Linked Life Data

18282634

Source:http://linkedlifedata.com/resource/pubmed/id/18282634

Statements in which the resource exists as a subject.
PredicateObject
rdf:type
lifeskim:mentions
pubmed:issue
3
pubmed:dateCreated
2008-4-28
pubmed:abstractText
Accurate measurement of pupil size is essential for pupillary light reflex (PLR) analysis in clinical diagnosis and vision research. Low pupil-iris contrast, corneal reflection, artifacts and noises in infrared eye imaging pose challenges for automated pupil detection and measurement. This paper describes a computerized method for pupil detection or identification. After segmentation by a region-growing algorithm, pupils are detected by an iterative randomized Hough transform (IRHT) with an elliptical model. The IRHT iteratively suppresses the effects of extraneous structures and noise, yielding reliable measurements. Experimental results with 72 images showed a mean absolute difference of 3.84% between computerized and manual measurements. The inter-run variation for the computerized method (1.24%) was much smaller than the inter-observer variation for the manual method (7.45%), suggesting a higher level of consistency of the former. The computerized method could facilitate PLR analysis and other non-invasive functional tests that require pupil size measurements.
pubmed:language
eng
pubmed:journal
pubmed:citationSubset
IM
pubmed:status
MEDLINE
pubmed:month
Jun
pubmed:issn
0169-2607
pubmed:author
pubmed:issnType
Print
pubmed:volume
90
pubmed:owner
NLM
pubmed:authorsComplete
Y
pubmed:pagination
202-9
pubmed:meshHeading
pubmed:year
2008
pubmed:articleTitle
Computerized mouse pupil size measurement for pupillary light reflex analysis.
pubmed:affiliation
Department of Biological Engineering, University of Missouri, Columbia, MO 65211, USA.
pubmed:publicationType
Journal Article, Research Support, Non-U.S. Gov't, Evaluation Studies