Abstract
Purpose:
To develop a new strategy which provides interaction free electrical stimulation of retinal ganglion cells for improving temporal and spatial resolution of retina implant systems.
Methods:
This strategy includes two steps to provide high spatial and temporal resolution for retina implant systems. In first step of this strategy, for a given electrode neighborhood the most active ganglion outputs (from our artificial retina software) are analyzed and selected as stimulation data. By this way, high resolution image is mapped to low resolution electrode matrix. In the second step, local interleaved electrical pulses are generated to reduce electrical interactions between implanted electrodes during electrical stimulation of retina. To evaluate the contribution of this strategy to the perceived image quality, computer based quantitative simulation studies and visual evaluation tests on normal seeing people were performed. For quantitative evaluation, the outputs for the classical method and the developed strategy were compared based on the Mean Squared Error (MSE), the Histogram Similarity Ratio (HSR), and Incorrectly Detected Edge Pixel (IDEP) parameters. In visual tests, performance of the method was evaluated in terms of contrast discrimination, pattern recognition,and object counting.
Results:
In simulation studies on normal seeing subjects, our strategy yielded higher scores than classical stimulation method as %10.5, %16.1, and %14.1 for the contrast discrimination, pattern recognition, and object counting tests, respectively. In the quantitative comparison, developed strategy introduced 32.8% lower MSE value, 3% higher HSR value, and %11.34 lower IDEP value than the classical method.
Conclusions:
By considering quantitative and visual tests’ results, it is concluded that this strategy is useful for stimulation of electrode matrix and can contribute to the improvement of visual prosthesis systems, especially for future’s high resolution retina implant systems. Acknowledgement: This work was supported by a grant from TUBITAK(113E181)
Keywords: 549 image processing