Method for identification of retinal cell types intrinsic properties
Retinal prostheses are configured to apply retinal stimulus signals based on retinal cell classification such as cell types or cell clusters identified based on retinal cell signals. Cell types are identified based on spontaneous or induced retinal electrical signals and analyzed based on temporal patterns of electrical activity, spatio-temporal voltage waveforms, and signal correlations that can be obtained from temporal spike patterns. Cell type indications are stored, and a signals are applied to the retina based on the stored cell type indications.
1. A method, comprising:
receiving electrical signals corresponding to a plurality of retinal locations;
processing the electrical signals using a signal analyzer, the processing including identifying cell types of retinal cells associated with the plurality of retinal locations;
storing indications of the cell types within a memory wherein each of the cell types corresponds to at least one of a plurality of different cell types;
applying, by a retinal driver, retinal stimulus signals to the plurality of retinal locations based on the identified cell types; and
adjusting timing of ones of the retinal stimulus signals based on the identified cell types.
2. The method of claim 1 , wherein the identifying the cell types includes recording temporal spike patterns at the plurality of retinal locations and processing the spike patterns to determine cell types.
3. The method of claim 2 , wherein the spike patterns are processed based on spike pattern clustering to determine cell types.
4. The method of claim 1 , wherein the identifying the cell types includes obtaining retinal signals from a plurality of locations, and estimating retinal cell types at the plurality of locations based on cross-correlations of the obtained retinal signals.
5. The method of claim 4 , wherein the retinal cell types are estimated based on clustering of correlations.
6. The method of claim 1 , wherein the cell types are identified based on cell spatio-temporal voltage patterns and wherein at least one of the indications is indicative of proportions of different types of cells associated with one of the plurality of retinal locations.
7. The method of claim 6 , wherein the cell types are identified by clustering based on cell spatio-temporal voltage patterns.
8. The method of claim 1 , wherein the cell types are identified based on clustering of one or more of retinal signal correlations, temporal spike patterns, or spatio-temporal waveforms or combinations thereof.
9. A method, comprising:
receiving electrical signals from a plurality of retinal electrodes;
processing, by a signal analyzer, the electrical signals wherein the processing includes identifying locations of light sensitive regions of a plurality of retinal cells associated with respective retinal electrodes;
determining portions of an image to be associated with the locations of the light sensitive regions; and
applying retinal stimulus signals to the retinal electrodes wherein each of the retinal stimulus signals is associated with one of the portions of the image wherein the retinal stimulus signals are based on a current image and at least one prior image.
10. The method of claim 9 , wherein the retinal stimulus signals are based on retinal cell classifications at respective electrodes.
11. The method of claim 10 , wherein the retinal cell classifications include at least one of cell types and cell clusters.
12. An apparatus, comprising:
a signal conditioner configured to receive electrical signals corresponding to a plurality of retinal locations;
a signal analyzer coupled to the signal conditioner and configured for processing the electrical signals, the processing including identifying cell types of retinal cells associated with the plurality of retinal locations;
a memory for storing indications of the cell types wherein each of the cell types corresponds to at least one of a plurality of different cell types; and
a retinal driver configured to apply retinal stimulus signals to the plurality of retinal locations based on the identified cell types wherein timing of ones of the retinal stimulus signals is adjusted based on the identified cell types.
13. The apparatus of claim 12 wherein the signal analyzer is further configured to record temporal spike patterns at the plurality of retinal locations and process the spike patterns to determine cell types.
14. The apparatus of claim 13 wherein the signal analyzer is further configured to process the spike patterns based on spike pattern clustering to determine cell types.
15. The apparatus of claim 12 wherein the signal analyzer is further configured to obtain retinal signals from a plurality of locations and to estimate retinal cell types at the plurality of locations based on cross-correlations of the obtained retinal signals.
16. The apparatus of claim 15 wherein the signal analyzer is further configured to estimate the retinal cell types based on clustering of correlations.
17. The apparatus of claim 12 wherein the signal analyzer is further configured to identify the cell types based on cell spatio-temporal voltage patterns and wherein at least one of the indications is indicative of proportions of different types of cells associated with one of the plurality of retinal locations.
18. The apparatus of claim 12 wherein the signal analyzer is further configured to identify the cell types based on clustering of one or more of retinal signal correlations, temporal spike patterns, or spatio-temporal waveforms or combinations thereof.
19. An apparatus, comprising:
a signal conditioner configured to receive electrical signals from a plurality of retinal electrodes;
a signal analyzer coupled to the signal conditioner and configured for (i) processing the electrical signals wherein the processing includes identifying locations of light sensitive regions of a plurality of retinal cells associated with respective retinal electrodes, and (ii) determining portions of an image to be associated with the locations of the light sensitive regions; and
a retinal driver configured to apply retinal stimulus signals to the retinal electrodes wherein each of the retinal stimulus signals is associated with one of the portions of the image and wherein the retinal stimulus signals are based on a current image and at least one prior image.