Bionic visual perception method based on "light + neurotransmitter" dual regulation

Abstract

Based on the formation mechanism of optic nerves in which organisms perceive light and neurotransmitters at the same time, photosynaptic neurotransmitter sensors are used to perceive light and neurotransmitters at the same time, and a bionic visual perception research model regulated by the joint action of light and neurotransmitters is established. Neurotransmitters are used to regulate the function of photosynapses to simulate abnormal optic nerve conduction, simulate the two abnormal conditions of "amblyopia" and "color blindness" in vision, and simulate the individual visual differences of bionic organisms. Combined with the conditions for photosynaptic neurotransmitter sensors to perceive light, the ability to capture moving light sources and identify object contours in vision is simulated, and the bionic visual perception functions of optic nerves under abnormal and normal conditions are compared. This method combines biochemical reactions with photoelectric stimulation, providing a more realistic bionic visual perception method from the perspective of biological optic nerve conduction.

Description

Technical Field

[0001] The present invention relates to the field of neurotransmitter sensors and bionic vision technology, and in particular to a bionic vision perception method based on "light+neurotransmitter" dual regulation. Background Art

[0002] The 21st century is known as the century of brain science, including the structure and function of the brain. Neurobiochemistry believes that the transmission of neurotransmitters between neurons and synapses is the molecular biological basis for the coordinated work of organisms. In biological synapses, the electrical activity of the previous neuron in the process of biological information transmission prompts the presynaptic membrane to secrete specific chemicals (i.e. neurotransmitters), which act on the postsynaptic membrane and are transmitted to the next neuron, thereby causing changes in the electrical activity of the next neuron. In order to simulate this synaptic transmission behavior, a large number of new neuromorphic synaptic devices have begun to appear, which can realize basic synaptic plasticity behavior or more complex neural reflex behavior.

[0003] Early visual perception usually relied on image sensors (such as CCD) and large-scale integrated circuit technology to simulate the visual tracking process of humans and animals. With the rapid development of neuromorphic devices, new vitality has been injected into artificial vision technology. Compared with general photodetectors, they have inherent advantages in simulating the brain's efficient and low-power neural learning process. However, modern neuroscience points out that human vision is an extremely complex neural response process. As the most important sensory organ of the human body, the human eye can receive most of the external optical information, and the formation of vision is closely related to neurotransmitters. At present, most of the technologies for bionic visual perception using neuromorphic devices are regulated by photoelectric signals. No research reports on visual perception technology based on neurotransmitter regulation have been found, which leads to differences between the simulation results and the real neurochemical reactions of the human body. How to combine the biochemical reactions between neurotransmitters and neural receptors in the visual neuron system with photoelectric stimulation through a more realistic and reliable biological method, and trigger synaptic plasticity through dual mechanisms to simulate the visual perception function of biological organisms is still a challenge.

[0004] Therefore, the present invention provides a bionic visual perception method based on "light + neurotransmitter" dual regulation, which is an advanced technology for detecting neurotransmitters from a biological perspective and simulating their neural regulation behavior. By establishing a photosynaptic neurotransmitter sensitive mechanism, the optical image is converted into a synaptic photocurrent regulated by neurotransmitters. This is a new method for simulating visual perception from the perspective of biological optic nerve conduction, which will create a truly "bionic" visual perception. Summary of the invention

[0005] The purpose of the present invention is to provide a bionic visual perception method based on "light + neurotransmitter" dual regulation, which uses light and neurotransmitters to regulate neuromorphic devices, combines biochemical reactions with photoelectric stimulation, and more realistically simulates the formation mechanism of the human visual system. It solves the problem of the single perception mechanism of existing bionic visual perception technology, provides a new method for simulating visual perception from the perspective of biological optic nerve conduction, and creates a truly "bionic" visual perception.

[0006] To achieve the above object, the present invention provides a bionic visual perception method based on "light + neurotransmitter" dual regulation, comprising the following steps:

[0007] Step S1, based on the optic nerve formation mechanism of organisms that simultaneously perceive light and neurotransmitters, a bionic visual perception research model regulated by the combined action of light and neurotransmitters is established;

[0008] Step S2, according to the established research model, by adding neurotransmitter modulation, two abnormal conditions of "amblyopia" and "color blindness" in the vision of the organism are simulated to imitate the individual visual differences of the organism;

[0009] Step S3, using the lighting conditions to perform regulation to simulate the two visual perception functions of capturing moving light sources and identifying object contours in biological vision;

[0010] Step S4, dual regulation is performed using both illumination conditions and neurotransmitter conditions, to simulate the visual perception function of individual vision of bionic organisms, and to compare the difference in visual perception function between the two.

[0011] Furthermore, the step S1 comprises:

[0012] Step S11, using a neurotransmitter receptor modified optical synaptic neurotransmitter sensor, adding different types of neurotransmitters on the surface, and using the same lighting conditions to study the synaptic photocurrent output by the sensor, and establish a relationship model between the synaptic photocurrent and the type of neurotransmitter;

[0013] Step S12, using a photosynaptic neurotransmitter sensor modified with a neurotransmitter receptor, changing the type of modified neurotransmitter receptor, and adding neurotransmitters of the same conditions on the surface under the same illumination conditions, studying the synaptic photocurrent output by the sensor, and establishing a relationship model between the synaptic photocurrent and the type of neurotransmitter receptor;

[0014] Step S13, using a neurotransmitter receptor modified optical synaptic neurotransmitter sensor, adding neurotransmitters of different concentrations on the surface, and using the same lighting conditions to study the synaptic photocurrent output by the sensor, and establish a relationship model between the synaptic photocurrent and the neurotransmitter concentration;

[0015] Step S14, using a neurotransmitter receptor modified optical synaptic neurotransmitter sensor, adding neurotransmitters under the same conditions on the surface, using excitation light of different wavelengths to study the synaptic photocurrent output by the sensor, and establishing a relationship model between the synaptic photocurrent and the wavelength of light;

[0016] Step S15, using a neurotransmitter receptor modified optical synaptic neurotransmitter sensor, adding neurotransmitters under the same conditions to the surface, using excitation light of the same wavelength and different light intensity to study the synaptic photocurrent output by the sensor, and establishing a relationship model between the synaptic photocurrent and light intensity;

[0017] Step S16, finally using a neurotransmitter receptor modified optical synaptic neurotransmitter sensor to establish a relationship model between synaptic photocurrent and neurotransmitter type, neurotransmitter receptor type, neurotransmitter concentration, excitation light wavelength and light intensity.

[0018] Furthermore, the step S2 comprises:

[0019] Step S21, using a photosynaptic neurotransmitter sensor modified with a neurotransmitter receptor, controlling the type of neurotransmitter, the type of neurotransmitter receptor, the wavelength of the excitation light and the intensity of the light to remain unchanged, reducing the surface neurotransmitter concentration, and simulating the "amblyopia" situation in the vision of an organism;

[0020] Step S22, studying the synaptic photocurrent output by the sensor under the change of neurotransmitter concentration, and the individual visual difference of the bionic organism under normal vision and "amblyopia" conditions;

[0021] Step S23, using an optical synaptic neurotransmitter sensor, controlling the type of neurotransmitter, neurotransmitter concentration, excitation light wavelength and light intensity to remain unchanged, changing the type of modified neurotransmitter receptor, and simulating the "color blindness" situation in the vision of an organism;

[0022] Step S24, studying the synaptic photocurrent output by the sensor under the change of the modified neurotransmitter receptor, and the individual visual differences between normal vision and "color blindness" of the bionic organism.

[0023] Furthermore, the step S3 comprises:

[0024] Step S31, using the light addressable function of the optical synaptic neurotransmitter sensor, adding spatial variables to the original research model and making one-to-one correspondence with the sensitive points in the sensor chip array.

[0025] Step S32, adjusting the excitation light source of the optical synapse neurotransmitter sensor to a moving light source, using time and space to describe the movement trajectory of the light source, and tracking the position change of the point light source according to the output current intensity of the sensor at each sensitive site during the movement of the light source;

[0026] Step S33, randomly select an object, apply uniform parallel light to the left side of the object, so that a shadow of the object is formed on the sensor array on the right side, and the contour of the object is recognized by the output current intensity of the sensor at each sensitive site.

[0027] Furthermore, the step S4 comprises:

[0028] Step S41, changing the types of neurotransmitters dripped on the surface of the optical synaptic neurotransmitter sensor and the modified neurotransmitter receptors to simulate the visual "amblyopia", "color blindness" and normal conditions of the organism;

[0029] Step S42, in three different visual states, using the output current caused by the sensor at each sensitive site when the light source position changes as training data, and using an artificial neural network and a simulated hardware artificial neural network model to train the motion light source capture capability;

[0030] Step S43, in three different visual states, uniform parallel light is applied to the left side of the object so that a shadow of the object is formed on the right sensor array, and the output current caused by the sensor at each sensitive site due to different object contours is used as training data, and the motion light source capture capability is trained using an artificial neural network and a simulated hardware artificial neural network model;

[0031] Step S44, training two visual perception function models under three visual states, and comparing the differences in visual perception ability caused by individual visual differences by comparing the three factors of judgment accuracy, running time, and number of iterations of the six models.

[0032] In summary, by adopting the above technical solution, the beneficial effects of the present invention are:

[0033] It can sense light and neurotransmitters at the same time, and use the concentration of neurotransmitters and the type of neurotransmitter receptors to regulate the current of the optical synapse sensor, truly simulating the visual perception ability of the human eye under abnormal and normal conditions, and providing a new method for bionic vision to simulate visual perception from the perspective of biological optic nerve conduction. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a technical roadmap for bionic visual perception methods;

[0035] Figure 2 It is a research model roadmap;

[0036] Figure 3 It is a schematic diagram of light source motion capture;

[0037] Figure 4 It is a schematic diagram of object contour judgment. DETAILED DESCRIPTION

[0038] Embodiments of the present invention are given below in conjunction with the accompanying drawings and described in detail.

[0039] A bionic visual perception method based on "light + neurotransmitter" dual regulation, such as Figure 1 As shown, the following steps are included:

[0040] Step S1, based on the optic nerve formation mechanism of organisms that simultaneously perceive light and neurotransmitters, a bionic visual perception research model regulated by the combined action of light and neurotransmitters is established, such as Figure 2 As shown; step S1 specifically includes:

[0041] Step S11, using a neurotransmitter receptor modified optical synaptic neurotransmitter sensor, adding different types of neurotransmitters on the surface, and using the same lighting conditions to study the synaptic photocurrent (Is) output by the sensor, and establish a relationship model between the synaptic photocurrent and the type of neurotransmitter.

[0042] Specifically, glutamate (Glu) is a common neurotransmitter in the central nervous system and is widely distributed in the brain. In terms of visual function transmission, Glu has been proven to be the main excitatory neurotransmitter in the axon terminals projecting from the retina to the lateral geniculate body and then to the visual cortex, and is closely related to the excitatory transmission of the visual pathway and the normal performance of visual function.

[0043] AMPA receptor is a common subtype of Glu receptor. We selected an AMPA receptor-modified optical synaptic neurotransmitter sensor, added different types of neurotransmitters Glu and GABA on the surface, and used the same lighting conditions to study the synaptic photocurrent output by the sensor.

[0044] Step S12, using a photosynaptic neurotransmitter sensor modified with neurotransmitter receptors, changing the type of modified neurotransmitter receptors, and adding neurotransmitters of the same conditions on the surface under the same lighting conditions, studying the synaptic photocurrent output by the sensor, and establishing a relationship model between the synaptic photocurrent and the type of neurotransmitter receptors.

[0045] Specifically, the study found that the biological function of AMPA receptors is mainly to mediate the transmission of excitatory neurotransmitters. The expression and activation of different subunits can determine the selective permeability of synaptic membranes to different ions, and their roles in the regulation mechanism of visual neuron plasticity may be different.

[0046] Therefore, receptors containing GluR2 subunits and receptors without GluR2 subunits were selected as research objects, different receptors were modified in optical synaptic neurotransmitter sensors, and the synaptic photocurrent output by the sensors was studied.

[0047] Step S13, using a neurotransmitter receptor modified optical synaptic neurotransmitter sensor, adding neurotransmitters of different concentrations on the surface, and using the same lighting conditions to study the synaptic photocurrent output by the sensor, and establish a relationship model between the synaptic photocurrent and the neurotransmitter concentration.

[0048] Specifically, an AMPA receptor-modified optical synaptic neurotransmitter sensor was selected, and 20%, 40%, 60%, 80%, and 100% Glu were added to the surface to study the synaptic photocurrent output by the sensor.

[0049] Step S14, using a neurotransmitter receptor modified optical synaptic neurotransmitter sensor, adding neurotransmitters under the same conditions on the surface, using excitation light of different wavelengths to study the synaptic photocurrent output by the sensor, and establishing a relationship model between the synaptic photocurrent and the wavelength of light.

[0050] Step S15, using a photosynaptic neurotransmitter sensor modified with a neurotransmitter receptor, adding neurotransmitters under the same conditions to the surface, using excitation light of the same wavelength and different light intensities to study the synaptic photocurrent output by the sensor, and establishing a relationship model between the synaptic photocurrent and light intensity.

[0051] Step S2, according to the established research model, by adding neurotransmitter modulation, two abnormal conditions of "amblyopia" and "color blindness" in biological vision are simulated to imitate the individual visual differences of biological bodies; Step S2 specifically includes:

[0052] Step S21, using a photosynaptic neurotransmitter sensor modified with a neurotransmitter receptor, controlling the type of neurotransmitter, the type of neurotransmitter receptor, the wavelength of the excitation light and the intensity of the light to remain unchanged, reducing the surface neurotransmitter concentration, and simulating the "amblyopia" situation in the vision of an organism.

[0053] Specifically, under the same lighting conditions, an AMPA receptor-modified optical synaptic neurotransmitter sensor is selected to change the concentration of the neurotransmitter Glu added to the surface. A Glu concentration of 100% indicates normal vision, and a Glu concentration of 50% indicates an abnormal "amblyopia". Step S22, under the change of neurotransmitter concentration, the synaptic photocurrent output by the sensor is studied to simulate the individual visual differences under normal vision and "amblyopia" conditions of the bionic organism.

[0054] Step S23, using an optical synaptic neurotransmitter sensor, controlling the type of neurotransmitter, neurotransmitter concentration, excitation light wavelength and light intensity to remain unchanged, changing the type of modified neurotransmitter receptor, and simulating the "color blindness" situation in the vision of an organism.

[0055] Specifically, under the same lighting conditions, receptors containing the GluR2 subunit and receptors without the GluR2 subunit are selected to represent normal vision and abnormal "color blindness", respectively.

[0056] Step S24, studying the synaptic photocurrent output by the sensor under the change of the modified neurotransmitter receptor, and the individual visual differences between normal vision and "color blindness" of the bionic organism.

[0057] Step S3, using the lighting conditions to control, simulating the two visual perception functions of capturing moving light sources and identifying object contours in biological vision; Step S3 specifically includes:

[0058] Step S31, using the light addressable function of the optical synaptic neurotransmitter sensor, adding spatial variables to the original research model and making one-to-one correspondence with the sensitive points in the sensor chip array.

[0059] Step S32, adjust the excitation light source of the optical synaptic neurotransmitter sensor to a moving light source, use time and space to describe the movement trajectory of the light source, and track the position change of the point light source according to the output current intensity of the sensor at each sensitive site during the movement of the light source. Figure 3 shown.

[0060] Step S33, take any object, apply uniform parallel light to the left side of the object, so that the shadow of the object is formed on the sensor array on the right side, and the contour of the object is recognized by the output current intensity of the sensor at each sensitive site, such as Figure 4 shown.

[0061] Step S4, using both illumination conditions and neurotransmitter conditions to perform dual regulation, to simulate the visual perception function of individual vision of bionic organisms, and to compare the difference in visual perception function between the two; Step S4 specifically includes:

[0062] Step S41, changing the types of neurotransmitters dripped on the surface of the optical synaptic neurotransmitter sensor and the modified neurotransmitter receptors to simulate the visual "amblyopia", "color blindness" and normal conditions of the organism.

[0063] Specifically, an AMPA receptor-modified optical synaptic neurotransmitter sensor is selected to change the concentration of the neurotransmitter Glu added to the surface. A Glu concentration of 100% indicates normal vision, and a Glu concentration of 50% indicates an abnormal "amblyopia" condition.

[0064] Receptors containing GluR2 subunits and receptors without GluR2 subunits are selected to represent normal vision and abnormal "color blindness" conditions, respectively.

[0065] Step S42, in three different visual states, using the output current caused by the sensor at each sensitive site when the light source position changes as training data, and using artificial neural network and simulated hardware artificial neural network model to train the motion light source capture capability.

[0066] Step S43, in three different visual states, uniform parallel light is applied to the left side of the object to form a shadow of the object on the right sensor array, and the output current caused by the sensors at each sensitive site due to different object contours is used as training data, and the object contour recognition capability is trained using artificial neural networks and simulated hardware artificial neural network models.

[0067] Step S44, training two visual perception function models under three visual states, and comparing the differences in visual perception ability caused by individual visual differences by comparing the three factors of judgment accuracy, running time, and number of iterations of the artificial neural network models of six simulated hardware.

[0068] Specifically, model one is a training model for the ability to capture moving light sources under normal vision, model two is a training model for the ability to capture moving light sources under "amblyopia", model three is a training model for the ability to capture moving light sources under "color blindness", model four is a training model for the ability to recognize object contours under normal vision, model five is a training model for the ability to recognize object contours under "amblyopia", and model six is ​​a training model for the ability to recognize object contours under "color blindness".

Claims

1. A bionic visual perception method based on "light + neurotransmitter" dual regulation, characterized in that: The following steps are involved: Step S1, based on the optic nerve formation mechanism of organisms that simultaneously perceive light and neurotransmitters, a bionic visual perception research model regulated by the combined action of light and neurotransmitters is established, comprising the following steps: Step S11, using a neurotransmitter receptor modified optical synaptic neurotransmitter sensor, adding different types of neurotransmitters on the surface, and using the same lighting conditions to study the synaptic photocurrent output by the sensor, and establish a relationship model between the synaptic photocurrent and the type of neurotransmitter; Step S12, using a photosynaptic neurotransmitter sensor modified with a neurotransmitter receptor, changing the type of modified neurotransmitter receptor, and adding neurotransmitters of the same conditions on the surface under the same illumination conditions, studying the synaptic photocurrent output by the sensor, and establishing a relationship model between the synaptic photocurrent and the type of neurotransmitter receptor; Step S13, using a neurotransmitter receptor modified optical synaptic neurotransmitter sensor, adding neurotransmitters of different concentrations on the surface, and using the same lighting conditions to study the synaptic photocurrent output by the sensor, and establish a relationship model between the synaptic photocurrent and the neurotransmitter concentration; Step S14, using a neurotransmitter receptor modified optical synaptic neurotransmitter sensor, adding neurotransmitters under the same conditions on the surface, using excitation light of different wavelengths to study the synaptic photocurrent output by the sensor, and establishing a relationship model between the synaptic photocurrent and the wavelength of light; Step S15, using a neurotransmitter receptor modified optical synaptic neurotransmitter sensor, adding neurotransmitters under the same conditions to the surface, using excitation light of the same wavelength and different light intensity to study the synaptic photocurrent output by the sensor, and establishing a relationship model between the synaptic photocurrent and light intensity; Step S16, finally using a neurotransmitter receptor modified optical synaptic neurotransmitter sensor to establish a relationship model between synaptic photocurrent and neurotransmitter type, neurotransmitter receptor type, neurotransmitter concentration, excitation light wavelength and light intensity; Step S2, according to the established research model, by adding neurotransmitter modulation, two abnormal conditions of "amblyopia" and "color blindness" in the vision of the organism are simulated to imitate the individual visual differences of the organism; Step S3, using the lighting conditions to perform regulation to simulate the two visual perception functions of capturing moving light sources and identifying object contours in biological vision; Step S4, dual regulation is performed using both illumination conditions and neurotransmitter conditions, to simulate the visual perception function of individual vision of bionic organisms, and to compare the difference in visual perception function between the two.

2. According to claim 1, a bionic visual perception method based on "light + neurotransmitter" dual regulation is characterized in that: The step S2 comprises the following steps: Step S21, using a photosynaptic neurotransmitter sensor modified with a neurotransmitter receptor, controlling the type of neurotransmitter, the type of neurotransmitter receptor, the wavelength of the excitation light and the intensity of the light to remain unchanged, reducing the surface neurotransmitter concentration, and simulating the "amblyopia" situation in the vision of an organism; Step S22, studying the synaptic photocurrent output by the sensor under the change of neurotransmitter concentration, and the individual visual difference of the bionic organism under normal vision and "amblyopia" conditions; Step S23, using an optical synaptic neurotransmitter sensor, controlling the type of neurotransmitter, neurotransmitter concentration, excitation light wavelength and light intensity to remain unchanged, changing the type of modified neurotransmitter receptor, and simulating the "color blindness" situation in the vision of an organism; Step S24, under the change of the modified neurotransmitter receptor, the synaptic photocurrent output by the sensor is studied to simulate the individual visual differences of the bionic organism under normal vision and "color blindness" conditions.

3. According to claim 1, a bionic visual perception method based on "light + neurotransmitter" dual regulation is characterized in that: The step S3 comprises the following steps: Step S31, using the light addressable function of the optical synaptic neurotransmitter sensor, adding spatial variables to the original research model, and corresponding them one by one with the sensitive points in the sensor chip array; Step S32, adjusting the excitation light source of the optical synapse neurotransmitter sensor to a moving light source, using time and space to describe the movement trajectory of the light source, and tracking the position change of the point light source according to the output current intensity of the sensor at each sensitive site during the movement of the light source; Step S33, randomly select an object, apply uniform parallel light to the left side of the object, so that a shadow of the object is formed on the sensor array on the right side, and the contour of the object is recognized by the output current intensity of the sensor at each sensitive site.

4. The bionic visual perception method based on "light + neurotransmitter" dual regulation according to claim 1, characterized in that: The step S4 comprises the following steps: Step S41, changing the types of neurotransmitters dripped on the surface of the optical synaptic neurotransmitter sensor and the modified neurotransmitter receptors to simulate the visual "amblyopia", "color blindness" and normal conditions of the organism; Step S42, in three different visual states, using the output current caused by the sensor at each sensitive site when the light source position changes as training data, and using an artificial neural network and a simulated hardware artificial neural network model to train the motion light source capture capability; Step S43, in three different visual states, uniform parallel light is applied to the left side of the object so that a shadow of the object is formed on the right sensor array, and the output current caused by the sensor at each sensitive site due to different object contours is used as training data, and the motion light source capture capability is trained using an artificial neural network and a simulated hardware artificial neural network model; Step S44, training two visual perception function models under three visual states, and comparing the differences in visual perception ability caused by individual visual differences by comparing the three factors of judgment accuracy, running time, and number of iterations of the six models.

Images