• 대한전기학회논문지:전기물성ㆍ응용부문C
• /
• 제48권4호
• /
• pp.247-252
• /
• 1999

In this paper, in order to investigate availability of the OW-PD measurement method which has been proposed as an alternative of AC-PD measurement method to an after laying test and/or diagnosis for the power cable system, partial discharges owing to the needle-type defect integrated into the cable have been measured using AC and OW(Oscillating Wave) voltages. In the AC-PD measurement, the magnitude, phase and pulse number of partial discharges have been changed with the duration of voltage application, which can be analyzed through the relation with the process of the electrical tree initiation and propagation. In addition, the characteristics of partial discharges using OW voltage are appeared to be similar to those in case of AC-PD measurement and to be different with the shapes of electrical tree. From these results, it is concluded that the OW-PD measurement method is available to the tests for the cable system.

• 한국멀티미디어학회논문지
• /
• 제16권3호
• /
• pp.269-280
• /
• 2013

본 논문에서 망막 질환 요인간의 의존도 분석을 위해 효율적인 분류기를 활용할 수 있는 방안을 제시하였다. 먼저 여러 베이지안 네트워크 중에서 TAN (Tree-Augmented Naive Bayesian Network), GBN(General Bayesian Network)과 Markov Blanket으로 특징축소된 GBN과의 분류성능과 예측정확률을 비교분석하였다. 그리고 처음으로, 높은 성능을 보인 TAN을 망막 질환 임상데이터의 의존도 분석에 적용하였다. 의존도 분석 결과, 망막 질환의 진단과 예후 예측에 활용의 가능성을 보였다.

• International Journal of Internet, Broadcasting and Communication
• /
• 제15권3호
• /
• pp.32-42
• /
• 2023

There are serious problems worldwide, such as a pandemic due to an unprecedented infection caused by COVID-19. On previous approaches, they invented medical vaccines and preemptive testing tools for medical engineering. However, it is difficult to access poor medical systems and medical institutions due to disparities between countries and regions. In advanced nations, the damage was even greater due to high medical and examination costs because they did not go to the hospital. Therefore, from a software engineering-based perspective, we propose a learning model for determining coronavirus infection through symptom data-based software prediction models and tools. After a comparative analysis of various models (decision tree, Naive Bayes, KNN, multi-perceptron neural network), we decide to choose an appropriate decision tree model. Due to a lack of data, additional survey data and overseas symptom data are applied and built into the judgment model. To protect from thiswe also adapt human normalization approach with traditional Korean medicin approach. We expect to be possible to determine coronavirus, flu, allergy, and cold without medical examination and diagnosis tools through data collection and analysis by applying decision trees.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제18권4호
• /
• pp.860-880
• /
• 2024

Breast cancer is a major health concern affecting women and men globally. Early detection and accurate classification of breast cancer are vital for effective treatment and survival of patients. This study addresses the challenge of accurately classifying breast tumors using machine learning classifiers such as MLP, AdaBoostM1, logit Boost, Bayes Net, and the J48 decision tree. The research uses a dataset available publicly on GitHub to assess the classifiers' performance and differentiate between the occurrence and non-occurrence of breast cancer. The study compares the 10-fold and 5-fold cross-validation effectiveness, showing that 10-fold cross-validation provides superior results. Also, it examines the impact of varying split percentages, with a 66% split yielding the best performance. This shows the importance of selecting appropriate validation techniques for machine learning-based breast tumor classification. The results also indicate that the J48 decision tree method is the most accurate classifier, providing valuable insights for developing predictive models for cancer diagnosis and advancing computational medical research.

• 모자간호학회지
• /
• 제4권1호
• /
• pp.3-11
• /
• 1994

Presently there is an increasing demand for home health care services due to changes in the demographic structure as a result of an increasing elderly population, socio-economic improvements, and changes in the family structure, as well as the growing number of people with degenerative diseases. In addition to these reasons, rising medical costs and there a shortage of patient beds space in the hospital, particularly since introduction of national medical insurance. There has been an increasing demand for health care health care services. This study was done to identify the basic data for home health care management. It focused on developing client selection criteria, assessment tools, and recording methods. This was accomplished by the researchers visiting the patients in their homes. The research process included preparation investigation, tool development, training of the project researcher, and visiting the clients in their homes. The research tools are as follows : 1. Record development : a) The selection criteria tool for home health care of postpartum women was a structured tool and consisted of four parts. b) The structured assessment tool consisted of a general items, obstetric history, past medical history, methods of feeding, medications taken before admission, laboratory test results, discharge instructions, discharge medications, family tree, economic status, environmental status, a map, health assessment of postpartum women and their newborns. c) The visit note I consisted of the frequency of visits. Visit note II consisted of the date ; nursing problems ; nursing process including the initial assessment ; nursing goal ; visit plan ; postpartum women and their neonate health status, diagnosis, goal, implementation, evaluation, summary, next plan, for visit revision. d) Problem note consisted of the date, problem numbers, nursing diagnosis, problem appearance date problem resolution date. The research results are as follows : 1. Nursing problems : The nursing problems of the postpartum women and their neonates were evaluated by the number of nursing diagnoses and the change in the pattern of nursing diagnosis related to the number of visits. a) Nursing diagnosis The nursing diagnosis was classified according to physical function, psychosocial function, family system maintained function. b) The changes of nursing diagnosis related to the number of visits. As the type of nursing diagnosis changed related to the number of visits the number of nursing diagnoses decreased. 2. Contents of home health care : The content was categorized according to assessment, direct care, counseling, education, family care, reporting to with the attending doctor. The recommendations based on the research results are as follows : 1. Tool development Replication of this study is needed to test the validity of the assessment tools used. 2. Home visit a) Home health care nurses should be licensed and qualified. A referral form from the attending doctor is needed for legal protection of nurses. b) The first home visit need to be within 24 hours of discharge from the hospital to decrease the anxiety of frightened postpartum women. c) When the changes occur in the newborn's status, home health care nurses should consult a pediatrician. Communication within the home healthcare team is essential and needs to consistent and done smoothly. 3. Home health care A Study is required to develop protocols for education of staff and for operation of all aspects of this program.

• Asian Pacific Journal of Cancer Prevention
• /
• 제17권2호
• /
• pp.835-838
• /
• 2016

Background: Lung cancer remains one of the most common cancers in the world, both in terms of new cases (about 13% of total per year) and deaths (nearly one cancer death in five), because of the high case fatality. Errors in lung cancer type or malignant growth determination lead to degraded treatment efficacy, because anticancer strategy depends on tumor morphology. Materials and Methods: We have made an attempt to evaluate effectiveness of machine learning algorithms in the task of lung cancer classification based on gene expression levels. We processed four publicly available data sets. The Dana-Farber Cancer Institute data set contains 203 samples and the task was to classify four cancer types and sound tissue samples. With the University of Michigan data set of 96 samples, the task was to execute a binary classification of adenocarcinoma and non-neoplastic tissues. The University of Toronto data set contains 39 samples and the task was to detect recurrence, while with the Brigham and Women's Hospital data set of 181 samples it was to make a binary classification of malignant pleural mesothelioma and adenocarcinoma. We used the k-nearest neighbor algorithm (k=1, k=5, k=10), naive Bayes classifier with assumption of both a normal distribution of attributes and a distribution through histograms, support vector machine and C4.5 decision tree. Effectiveness of machine learning algorithms was evaluated with the Matthews correlation coefficient. Results: The support vector machine method showed best results among data sets from the Dana-Farber Cancer Institute and Brigham and Women's Hospital. All algorithms with the exception of the C4.5 decision tree showed maximum potential effectiveness in the University of Michigan data set. However, the C4.5 decision tree showed best results for the University of Toronto data set. Conclusions: Machine learning algorithms can be used for lung cancer morphology classification and similar tasks based on gene expression level evaluation.

• 한국전산구조공학회논문집
• /
• 제17권2호
• /
• pp.131-140
• /
• 2004

철근콘크리트 구조물의 보수ㆍ보강 등의 유지관리를 위해서는 내구성과 내하성을 동시에 고려한 건전성평가의 의사결정기준이 절실히 요구된다. 본 논문은 CART-ANFIS을 사용하는 철근콘크리트 구조물에 대하여 효율적인 모델을 나타내었다. 철근콘크리트 구조물의 손상과 진단 등에 활용되어온 분류형 전문가시스템의 일종인 퍼지이론을 이용한 결정목 구조와 기존의 인공신경망을 이용한 결정목 구조의 건전성평가를 비교 분석한다. 손상된 철근콘크리트의 내구성 회복을 위한 보강설계 이론과 내하력 증가를 위한 보장설계 이론을 정립시켜 손상검출의 산정식을 유도하였다. 본 연구의 건전성 평가시스템 모델을 이용함으로서 보다 효율적인 철근콘크리트 유지관리 뿐만 아니라 생애주기비용 예측을 수행 할 수 있다.

• Journal of Korean Biological Nursing Science
• /
• 제21권1호
• /
• pp.85-97
• /
• 2019

Purpose: This study was performed to analyze the levels of blood pressure and to identify good or poor blood pressure control (BPC) groups among hypertension patients. The study was based on the Korea National Health and Nutrition Examination Survey (KNHANES VI and VII) conducted from 2013 to 2016. Methods: The sociodemographic and clinical data of 4,151 Korean hypertension patients aged 20-79 years and who were taking antihypertensive medications was extracted from the KNHANES VI and VII database. Descriptive statistics for complex samples and a decision-tree analysis were performed using the SPSS WIN 24.0 program. Results: The mean age was $62.46{\pm}0.21years$. The mean systolic blood pressure (SBP) was $128.07{\pm}0.28mmHg$, and the diastolic blood pressure (DBP) was $76.99{\pm}0.21mmHg$. 71.9% of participants showed normal blood pressure (SBP < 140mmHg and DBP < 90mmHg). From the decisiontrees analysis, the characteristics of participants related to good BPC group were presented with 9 different pathways same as those from the poor BPC group. Good or poor BPC groups were classified according to the patients' characteristics such as age, living status, occupation, education, hypertension diagnosis period, numbers of comorbidity, perceived health status, total cholesterol, high density lipoprotein-cholesterol, alcohol drinking per month, and depressive mood. Total cholesterol level (< 201mg/dL or ${\geq}201mg/dL$ cutoff point) was the most significant predictor of the participants' BPC group. Conclusion: This decision-tree model with the 18 different pathways can form a basis for the screening of hypertension patients with good or poor BPC in either clinical or community settings.

• International Journal of Computer Science & Network Security
• /
• 제23권9호
• /
• pp.150-156
• /
• 2023

For a doctor, diagnosing a patient's heart disease is not easy. It takes the ability and experience with high flying hours to be able to accurately diagnose the type of patient's heart disease based on the existing factors in the patient. Several studies have been carried out to develop tools to identify types of heart disease in patients. However, most only focus on the results of patient answers and lab results, the rest use only echocardiography data or electrocardiogram results. This research was conducted to test how accurate the results of the classification of heart disease by using two medical data, namely echocardiography and electrocardiogram. Three treatments were applied to the two medical data and analyzed using the decision tree approach. The first treatment was to build a classification model for types of heart disease based on echocardiography and electrocardiogram data, the second treatment only used echocardiography data and the third treatment only used electrocardiogram data. The results showed that the classification of types of heart disease in the first treatment had a higher level of accuracy than the second and third treatments. The accuracy level for the first, second and third treatment were 78.95%, 73.69% and 50%, respectively. This shows that in order to diagnose the type of patient's heart disease, it is advisable to look at the records of both the patient's medical data (echocardiography and electrocardiogram) to get an accurate level of diagnosis results that can be accounted for.

• Journal of Korean Biological Nursing Science
• /
• 제25권2호
• /
• pp.131-142
• /
• 2023

Purpose: The aim of this study was to assess lipid levels and to identify groups with poor lipid control group among patients with dyslipidemia. Methods: Data from 1,399 Korean patients with dyslipidemia older than 20 years were extracted from the Korea National Health and Nutrition Examination Survey. Complex sample analysis and decision-tree analysis were conducted with using SPSS for Windows version 27.0. Results: The mean levels of total cholesterol (TC), triglyceride (TG), low density lipoprotein-cholesterol (LDL-C), and high density lipoprotein cholesterol were 211.38±1.15 mg/dL, 306.61±1.15 mg/dL, 118.48±1.08 mg/dL, and 42.39±1.15 mg/dL, respectively. About 61% of participants showed abnormal lipid control. Poor glycemic control groups (TC ≥ 200 mg/dL or TG ≥ 150 mg/dL or LDL-C ≥ 130 mg/dL) were identified through seven different pathways via decision-tree analysis. Poor lipid control groups were categorized based on patients' characteristics such as gender, age, education, dyslipidemia medication adherence, perception of dyslipidemia, diagnosis of myocardial infarction or angina, diabetes mellitus, perceived health status, relative hand grip strength, hemoglobin A1c, aerobic exercise per week, and walking days per week. Dyslipidemia medication adherence was the most significant predictor of poor lipid control. Conclusion: The findings demonstrated characteristics that are predictive of poor lipid control and can be used to detect poor lipid control in patients with dyslipidemia.