According to the data of A Christianson (1992 Acta Neurologica Scandinavica 86 545 – 554) the testing of half of the visual field is useful when testing patients who suffer from visual perception disorders (VPDs). With this in mind, a new computer interactive system was developed for diagnosis, testing, and rehabilitation of VPDs. Our computer interactive system was composed of several subsystems: test procedure, test records database, and machine learning algorithm. The interactive test procedure was controlled by several parameters. Results (individual tests, averages) were presented in stereo 3-D form. Individual test records were preserved and formed a database. This database was used by the machine learning algorithm (A Raskinis, 1980, FTEPI Technical Report) to create rules for VPD prediction. Medical computer-aided learning experiments with persons with some kind of VPD have been carried out in the Neurosurgical Centre of Kaunas Academic Clinics with the use of a PC (IBM AT 486). So far we have tested 52 persons with damaged right and left temporal lobes. Our objectives were: (1) to diagnose various kinds of VPD; (2) to train and rehabilitate persons with various kinds of VPD; (3) to collect and analyse data on persons who can be treated and trained; and (4) to develop efficient helping measures and new skills from the effects of intensive training.
