The aim of the proposed research is to demonstrate the feasibility of applying perceptual-scaling methods to derive and identify visually perceptible features of in vivo cataracts. This is a step towards the National Advisory Eye Council's goal of developing """"""""an objective, reproducible, standardized classification for both type and severity of [in vivo] cataract ...."""""""" A classification system is urgently needed in drug research to determine the efficacy of proposed anti-cataract drugs, and in epidemiology to determine the role of various risk factors in development of cataract. The methods to be employed in identifying visually perceptible features of cataracts are multidimensional scaling and hierarchical cluster analysis. These psychological methods, applied successfully to other areas of medical imagery, depend on mathematical analysis of judgments of similarity made between pairs of cataractous lenses. Development of an adequate measurement system is a three stage process: (1) identification of a full set of visual features, (2) development of subjective measurement or rating scales for all features, and (3) determination that feature assessment is sufficiently reproducible both within and between observers. This pilot effort addresses the first of these steps.
