서브메뉴
검색
Safe and Viable AI Systems for Biomedicine
Safe and Viable AI Systems for Biomedicine
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260202105609
- ISBN
- 9798265426901
- DDC
- 519.5
- 저자명
- Wu, Kevin.
- 서명/저자
- Safe and Viable AI Systems for Biomedicine
- 발행사항
- [Sl] : Stanford University, 2025
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2025
- 형태사항
- 139 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 87-05, Section: B.
- 주기사항
- Advisor: Zou, James;Ho, Daniel.
- 학위논문주기
- Thesis (Ph.D.)--Stanford University, 2025.
- 초록/해제
- 요약Recent years have brought a remarkable surge in the capabilities of artificial intelligence. In biomedicine, these advances have heralded groundbreaking claims: algorithms can now diagnose rare diseases from subtle clinical cues, accelerate the discovery of novel therapeutics, and automate swaths of physician workload that once demanded painstaking manual effort. It is now common to see new headlines announcing that AI has surpassed yet another milestone in clinical performance.Yet, paradoxically, the healthcare system in which these technologies operate remains weighed down by the same systemic dysfunction. Costs continue their relentless climb, outpacing inflation and burdening both patients and providers. Staffing shortages strain hospital wards and outpatient clinics alike, while administrative complexity grows rather than diminishes. The juxtaposition is striking: an era of unprecedented technological potential alongside deepening systemic problems.A simple question emerges: Why hasn't AI "fixed" healthcare yet? The answer to this paradox tend to fall into two broad hypotheses. The first is that the technology itself is still insufficient. That despite the hype and resources given, AI is not yet capable enough to produce system-wide transformation. This line of research has historically focused on benchmarking model performance against fixed datasets and making changes to algorithms and data to ascend the leaderboard. The second hypothesis is that the true problem lies elsewhere: in the way AI is integrated into the lived reality of healthcare delivery.To date, the overwhelming majority of attention, both financially and and academically, has been directed toward the first hypothesis. Vast sums have been devoted to refining algorithms, expanding datasets, and improving predictive accuracy. By contrast, the second question - how these tools are deployed within the social, economic, and institutional fabric of healthcare - has been less studied. The primary aim of my research is to shine light on this imbalance, as it may hold the key to understanding why the promise of AI has, for now, been unfulfilled.
- 일반주제명
- Use statistics
- 일반주제명
- Medical equipment
- 일반주제명
- Federal regulation
- 일반주제명
- Medicine
- 기타저자
- Stanford University.
- 기본자료저록
- Dissertations Abstracts International. 87-05B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260126s2025 us c eng d■001000017360711
■00520260202105609
■006m o d
■007cr#unu||||||||
■020 ▼a9798265426901
■035 ▼a(MiAaPQ)AAI32316375
■035 ▼a(MiAaPQ)Stanfordch828dx1037
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a519.5
■1001 ▼aWu, Kevin.
■24510▼aSafe and Viable AI Systems for Biomedicine
■260 ▼a[Sl]▼bStanford University▼c2025
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2025
■300 ▼a139 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 87-05, Section: B.
■500 ▼aAdvisor: Zou, James;Ho, Daniel.
■5021 ▼aThesis (Ph.D.)--Stanford University, 2025.
■520 ▼aRecent years have brought a remarkable surge in the capabilities of artificial intelligence. In biomedicine, these advances have heralded groundbreaking claims: algorithms can now diagnose rare diseases from subtle clinical cues, accelerate the discovery of novel therapeutics, and automate swaths of physician workload that once demanded painstaking manual effort. It is now common to see new headlines announcing that AI has surpassed yet another milestone in clinical performance.Yet, paradoxically, the healthcare system in which these technologies operate remains weighed down by the same systemic dysfunction. Costs continue their relentless climb, outpacing inflation and burdening both patients and providers. Staffing shortages strain hospital wards and outpatient clinics alike, while administrative complexity grows rather than diminishes. The juxtaposition is striking: an era of unprecedented technological potential alongside deepening systemic problems.A simple question emerges: Why hasn't AI "fixed" healthcare yet? The answer to this paradox tend to fall into two broad hypotheses. The first is that the technology itself is still insufficient. That despite the hype and resources given, AI is not yet capable enough to produce system-wide transformation. This line of research has historically focused on benchmarking model performance against fixed datasets and making changes to algorithms and data to ascend the leaderboard. The second hypothesis is that the true problem lies elsewhere: in the way AI is integrated into the lived reality of healthcare delivery.To date, the overwhelming majority of attention, both financially and and academically, has been directed toward the first hypothesis. Vast sums have been devoted to refining algorithms, expanding datasets, and improving predictive accuracy. By contrast, the second question - how these tools are deployed within the social, economic, and institutional fabric of healthcare - has been less studied. The primary aim of my research is to shine light on this imbalance, as it may hold the key to understanding why the promise of AI has, for now, been unfulfilled.
■590 ▼aSchool code: 0212.
■650 4▼aUse statistics
■650 4▼aMedical equipment
■650 4▼aFederal regulation
■650 4▼aMedicine
■690 ▼a0800
■690 ▼a0564
■71020▼aStanford University.
■7730 ▼tDissertations Abstracts International▼g87-05B.
■790 ▼a0212
■791 ▼aPh.D.
■792 ▼a2025
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17360711▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
