서브메뉴
검색
Finding Structures in High-Dimensional Biomedical Data
Finding Structures in High-Dimensional Biomedical Data
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20250211151956
- ISBN
- 9798384463313
- DDC
- 621.3
- 서명/저자
- Finding Structures in High-Dimensional Biomedical Data
- 발행사항
- [Sl] : Princeton University, 2024
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2024
- 형태사항
- 139 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 86-04, Section: B.
- 주기사항
- Advisor: Fleischer, Jason W.
- 학위논문주기
- Thesis (Ph.D.)--Princeton University, 2024.
- 초록/해제
- 요약High-dimensional datasets often contain unlabeled structures with insightful knowledge. Neighbor embedding algorithms are unsupervised algorithms that identify groups of related data to visualize this structure. The algorithms are regarded as nonlinear dimensionality reduction methods, as they allow visualization of high-dimensional data in a low-dimensional space (typically 2-D). A popular neighbor embedding algorithm is Uniform Manifold Approximation and Projection (UMAP). UMAP utilizes a k-nearest neighbor (k-NN) graph to establish a pairwise metric in a high-dimensional space, which it uses to align a lower-dimensional representation. This dissertation explores techniques for improving UMAP and utilizing it to design new algorithms. We analyze the UMAP algorithm and better explain its optimization scheme and cluster formation, enhance the consistency of embeddings with respect to initialization, and improve its out-of-sample embedding. Then, we apply UMAP for aligning manifolds and analyzing large biomedical image datasets. In particular, we analyze chest x-rays and show that dimensionality reduction can discover 1) different phenotypes of COVID-19 response and 2) outliers in image datasets.Overall, the methodologies presented in this dissertation provide tools to analyze any iterative dimensionality reduction algorithms to demystify their inner workings and design methods for discovering unlabeled patterns.
- 일반주제명
- Electrical engineering
- 일반주제명
- Computer science
- 일반주제명
- Medical imaging
- 키워드
- X-ray
- 키워드
- COVID-19
- 키워드
- Biomedical image
- 기타저자
- Princeton University Electrical and Computer Engineering
- 기본자료저록
- Dissertations Abstracts International. 86-04B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008250123s2024 us c eng d■001000017162301
■00520250211151956
■006m o d
■007cr#unu||||||||
■020 ▼a9798384463313
■035 ▼a(MiAaPQ)AAI31329161
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a621.3
■1001 ▼aIslam, Mohammad Tariqul.▼0(orcid)0000-0003-0541-9110
■24510▼aFinding Structures in High-Dimensional Biomedical Data
■260 ▼a[Sl]▼bPrinceton University▼c2024
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2024
■300 ▼a139 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 86-04, Section: B.
■500 ▼aAdvisor: Fleischer, Jason W.
■5021 ▼aThesis (Ph.D.)--Princeton University, 2024.
■520 ▼aHigh-dimensional datasets often contain unlabeled structures with insightful knowledge. Neighbor embedding algorithms are unsupervised algorithms that identify groups of related data to visualize this structure. The algorithms are regarded as nonlinear dimensionality reduction methods, as they allow visualization of high-dimensional data in a low-dimensional space (typically 2-D). A popular neighbor embedding algorithm is Uniform Manifold Approximation and Projection (UMAP). UMAP utilizes a k-nearest neighbor (k-NN) graph to establish a pairwise metric in a high-dimensional space, which it uses to align a lower-dimensional representation. This dissertation explores techniques for improving UMAP and utilizing it to design new algorithms. We analyze the UMAP algorithm and better explain its optimization scheme and cluster formation, enhance the consistency of embeddings with respect to initialization, and improve its out-of-sample embedding. Then, we apply UMAP for aligning manifolds and analyzing large biomedical image datasets. In particular, we analyze chest x-rays and show that dimensionality reduction can discover 1) different phenotypes of COVID-19 response and 2) outliers in image datasets.Overall, the methodologies presented in this dissertation provide tools to analyze any iterative dimensionality reduction algorithms to demystify their inner workings and design methods for discovering unlabeled patterns.
■590 ▼aSchool code: 0181.
■650 4▼aElectrical engineering
■650 4▼aComputer science
■650 4▼aMedical imaging
■653 ▼aDimensionality reduction
■653 ▼aNeighbor embedding algorithms
■653 ▼aX-ray
■653 ▼aCOVID-19
■653 ▼aBiomedical image
■690 ▼a0544
■690 ▼a0984
■690 ▼a0574
■71020▼aPrinceton University▼bElectrical and Computer Engineering.
■7730 ▼tDissertations Abstracts International▼g86-04B.
■790 ▼a0181
■791 ▼aPh.D.
■792 ▼a2024
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17162301▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
