This paper presents a platform independent system, RetinaView, that is designed for manual analysis as well as automatic segmentation and characterization of multi-modality retinal images and videos. A user-friendly interface that allows display of individual retinal images in four or more windows has been designed for simultaneous viewing of multiple modalities in research-based retinal studies. A variety of tools have been developed for RetinaView, including automatic segmentation of retinal vasculature, segmentation of small vascular abnormalities found in early stages of diabetic retinopathy (microaneurysms), segmentation of drusen, lesions pathoneumonic of age related macular degeneration (ARMD), analysis of fluorescein and indocyanine green videos, and manual annotation capabilities and fovea location. This system will enable ophthalmological researchers, clinicians, ophthalmologists to study, diagnose, and monitor the progression of pathologies in retinal images with much greater precision than is possible with totally manual techniques. This paper presents an overview of the RetinaView system and gives examples of its application in segmentation of normal anatomical features as well as pathological lesions.
[1]
Stephen A Gibson.
An optimal FFT-based algorithm for mosaicking images
,
1999
.
[2]
R. Klein,et al.
Ten-year incidence and progression of age-related maculopathy: The Beaver Dam eye study.
,
2002,
Ophthalmology.
[3]
R. Klein,et al.
The five-year incidence and progression of age-related maculopathy: the Beaver Dam Eye Study.
,
1997,
Ophthalmology.
[4]
Sunanda Mitra,et al.
Full automation of morphological segmentation of retinal images: a comparison with human-based analysis
,
2003,
SPIE Medical Imaging.
[5]
P F Sharp,et al.
An image-processing strategy for the segmentation and quantification of microaneurysms in fluorescein angiograms of the ocular fundus.
,
1996,
Computers and biomedical research, an international journal.