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

FindBook

Google Book

Amazon

博客來

Design of a novel paroxysmal atrial fibrillation identification system.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Design of a novel paroxysmal atrial fibrillation identification system./
作者:	Lynn, Ke-Shiuan.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2004,
面頁冊數:	328 p.
附註:	Source: Dissertations Abstracts International, Volume: 65-11, Section: B.
Contained By:	Dissertations Abstracts International65-11B.
標題:	Electrical engineering. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3114578
ISBN:	9780496620708

Design of a novel paroxysmal atrial fibrillation identification system.
Lynn, Ke-Shiuan.

Design of a novel paroxysmal atrial fibrillation identification system. - Ann Arbor : ProQuest Dissertations & Theses, 2004 - 328 p.

Source: Dissertations Abstracts International, Volume: 65-11, Section: B.

Thesis (Ph.D.)--Cornell University, 2004.

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

We design a novel two-stage Paroxysmal Atrial Fibrillation (PAF) identification system to perform short-term prediction of the onset of PAF based on half-hour heart rate variability (HRV) signals. The two-stage PAF identification system identify PAF HRV sequences with significant level of PAF-related abnormal beats at the first stage and identify other PAF HRV sequences at the second stage. We formulate the design of the identification system as a constrained, combinatorial, multiple-objective optimization problem. To construct the stage-1 PAF identification subsystem, we first develop beat-correlated features to capture different interbeat-to-interbeat correlations in HRV sequences. We then develop a two-phase solution algorithm to select critical beat-correlated features and to train a binary classifier to identify PAF HRV sequences with significant level of PAF-related abnormal beats. To construct the stage-2 PAF identification subsystem, we first develop nonlinear features to capture different nonlinear properties in HRV sequences. We then develop a three-phase solution algorithm to select critical nonlinear features and to train a feedforward neural network to identify PAF HRV sequences that containing no significant level of PAF-related abnormal beats. Distinguish features of the two-stage PAF identification system include (i) it achieves high sensitivity with acceptable specificity, (ii) it requires only short duration of measurements, say 30 minutes of HRV signals, (iii) it has excellent generalization capability, (iv) it is robust, and (v) it performs fast identification.

ISBN: 9780496620708Subjects--Topical Terms:

649834
Electrical engineering.
Subjects--Index Terms:

Atrial fibrillation

Design of a novel paroxysmal atrial fibrillation identification system.
LDR:02803nmm a2200361 4500 001 2352895
005 20221205085635.5
008 241004s2004 ||||||||||||||||| ||eng d
020 $a 9780496620708
035 $a (MiAaPQ)AAI3114578
035 $a AAI3114578
040 $a MiAaPQ $c MiAaPQ
100 1 $a Lynn, Ke-Shiuan. $3 3692562
245 1 0 $a Design of a novel paroxysmal atrial fibrillation identification system.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2004
300 $a 328 p.
500 $a Source: Dissertations Abstracts International, Volume: 65-11, Section: B.
500 $a Publisher info.: Dissertation/Thesis.
500 $a Advisor: Chiang, Hsiao-Dong.
502 $a Thesis (Ph.D.)--Cornell University, 2004.
506 $a This item must not be sold to any third party vendors.
506 $a This item must not be added to any third party search indexes.
520 $a We design a novel two-stage Paroxysmal Atrial Fibrillation (PAF) identification system to perform short-term prediction of the onset of PAF based on half-hour heart rate variability (HRV) signals. The two-stage PAF identification system identify PAF HRV sequences with significant level of PAF-related abnormal beats at the first stage and identify other PAF HRV sequences at the second stage. We formulate the design of the identification system as a constrained, combinatorial, multiple-objective optimization problem. To construct the stage-1 PAF identification subsystem, we first develop beat-correlated features to capture different interbeat-to-interbeat correlations in HRV sequences. We then develop a two-phase solution algorithm to select critical beat-correlated features and to train a binary classifier to identify PAF HRV sequences with significant level of PAF-related abnormal beats. To construct the stage-2 PAF identification subsystem, we first develop nonlinear features to capture different nonlinear properties in HRV sequences. We then develop a three-phase solution algorithm to select critical nonlinear features and to train a feedforward neural network to identify PAF HRV sequences that containing no significant level of PAF-related abnormal beats. Distinguish features of the two-stage PAF identification system include (i) it achieves high sensitivity with acceptable specificity, (ii) it requires only short duration of measurements, say 30 minutes of HRV signals, (iii) it has excellent generalization capability, (iv) it is robust, and (v) it performs fast identification.
590 $a School code: 0058.
650 4 $a Electrical engineering. $3 649834
650 4 $a Biomedical research. $3 3433833
650 4 $a Biomedical engineering. $3 535387
653 $a Atrial fibrillation
653 $a Heart rate
653 $a Paroxysmal
690 $a 0544
690 $a 0541
710 2 $a Cornell University. $3 530586
773 0 $t Dissertations Abstracts International $g 65-11B.
790 $a 0058
791 $a Ph.D.
792 $a 2004
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3114578