인문학
사회과학
자연과학
공학
의약학
농수해양학
예술체육학
복합학
DBpia AI
AI 에이전트
AI 아이디어
AI 리더
AI 뷰어
Citeasy
크롬 서지관리 다운로드
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
- 학술저널
- 저자정보
- 저널정보
- 대한의용생체공학회 Biomedical Engineering Letters (BMEL) Biomedical Engineering Letters (BMEL) Vol.8 No.1
- 발행연도
- 2018.1
- 수록면
- 41 - 57 (17page)
이용수
초록· 키워드
The high-pace rise in advanced computing andimaging systems has given rise to a new research dimensioncalled computer-aided diagnosis (CAD) system forvarious biomedical purposes. CAD-based diabeticretinopathy (DR) can be of paramount significance toenable early disease detection and diagnosis decision. Considering the robustness of deep neural networks(DNNs) to solve highly intricate classification problems, inthis paper, AlexNet DNN, which functions on the basis ofconvolutional neural network (CNN), has been applied toenable an optimal DR CAD solution. The DR modelapplies a multilevel optimization measure that incorporatespre-processing, adaptive-learning-based Gaussian mixturemodel (GMM)-based concept region segmentation, connectedcomponent-analysis-based region of interest (ROI)localization, AlexNet DNN-based highly dimensional featureextraction, principle component analysis (PCA)- andlinear discriminant analysis (LDA)-based feature selection,and support-vector-machine-based classification to ensureoptimal five-class DR classification. The simulation resultswith standard KAGGLE fundus datasets reveal that theproposed AlexNet DNN-based DR exhibits a better performancewith LDA feature selection, where it exhibits aDR classification accuracy of 97.93% with FC7 features,whereas with PCA, it shows 95.26% accuracy. Comparativeanalysis with spatial invariant feature transform (SIFT)technique (accuracy—94.40%) based DR feature extractionalso confirms that AlexNet DNN-based DR outperformsSIFT-based DR.
상세정보 수정요청해당 페이지 내 제목·저자·목차·페이지정보가 잘못된 경우 알려주세요!
목차
등록된 정보가 없습니다.
참고문헌
참고문헌 신청최근 본 자료
