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

Developing Machine Learning Methodol...

Jiang, Xiaotong.

FindBook

Google Book

Amazon

博客來

Developing Machine Learning Methodology for Precision Health.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Developing Machine Learning Methodology for Precision Health./
作者:	Jiang, Xiaotong.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2020,
面頁冊數:	141 p.
附註:	Source: Dissertations Abstracts International, Volume: 82-01, Section: B.
Contained By:	Dissertations Abstracts International82-01B.
標題:	Biostatistics. -
電子資源:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=27836497
ISBN:	9798635265857

Developing Machine Learning Methodology for Precision Health.
Jiang, Xiaotong.

Developing Machine Learning Methodology for Precision Health. - Ann Arbor : ProQuest Dissertations & Theses, 2020 - 141 p.

Source: Dissertations Abstracts International, Volume: 82-01, Section: B.

Thesis (Ph.D.)--The University of North Carolina at Chapel Hill, 2020.

This item must not be sold to any third party vendors.

Precision health has been an increasingly popular solution to improve health care quality and guide the decision making process. This includes precision medicine (at the individual level) and precision public health (at the population level such as communities and institutions). By learning from the available medical data with advanced analytical tools, precision health recommends the treatments that are individualized to each patient or entity to maximize clinical outcomes for each individual. We extend and develop three machine learning methods to improve the estimation of optimal individualized treatment regimes in precision health: the jackknife estimator of value function of precision medicine models compared with zero-order models, doubly robust outcome-weighted estimators with deep neural network structures for complex and large data, and risk-adjusted adverse event monitoring for survival data. First, motivated by a knee osteoarthristis trial, we estimate value functions and select the optimal treatment with the jackknife method whose consistency is established under weak assumptions. Next, we implement deep learning architecture in augmented outcome-weighted learning to increase model flexibility and computation efficiency, especially for high-dimensional data such as medical imaging. Lastly, we develop a risk-adjusted survival model to monitor adverse events and estimate its variance for hierarchical, right-censored data with recurrent events. All three methodologies aim to solve practical, health-related challenges and provide data-driven decision support and operations.

ISBN: 9798635265857Subjects--Topical Terms:

1002712
Biostatistics.
Subjects--Index Terms:

Biostatistics

Developing Machine Learning Methodology for Precision Health.
LDR:02770nmm a2200361 4500 001 2279038
005 20210730131525.5
008 220723s2020 ||||||||||||||||| ||eng d
020 $a 9798635265857
035 $a (MiAaPQ)AAI27836497
035 $a AAI27836497
040 $a MiAaPQ $c MiAaPQ
100 1 $a Jiang, Xiaotong. $3 3557450
245 1 0 $a Developing Machine Learning Methodology for Precision Health.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2020
300 $a 141 p.
500 $a Source: Dissertations Abstracts International, Volume: 82-01, Section: B.
500 $a Advisor: Kosorok, Michael R.
502 $a Thesis (Ph.D.)--The University of North Carolina at Chapel Hill, 2020.
506 $a This item must not be sold to any third party vendors.
520 $a Precision health has been an increasingly popular solution to improve health care quality and guide the decision making process. This includes precision medicine (at the individual level) and precision public health (at the population level such as communities and institutions). By learning from the available medical data with advanced analytical tools, precision health recommends the treatments that are individualized to each patient or entity to maximize clinical outcomes for each individual. We extend and develop three machine learning methods to improve the estimation of optimal individualized treatment regimes in precision health: the jackknife estimator of value function of precision medicine models compared with zero-order models, doubly robust outcome-weighted estimators with deep neural network structures for complex and large data, and risk-adjusted adverse event monitoring for survival data. First, motivated by a knee osteoarthristis trial, we estimate value functions and select the optimal treatment with the jackknife method whose consistency is established under weak assumptions. Next, we implement deep learning architecture in augmented outcome-weighted learning to increase model flexibility and computation efficiency, especially for high-dimensional data such as medical imaging. Lastly, we develop a risk-adjusted survival model to monitor adverse events and estimate its variance for hierarchical, right-censored data with recurrent events. All three methodologies aim to solve practical, health-related challenges and provide data-driven decision support and operations.
590 $a School code: 0153.
650 4 $a Biostatistics. $3 1002712
653 $a Biostatistics
653 $a Decision making
653 $a Deep learning
653 $a Machine learning
653 $a Precision health
653 $a Survival analysis
690 $a 0308
710 2 $a The University of North Carolina at Chapel Hill. $b Biostatistics. $3 1023527
773 0 $t Dissertations Abstracts International $g 82-01B.
790 $a 0153
791 $a Ph.D.
792 $a 2020
793 $a English
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=27836497