서브메뉴
검색
Uncovering Strategies for Personalized Treatment Selection Using Large Language Models
Uncovering Strategies for Personalized Treatment Selection Using Large Language Models
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20250211151347
- ISBN
- 9798382813813
- DDC
- 574
- 저자명
- Miao, Brenda.
- 서명/저자
- Uncovering Strategies for Personalized Treatment Selection Using Large Language Models
- 발행사항
- [Sl] : University of California, San Francisco, 2024
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2024
- 형태사항
- 168 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 85-12, Section: A.
- 주기사항
- Includes supplementary digital materials.
- 주기사항
- Advisor: Butte, Atul.
- 학위논문주기
- Thesis (Ph.D.)--University of California, San Francisco, 2024.
- 초록/해제
- 요약Healthcare data has never been so accessible to patients and physicians, from smartphones and other remote monitoring devices to improved access for patients to their own Electronic Medical Record (EMR) history and clinical notes. Despite the ubiquity of healthcare data collection and distribution, there remains a significant gap in understanding the impacts of this data on clinical care. Insights from these digital health tools and downstream clinical decision-making processes are often only captured in medical notes, which are complex, sparse, unstructured, and difficult to model even with traditional deep learning methods. Only recently have large language models (LLMs) emerged that are capable of zero- or few-shot clinical language, without the need for large, manually annotated datasets. In this dissertation, I develop methods to adapt LLMs to healthcare tasks, particularly for identifying points of actionable insights for both digital and pharmaceutical therapeutics. These approaches demonstrate the ways in which digital health products can impact clinical care, as well as provide methods to identify reasons for medication class switching that consider the complexities of patient care beyond lab values and diagnosis codes. While careful, rigorous research is needed to ensure that these approaches are effective in facilitating patient care improvements and to reduce any potential for harm, the rapid pace of language model development provides an extraordinary opportunity to transform clinical practice. These new methods allow us to take an unprecedented look at the conversations, decisions, and medical expertise captured in billions of clinical notes and other clinical text, and to learn from this shared knowledge to accelerate medical research, improve clinical guidelines, and personalize patient care.
- 일반주제명
- Bioinformatics
- 일반주제명
- Computer science
- 일반주제명
- Language
- 기타저자
- University of California, San Francisco Biological and Medical Informatics
- 기본자료저록
- Dissertations Abstracts International. 85-12A.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008250123s2024 us c eng d■001000017161367
■00520250211151347
■006m o d
■007cr#unu||||||||
■020 ▼a9798382813813
■035 ▼a(MiAaPQ)AAI31242717
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a574
■1001 ▼aMiao, Brenda.▼0(orcid)0000-0002-3393-9837
■24510▼aUncovering Strategies for Personalized Treatment Selection Using Large Language Models
■260 ▼a[Sl]▼bUniversity of California, San Francisco▼c2024
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2024
■300 ▼a168 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 85-12, Section: A.
■500 ▼aIncludes supplementary digital materials.
■500 ▼aAdvisor: Butte, Atul.
■5021 ▼aThesis (Ph.D.)--University of California, San Francisco, 2024.
■520 ▼aHealthcare data has never been so accessible to patients and physicians, from smartphones and other remote monitoring devices to improved access for patients to their own Electronic Medical Record (EMR) history and clinical notes. Despite the ubiquity of healthcare data collection and distribution, there remains a significant gap in understanding the impacts of this data on clinical care. Insights from these digital health tools and downstream clinical decision-making processes are often only captured in medical notes, which are complex, sparse, unstructured, and difficult to model even with traditional deep learning methods. Only recently have large language models (LLMs) emerged that are capable of zero- or few-shot clinical language, without the need for large, manually annotated datasets. In this dissertation, I develop methods to adapt LLMs to healthcare tasks, particularly for identifying points of actionable insights for both digital and pharmaceutical therapeutics. These approaches demonstrate the ways in which digital health products can impact clinical care, as well as provide methods to identify reasons for medication class switching that consider the complexities of patient care beyond lab values and diagnosis codes. While careful, rigorous research is needed to ensure that these approaches are effective in facilitating patient care improvements and to reduce any potential for harm, the rapid pace of language model development provides an extraordinary opportunity to transform clinical practice. These new methods allow us to take an unprecedented look at the conversations, decisions, and medical expertise captured in billions of clinical notes and other clinical text, and to learn from this shared knowledge to accelerate medical research, improve clinical guidelines, and personalize patient care.
■590 ▼aSchool code: 0034.
■650 4▼aBioinformatics
■650 4▼aComputer science
■650 4▼aLanguage
■653 ▼aClinical text processing
■653 ▼aLarge language models
■653 ▼aTreatment strategy optimization
■653 ▼aElectronic Medical Record history
■653 ▼aDeep learning methods
■690 ▼a0715
■690 ▼a0984
■690 ▼a0679
■690 ▼a0769
■71020▼aUniversity of California, San Francisco▼bBiological and Medical Informatics.
■7730 ▼tDissertations Abstracts International▼g85-12A.
■790 ▼a0034
■791 ▼aPh.D.
■792 ▼a2024
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17161367▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
