인문학
사회과학
자연과학
공학
의약학
농수해양학
예술체육학
복합학
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
- 2019.11
- 수록면
- 58 - 61 (4page)
이용수
초록· 키워드
Recently, several high dimensional classification methods have been proposed to automatically discriminate between patients with Alzheimer"s disease (AD) and elderly controls (CN) based on T1-weighted MRI images. Moreover, many works have validated that multiple neuroimaging tools and biological biomarkers contain complementary information for early prognosis of AD. In this paper, we proposed an automatic classification by using multiple neuroimaging tools and finally combining them to classify AD vs HC. In total 180 subjects were downloaded from Open Access Series of Imaging Studies (OASIS) homepage, form which 90 subjects belong to AD and 90 subjects belong to (HC). We have used three types of feature extraction process which are NiftyReg, Freesurfer (6) and MALPEM (Multi-Atlas Labeled Propagation with EM refinement). For each image, we have extracted 246 Region of interest (ROI) using Brainnetome atlas using NiftyReg, 138 Region of Interest (ROI) using multi-labeled atlas in MALPEM and 68 regions using Freesurfer Destrieux atlas. According to KNN classifier result, multimodal feature achieved a good result for binary classification problem (AD and HC). Result, AUC of ROC for each of neuroimaging tools are 71%, 80%, 84% (NiftyReg, Freesurfer, MALPEM) and our method achieved 90% (AD and HC). Moreover, we have compared our result with the latest published state-of-the-art multimodal method.
상세정보 수정요청해당 페이지 내 제목·저자·목차·페이지정보가 잘못된 경우 알려주세요!
목차
- Abstract
- I. INTRODUCTION
- II. MATERIAL AND METHOD
- III. EXPERIMENT RESULT
- IV. CONCLUSION
- REFERENCE
참고문헌
참고문헌 신청최근 본 자료
UCI(KEPA) : I410-ECN-0101-2020-567-000084819