• The Journal of the Acoustical Society of Korea
• /
• v.26 no.5
• /
• pp.194-198
• /
• 2007

In this paper, we apply a discriminative weight training to a statistical model-based voice activity detection(VAD). In our approach, the VAD decision rule is expressed as the geometric mean of optimally weighted likelihood ratios(LRs) based on a minimum classification error(MCE) method which is different from the previous works in that different weights are assigned to each frequency bin which is considered more realistic. According to the experimental results, the proposed approach is found to be effective for the statistical model-based VAD using the LR test.

• Nuclear Engineering and Technology
• /
• v.51 no.4
• /
• pp.1169-1175
• /
• 2019

The minimum detectable activity (MDA) value was derived according to the flow rate of the sample and degree of amplification of the device by sending the sample directly from the collection site to the detection part through a pump. This method can lead to reduction in time and cost compared to the existing measurement method that uses a pre-treatment process. In this study, experiments were conducted on $^3H$ and $^{90}Sr$, which are the major pure beta-emitting radionuclides, by setting the sample flow rate and the amplification gain as factors. The MDA values were derived according to the flow rates, considering that the flow rate can affect the MDA values. There were no change in the MDA under different flow rates of 0, 600, 800, and 1000 mL/min. Therefore, it was confirmed that the flow rate may not be considered when collecting samples for monitoring in actual field. As the degree of amplification of the amplifier increased, the time required to reach the target MDA decreased. When the amplification was quadrupled, the detection efficiency increased by approximately 23.4 times, and the time to reach the MDA decreased to approximately 1/550 times. This method offers the advantage of real-time on-site monitoring.

• Journal of Radiation Protection and Research
• /
• v.24 no.3
• /
• pp.171-184
• /
• 1999

In this paper, using statistical theory the basic concepts and the formulars of lower limit of detection were deasribed. Assuming that risks of 5% are acceptable (95% confidence level), lower limits of detection were calculated on the measuring apparatus for alpha, beta and gamma nuclides. Also, on the basis concepts of lower limit of detection, the MDA values were calculated for the radionuclides detected easily in the environment. These results make it possible to evaluate confidence limits on the radioanalytical results in the environmental sample.

• Safety and Health at Work
• /
• v.10 no.3
• /
• pp.389-392
• /
• 2019

When conducting an exposure assessment, the primary goal of the industrial hygienist is to fully characterize the worker's exposure during a work shift to compare it with an occupational exposure limit. This applies regardless of the duration of the work activity as an activity that is relatively short in duration can still present exposure in excess of the occupational exposure limit even when normalized over an 8-hr shift. This goal, however, is often impeded by the specification of a minimum sample volume in the published sampling method, which may prevent the sample from being collected or submitted for analysis. Removing the specification of minimum sample volume (or adjusting it from a requirement to a recommendation), in contrast, allows for a broader assessment of jobs that consist of short-duration and high-exposure activities and also eliminates the unnecessary practice of running sampling pumps in clean air to collect a specified, minimum volume.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.45 no.4
• /
• pp.100-106
• /
• 2008

In this paper we apply a discriminative weight training employing power spectral flatness measure (PSFM) to a statistical model-based voice activity detection (VAD) in various noise environments. In our approach, the VAD decision rule is expressed as the geometric mean of optimally weighted likelihood ratio test (LRT) based on a minimum classification error (MCE) method which is different from the previous works in th at different weights are assigned to each frequency bin and noise environments depending on PSFM. According to the experimental results, the proposed approach is found to be effective for the statistical model-based VAD using the LRT.

• The Journal of the Acoustical Society of Korea
• /
• v.27 no.8
• /
• pp.443-449
• /
• 2008

One of the key issues in practical speech processing is to achieve robust Voice Activity Deteciton (VAD) against the background noise. Most of the statistical model-based approaches have tried to employ equally weighted likelihood ratios (LRs), which, however, deviates from the real observation. Furthermore voice activities in the adjacent frames have strong correlation. In other words, the current frame is highly correlated with previous frame. In this paper, we propose the effective VAD approach based on a minimum classification error (MCE) method which is different from the previous works in that different weights are assigned to both the likelihood ratio on the current frame and the decision statistics of the previous frame.

• Analytical Science and Technology
• /
• v.14 no.1
• /
• pp.64-71
• /
• 2001

In this paper, the formula of lower limit of detection considering the measuring time of sample and background was derived using the basic concepts of lower limit of detection. Among parameters affecting the determination of MDA value, the MDA values were calculated with variation of amount of sample and measuring time of sample and background. The results of adequate division between the measuring time of sample and that of background studied in this research make it possible to evaluate confidence limits on the radio analytical results in the environmental sample.

• Journal of the Korean Society of Radiology
• /
• v.5 no.1
• /
• pp.5-10
• /
• 2011

Based on basic concept of detection limit, sample measurement time & background measurement time was considered, and MDA values according to background measurement time and sample measurement time in land samples(river soil, surface soil, drinking water, underground water, surface water, pine leaf, mugwort) analysis among environmental samples were compared. Seeing the water sample analysis result, it was shown that most of the samples were not detected, and most of the samples in land specimen analysis showed to be below the detection limit of "Ministry of Education, Science and Technology Announcement Je-2008-28-ho", but $^{137}Cs$ which is one of artificial radioactive nuclide was detected in some samples. It can be traced back to 1950s and 1960s when nuclear tests were carried out in atmosphere and catastrophic Chernobyl atomic power station accident that caused fallouts in the sky, and this is common level of detection that can be observed worldwide. Seeing the result that the $^{134}Cs$(which is a isotope of $^{137}Cs$, and it has relatively short half life) was not detected in all samples, it can be considered it doesn't affect to the operation of atomic power station.

• Journal of Ginseng Research
• /
• v.41 no.2
• /
• pp.180-187
• /
• 2017

Background: The incidence of halitosis has a prevalence of 22-50% throughout the world and is generally caused by anaerobic oral microorganisms, such as Fusobacterium nucleatum, Clostridium perfringens, and Porphyromonas gingivalis. Previous investigations on the structure-activity relationships of ginsenosides have led to contrasting results. Particularly, the antibacterial activity of less polar ginsenosides against halitosis-related bacteria has not been reported. Methods: Crude saponins extracted from the Panax quinquefolius leaf-stem (AGS) were treated at $130^{\circ}C$ for 3 h to obtain heat-transformed saponins (HTS). Five ginsenoside-enriched fractions (HTS-1, HTS-2, HTS-3, HTS-4, and HTS-5) and less polar ginsenosides were separated by HP-20 resin absorption and HPLC, and the antimicrobial activity and mechanism were investigated. Results: HPLC with diode-array detection analysis revealed that heat treatment induced an extensive conversion of polar ginsenosides (-Rg1/Re, -Rc, -Rb2, and -Rd) to less polar compounds (-Rg2, -Rg3, -Rg6, -F4, -Rg5, and -Rk1). The antimicrobial assays showed that HTS, HTS-3, and HTS-4 were effective at inhibiting the growth of F. nucleatum, C. perfringens, and P. gingivalis. Ginsenosides-Rg5 showed the best antimicrobial activity against the three bacteria, with the lowest values of minimum inhibitory concentration and minimum bactericidal concentration. One major reason for this result is that less polar ginsenosides can more easily damage membrane integrity. Conclusion: The results indicated that the less polar ginsenoside-enriched fraction from heat transformation can be used as an antibacterial agent to control halitosis.

• Journal of Veterinary Science
• /
• v.22 no.6
• /
• pp.41.1-41.14
• /
• 2021

Background: Our previously prepared ceftiofur (CEF) hydrochloride oily suspension shows potential wide applications for controlling swine Streptococcus suis infections, while the irrational dose has not been formulated. Objectives: The rational dose regimens of CEF oily suspension against S. suis were systematically studied using a pharmacokinetic-pharmacodynamic model method. Methods: The healthy and infected pigs were intramuscularly administered CEF hydrochloride oily suspension at a single dose of 5 mg/kg, and then the plasma and pulmonary epithelial lining fluid (PELF) were collected at different times. The minimum inhibitory concentration (MIC), minimal bactericidal concentration, mutant prevention concentration (MPC), post-antibiotic effect (PAE), and time-killing curves were determined. Subsequently, the area under the curve by the MIC (AUC_0-24h/MIC) values of desfuroylceftiofur (DFC) in the PELF was obtained by integrating in vivo pharmacokinetic data of the infected pigs and ex vivo pharmacodynamic data using the sigmoid E_max (Hill) equation. The dose was calculated based on the AUC_0-24h/MIC values for bacteriostatic action, bactericidal action, and bacterial elimination. Results: The peak concentration, the area under the concentration-time curve, and the time to peak for PELF's DFC were 24.76 ± 0.92 ㎍/mL, 811.99 ± 54.70 ㎍·h/mL, and 8.00 h in healthy pigs, and 33.04 ± 0.99 ㎍/mL, 735.85 ± 26.20 ㎍·h/mL, and 8.00 h in infected pigs, respectively. The MIC of PELF's DFC against S. suis strain was 0.25 ㎍/mL. There was strong concentration-dependent activity as determined by MPC, PAE, and the time-killing curves. The AUC_0-24h/MIC values of PELF's DFC for bacteriostatic activity, bactericidal activity, and virtual eradication of bacteria were 6.54 h, 9.69 h, and 11.49 h, respectively. Thus, a dosage regimen of 1.94 mg/kg every 72 h could be sufficient to reach bactericidal activity. Conclusions: A rational dosage regimen was recommended, and it could assist in increasing the treatment effectiveness of CEF hydrochloride oily suspension against S. Suis infections.