Computer-assisted, interactive fundus image processing for macular drusen quantitation

Purpose: To design and validate a software package to quantitate the area s ubtended by drusen in color fundus photographs for the conduct of efficient , accurate clinical trials in age-related macular degeneration. Design: Algorithm and software development. Comparisons with manual methodo logies. Participants: Evaluation and testing on color fundus photographs from patie nt records and from eyes enrolled in the Choroidal Neovascularization Preve ntion Trial. Methods: Fundus photographs of eyes with drusen were digitized, The green c hannel was selected for maximum contrast and preprocessed with filtering an d shade correction to minimize noise, improve contrast, and correct for ill umination and background inhomogeneities. Local thresholding and region-gro wing algorithms identified drusen, Multiple levels of supervision were inco rporated to maximize robustness, accuracy, and validity. Validation studies compared computer-assisted with manual grading by an experienced grader. I ntraclass correlation coefficients were calculated as a measure of the conc ordance between manual and computer-assisted fundus gradings, Main Outcome Measures: Drusen area and concordance with manual grading. Results: Automated supervised image analysis offers extreme robustness and accuracy, Most images were segmented with little or no supervision, with pr ocessing times on the order of 5 seconds. More complicated images required supervision and a total analysis time varying from 20 seconds to 5 minutes, with most of this time devoted to inspection and comparison. Interactive i mage processing affords arbitrarily close concordance with manual drusen id entification, with calculated intraclass correlation coefficients of 0.92 a nd 0.93 for comparison of manual with automated, supervised grading by two observers. Conclusions: Automated supervised fundus image analysis is an efficient, ro bust, valid technique for drusen quantitation from color fundus photographs . This approach should prove useful in the conduct of efficient accurate cl inical trials for age-related macular degeneration.