지원사업
학술연구/단체지원/교육 등 연구자 활동을 지속하도록 DBpia가 지원하고 있어요.
커뮤니티
연구자들이 자신의 연구와 전문성을 널리 알리고, 새로운 협력의 기회를 만들 수 있는 네트워킹 공간이에요.
논문 기본 정보
- 자료유형
- 학위논문
- 저자정보
- 지도교수
- 이제영
- 발행연도
- 2019
- 저작권
- 영남대학교 논문은 저작권에 의해 보호받습니다.
이용수8
이 논문의 연구 히스토리 (3)
- 이 논문의 후속연구가 궁금하신가요?
- 연관 학술논문 또는 학술발표를 통해 보다 발전된 연구결과를 확인하실 수 있습니다.
- 이 논문의 연구 히스토리 확인하기
초록· 키워드
오류제보하기
In this study, we propose the nomogram construction using logistic regression analysis and the naive Bayesian classifier for chronic diseases, and propose each nomogram using the Korean National Health and Nutrition Examination Survey (KNHANES) data. We identify risk factors for dyslipidemia and chronic obstructive pulmonary disease (COPD), and propose logistic nomogram, Bayesian nomogram to predict the incidence of dyslipidemia and COPD. The proposed nomogram is verified using AUC of ROC curve and calibration plot. In addition, we compare the logistic and the Bayesian nomograms.
The high risk factor of dyslipidemia was obesity, followed by diabetes, 56-69 years of age group, hypertension and heart disease (myocardial infarction and angina pectoris) in the order. On the other hand, there was diabetes in Bayesian nomograms, followed by heart disease, hypertension, and education level below elementary school level in the order. In COPD, asthma was the high risk factor in the logistic nomogram, followed by over 65 years of age group, men, and education level below middle school in the order. Asthma
was also the high risk factor in the Bayesian nomogram, followed by over 65 years of age group, tuberculosis, and current smoker in the order. The AUC of the ROC curve was statistically significant, 0.741, 0.737, 0.807, and 0.802, respectively. Finally, the coefficient of determination of the calibration plot (R2) was high as 0.931, 0.815, 0.900, 0.884. Using the nomograms suggested in this study, individual and health worker can easily identify the risk factors for chronic diseases and predict the incidence, which can be a great help in the prevention of chronic diseases.
The high risk factor of dyslipidemia was obesity, followed by diabetes, 56-69 years of age group, hypertension and heart disease (myocardial infarction and angina pectoris) in the order. On the other hand, there was diabetes in Bayesian nomograms, followed by heart disease, hypertension, and education level below elementary school level in the order. In COPD, asthma was the high risk factor in the logistic nomogram, followed by over 65 years of age group, men, and education level below middle school in the order. Asthma
was also the high risk factor in the Bayesian nomogram, followed by over 65 years of age group, tuberculosis, and current smoker in the order. The AUC of the ROC curve was statistically significant, 0.741, 0.737, 0.807, and 0.802, respectively. Finally, the coefficient of determination of the calibration plot (R2) was high as 0.931, 0.815, 0.900, 0.884. Using the nomograms suggested in this study, individual and health worker can easily identify the risk factors for chronic diseases and predict the incidence, which can be a great help in the prevention of chronic diseases.
목차
1. 서론 12. 통계적 연구방법 52.1. Logistic regression model을 통한 노모그램 52.1.1. Logistic regression model 52.1.2. Logistic 노모그램 구축 절차 소개 82.2. Naive Bayesian classifier model을 통한 노모그램 92.2.1. Naive Bayesian classifier model 92.2.2. Bayesian 노모그램 구축 절차 소개 112.3. ROC curve와 calibration plot을 통한 노모그램 검증 133. 적용 결과 143.1. 이상지질혈증의 노모그램 143.1.1. 이상지질혈증의 logistic 노모그램 173.1.2. 이상지질혈증의 Bayesian 노모그램 223.2. COPD의 노모그램 253.2.1. COPD의 logistic 노모그램 263.2.2. COPD의 Bayesian 노모그램 283.3. Logistic 노모그램과 Bayesian 노모그램의 비교 293.4. 이상지질혈증과 COPD의 노모그램 검증 313.4.1. ROC curve 323.4.2. Calibration plot 334. 결론 및 토의 34참고문헌 37
최근 본 자료
댓글(0)
0