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

Group sequential designs and inferen...

Shu, Yu.

FindBook

Google Book

Amazon

博客來

Group sequential designs and inference of a medical diagnostic test with binary outcomes.

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	Group sequential designs and inference of a medical diagnostic test with binary outcomes./
作者:	Shu, Yu.
面頁冊數:	116 p.
附註:	Advisers: Zhaohai Li; Aiyi Liu.
Contained By:	Dissertation Abstracts International68-08B.
標題:	Biology, Biostatistics. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3275156
ISBN:	9780549162933

Group sequential designs and inference of a medical diagnostic test with binary outcomes.
Shu, Yu.

Group sequential designs and inference of a medical diagnostic test with binary outcomes. - 116 p.

Advisers: Zhaohai Li; Aiyi Liu.

Thesis (Ph.D.)--The George Washington University, 2007.

Sequential designs have been well developed in the area of clinical trial research for the past thirty years. The methods of estimating of parameters followed by a sequential design have also attracted more attention in the past decade. However, the application of sequential design methods has not been well recognized in the development of medical diagnostic tests research. This dissertation develops group sequential designs in the early research of developing medical diagnostic tests with binary outcomes.

ISBN: 9780549162933Subjects--Topical Terms:

1018416
Biology, Biostatistics.

Group sequential designs and inference of a medical diagnostic test with binary outcomes.
LDR:03620nam 2200313 a 45 001 962610
005 20110830
008 110831s2007 ||||||||||||||||| ||eng d
020 $a 9780549162933
035 $a (UMI)AAI3275156
035 $a AAI3275156
040 $a UMI $c UMI
100 1 $a Shu, Yu. $3 1285670
245 1 0 $a Group sequential designs and inference of a medical diagnostic test with binary outcomes.
300 $a 116 p.
500 $a Advisers: Zhaohai Li; Aiyi Liu.
500 $a Source: Dissertation Abstracts International, Volume: 68-08, Section: B, page: 4925.
502 $a Thesis (Ph.D.)--The George Washington University, 2007.
520 $a Sequential designs have been well developed in the area of clinical trial research for the past thirty years. The methods of estimating of parameters followed by a sequential design have also attracted more attention in the past decade. However, the application of sequential design methods has not been well recognized in the development of medical diagnostic tests research. This dissertation develops group sequential designs in the early research of developing medical diagnostic tests with binary outcomes.
520 $a Human samples and tissues are generally needed in medical diagnostic tests, and new diagnostic tests are often evaluated through experiments and survey in which data accumulate over a period of time. Like clinical trials, it is natural to monitor results as they occur with a view to terminating early or making modification of the design for ethical, administrative and economic reasons. The sequential designs to evaluate the accuracy of a single medical test in its early development phases from case-control studies are proposed.
520 $a Sensitivity and specificity of a medical test are simultaneously tested by two group sequential approaches: one method adopts the error spending method by Lan and DeMets (1983) to stop a trial early because of high accuracy of the test; the other one minimizes the average sample size of the study when the accuracy of the medical test is under expectation. In some studies, it is important and convenient to test a one-dimension summary index of diagnostic tests. We then propose a group sequential method to evaluate a summary index that is a linear combination of sensitivity and specificity. The limitations in early stages of research development make two-stage designs more appealing than other multiple stage designs. The two-stage designs illustrated here generally requires more subjects per group than the fixed size designs with the same probabilities of type I and II errors. Yet, the advantages of sequential methods are to allow early termination and the reduction of the average sample sizes under the hypotheses.
520 $a The point and interval estimates of sensitivity and specificity followed by a group sequential design are challenging because of the multiple endpoints. We compared the bias, and mean squared errors of the maximum likelihood estimator and the Rao-Blackwell unbiased estimator of sensitivity and specificity in two-stage designs. The coverage probability and the expected length of confidence intervals for the parameters are compared by simulation studies. The results show that the normal approximation theories do not apply accurately to small sample size sequential designs.
590 $a School code: 0075.
650 4 $a Biology, Biostatistics. $3 1018416
690 $a 0308
710 2 $a The George Washington University. $3 1017405
773 0 $t Dissertation Abstracts International $g 68-08B.
790 $a 0075
790 1 0 $a Li, Zhaohai, $e advisor
790 1 0 $a Liu, Aiyi, $e advisor
791 $a Ph.D.
792 $a 2007
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3275156