東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Sampling Designs for Resource Effici...

Tan, Wei Ling Katherine.

FindBook

Google Book

Amazon

博客來

Sampling Designs for Resource Efficient Collection of Outcome Labels for Machine-Learning, with Application to Electronic Medical Records.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Sampling Designs for Resource Efficient Collection of Outcome Labels for Machine-Learning, with Application to Electronic Medical Records./
作者:	Tan, Wei Ling Katherine.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2018,
面頁冊數:	199 p.
附註:	Source: Dissertation Abstracts International, Volume: 80-07(E), Section: B.
Contained By:	Dissertation Abstracts International80-07B(E).
標題:	Biostatistics. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10981508
ISBN:	9780438871168

Sampling Designs for Resource Efficient Collection of Outcome Labels for Machine-Learning, with Application to Electronic Medical Records.
Tan, Wei Ling Katherine.

Sampling Designs for Resource Efficient Collection of Outcome Labels for Machine-Learning, with Application to Electronic Medical Records. - Ann Arbor : ProQuest Dissertations & Theses, 2018 - 199 p.

Source: Dissertation Abstracts International, Volume: 80-07(E), Section: B.

Thesis (Ph.D.)--University of Washington, 2018.

In leveraging data from large-scale electronic medical record systems for research, an important step is the accurate identification of key clinical outcomes. Some outcomes must be derived or predicted from both structured and unstructured data, for example using statistical machine-learning classification. Classification requires the collection of labeled data, which is a sample where actual outcome statuses are manually coded by human clinical experts. For rare outcomes, simple random sampling (SRS) for labeled data collection results in very few cases in the sample. Such outcome class imbalance results in insufficient information for classifier modeling, yet additional abstraction is often expensive and time-consuming. In this dissertation, we propose sampling designs for labeled data collection towards machine-learning, targeting the rare outcome scenario. Our proposed designs are resource efficient, requiring a smaller sample size for modeling goals compared to SRS, yet design impacts on model development and validation can be statistically characterized to be "valid". We first introduce a stratified sampling procedure based on values of enrichment surrogates, which are summaries of structured data related to the clinical outcome requiring abstraction. Next, motivated by radiology reports with multiple co-occurring findings, we discuss extensions to the multi-label setting. Finally, for scenarios where a previously developed "source" model is to be externally transferred, we propose a framework for such "new'' labeled data collection.

ISBN: 9780438871168Subjects--Topical Terms:

1002712
Biostatistics.

Sampling Designs for Resource Efficient Collection of Outcome Labels for Machine-Learning, with Application to Electronic Medical Records.
LDR:02584nmm a2200301 4500 001 2202309
005 20190513114647.5
008 201008s2018 ||||||||||||||||| ||eng d
020 $a 9780438871168
035 $a (MiAaPQ)AAI10981508
035 $a (MiAaPQ)washington:19428
035 $a AAI10981508
040 $a MiAaPQ $c MiAaPQ
100 1 $a Tan, Wei Ling Katherine. $3 3429053
245 1 0 $a Sampling Designs for Resource Efficient Collection of Outcome Labels for Machine-Learning, with Application to Electronic Medical Records.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2018
300 $a 199 p.
500 $a Source: Dissertation Abstracts International, Volume: 80-07(E), Section: B.
500 $a Adviser: Patrick J. Heagerty.
502 $a Thesis (Ph.D.)--University of Washington, 2018.
520 $a In leveraging data from large-scale electronic medical record systems for research, an important step is the accurate identification of key clinical outcomes. Some outcomes must be derived or predicted from both structured and unstructured data, for example using statistical machine-learning classification. Classification requires the collection of labeled data, which is a sample where actual outcome statuses are manually coded by human clinical experts. For rare outcomes, simple random sampling (SRS) for labeled data collection results in very few cases in the sample. Such outcome class imbalance results in insufficient information for classifier modeling, yet additional abstraction is often expensive and time-consuming. In this dissertation, we propose sampling designs for labeled data collection towards machine-learning, targeting the rare outcome scenario. Our proposed designs are resource efficient, requiring a smaller sample size for modeling goals compared to SRS, yet design impacts on model development and validation can be statistically characterized to be "valid". We first introduce a stratified sampling procedure based on values of enrichment surrogates, which are summaries of structured data related to the clinical outcome requiring abstraction. Next, motivated by radiology reports with multiple co-occurring findings, we discuss extensions to the multi-label setting. Finally, for scenarios where a previously developed "source" model is to be externally transferred, we propose a framework for such "new'' labeled data collection.
590 $a School code: 0250.
650 4 $a Biostatistics. $3 1002712
650 4 $a Artificial intelligence. $3 516317
690 $a 0308
690 $a 0800
710 2 $a University of Washington. $b Biostatistics (Public Health). $3 3429054
773 0 $t Dissertation Abstracts International $g 80-07B(E).
790 $a 0250
791 $a Ph.D.
792 $a 2018
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10981508