Language:
繁體中文
English
日文
說明(常見問題)
南開科技大學
圖書館首頁
編目中圖書申請
登入
回首頁
切換:
標籤
|
MARC模式
|
ISBD
Spatiotemporal independent component...
~
Barriga, Eduardo S.
Spatiotemporal independent component analysis with applications to optical imaging.
紀錄類型:
書目-電子資源 : 單行本
正題名/作者:
Spatiotemporal independent component analysis with applications to optical imaging./
作者:
Barriga, Eduardo S.
面頁冊數:
110 p.
附註:
Source: Dissertation Abstracts International, Volume: 67-09, Section: B, page: 5270.
Contained By:
Dissertation Abstracts International67-09B.
標題:
Engineering, Biomedical. -
電子資源:
Download PDF (下載PDF全文)
ISBN:
9780542849046
Spatiotemporal independent component analysis with applications to optical imaging.
Barriga, Eduardo S.
Spatiotemporal independent component analysis with applications to optical imaging.
- 110 p.
Source: Dissertation Abstracts International, Volume: 67-09, Section: B, page: 5270.
Thesis (Ph.D.)--The University of New Mexico, 2006.
Independent Component Analysis (ICA) refers to a set of statistical processing techniques that estimate statistically independent sources mixed by a linear mixing matrix. The sources are estimated using only the measures available by the output of an unknown system. In this dissertation, we develop a new spatiotemporal ICA model and a new ICA algorithm (ICA-P). The new algorithm and spatiotemporal model are used to model and then measure physiological responses in the retina to an optical stimulation from a new, non-invasive optical imaging device. The new device was designed to provide early detection of retinal damage from glaucoma and other retinal diseases.
ISBN: 9780542849046Subjects--Topical Terms:
1000005515
Engineering, Biomedical.
Spatiotemporal independent component analysis with applications to optical imaging.
LDR
:03244nmm 2200313 4500
001
1000004921
005
20070601084733.5
008
070601s2006 eng d
020
$a
9780542849046
035
$a
(UnM)AAI3230562
035
$a
AAI3230562
040
$a
UnM
$c
UnM{me_controlnum}
100
1
$a
Barriga, Eduardo S.
$3
1000006076
245
1 0
$a
Spatiotemporal independent component analysis with applications to optical imaging.
300
$a
110 p.
500
$a
Source: Dissertation Abstracts International, Volume: 67-09, Section: B, page: 5270.
500
$a
Adviser: Marios Pattichis.
502
$a
Thesis (Ph.D.)--The University of New Mexico, 2006.
520
$a
Independent Component Analysis (ICA) refers to a set of statistical processing techniques that estimate statistically independent sources mixed by a linear mixing matrix. The sources are estimated using only the measures available by the output of an unknown system. In this dissertation, we develop a new spatiotemporal ICA model and a new ICA algorithm (ICA-P). The new algorithm and spatiotemporal model are used to model and then measure physiological responses in the retina to an optical stimulation from a new, non-invasive optical imaging device. The new device was designed to provide early detection of retinal damage from glaucoma and other retinal diseases.
520
$a
The new spatiotemporal model is used to develop realistic one- and three-dimensional, synthetic source simulations that closely resemble processes found during visual stimulation of the retina. For the synthetic simulations, the performances of some of the most widely used ICA algorithms: JADE, Infomax, SOBI, ESD, and Fast-ICA are evaluated and compared establish for a wide variety of Signal to Noise Ratios (SNRs) in order to establish the limits of each approach.
520
$a
The proposed ICA algorithm (ICA-P) uses prior information on the stimulus to provide improved detection performance. In comparison with other considered methods (JADE, Infomax, SOBI, ESD, and Fast ICA), ICA-P is shown to provide significantly improved performance in detecting the response to the stimuli at very low SNRs. The results from the hybrid simulations show that we can estimate changes as small as 0.01% (-40 dB) of the total intensity levels in the images, far lower in magnitude than the ones present in the live data recordings.
520
$a
All algorithms are tested on optical video recordings from live cat retinal videos and live human data. The results show that it is possible to detect functional responses to visual stimulation in the retina. For the cat retinal data, the proposed spatiotemporal models can be used to detect the response in 86% of 60 live video recordings. For human retinal data, our approach can be used to detect the retinal responses in videos with low SNR. The proposed spatiotemporal approach is general enough and could be applied to other physiological data, such as fMRI.
590
$a
School code: 0142.
650
4
$a
Engineering, Biomedical.
$3
1000005515
650
4
$a
Engineering, Electronics and Electrical.
$3
170927
690
$a
0541
690
$a
0544
710
2 0
$a
The University of New Mexico.
$3
1000005571
773
0
$t
Dissertation Abstracts International
$g
67-09B.
790
1 0
$a
Pattichis, Marios,
$e
advisor
790
$a
0142
791
$a
Ph.D.
792
$a
2006
856
4 0
$u
http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3230562
$z
Download PDF (下載PDF全文)
0 筆讀者評論
館藏地:
全部
線上資料庫
出版年:
卷號:
館藏
1 筆 • 頁數 1 •
1
條碼號
典藏地名稱
館藏流通類別
資料類型
索書號
使用類型
借閱狀態
預約人數
備註欄
附件
OE0000896
線上資料庫
線上資源
線上電子書
OE
一般(Normal)
在架
0
1 筆 • 頁數 1 •
1
多媒體
評論
新增評論
分享你的心得
建立或儲存個人書籤
書目轉出
取書館別
處理中
...
變更密碼
登入