東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Similarity measures for time-varying...

Gaskin, Nicole Frances.

Linked to FindBook

Google Book

Amazon

博客來

Similarity measures for time-varying rheumatoid arthritis patient data.

Record Type:	Language materials, printed : Monograph/item
Title/Author:	Similarity measures for time-varying rheumatoid arthritis patient data./
Author:	Gaskin, Nicole Frances.
Description:	131 p.
Notes:	Advisers: Bradford G. Nickerson; Monique F. Frize.
Contained By:	Masters Abstracts International41-02.
Subject:	Biology, Biostatistics. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=MQ72476
ISBN:	9780612724761

Similarity measures for time-varying rheumatoid arthritis patient data.
Gaskin, Nicole Frances.

Similarity measures for time-varying rheumatoid arthritis patient data. - 131 p.

Advisers: Bradford G. Nickerson; Monique F. Frize.

Thesis (M.Sc.E.)--University of New Brunswick (Canada), 2001.

The two techniques give reasonable results yet point to the need for further investigation of a robust time-varying similarity measure for medical data.

ISBN: 9780612724761Subjects--Topical Terms:

1018416
Biology, Biostatistics.

Similarity measures for time-varying rheumatoid arthritis patient data.
LDR:02751nam 2200325 a 45 001 965575
005 20110906
008 110906s2001 eng d
020 $a 9780612724761
035 $a (UnM)AAIMQ72476
035 $a AAIMQ72476
040 $a UnM $c UnM
100 1 $a Gaskin, Nicole Frances. $3 1288348
245 1 0 $a Similarity measures for time-varying rheumatoid arthritis patient data.
300 $a 131 p.
500 $a Advisers: Bradford G. Nickerson; Monique F. Frize.
500 $a Source: Masters Abstracts International, Volume: 41-02, page: 0478.
502 $a Thesis (M.Sc.E.)--University of New Brunswick (Canada), 2001.
520 $a The two techniques give reasonable results yet point to the need for further investigation of a robust time-varying similarity measure for medical data.
520 $a This thesis investigates two methods, the area difference (AD) method and the cross correlation coefficient (CCC) method, to quantify the similarity of the temporal changes in a disease variable between two patients.
520 $a A database of 816 Rheumatoid Arthritis (RA) patients with demographic and time-varying data was revised to include only the patients with meaningful time-varying data. A subset of thirty patients (test cases) was selected at random from the remaining 523 patients to experimentally evaluate the similarity measures for the Lansbury score (an overall measure of RA disease activity). Eight match cases (the three most similar and the least similar patients as determined for both the AD and CCC methods) were extracted for each test case. The graphically depicted results were randomized for each test case and presented in a 150-page document for blind evaluation by a rheumatologist. Each match case was rated from 5 (highest) to 1 (lowest) for degree of match to the test case and the three best matches were recorded.
520 $a Results show that for the 90 best matches (top 3 for each of 30 test cases) selected by the rheumatologist, the CCC measure may prove to be better than the AD measure, though when tested at the 99% confidence level the CCC measure wasn't significantly better. It appeared, through comments supplied by the rheumatologist in the evaluation, that the short duration of some cases (test or match) may have biased some of the experimental rating and ranking. In three of the 90 results, the rheumatologist chose the least similar patient as one of the top three matches.
590 $a School code: 0823.
650 4 $a Biology, Biostatistics. $3 1018416
650 4 $a Health Sciences, Medicine and Surgery. $3 1017756
690 $a 0308
690 $a 0564
710 2 0 $a University of New Brunswick (Canada). $3 1057458
773 0 $t Masters Abstracts International $g 41-02.
790 $a 0823
790 1 0 $a Frize, Monique F., $e advisor
790 1 0 $a Nickerson, Bradford G., $e advisor
791 $a M.Sc.E.
792 $a 2001
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=MQ72476