was stained unclearly in antigen of 2 weeks worm, but changed to concentrated state in antigen of 5 weeks worm. 35Kd band was found in antigen of 2 weeks worm, however this band was disappeared in antigen of 5 weeks worm. 22Kd band also lost its staining property gradually. 2) In spite of differences in antigenic profile, there was no differences in the data obtained by microplate ELISA using each antigen preparation. Absorbance value began to rise in between 2 to 3 weeks after infection. 3) By EITB. serum antibody recognized major protein bands with molecular weight of 91, 85, 63, 46, 40, 33, 24, 14 and 10Kd hand respectively. Among them 66, 33, 17 and 14Kd bands were observed as non-specific band because they reacted even in normal control sera. Generally, gradual increase of positive reactions were observed as the infection period of C. sinensis was prolonged. In other hand, the reaction of 10Kd hand did not occurred when 26th week sera was tested. 4) The positive reactions using antigens of 2 weeks worm, especially on 40 and 24Kd bands, were most strong and sharply demarcated compared to those of 3~5 weeks worm antigen.

• Journal of Preventive Medicine and Public Health
• /
• v.36 no.2
• /
• pp.101-107
• /
• 2003

Objectives : Dementia has emerged as a leading public health problem in elderly persons, and its early detection is important for the treatment of curable cases, and in the educational support for other family members. Although dementia screening tests are available, they have not gained widespread use in community or primary care settings. Our goal was to validate the Tine and Change (T&C) Test, -including its validity and reliability in patients, and to assess it as a simple, standardized method for the screening of dementia in the rural elderly. Methods : The participants in this study comprised of 59 patients from an urban hospital and 405 persons from a rural community aged 65 years or older. The time test evaluated the understanding of clock hands indicating 11:10, and the change test the ability to make 1,000 Won from a group of coins, consisting of one 500, seven 100, and seven 50 Won coins. The T&C ratings were validated against a reference standard based on the physician's diagnosis of the patients. The convergent validity in relation to other cognitive measure, test-retest agreement, and inter-observer reliability were assessed. To assess the relationship between the Korean Mini-Mental State Exam (K-MMSE) and the T&C Test, the mean K-MMSE scores were compared with the results of the T&C Test in the elderly from a rural community. Results The T&C Test had a sensitivity and specificity of 73.0, and 90.9%, and positive and negative predictive values of 93.1, and 66.7%, respectively. The test-retest and inter-observer agreement rates were both 95%. The K-MMSE scores and T&C Test were significantly related in the elderly from a rural community (p<0.01), The T&C Test was not influenced by the educational status. The Time and Change Tests took a mean of 6.3 and 12.7 seconds, respectively, to complete Conclusion : The T&C Test is a simple, accurate and reliable, performance-based tool in the screening for dementia. Because it is quick, and easy-to-use, it is hoped the T&C Test will be used for the widespread cognitive screening of aging populations.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.5
• /
• pp.81-88
• /
• 2021

Structural health monitoring (SHM) systems have attracted considerable interest owing to the frequent earthquakes over the last decade. Smart concrete is a technology that can analyze the state of structures based on their electro-mechanical behavior. On the other hand, most research on the self-sensing response of smart concrete generally investigated the electro-mechanical behavior of smart concrete under a static loading rate, even though the loading rate under an earthquake would be much faster than the static rate. Thus, this study evaluated the electro-mechanical behavior of smart ultra-high-performance concrete (S-UHPC) at three different loading rates (1, 4, and 8 mm/min) using a Universal Testing Machine (UTM). The stress-sensitive coefficient (SC) at the maximum compressive strength of S-UHPC was -0.140 %/MPa based on a loading rate of 1 mm/min but decreased by 42.8% and 72.7% as the loading rate was increased to 4 and 8 mm/min, respectively. Although the sensing capability of S-UHPC decreased with increased load speed due to the reduced deformation of conductive materials and increased microcrack, it was available for SHM systems for earthquake detection in structures.

• Journal of the Korean Chemical Society
• /
• v.65 no.6
• /
• pp.433-440
• /
• 2021

Surfactants are organic compounds that have both hydrophilic and non-polar parts in one molecule, classified as non-ion, anion, cation, and amphoteric surfactants according to the charge of hydrophilic parts in aqueous state. A trace amounts may remain when vegetables and fruits are washed using type1 detergent (Vegetable and fruit detergent), and there is a possibility of exposure to the human body through ingestion. This study developed the simultaneous analysis method for 5 surfactants with LC-MS/MS for analysis of detergent residues after washing vegetables and fruits with detergent. The mobile phase used distilled water and acetonitrile containing 50 mM ammonium formate and 0.1% formic acid and was analyzed using a gradient method using XBridge BEH C8 column. The accuracy of the established method was 83.9-112.1%, and the precision was less than 20%. The detection limit was 7.0 (SLS) to 29.0 (SLES-N3) ㎍/L, and the correlation coefficient (r²) of calibration line regression was greater than 0.99, it is considered suitable for the analysis of trace amounts of surfactant components remaining in vegetables and fruits.

• Journal of the Korean Society of Radiology
• /
• v.14 no.6
• /
• pp.773-780
• /
• 2020

With the development of the 4th industrial, research is being conducted to prevent diseases and reduce damage in various fields of science and technology such as medicine, health, and bio. As a result, artificial intelligence technology has been introduced and researched for image analysis of radiological examinations. In this paper, we will directly apply a deep learning model for classification and detection of pneumonia using chest X-ray images, and evaluate whether the deep learning model of the Inception series is a useful model for detecting pneumonia. As the experimental material, a chest X-ray image data set provided and shared free of charge by Kaggle was used, and out of the total 3,470 chest X-ray image data, it was classified into 1,870 training data sets, 1,100 validation data sets, and 500 test data sets. I did. As a result of the experiment, the result of metric evaluation of the Inception V3 deep learning model was 94.80% for accuracy, 97.24% for precision, 94.00% for recall, and 95.59 for F1 score. In addition, the accuracy of the final epoch for Inception V3 deep learning modeling was 94.91% for learning modeling and 89.68% for verification modeling for pneumonia detection and classification of chest X-ray images. For the evaluation of the loss function value, the learning modeling was 1.127% and the validation modeling was 4.603%. As a result, it was evaluated that the Inception V3 deep learning model is a very excellent deep learning model in extracting and classifying features of chest image data, and its learning state is also very good. As a result of matrix accuracy evaluation for test modeling, the accuracy of 96% for normal chest X-ray image data and 97% for pneumonia chest X-ray image data was proven. The deep learning model of the Inception series is considered to be a useful deep learning model for classification of chest diseases, and it is expected that it can also play an auxiliary role of human resources, so it is considered that it will be a solution to the problem of insufficient medical personnel. In the future, this study is expected to be presented as basic data for similar studies in the case of similar studies on the diagnosis of pneumonia using deep learning.

• Journal of Preventive Medicine and Public Health
• /
• v.22 no.3 s.27
• /
• pp.328-336
• /
• 1989

27 shipyard welders were diagnosed as pneumoconiosis and suspected pneumoconiosis(1976-1988) by chest radiographs and were observed over three years. 27 welders were divided into three groups by the state of exposure to welding fume i.e. cessation, decresase or continuity of exposure. And we observed the changing pattern of the chest radiographs of 27 welders with the passage of time. The results were as follows; 1. Grour I (ceased exposure to welding fume) were 10 cases(3 cases: suspected pneumoconiosis,7 cases: pneumoconiosis). Chest radiographs of all cases were improved. The shape and size of small opacities was improved in 6 cases(85.7%) and did not changed in 1 case(14.3%) out of 7 pneumoconiosis welders. 2. Group II (decreased exposure to welding fume) were S cases(2 cases: suspected pneumoconiosis, 3 cases: pneumoconiosis). Chest radiographs were progressed in 2 cases(40%), did not changed in 1 case(20%), were improved in 2 cases(40%) out of 5 cases. The shape and size of small opacities was progressed in 1 case(33.3%) and was improved in 2 cases(66.7%) out of 3 pneumoconiosis welders. 3. Group III(continued expoxsure to welding fume) were 12 cases(1 case: suspected pneumoconiosis, 11 cases: pneumoconiosis). Chest radiographs were progressed in 9 cases(75%), did not changed in 3 cases(25%) out of 12 cases. The shape and size of small opacities was progressed in 1 case(9.1%) and did not changed in 10 cases(90.9%) out of 11 pneumoconiosis welders. 4. The average duration for development into suspected pneumoconiosis was 6.6 years and for progression of each one category after that was 2.2 years(p<0.01). The radiological appearance of pneumoconiosis had disappeared or decreased after cessation of exposure to the welding fume. So that, early detection and control e.g., change of department of pneumoconiosis of welders by screening program will be important for medical surveillance of welders.

• Journal of Society of Preventive Korean Medicine
• /
• v.28 no.2
• /
• pp.55-65
• /
• 2024

Background : Tuberculosis (TB) remains a significant public health issue worldwide, particularly among healthcare workers (HCWs) at high risk of exposure. Latent tuberculosis infection (LTBI) is a state where individuals are infected with Mycobacterium tuberculosis but do not show clinical symptoms. Early detection and treatment of LTBI are crucial to prevent progression to active TB. This study aimed to investigate the prevalence and risk factors of LTBI among Korean Medicine (KM) workers in Seoul, South Korea. Methods : This study analyzed 368 adults aged 19 and over working in Korean medicine institutions in Seoul by September 2023. Participants underwent a tuberculin skin test (TST) and completed a survey collecting demographic information, occupation, work duration, smoking status, BCG vaccination, TB history, and comorbidities. Data were analyzed using descriptive statistics and chi-square tests, with significance set at p < 0.05. Results : The average age of participants was 43.1 years, with an LTBI prevalence rate of 3.5%. Significant risk factors included age and history of TB, Older age and a history of TB were associated with higher LTBI positivity. Conclusion : The study identified the prevalence and risk factors of LTBI among Korean medicine workers in Seoul. The findings highlight the need for targeted LTBI screening and preventive measures, especially for older workers and those with a history of TB. While the prevalence was lower than in other healthcare settings, the results emphasize the importance of regular LTBI testing and prevention education for KM workers. Future large-scale studies are needed to confirm these findings and further understand the relationship between various risk factors and LTBI in KM settings.