Computerized image processing in Ophthalmology

Any interpretation of information provided in image form will be mainly subjective and qualitative and is frequently complicated by the numerous geometrical and photometric distortions inlroduced during the image formation process. There is a general desire to introduce modern image analysis and processing techniques into ophtalmology. These techniques are now widely used in other disciplines and reduce the difficulties inherent in analysis, rendering the clinical investigation more objective by increasing the possibility of making quantitative as well as qualitative evaluations. It should be remebered that the introduction of large scale integration techniques (VLSI), and of microcomputers and digital systems are producing continual progresses in three main types of applications: (a) instrument checking; (b) data processing; (c) image and signal processing. With reference to the first of these applications, in the exhibition which was held during the congress, there were a number of microprocessors integrated into ophthalmologic instruments to execute checking and partial data processing operations. Examples we can mention here are the most rececnt generations of autokeratometers, echobiometers, visual field analyzers, fundus cameras, autorefractometers, and devices for the acquisition and processing of evoked potentials and electroretinograms. In devices of this type, the most the microprocessor can do is to process one-dimensional signals or to analyze relatively simple data structures, such as, for example, measuring perimeters or determining evoked potentials.