Exploration of a dialog-based tunable retina encoder for retina implants

We propose a learning retina encoder (RE) with tunable spatiotemporal funct ions and corresponding perception-based dialog procedure for a retina impla nt as visual prosthesis. A learning DSP-based RE was developed for real tim e implementation of about 200 individually tunable spatiotemporal receptive field (RF)-filters approximating M- or P-ganglion cells to map visual patt erns onto pulse trains as stimulation inputs for implanted microcontact foi ls. An experimental environment providing visual feedback for humans with n ormal vision was developed to test the adaptive dialog module. With the pre sented tuning schema, interaction of RE and dialog module becomes possible even before tests with blind subjects. (C) 1999 Elsevier Science B.V. All r ights reserved.