• Journal of Preventive Medicine and Public Health
• /
• v.23 no.3 s.31
• /
• pp.324-337
• /
• 1990

To assess the effectiveness of the interventions in working environment and personal hygiene for the occupational exposure to the lead, 156 workers (116 exposed subjects and 40 controls) of a newly established battery factory were examined for their blood lead concentration (Pb-B) in every 3 months up to 18 months. Air lead concentration (Pb-A) of the workplaces was also checked for 3 times in 6 months interval from August 1987. Environmental intervention included the local exhaust ventilation and vacuum cleaning of the floor. Intervention of the personal hygiene included the daily change of clothes, compulsory shower after work and hand washing before meal, prohibition of cigarette smoking and food consumption at the work site and wearing mask. Mean Pb-B of the controls was $21.97{\pm}3.36{\mu}g/dl$ at the preemployment examination and slightly increased to $22.75{\pm}3.38{\mu}g/dl$ after 6 months. Mean Pb-B of the workers who were employed before the factory was in operation (Group A) was $20.49{\pm}3.84{\mu}g/dl$ on employment and it was increased to $23.90{\pm}5.30{\mu}g/dl$ after 3 months (p<0.01). Pb-B was increased to $28.84{\pm}5.76{\mu}g/dl$ 6 months after the employment which was 1 month after the initiation of intervention program. It did not increase thereafter and ranged between $26.83{\mu}g/dl\;and\;28.28{\mu}g/dl$ in the subsequent 4 tests. Mean Pb-B of the workers who were employed after the factory had been in operation but before the intervention program was initiated (Group B) was $16.58{\pm}4/53{\mu}g/dl$ before the exposure and it was increased to $28.82{\pm}5.66{\mu}g/dl$(P<0.01) in 3 months later (1 month after the intervention). The values of subsequent 4 tests remained between 26.46 and $28.54{\mu}g/dl$. Mean Pb-B of the workers who were employed after intervention program had been started (Group C) was $19.45{\pm}3.44{\mu}g/dl$ at the preemployment examination and gradually increased to $22.70{\pm}4.55{\mu}g/dl$ after 3 months(P<0.01), $23.68{\pm}4.18{\mu}g/dl$ after 6 months, and $24.42{\pm}3.60{\mu}g/dl$ after 9 months. Work stations were classified into 4 parts according to Pb-A. The Pb-A of part I, the highest areas, were $0.365mg/m^3$, and after the intervention the levels were decreased to $0.216mg/m^3\;and\;0.208mg/m^3$ in follow-up tests. The Pb-A of part II was decreased from $0.232mg/m^3\;to\;0.148mg/m^3,\;and\;0.120mg/m^3$ after the intervention. Pb-A of part III and W was tested only after intervention and the Pb-A of part III were $0.124mg/m^3$ in Jannuary 1988 and $0.081mg/m^3$ in August 1988. The Pb-A of part IV not stationed at one place but moving around, was $0.110mg/m^3$ in August 1988. There was no consistent relationship between Pb-B and Pb-A. Pb-B of the group A and B workers in the part of the highest Pb-A were lower than those of the workers in the parts of lower Pb-A. Pb-B of the workers in the part of the lowest Pb-A incerased more rapidly. Pb-B of group C workers was the highest in part I and the lowest in part IV. These findings suggest that Pb-B is more valid method than Pb-A for monitoring the health of lead workers and intervention in personal hygiene is more effective than environmental intervention.

• Tuberculosis and Respiratory Diseases
• /
• v.44 no.4
• /
• pp.785-795
• /
• 1997

Background : Tumor markers have been used in diagnosis, predicting the extent of disease, monitoring recurrence after therapy and prediction of prognosis. But the utility of markers in lung cancer has been limited by low sensitivity and specificity. TPA-M is recently developed marker using combined monoclonal antibody of Cytokeratin 8, 18, and 19. This study was conducted to evaluate the efficacy of new tumor marker, TPA-M by comparing the estabilished markers SCC, CEA, Cyfra21-1 in lung cancer. Method : An immunoradiometric assay of serum CEA, sec, Cyfra21-1, and TPA-M was performed in 49 pathologically confirmed lung cancer patients who visited Keimyung University Hospital from April 1996 to August 1996, and 29 benign lung diseases. Commercially available kits, Ab bead CEA (Eiken) to CEA, SCC RIA BEAD (DAINABOT) to SCC, CA2H (TFB) to Cyfra2H. and TPA-M (DAIICHI) to TPA-M were used for this study. Results : The mean serum values of lung cancer group and control group were $10.05{\pm}38.39{\mu}/L$, $1.59{\pm}0.94{\mu}/L$ in CEA, $3.04{\pm}5.79{\mu}/L$, $1.58{\pm}2.85{\mu}/L$ in SCC, $8.27{\pm}11.96{\mu}/L$, $1.77{\pm}2.72{\mu}/L$ in Cyfra21-1, and $132.02{\pm}209.35\;U/L$, $45.86{\pm}75.86\;U/L$ in TPA-M respectively. Serum values of Cyfra21-1 and TPA-M in lung cancer group were higher than control group (p<0.05). Using cutoff value recommended by the manufactures, that is $2.5{\mu}/L$ in CEA, $3.0{\mu}/L$ in Cyfra21-1, 70.0 U/L in TPA-M, and $2.0{\mu}/L$ in SCC, sensitivity and specificity of lung cancer were 33.3%, 78.6% in CEA, 50.0%, 89.7% in Cyfra21-1, 52.3%, 89.7% in TPA-M, 23.8%, 89.3% in SCC. Sensitivity and specificity of nonsmall cell lung cancer were 36.1%, 78.1% in CEA, 50.1%, 89.7% in Cyfra21-1, 53.1%, 89.7% in TPA-M, 33.8%, 89.3% in SCC. Sensitivity and specificity of small cell lung cancer were 25.0%, 78.5% in CEA, 50.0%, 89.6% in Cyfra21-1, 50.0%, 89.6% in TPA-M, 0%, 89.2% in SCC. Cutoff value according to ROC(Receiver operating characteristics) curve was $1.25{\mu}/L$ in CEA, $1.5{\mu}/L$ in Cyfra2-1, 35 U/L in TPA-M, $0.6{\mu}/L$ in SCC. With this cutoff value, sensitivity, specificity, accuracy and kappa index of Cyfra21-1 and TPA-M were better than CEA and SCC. SCC only was related with statistic significance to TNM stages, dividing to operable stages(TNM stage I to IIIA) and inoperable stages (IIIB and IV) (p<0.05). But no tumor markers showed any correlation with significance with tumor size(p>0.05). Conclusion : Serum TPA-M and Cyfra21-1 shows higher sensitivity and specificity than CEA and SCC in overall lung cancer and nonsmall cell lung cancer those were confirmed pathologically. SCC has higher specificity in nonsmall cell lung cancer. And the level of serum sec are signiticantly related with TNM staging.

• Journal of Intelligence and Information Systems
• /
• v.26 no.4
• /
• pp.127-148
• /
• 2020

The data center is a physical environment facility for accommodating computer systems and related components, and is an essential foundation technology for next-generation core industries such as big data, smart factories, wearables, and smart homes. In particular, with the growth of cloud computing, the proportional expansion of the data center infrastructure is inevitable. Monitoring the health of these data center facilities is a way to maintain and manage the system and prevent failure. If a failure occurs in some elements of the facility, it may affect not only the relevant equipment but also other connected equipment, and may cause enormous damage. In particular, IT facilities are irregular due to interdependence and it is difficult to know the cause. In the previous study predicting failure in data center, failure was predicted by looking at a single server as a single state without assuming that the devices were mixed. Therefore, in this study, data center failures were classified into failures occurring inside the server (Outage A) and failures occurring outside the server (Outage B), and focused on analyzing complex failures occurring within the server. Server external failures include power, cooling, user errors, etc. Since such failures can be prevented in the early stages of data center facility construction, various solutions are being developed. On the other hand, the cause of the failure occurring in the server is difficult to determine, and adequate prevention has not yet been achieved. In particular, this is the reason why server failures do not occur singularly, cause other server failures, or receive something that causes failures from other servers. In other words, while the existing studies assumed that it was a single server that did not affect the servers and analyzed the failure, in this study, the failure occurred on the assumption that it had an effect between servers. In order to define the complex failure situation in the data center, failure history data for each equipment existing in the data center was used. There are four major failures considered in this study: Network Node Down, Server Down, Windows Activation Services Down, and Database Management System Service Down. The failures that occur for each device are sorted in chronological order, and when a failure occurs in a specific equipment, if a failure occurs in a specific equipment within 5 minutes from the time of occurrence, it is defined that the failure occurs simultaneously. After configuring the sequence for the devices that have failed at the same time, 5 devices that frequently occur simultaneously within the configured sequence were selected, and the case where the selected devices failed at the same time was confirmed through visualization. Since the server resource information collected for failure analysis is in units of time series and has flow, we used Long Short-term Memory (LSTM), a deep learning algorithm that can predict the next state through the previous state. In addition, unlike a single server, the Hierarchical Attention Network deep learning model structure was used in consideration of the fact that the level of multiple failures for each server is different. This algorithm is a method of increasing the prediction accuracy by giving weight to the server as the impact on the failure increases. The study began with defining the type of failure and selecting the analysis target. In the first experiment, the same collected data was assumed as a single server state and a multiple server state, and compared and analyzed. The second experiment improved the prediction accuracy in the case of a complex server by optimizing each server threshold. In the first experiment, which assumed each of a single server and multiple servers, in the case of a single server, it was predicted that three of the five servers did not have a failure even though the actual failure occurred. However, assuming multiple servers, all five servers were predicted to have failed. As a result of the experiment, the hypothesis that there is an effect between servers is proven. As a result of this study, it was confirmed that the prediction performance was superior when the multiple servers were assumed than when the single server was assumed. In particular, applying the Hierarchical Attention Network algorithm, assuming that the effects of each server will be different, played a role in improving the analysis effect. In addition, by applying a different threshold for each server, the prediction accuracy could be improved. This study showed that failures that are difficult to determine the cause can be predicted through historical data, and a model that can predict failures occurring in servers in data centers is presented. It is expected that the occurrence of disability can be prevented in advance using the results of this study.