Automatic visual field: a software diagnostic procedure.

A recently developed technique for the automatic acquisition of data on visual field losses was tested, with reliable neuro-ophthalmological parameters, on 300 subjects with and without eye disorders. Algorithms for acquisition and processing of the central visual field were implemented on an IBM-AT personal computer with standard peripherals. The contours of scotomata (visual field losses) were best estimated by probabilistic adaptive enhancement of data sampling in those areas with greater visual variability. Image processing tools were specially designed to extract information concerning the size, shape, position and number of scotomata in the visual field. The diagnosis of wider campimetric lesions required parameters such as symmetry or specularity coefficients, obtainable by analysing the visual field of both eyes. Data obtained were correlated to the pathology involved by univariate statistical tools.