Feature analysis in slit-lamp image for nuclear cataract diagnosis

Huiqi Li, Joo Hwee Lim, Jiang Liu, Damon Wing, Kee Wong, Tien Yin Wong

Research output: Chapter in Book/Conference proceedingConference contributionpeer-review

11 Citations (Scopus)

Abstract

Nuclear cataract is the most common type of age-related cataract and it is clinically diagnosed using slit-lamp images. Objective measurement of the features in slit-lamp image is investigated using a computerized software system. The correlation between the features and the nuclear cataract grades is analyzed. Experimental results show that intensity of sulcus, color in the lens and nucleus region, intensity of nucleus, and color in posterior reflex region are the key features for grading nuclear cataract. This study of feature analysis can benefit clinical cataract diagnosis and clinical research. The feature analysis can also be utilized to investigate the performance of different graders and be employed in training of new graders.

Original languageEnglish
Title of host publicationProceedings - 2010 3rd International Conference on Biomedical Engineering and Informatics, BMEI 2010
Pages253-256
Number of pages4
DOIs
Publication statusPublished - 2010
Externally publishedYes
Event3rd International Conference on BioMedical Engineering and Informatics, BMEI 2010 - Yantai, China
Duration: 16 Oct 201018 Oct 2010

Publication series

NameProceedings - 2010 3rd International Conference on Biomedical Engineering and Informatics, BMEI 2010
Volume1

Conference

Conference3rd International Conference on BioMedical Engineering and Informatics, BMEI 2010
Country/TerritoryChina
CityYantai
Period16/10/1018/10/10

Keywords

  • Computer-aided diagnosis
  • Feature analysis
  • Nuclear cataract
  • Slit-lamp image

ASJC Scopus subject areas

  • Biomedical Engineering
  • Health Informatics
  • Health Information Management

Fingerprint

Dive into the research topics of 'Feature analysis in slit-lamp image for nuclear cataract diagnosis'. Together they form a unique fingerprint.

Cite this