서브메뉴
검색
Machine Learning and Statistical Approaches for Efficient Disease Screening.
Machine Learning and Statistical Approaches for Efficient Disease Screening.
상세정보
- 자료유형
- 학위논문
- Control Number
- 0017163723
- International Standard Book Number
- 9798342109321
- Dewey Decimal Classification Number
- 519
- Main Entry-Personal Name
- Landolfi, Nicholas Charles.
- Publication, Distribution, etc. (Imprint
- [S.l.] : Stanford University., 2024
- Publication, Distribution, etc. (Imprint
- Ann Arbor : ProQuest Dissertations & Theses, 2024
- Physical Description
- 155 p.
- General Note
- Source: Dissertations Abstracts International, Volume: 86-04, Section: B.
- General Note
- Advisor: Lall, Sanjay;Sadigh, Dorsa.
- Dissertation Note
- Thesis (Ph.D.)--Stanford University, 2024.
- Summary, Etc.
- 요약We study probabilistic models with algebraic group symmetry and associated algorithms. We are motivated by pooled testing, which is widely used to conserve resources when performing large-scale population screening of infectious diseases such as that caused by SARS-CoV-2. Looking forward, such protocols are also a promising route to increase public access to novel and expensive biological assays in areas like early cancer detection. Classical approaches, however, are based on single-parameter models of disease prevalence and justified under assumptions of independence. We analyze symmetric probabilistic models that allow for correlation in test statuses and accommodate side information. By leveraging the probabilistic symmetries, we give efficient algorithms for learning these models from data and computing the optimal testing designs. When applied to data from the COVID-19 pandemic, these methods indicate more efficient test designs and help explain the unexpectedly high empirical efficiency observed by the original investigators.
- Subject Added Entry-Topical Term
- Linear programming.
- Subject Added Entry-Topical Term
- Dynamic programming.
- Subject Added Entry-Topical Term
- Medical screening.
- Subject Added Entry-Topical Term
- Optimization techniques.
- Subject Added Entry-Topical Term
- Visualization.
- Subject Added Entry-Topical Term
- Orbits.
- Subject Added Entry-Topical Term
- Symmetry.
- Subject Added Entry-Topical Term
- COVID-19.
- Subject Added Entry-Topical Term
- Computer science.
- Added Entry-Corporate Name
- Stanford University.
- Host Item Entry
- Dissertations Abstracts International. 86-04B.
- Electronic Location and Access
- 로그인을 한후 보실 수 있는 자료입니다.
- Control Number
- joongbu:655068
MARC
008250224s2024 us ||||||||||||||c||eng d■001000017163723
■00520250211152745
■006m o d
■007cr#unu||||||||
■020 ▼a9798342109321
■035 ▼a(MiAaPQ)AAI31520277
■035 ▼a(MiAaPQ)Stanfordhb623pm4894
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a519
■1001 ▼aLandolfi, Nicholas Charles.
■24510▼aMachine Learning and Statistical Approaches for Efficient Disease Screening.
■260 ▼a[S.l.]▼bStanford University. ▼c2024
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2024
■300 ▼a155 p.
■500 ▼aSource: Dissertations Abstracts International, Volume: 86-04, Section: B.
■500 ▼aAdvisor: Lall, Sanjay;Sadigh, Dorsa.
■5021 ▼aThesis (Ph.D.)--Stanford University, 2024.
■520 ▼aWe study probabilistic models with algebraic group symmetry and associated algorithms. We are motivated by pooled testing, which is widely used to conserve resources when performing large-scale population screening of infectious diseases such as that caused by SARS-CoV-2. Looking forward, such protocols are also a promising route to increase public access to novel and expensive biological assays in areas like early cancer detection. Classical approaches, however, are based on single-parameter models of disease prevalence and justified under assumptions of independence. We analyze symmetric probabilistic models that allow for correlation in test statuses and accommodate side information. By leveraging the probabilistic symmetries, we give efficient algorithms for learning these models from data and computing the optimal testing designs. When applied to data from the COVID-19 pandemic, these methods indicate more efficient test designs and help explain the unexpectedly high empirical efficiency observed by the original investigators.
■590 ▼aSchool code: 0212.
■650 4▼aLinear programming.
■650 4▼aDynamic programming.
■650 4▼aMedical screening.
■650 4▼aOptimization techniques.
■650 4▼aVisualization.
■650 4▼aOrbits.
■650 4▼aSymmetry.
■650 4▼aCOVID-19.
■650 4▼aComputer science.
■690 ▼a0984
■71020▼aStanford University.
■7730 ▼tDissertations Abstracts International▼g86-04B.
■790 ▼a0212
■791 ▼aPh.D.
■792 ▼a2024
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17163723▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
