• DocumentCode
    3138588
  • Title

    Real Time Pupil Size Monitoring As a Screening Method for Diabetic Retinopathy

  • Author

    Long, Xindian ; Tonguz, Ozan K. ; Kidernan, A.

  • Author_Institution
    Carnegie Mellon Univ., Pittsburgh
  • fYear
    2007
  • fDate
    27-30 Nov. 2007
  • Firstpage
    215
  • Lastpage
    218
  • Abstract
    This paper reports an accurate pupil size monitoring and analyzing system designed for a new method for screening diabetic retinopathy. The system monitors the pupil size change under stimulation of circular or annular objects of different sizes and brightness produced on a LCD screen facing the subject. Pupil response is captured by digital cameras and the eye images acquired are processed in real time. A novel algorithm is designed to calculate the pupil size to provide data input for the screening procedure of diabetic retinopathy. To distinguish real pupil size changes from artifacts like blinking, eyelid drop or reflections, we use several image processing techniques to estimate the pupil size when the pupil shape is partially occluded.
  • Keywords
    diseases; eye; medical image processing; patient diagnosis; LCD screen; annular objects; brightness; circular objects; diabetic retinopathy; digital cameras; eye images; image processing; real time pupil size monitoring; screening method; Algorithm design and analysis; Brightness; Computer displays; Diabetes; Digital cameras; Eyelids; Image processing; Monitoring; Reflection; Retinopathy; Diabetic retinopathy; ellipse detection; image processing; pupil size;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Biomedical Circuits and Systems Conference, 2007. BIOCAS 2007. IEEE
  • Conference_Location
    Montreal, Que.
  • Print_ISBN
    978-1-4244-1524-3
  • Electronic_ISBN
    978-1-4244-1525-0
  • Type

    conf

  • DOI
    10.1109/BIOCAS.2007.4463347
  • Filename
    4463347