• Journal of Preventive Medicine and Public Health
• /
• v.29 no.3 s.54
• /
• pp.639-655
• /
• 1996

Cerebrovascular disease and coronary heart disease are the first and the fourth common causes of death among adults in Korea. Reported risk factors of these diseases are mostly alike. But some risk factors of one of these diseases may prevent other diseases. Therefore, we tried to compare and discriminate the risk factors of these diseases. We recruited four case groups and four control groups among the inpatients who were admitted to Wonju Christian Hospital from March, 1994 to November, 1995. Four control groups were matched with each of four case groups by age and sex. The number of patients in each of four case and control groups were 106 and 168 for acute myocardial infarction(AMI), 84 and 133 for subarachnoid hemorrhage(SAH), 102 and 148 for intracerebral hemorrhage(ICH), and 91 and 182 for ischemic stroke(IS) respectively. Factors whose levels were significantly higher in AMI and IS than in responding control group (RCG) were education, economic status, and triglyceride. Factors whose levels were significantly lower in hemorrhagic stroke than in RCG were age of monarch, and prothrombin time. The factor whose level was higher in AMI than ill RCG was uric acid. The factor whose level was higher in AMI, ICH, and SAM than in RCG was blood sugar. Factors whose levels were significantly higher in all the case groups than in RCG were earlobe crease, Quetelet index, white blood cell count, hemoglobin, hematocrit, and total cholesterol. The list of risk factors were somewhat different among the four diseases, though none of the risk factors to the one disease except prothrombin time acted as a preventive factor to the other diseases. The percent of grouped cases correctly classified was higher in the discrimination of ischemic diseases(AMI and IS) from hemorrhagic diseases(SAM and ICH) than in the discrimination of cerebrovascular disease from AMI. The factors concerned in the discrimination of ischemic diseases from hemorrhagic diseases were prothrombin time, earlobe crease, gender, age, uric acid, education, albumin, hemoglobin, the history of taking steroid, total cholesterol, and hematocrit according to the selection order through forward selection.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.57 no.6
• /
• pp.1058-1062
• /
• 2008

In clinical decision support system(CDSS), unlike rule-based expert method, appropriate data-driven machine learning method can easily provide the information of individual feature(clinical test) for disease classification. However, currently developed methods focus on the improvement of the classification accuracy for diagnosis. With the analysis of feature importance in classification, one may infer the novel clinical test sets which highly differentiate the specific diseases or disease states. In this background, we introduce a novel CDSS that integrate a classifier and feature selection module together. Random forest algorithm is applied for the classifier and the feature importance measure. The system selects the significant clinical tests discriminating the diseases by examining the classification error during backward elimination of the features. The superior performance of random forest algorithm in clinical classification was assessed against artificial neural network and decision tree algorithm by using breast cancer, diabetes and heart disease data in UCI Machine Learning Repository. The test with the same data sets shows that the proposed system can successfully select the significant clinical test set for each disease.

• Clinical and Experimental Pediatrics
• /
• v.62 no.9
• /
• pp.325-333
• /
• 2019

Allergic diseases, including allergic rhinitis, asthma, and atopic dermatitis, are common heterogeneous diseases that encompass diverse phenotypes and different pathogeneses. Phenotype studies of allergic diseases can facilitate the identification of risk factors and their underlying pathophysiology, resulting in the application of more effective treatment, selection of better treatment responses, and prediction of prognosis for each phenotype. In the early phase of phenotype studies in allergic diseases, artificial classifications were usually performed based on clinical features, such as triggering factors or the presence of atopy, which can result in the biased classification of phenotypes and limit the characterization of heterogeneous allergic diseases. Subsequent phenotype studies have suggested more diverse phenotypes for each allergic disease using relatively unbiased statistical methods, such as cluster analysis or latent class analysis. The classifications of phenotypes in allergic diseases may overlap or be unstable over time due to their complex interactions with genetic and encountered environmental factors during the illness, which may affect the disease course and pathophysiology. In this review, diverse phenotype classifications of allergic diseases, including atopic dermatitis, asthma, and wheezing in children, allergic rhinitis, and atopy, are described. The review also discusses the applications of the results obtained from phenotype studies performed in other countries to Korean children. Consideration of changes in the characteristics of each phenotype over time in an individual's lifespan is needed in future studies.

• International Journal of Computer Science & Network Security
• /
• v.23 no.5
• /
• pp.148-162
• /
• 2023

Classification systems can significantly assist the medical sector by allowing for the precise and quick diagnosis of diseases. As a result, both doctors and patients will save time. A possible way for identifying risk variables is to use machine learning algorithms. Non-surgical technologies, such as machine learning, are trustworthy and effective in categorizing healthy and heart-disease patients, and they save time and effort. The goal of this study is to create a medical intelligent decision support system based on machine learning for the diagnosis of heart disease. We have used a mixed feature creation (MFC) technique to generate new features from the UCI Cleveland Cardiology dataset. We select the most suitable features by using Least Absolute Shrinkage and Selection Operator (LASSO), Recursive Feature Elimination with Random Forest feature selection (RFE-RF) and the best features of both LASSO RFE-RF (BLR) techniques. Cross-validated and grid-search methods are used to optimize the parameters of the estimator used in applying these algorithms. and classifier performance assessment metrics including classification accuracy, specificity, sensitivity, precision, and F1-Score, of each classification model, along with execution time and RMSE the results are presented independently for comparison. Our proposed work finds the best potential outcome across all available prediction models and improves the system's performance, allowing physicians to diagnose heart patients more accurately.

• Journal of Korean Medicine Rehabilitation
• /
• v.26 no.3
• /
• pp.129-141
• /
• 2016

Objectives As awareness of the public about Korean medicine health care and the social demand about improvement for quality of health care service is constantly rising, the quality evaluation of Korean medicine health care service is needed to improve the quality. Through trial of Delphi method, we tried to set the priority in short, medium, long term among the disease which is the subject of quality assessment. Methods Carrying out the delphi survey to 50 experts of korean medicine who were recommended by the 41 member societies of Korean medicine and related organizations, we selected final candidates for quality assessment. It is composed with total 2 rounds, and we investigated the priority in three aspects; the importance of the matter, possibility quality assessment, potential about if there's any chance of improvement. Results By delphi method, we set the priority of quality assessment. Base on the result of the second round, we classified importance of the questions into above average, average, below average, and categorized items as short, medium, long term according on the final priority. We classified of musculoskeletal diseases and diseases of connective tissues and musculoskeletal injury as short term and cerebrovascular disease and disease of nerve system and malignant neoplasm as medium term, disease of digestive organs and diseases, symptoms and abnormal findings in clinical field or inspections which are not categorized as long term. Conclusions We set the subjects of quality assessment by delphi survey by experts, and classified into short, medium, long term. Further research is necessary for execution the Quality Assessment to each of the candidate. Also, we can send feedback to medical institution base on the result of Quality Assessment. then it would be able to induce the improvement in quality of medical institution by itself.

• Korean Journal of Acupuncture
• /
• v.38 no.2
• /
• pp.100-109
• /
• 2021

Objectives : Source acupoint is one of the representative acupoints to treat various diseases in each meridian. We aimed to identify the patterns of selection of Source acupoints and their associations with diseases using clinical trials data. Methods : We extracted the frequency of Source acupoints across 30 diseases from clinical trials database. Acupuncture treatment regimens were retrieved from the Cochrane Database of Systematic Reviews. The frequency of Source acupoint use was calculated as the number of studies using a certain acupoint divided by the total number of included studies. Using hierarchical clustering and multidimensional scaling, the characteristics of Source acupoints were analyzed based on the similarity of the relationships between the Source acupoints and the diseases. Results : A total of 421 clinical trials were included for this analysis. LR3, HT7, KI3, and LI4 acupoints were most frequently used for the treatment of 30 diseases. Cluster analysis showed that LR3 and LI4 acupoints were grouped together and HT7 and KI3 acupoints were grouped together. Multidimensional scaling revealed that LR3, LI4, HT7, and KI3 acupoints have intrinsic properties in the two-dimensional space. Conclusions : The present study identified the selection patterns of the Source acupoints using clinical trials data. Our finding will provide the understanding of the characteristics of Source acupoints.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.6
• /
• pp.849-855
• /
• 2008

The Diagnostic Unit of CLINAID can infer working diagnoses for general diseases from the information provided by a user. This user-provided information in the form of signs and symptoms, however, is usually not sufficient to make a final decision on a working diagnosis. In order for the Diagnostic Unit to reach a diagnostic conclusion, it needs to select suitable clinical investigations for the patients. Because different investigations can be selected for the same patient, we need a process that can optimize the selection procedure employed by the Diagnostic Unit. This process, called a selection algorithm, must work with the fuzzy relational method because CLINAID uses fuzzy relational structures extensively for its knowledge bases and inference mechanism. In this paper we present steps of the selection algorithm along with simulation results on this algorithm using fuzzy relational products, both harsh product and mean product. The computation results of applying several different fuzzy implication operators are compared and analyzed.

• Tuberculosis and Respiratory Diseases
• /
• v.75 no.4
• /
• pp.135-139
• /
• 2013

Endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) is becoming a standard method for invasive mediastinal staging and for the diagnosis of paratracheal and peribronchial lesions. It is essential to understand the technical aspects of EBUS-TBNA to ensure safe and efficient procedures. In this review, we discuss the practical aspects to be considered during EBUS-TBNA, including anesthesia, manipulation of equipment, understanding mediastinal ultrasound images, target selection, number of aspirations needed per target, sample handling, and complications.

• Clinical and Experimental Pediatrics
• /
• v.50 no.2
• /
• pp.127-131
• /
• 2007

Fever of unknown origin (FUO) has been a convenient term used to classify patients who warrant a particular systemic approach to diagnostic evaluation and management. The greatest clinical concern in evaluating FUO is identifying patients whose fever has a serious or life-threatening cause when a delay in diagnosis could jeopardize successful intervention. Thorough history and complete physical examination are critical to uncover the etiologic diagnosis. Most cases of FUO in children are caused by atypical presentations of common diseases rather than by typical manifestations of rare disorders. Selection of diagnostic tests and speed of investigation should be guided by a knowledge of the disease severity, patient age, epidemiologic and geographic information, and any positive findings from a detailed history and physical examination. The three most common causes of FUO in children are infectious diseases, connective tissue diseases, and malignancy. In general, the prognosis of FUO in children is better than that of adults. Although the outcome is dependent on the primary disease process, fever abates spontaneously in most cases in whom the cause of fever remains unclear.

• Journal of Multimedia Information System
• /
• v.3 no.2
• /
• pp.1-8
• /
• 2016

In this paper, we propose a method to automatically diagnose various diseases. The input data consists of electrocardiograph (ECG) recordings. We extract R-to-R interval (RRI) signals from ECG recordings, which are preprocessed to remove trends and ectopic beats, and to keep the signal stationary. After that, we perform some prospective analysis to extract time-domain parameters, frequency-domain parameters, and nonlinear parameters of the signal. Those parameters are unique for each disease and can be used as the statistical symptoms for each disease. Then, we perform feature selection to improve the performance of the diagnosis classifier. We utilize the selected features to diagnose various diseases using machine learning. We subsequently measure the performance of the machine learning classifier to make sure that it will not misdiagnose the diseases. The first two steps, which are R-to-R extraction and preprocessing, have been successfully implemented with satisfactory results.