If you want to design the most perfect, low-energy photodetector for future cameras and artificial retinas, use what is called “efficient coding theory” to arrange the sensors.
Or you can see mammals retina..
In two papers on retinal structure, neurobiologists at Duke University showed that natural selection and evolutionary rigor shape the retina of the eye, as this theory of optimization predicts. And it is far ahead of what ergonomics can achieve at this point in the retina.
In a previous treatise published in March last year NatureResearchers have shown that rat and monkey retinas are arranged in a pattern of sensitivities that mimics what efficient coding theory predicts.Different sets of retinal neurons are sensitive to individual stimuli: bright, dark, moving, etc., they are arranged in three dimensions mosaic Of cells that work to add images.
By the way, in the treatise that appears this week Minutes of the National Academy of SciencesJohn Pearson, an assistant professor of biostatistics and bioinformatics at the School of Medicine, said: “Mosaics not only overlap randomly, they do not overlap in a highly ordered way.”
Greg Field, an assistant professor of neurobiology at Duke University School of Medicine, said: “The monkey retina and our retina are almost indistinguishable,” he said. “The fact that we observed this in the monkey retina gives us incredible confidence that our retinas are similarly aligned.”
In the cross section of the retina, the body of the ganglion cells, the round spheres containing the nucleus, are layered together, but they have branched dendrites like trees. Thick layer It looks like the tangled roots of potted houseplants. It is this thick and stunningly complex layer that has a regular pattern of mosaics of varying sensitivities.
The ganglion cells beneath the dendrite layer basically only output 1s and 0s. Sensitivity comes from the mosaic itself. And the mosaic is not only optimally placed, but also adapted to the current situation.
“The retina is not one mosaic. It is the whole of the stacked mosaics. And each of these mosaics Field of view“Field said. The mammalian retina analyzes about 40 different visual features.
“The depth at which dendrites reach the retina is like an addressing scheme, and deeper can provide some information,” says Field. “If it’s shallow, you get another kind of information. In fact, the deep one gets the” off “signal and the shallow one gets the” on “signal. Therefore, you can use many detectors that sample the same location. It’s a visual world because it uses depth to carry different types of signals. “
One of the reasons arrays are so efficient is that they save energy by not responding to some stimuli. In a very dark room, the environment is “noisy” to the receptors, so they adjust most statics and react only to very bright ones.
“The louder the world is, the more noisy the cell can be about what it reacts to,” Pearson said. “And you’ll find that the more stringent they are, the less verbose they are, so you can deploy them in a way that doesn’t need to be duplicated.”
If there was no noise in the visual environment, the detector mosaics would overlap each other, explained Na Young Jun, a graduate student who was the first author of the treatise and co-author of another treatise. However, she computationally modeled 168 different noise conditions and found that the higher the noise, the greater the offset between the detectors.
The team found that in the retina of a living mammal, the mosaic was offset, as the theory predicts. That is, the retina is optimized to handle higher noise conditions.
If you’re a small, tasty forest creature like a mouse, “your survival doesn’t really depend on what’s easy to see,” Field said. “It depends on what is hard to see. Therefore, the retina is tuned to be optimized to detect what is really hard to see.”
“This is an important design feature for incorporation into any kind of artificial retina you want to build,” says Field. However, it can take some time to incorporate this idea into your smartphone. For one thing, the retina is alive and self-organizing and adapts and changes over time.
The energy consumption of the human retina is also orders of magnitude lower than the best smartphone sensors at the moment, Jun said. For example, 5 megapixels, 1/5NS 1 inch OmniVision OV5675 smartphone image sensor consumes 1.92×10-Ten Watts. The human retina is conservatively estimated to consume about 6 percent (1.27×10) of it.-11 Watts in bright light).In dim light, the energy consumption of the eyes is about 5.08×10-11However, it also captures a single photon, which was not possible with a smartphone camera.
The next feature of the system that the team wants to work on is the element of time. In other words, the difference in response time of retinal cells is added to form the sensation of movement, that is, the interpretation of moving images. Part of that depends on the speed at which the individual detectors fire, according to Jun.
Na Young Jun et al, scene statistics and noise determine the relative placement of receptive field mosaics, Minutes of the National Academy of Sciences (2021). DOI: 10.1073 / pnas.2105115118
Suva Roy et al, intermosaic coordination of retinal receptive fields, Nature (2021). DOI: 10.1038 / s41586-021-03317-5
Quote: Living Retina Achieves Sensitivity and Efficiency Engineers obtained from https://medicalxpress.com/news/2021-09-retina-sensitivevity-efficiency.html on September 28, 2021 (2021) September 28) I can only dream
This document is subject to copyright. No part may be reproduced without written permission, except for fair transactions for personal investigation or research purposes. The content is provided for informational purposes only.
Living retina achieves sensitivity and efficiency engineers can only dream about Source link Living retina achieves sensitivity and efficiency engineers can only dream about
The post Living retina achieves sensitivity and efficiency engineers can only dream about appeared first on California News Times.