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
- 학술대회자료
- 저자정보
- 발행연도
- 2013.5
- 수록면
- 310 - 315 (6page)
이용수
초록· 키워드
오류제보하기
Objective: The aim of this study was to develop a glaucoma diagnostic model applying the statistical classification method by various testing results for glaucoma. Background: Statistical diagnostic models applied with structural (e.g., fundus examination) and functional (e.g., visual field testing) testing results have been developed to diagnose glaucoma more accurately. The existing studies with regards to the glaucoma diagnostic model considered statistical techniques by various input data such as intraocular pressure and cup-to-disc ratio; however, constant exertions for improving the performance of a diagnostic model are necessary through applying unvalidated methods and input data. Method: Glaucoma testing results of 145 eyes (glaucomatous eyes = 74, normal eyes = 71) were used as input data. Five classification techniques (linear and Gaussian support vector machines [L-SVM and G-SVM], linear discriminant analysis [LDA], binary logistic regression [BLR], and classification and regression tree [CART]) were applied to develop the model. Used as input, the performance (accuracy, sensitivity, specificity and AUC) of the developed models were compared with among the developed models. Results: The PD/G-SVM model which applied with G-SVM with pattern deviation (PD) was determined as an optimal model (accuracy = 0.87; sensitivity = 0.81; specificity = 0.92; and AUC = 0.95). Conclusion: The developed PD/G-SVM model can contribute to making a diagnosis of glaucoma more precisely. Application: The developed model would be expected to apply to develop a comprehensive diagnostic system for glaucoma.
목차
ABSTRACT
1. Introduction
2. Materials and Methods
3. Results
4. Discussion
References
참고문헌 (0)
참고문헌 신청이 논문의 저자 정보
최근 본 자료
댓글(0)
0