• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.3
• /
• pp.570-590
• /
• 2024

Breast cancer ranks among the most prevalent forms of malignancy and foremost cause of death by cancer worldwide. It is not preventable. Early and precise detection is the only remedy for lowering the rate of mortality and improving the probability of survival for victims. In contrast to present procedures, thermography aids in the early diagnosis of cancer and thereby saves lives. But the accuracy experiences detrimental impact by low sensitivity for small and deep tumours and the subjectivity by physicians in interpreting the images. Employing deep learning approaches for cancer detection can enhance the efficacy. This study explored the utilization of thermography in early identification of breast cancer with the use of a publicly released dataset known as the DMR-IR dataset. For this purpose, we employed a novel approach that entails the utilization of a pre-trained MobileNetV2 model and fine tuning it through transfer learning techniques. We created three models using MobileNetV2: one was a baseline transfer learning model with weights trained from ImageNet dataset, the second was a fine-tuned model with an adaptive learning rate, and the third utilized early stopping with callbacks during fine-tuning. The results showed that the proposed methods achieved average accuracy rates of 85.15%, 95.19%, and 98.69%, respectively, with various performance indicators such as precision, sensitivity and specificity also being investigated.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.1
• /
• pp.149-158
• /
• 2009

Tsutsugamushi fever occupies more than 80% of total fall epidemic diseases and has an incubation period of 1 or 2 weeks as well. We have assumed that the incubation period distribution is gamma and therefore, reach an agreement that the infected distribution is normal with ${\hat{\mu}}=309.92$, ${\hat{\sigma}}=14.154$ by back calculation method. The infection cases are found severely large around the month of October. The infection case distribution demonstrates the incidence number increasing rapidly and progresses fast during the month of November. In this study, we have calculated the future prevalence number of maximum 1,200 people by inferred infection probability and incubation period distribution with some sort of limitation that the trend of increasing incidence number is not taking into an account. We considered the SIRS model which is also known as epidemic model, familiar to interaction between epidemiological classes. Our estimated parameters converged well with the initial parameter values.

• Journal of Trauma and Injury
• /
• v.30 no.4
• /
• pp.126-130
• /
• 2017

Purpose: Preventable Trauma Death Rate (PTDR) using Trauma and Injury Severity Score (TRISS) has been most widely used as a quality indicator in South Korea. However, this method has a small number of deaths corresponding to the denominator. Therefore, it is difficult to check the change of quality improvement for annual mortality, and there is a disadvantage that variation is severe. Therefore, we attempted to improve the quality of the mortality evaluation by reducing the variation by applying the PARK Index (preventable major trauma death rate, PMTDR) which can increase the number of denominator significantly. And the Save score (S-score) was also examined as another quality indicator. Methods: In the PARK Index, the denominator is number of all patients who have survival probability (Ps) larger than 0.25. Numerator is the number of deaths among these. The PARK Index includes only patients with ISS >15. The S-score is calculated in the same way as the W-score, but the S-score includes only patients with ISS >15, which is a difference from the W-score. Results: PARK Index decreased annually and was 12.9 (37/287) in 2014, 9.6 (33/343) in 2015, and 7.3 (52/709) in 2016. S-score increased annually and was -0.29 in 2014, 4.21 in 2015, and 8.75 in 2016. Conclusions: PARK Index and S-score improved annually. This shows that both quality indicators are improving year by year. PARK Index (PMTDR) has 9.5-fold increase in denominator overall compared to PTDR by TRISS. The S-score used only ISS >15 patients as a denominator. Therefore, there is an advantage that the numerical value change is larger than the W-score. In addition, S-score is not affected by the ratio of major trauma patients to minor trauma patients.

• Journal of Advanced Navigation Technology
• /
• v.27 no.6
• /
• pp.746-754
• /
• 2023

Aircraft accidents are characterized by a low probability of survival compared to other means of transportation, and the main causes appear to be human factors such as violation of regulations and communication. In order to activate the safety management system to prevent such accidents, an important key variable is to recognize the importance of safety culture and actively engage in safety behavior rather than simply emphasizing compliance with regulations to flight crew members. Even if there are well-established regulations, safety culture, The effectiveness varies depending on the safety atmosphere and level of safety behavior. In this study, the correlation between safety culture and safety behavior was verified through a survey of domestic flight crew members' awareness of safety culture. The results showed that fair culture and self-reporting were not activated enough to have a significant impact on safety behavior. We aim to improve the performance of the safety management system by confirming the characteristics of safety culture and safety behavior.

• Journal of Chest Surgery
• /
• v.24 no.6
• /
• pp.541-546
• /
• 1991

The Ionescu-Shiley bovine pericardial xenograft valve was the most common cardiac substitute valve at Seoul National University Hospital. Since the follow-up extended for longer than 10 years, a total of and consecutive 107 patients with double mitral and aortic valve replacement using this valve from May 1979 to June 1984 were studied for the long-term clinical results. Their ages were 34.0$\pm$11.9 years at surgery, and eight patients died within 30 days of surgery with operative mortality rate of 7.5%. Ninety-nine early survivors were followed up for a total of 488.1 patient-years[Mean$\pm$SD, 4.9$\pm$2.7 years], and nine died with the linearlized late mortality rate of 1.84% /patient-year[pt-yr]. They experienced major complications: thromboembolism, 0.615% /pt-yr bleeding, 0.205% /pt-yr; endocarditis, l. 639%/pt-yr; overall valve failure, 6.146% /pt-yr; and primary tissue failure, 1.639%/ pt-yr. The actuarial survival rates were 91.4$\pm$2.9% and 89.6$\pm$3.4% at postoperative 5 and 10 years, and the probability of freedom from thromboembolism was 95.8$\pm$2.5% at 10 years. The primary tissue failure began to occur from postoperative 6 years and the probabilities of freedom from structural valve failure were 80.2$\pm$7.9% and 62.3$\pm$12.7% at 8 and 10 years after surgery respectively. Although there was increasing number of patients with valve tissue failure after 6 years, the evidence of expected premature and accelerated valve degeneration among young population was not clear on the age-related analysis. And, no definite cumulative patient groups beyond the various age limits could be suggested for or against the use of this valve.

• International Journal of Internet, Broadcasting and Communication
• /
• v.13 no.2
• /
• pp.231-243
• /
• 2021

In Recent years the way we analyze the breast cancer has changed dramatically. Breast cancer is the most common and complex disease diagnosed among women. There are several subtypes of breast cancer and many options are there for the treatment. The most important is to educate the patients. As the research continues to expand, the understanding of the disease and its current treatments types, the researchers are constantly being updated with new researching techniques. Breast cancer survival rates have been increased with the use of new advanced treatments, largely due to the factors such as earlier detection, a new personalized approach to treatment and a better understanding of the disease. Many machine learning classification models have been adopted and modified to diagnose the breast cancer disease. In order to enhance the performance of classification model, our research proposes a model using A Hybrid Modified K-Means Clustering with Modified SVM (Support Vector Machine) Machine learning algorithm to create a new method which can highly improve the performance and prediction. The proposed Machine Learning model is to improve the performance of machine learning classifier. The Proposed Model rectifies the irregularity in the dataset and they can create a new high quality dataset with high accuracy performance and prediction. The recognized datasets Wisconsin Diagnostic Breast Cancer (WDBC) Dataset have been used to perform our research. Using the Wisconsin Diagnostic Breast Cancer (WDBC) Dataset, We have created our Model that can help to diagnose the patients and predict the probability of the breast cancer. A few machine learning classifiers will be explored in this research and compared with our Proposed Model "A Hybrid Modified K-Means with Modified SVM Machine Learning Algorithm to Enhance the Cancer Prediction" to implement and evaluated. Our research results show that our Proposed Model has a significant performance compared to other previous research and with high accuracy level of 99% which will enhance the Cancer Prediction.

• Asian-Australasian Journal of Animal Sciences
• /
• v.32 no.4
• /
• pp.519-526
• /
• 2019

Objective: We determined the risk factors limiting first service conception (FSC) rate in dairy cows and their economic impact. Methods: Data were collected from 790 lactations regarding cow parity, peri- and postpartum disorders, body condition score (BCS), reproductive performance, and expenses associated with reproductive management (treatment, culling, and others). Initially, we identified the risk factors limiting FSC rate in dairy cows. Various biological and environmental factors, such as herd, cow parity, BCS at 1 month postpartum and first artificial insemination (AI), resumption of cyclicity within 1 month of calving, year, AI season, insemination at detected estrus or timed AI, peri- and postpartum disorders, and calving to first AI interval, were evaluated. Next, we evaluated the economic impact of the success or failure of FSC by comparing the expense associated with reproductive management until conception between cows that did or did not conceive at their first service. Results: Cows with BCS <3.0 had a lower probability of conceiving at first insemination (odds ratio [OR] = 0.64, p<0.05) than cows with $BCS{\geq}3.0$. Cows inseminated during summer were less likely to conceive (OR = 0.44, p<0.001) than cows inseminated during spring. Cows with peri- or postpartum disorders were less likely to conceive (OR = 0.55, p<0.001) than cows without disorders. Survival curves generated using MedCalc showed an 81 day extension in the mean interval between calving and conception in cows that failed to conceive over those that did conceive at first insemination. Cows failing conceive required additional expenditure on reproductive treatment ($55.40) and other management ($567.00) than cows that conceived at first insemination. Conclusion: Lower BCS, hot weather at first insemination, and peri- and postpartum disorders are risk factors limiting FSC, which result in an economic loss of $622.40 per dairy cow.

• Journal of KIISE:Information Networking
• /
• v.34 no.6
• /
• pp.458-465
• /
• 2007

Wireless Sensor Network(WSN) is a special network that collects measured data by sensor nodes in the predefined sensor field and forwards them to the base station in a distance using their own routing scheme. WSN requires routing techniques to maximize energy efficiency because sensor nodes have non-rechargeable and thus limited energy. Characteristics of WSN are various according to applications, many of routing algorithms have been proposed. This paper proposes an algorithm called A-PEGASIS that basically bases on PEGASIS and enhances in two aspects - an elegant chain generation algorithm and periodical update of chains. We compare performance of the previous algorithm of LEACH, PEGASIS, PEDAP and PEDAP-PA with ours through simulation. It confirms that the A-PEGASIS is most superior in terms of average WSN lifetime and high probability of node survival rate during WSN life time.

• International Journal of Highway Engineering
• /
• v.13 no.4
• /
• pp.71-79
• /
• 2011

Korean national highway has been increased 2 times more for the past ten years because of many highway geometric improvements and new routes since 2000. According to the reasons, maintenance cost has been increased continuously. Deterioration of concrete bridge decks caused by asphalt pavement deformation occupies a high proportion of overall bridge management budget. The number of current highway bridges has reached over 7,800 in 2011, and It is difficult to determine to some future budget. This study predicted the remaining life of concrete bridge decks using the reliability analysis based on Weibull distribution. and The expected future maintenance cost was estimated.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.10
• /
• pp.82-88
• /
• 2018

The decoy defensive weapon system is a one-shot system. Reliability is maintained through periodic inspection and high reliability is required to confirm whether or not the functioning is normal after launch. The maintenance cycle of a decoy was set up without target reliability and reliability prediction during the development period. However, the number of operations in the military has been increasing, necessitating the optimization of the maintenance cycle. Reliability is analyzed using the maintenance data of a decoy operated for several decades and the optimal maintenance cycle is suggested. In chapter 2, data collection and classification methods are presented and analysis methodology is briefly introduced. In chapter 3, the data distribution analysis and fitness verification confirmed that applying the Weibull distribution is the most suitable for the maintenance data of the decoy. In chapter 4, we present the analysis result of percentile, survival probability and MTBF and the optimal maintenance cycle was derived from the reliability analysis. Finally, we suggest the application methods for this paper in the future.