• Journal of the Korean Data and Information Science Society
• /
• 제21권3호
• /
• pp.589-595
• /
• 2010

In discriminant analysis, we consider an intraclass correlation pattern by principal component analysis. We assume that the two populations are equally likely and the costs of misclassification are equal. In this situation, we consider two procedures, i.e., the test and proportion procedures, for selecting the principal components in classifica-tion. We compare the regular classification method and the proposed two procedures. We consider two methods for estimating error rate, i.e., the leave-one-out method and the bootstrap method.

• 동의신경정신과학회지
• /
• 제25권4호
• /
• pp.351-358
• /
• 2014

Objectives: This study was performed to evaluate the reliability and validity of the instrument of oriental medical evaluation for Hwa-byung. Methods: 97 participants diagnosed with Hwa-byung divided into 5 pattern identifications by the instrument on pattern identifications for Hwa-byung were tested by the instrument of oriental medical evaluation for Hwa-byung twice a week. Results: The intraclass correlation coefficient of the test-retest analysis of oriental medical evaluation for Hwa-byung had over moderate agreement. The intraclass correlation coefficient of test-retest analysis of evaluation index of symptoms had moderate agreement for the total symptoms, and fair agreement for the physical and mental symptoms. The validity analysis of the instrument represented a positive correlation in the pattern identification of stagnation of liver Gi, malfunction of gallbladder due to phlegm stagnation, and disharmony between heart and kidneys. Conclusions: The results showed that the reliability analysis of the instrument of oriental medical evaluation for Hwa-byung showed over moderate agreement, and validity analysis represented a positive correlation except for the pattern identification of flare-up of liver fire and deficiency of both Gi and blood.

• 한국통신학회논문지
• /
• 제15권8호
• /
• pp.678-688
• /
• 1990

본 논문에서는 pSDF(projection synthetic discriminant function) 공간불변 필터 개념을 적용하여 기준 이미지를 구현하고, 공간 평면 상관기인 이진 결합 변환 상관기를 이용하여 동일 클래스 인식을 수행하였다. 컴퓨터 시뮬레이션 결과, 이진 결합 변환 상관기가 기존의 결합 변환 상관기보다 상관 첨두치 세기, 상관 첨두치 세기대 부로브비, 신호대 잡음비, 상관폭 부분에서 뛰어난 상관 특성을 보였다. 이진 결합 변환상관기를 이용한 광 패턴 인식 실험을 한 결과, 동일 클래스 인식인 경우 4.1~9.6% 오차 범위내에서 상관 첨두치 세기가 일정하게 나타났다.

• 대한한방부인과학회지
• /
• 제32권2호
• /
• pp.29-49
• /
• 2019

Objectives: This study was a methodological study to verify the reliability and validity and to make a diagnosis of a diagnostic tool for climacteric and postmenopausal syndrome pattern identification (CaPSPI). Methods: This study was conducted from June 1, 2018 to October 18, 2018 with ${{\bigcirc}{\bigcirc}}$ University Korean Medicine Hospital IRB's approval (2018-3). To make a diagnosis using CaPSPI, we decided the cut-points for the tool. Three professors of ${{\bigcirc}{\bigcirc}}$Korean Medical University conducted pattern identification diagnosis. The result is marked from 0 to 3, 0 is 'No', 1 is 'Slightly Yes', 2 is 'Yes' and 3 is 'Very Yes'. And if two or more professors' diagnoses are the same, we took the diagnoses as a diagnostic criteria. The decision of pattern by three experts converted to 0-1 scores in two ways. In "method 1", if the diagnosis was zero points, the score was 0 (have no such identification), and the rest was 1 (have such identification). In "method 2", if the diagnosis was zero or 1 point, the score was 0, and the rest was 1. After that, intraclass correlation was calculated for experts agreement. And logistic regression was conducted. A response variable was the results of the experts' diagnosis and an explanatory variable was the results of the pattern identification diagnostic tool. Results: The diagnosis of the three experts showed excellent concordance of more than 0.794 and showed a significant correlation with the diagnostic tool. Both 'Method 1' and 'Method 2' showed statistically significant effects with the diagnosis of 3 experts and the results of the diagnostic tool. The frequency of cumulative pattern identification diagnosis in 'Method 1' and 'Method 2' were found to be 578 occasions and 203 occasions, respectively. The average number of pattern held by participants in 'Method 1' and 'Method 2' were found to be 5.26 and 1.85, respectively. In both "Method 1" and "Method 2," the yield frequency of liver qi depression was the highest, and the frequency of kidney yin deficiency and liver-kidney yin deficiency was relatively high. Conclusions: Based on the above results, it is thought that, in diagnosis using CaPSPI of menopausal women, 'method 1' could be used for the health diagnosis and prevention, 'method 2' could be used for the pattern diagnosis. On the conclusion, CaPSPI is thought to be available for pattern diagnosis of menopause women.

• 한국콘텐츠학회논문지
• /
• 제22권8호
• /
• pp.671-680
• /
• 2022

본 연구의 목적은 한의학 기반 건강상태 및 진단 분류 도구인 미병과 한열설문의 웹 기반과 종이 기반 간의 일치도를 평가하는 것이다. 일반인 72명을 대상으로 웹 기반 조사를 먼저 시행하고, 시간 간격을 두고 종이기반 설문을 진행하였다. 웹과 종이 기반 조사 방법 간의 동등성은 급내상관계수(Intraclass Correlation Coefficient, ICC)와 Bland-Altman 방법으로 설문 점수 간의 일치도를 평가하였다. 미병설문은 웹과 종이 조사 간에 높은 신뢰도를 보였고(ICC=0.95, 95% CI 0.92 - 0.97), 한열설문의 한증(ICC=0.98, 95% CI 0.96 - 0.99)과 열증(ICC=0.9, 95% CI 0.83 - 0.93) 모두 우수한 신뢰도를 나타냈다. 두 조사 방법 간의 평균 점수 차이는 미병설문이 -0.25점이고, 한열설문의 한증 -0.17점, 열증 0.11점으로 유사한 일치 양상을 나타냈다. 대상자의 84%가 웹 기반에 긍정적인 만족도를 보였고, 80%가 웹 조사를 선호하였다. 본 연구에서 미병설문과 한열설문의 웹 조사 방법 간의 신뢰도와 실행 가능성을 확인하였다. 이것은 장기간 걸쳐 진행되는 코호트 연구에서 대상자 추적 관찰에 유용한 도구로 고려된다.

• 대한한방내과학회지
• /
• 제41권6호
• /
• pp.1052-1065
• /
• 2020

Objectives: To evaluate the reliability of a pattern identification tool for benign prostatic hyperplasia and to examine the relationship between pattern identification tool readings and IPSS and uroflowmetry. Methods: We analyzed 56 patients diagnosed with benign prostatic hyperplasia from December 27th, 2017 to December 26th, 2018 by two different Korean medical doctors and followed with a pattern identification tool and by IPSS and uroflowmetry. One week later, the patients were retested to analyze the reliability of the pattern identification tool, determined with the intraclass correlation coefficient (ICC) using the test-retest method. The correlation between IPSS and uroflowmetry was analyzed with the Pearson coefficient. Result: The reliability of the pattern identification tool for benign prostatic hyperplasia was evaluated as "poor agreement beyond chance" (ICC=0.349). The reliability of each pattern identification score was evaluated as "good" for Yang Deficiency of Kidney, Yin Deficiency of Kidney, Deficiency of Middle Qi, and Dampness-heat of Lower Energizer. The internal consistency was evaluated as "good" for Yang Deficiency of Kidney, Yin Deficiency of Kidney, and Dampness-heat of Lower Energizer, and as "excellent" for Deficiency of Middle Qi. The correlation between pattern identification and IPSS was evaluated as a "moderate positive correlation" for all pattern identifications. The average flow rate and maximum flow rate using uroflowmetry was evaluated with "moderate negative correlation" for Yang Deficiency of Kidney and Dampness-heat of Lower Energizer. Conclusion: The reliability of a pattern identification tool for benign prostatic hyperplasia was evaluated as "poor agreement beyond chance." Further research is needed.

• 한방재활의학과학회지
• /
• 제29권3호
• /
• pp.113-128
• /
• 2019

Objectives To evaluate the pattern identification tool for knee osteoarthritis and to investigate the relationship between pattern identification tool and knee ROM (range of motion), VAS (visual analog scale) and WOMAC (Western Ontario & McMaster Universities osteoarthritis index). Methods We studied 50 patients who are diagnosed as degenerative osteoarthritis. With one patient dropping out, total of 49 patient went through pattern identification tool. Re-test of the pattern identification tool was implemented after about a week. The reliability was analyzed by calculating ICC (intraclass correlation coefficient). Also, reliability of each pattern identification score was calculated along with influence factor and correlation between knee ROM, VAS and WOMAC. Results Reliability of the pattern identification tool was evaluated as 'poor agreement beyond chance' with ICC value of 0.396. Reliablility of each pattern identification score was calculated and four out of five were measured 'good' with one 'moderate' reliability. Two of inter-item consistency were 'good', two were 'acceptable' and one was 'questionable'. Correlation between each pattern identification and knee ROM showed 'clear negative linear relationship' with two patterns. VAS showed 'clear positivie linear relationship' with all five patterns. WOMAC showed 'positivie linear relationship' with all five patterns. Conclusions According to the results, pattern identification tool for knee osteoarthritis seems to have meaningful relationship with other agnedas. Also, further research is needed to develop the tool.

• 대한예방한의학회지
• /
• 제19권3호
• /
• pp.11-19
• /
• 2015

Objective : This study aimed to develop and preliminary validate the Mibyeong Questionnaire (MQ) based on Deficiency-Stagnation pattern, which was report of a person's health condition that comes directly from the person. Method : The first phage of developing a MQ was to generate and exhaustive list of all MQ issue that are relevant to the domain of interest, using literature research and expert group discussions. Through those steps, we established MQ with 33 items divided into two parts: 21-item deficiency pattern and 12-item stagnation pattern. The second phage, we examined the preliminary tests of reliability and validity including the 16-item Deficiency of MQ (16D-MQ), with data (n=1,890) already collected on the Korean medicine data center in KIOM. Results : Exploratory factor analysis revealed three factors of the 16D-MQ. These factors were fatigue(Qi, 氣); psychic and physical elements(Shen, 神); and skin and hair(Jing, 精). Cronbach's coefficient alpha was 0.876 and the intraclass correlation coefficients was 0.368-0.538. In support of criteria validity, the 16D-MQ was weakly correlated with EQ-5D and physicians's opinion, but it was acceptable. Conclusion : The MQ shows that it has an appropriate level of internal consistency and validity. We think further study to reveal its reliability and validity, including stagnation pattern as well as deficiency pattern, is needed.

• Tuberculosis and Respiratory Diseases
• /
• 제84권4호
• /
• pp.274-281
• /
• 2021

Background: The Korea National Health and Nutrition Examination Survey (KNHANES) is a well-designed survey to collect national data, which many researchers have used for their studies. In KNHANES, although portable spirometry was used, its reliability has not been verified. Methods: We prospectively enrolled 58 participants from four Korean institutions. The participants were classified into normal pattern, obstructive pattern, and restrictive pattern groups according to their previous spirometry results. Lung function was estimated by conventional spirometry and portable spirometry, and the results were compared. Results: The intraclass correlation coefficients of forced vital capacity (FVC) (coefficient, 9.993; 95% confidence interval [CI], 0.988-0.996), forced expiratory volume in 1 second (FEV₁) (coefficient, 0.997; 95% CI, 0.995-0.998), FEV₁/FVC ratio (coefficient, 0.995; 95% CI, 0.992-0.997), and forced expiratory flow at 25-75% (FEF25-75%; coefficient, 0.991; 95% CI, 0.984-0.994) were excellent (all p<0.001). In the subgroup analysis, the results of the three parameters were similar in all groups. In the overall and subgroup analyses, Pearson's correlation of all the parameters was also excellent in the total (coefficient, 0.986-0.994; p<0.001) and subgroup analyses (coefficient, 0.915-0.995; p<0.001). In the paired t-test, FVC, FEV₁/FVC, and FEF25-75% estimated by the two instruments were statistically different. However, FEV₁ was not significantly different. Conclusion: Lung function estimated by portable spirometry was well-correlated with that estimated by conventional spirometry. Although the values had minimal differences between them, we suggest that the spirometry results from the KNHANES are reliable.

• 대한한방내과학회지
• /
• 제43권1호
• /
• pp.1-21
• /
• 2022

Purpose: The standard tool for the pattern identification is used for identifying patterns in patients using a questionnaire. The purpose of this study is to reorganize the standard tool for the pattern identification of gastroesophageal reflux disease (GERD) developed in 2017 and to analyze the reliability and validity of the standard tool for pattern identification by applying it to GERD patients. Methods: To reorganize the standard tool for the pattern identification of GERD developed in the previous study, we searched the literature in the main databases, OASIS (Oriental Medicine Advanced Searching Integrated System) and CNKI (China National Knowledge Infrastructure). We added the search results to the data used in the previous study and went through the reorganizing courses, such as evaluating the validity of the translation, the Delphi technique, and a small survey. After reorganization, the patients who visited the Kyunghee University Korean Medicine Center for GERD symptoms were provided the questionnaire, including the reorganized standard tool for pattern identification. We analyzed the survey results to evaluate their reliability and validity. Results: Fifty patients completed the questionnaire. Reliability analysis results showed a pattern identification match rate of 86%, Cronbach's α coefficient of 0.834, and intraclass correlation coefficient of 0.907. The Mann - Whitney U test and logistic regression were implemented to check the relations between the survey questions and pattern identification results; the Pearson correlation, compared with other scales, showed a moderate score. Conclusion: We reorganized the standard tool for the pattern identification of GERD to be updated on current issues and so that it is easily used. The analysis results of the questionnaire showed that the reorganized standard tool had high reliability and moderate validity.