论文信息 - Automated microaneurysm detection

Automated microaneurysm detection

A digital image processing system has been developed to quantify and monitor the presence of microaneurysms in retinal fluorescein angiograms. The system is fully automated (except for the initial digitisation), and includes stages for preprocessing, registration of same-eye images obtained at different visits, fovea detection for placement of regions-of-interest, and detection of microaneurysms and microaneurysm turnover. The system achieves 82% sensitivity for detecting microaneurysms with a specificity of 84%. Preliminary investigations indicate a high microaneurysm turnover for most patients with diabetic retinopathy, which may have significant clinical implications.

Peter F. Sharp | Michael J. Cree | John A. Olson | John V. Forrester | Kenneth C. McHardy

[1] 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.

[2] E M Kohner,et al. Does microaneurysm count reflect severity of early diabetic retinopathy? , 1986, Ophthalmology.

[3] W. Foulds,et al. Common causes of blindness and visual handicap in the west of Scotland. , 1983, The British journal of ophthalmology.

[4] Jean-Claude Klein,et al. Automatic Detection Of Microaneurysms In Retinopathy Fluoro-Angiogram , 1984, Optics + Photonics.

[5] Joachim H. Nagel,et al. Registration of high-resolution images of the retina , 1992, Medical Imaging.

[6] R. Klein,et al. The relationship of retinal microaneurysm counts to the 4-year progression of diabetic retinopathy. , 1989, Archives of ophthalmology.