Automatic visual stimulation analysis method based on target perception

Abstract

The invention discloses an automatic visual stimulation analysis method based on target perception. Comprising the following steps: a) acquiring two-photon imaging video data; b) neuron segmentation and activity information extraction: segmenting all active neurons in the two-photon imaging video, and respectively calculating activity curves of the active neurons; c) quantifying the response frequency and duration of a single neuron to visual stimulation in different directions by using the neuron activity curve obtained in the step b); and d) establishing an automatic analysis system, calculating DSI and OSI indexes, and accurately revealing the spatial tuning characteristics of neurons to visual stimulation in combination with the step c). According to the method, automatic analysis is carried out on existing visual stimulation experiment data, so that the problem of complexity and subjectivity of a visual stimulation analysis process in an existing biological experiment is solved, an activity mode of neurons under visual stimulation at different angles is analyzed in an efficient and accurate mode, a complex mechanism of brain visual information processing is disclosed, and the visual stimulation analysis efficiency is improved. The method is of great significance to influence behaviors and cognition of visual information.

Description

Technical Field

[0001] The invention belongs to the field of biological imaging / medical image processing, and in particular relates to an automatic visual stimulation analysis system based on target perception. Background Art

[0002] Visual information processing is one of the advanced functions of the brain, involving multiple levels from the retina to the cerebral cortex. Visual stimulation experiments explore the brain's visual information processing mechanism by precisely controlling visual stimulation, which is of great significance for in-depth exploration of the interaction mechanism between visual stimulation and cognitive function.

[0003] At present, the main challenges faced by visual stimulation experiment data processing are: 1. High complexity and multi-dimensional information have extremely high requirements for analysis methods and hardware; 2. The neural activity caused by visual stimulation is dynamic and real-time, which means that the data analysis method needs to be able to capture these dynamic changes, and further puts forward real-time requirements for the analysis method. Visual stimulation experiments based on target perception are powerful tools for exploring and analyzing how the brain processes visual information. How to analyze experimental data in an efficient and accurate manner, and then more accurately analyze the activity patterns of neurons under different visual stimulation conditions, is an urgent problem to be solved, which is of great significance to brain science research such as the impact of visual information on behavior and cognition. Summary of the invention

[0004] In order to solve the problems in the background technology, the present invention provides an automated analysis system for visual stimulation based on target perception to solve the cumbersomeness and subjectivity of the visual stimulation analysis process in existing biological experiments, and analyzes the activity patterns of neurons under visual stimulation at different angles in an efficient, accurate and automated manner, revealing the complex mechanism of visual information processing in the brain, which is of great significance to the impact of visual information on behavior and cognition.

[0005] To this end, the present invention adopts the following technical solution: a visual stimulus automatic analysis system based on target perception, and its specific steps are as follows:

[0006] 1) Two-photon imaging video data acquisition: For samples or living organisms labeled with fluorescent probes stimulated by grating visual stimulation, two-photon microscopic imaging is performed and video data is acquired;

[0007] 2) Neuron segmentation and activity information extraction: Segment all active neurons in the two-photon imaging video and calculate their activity curves (△F / F curves) respectively;

[0008] 3) Analysis of specific neuronal activity patterns: Using the neuronal activity curves obtained in step 2), quantify the frequency and duration of responses of individual neurons to visual stimuli in different directions;

[0009] 4) Constructing an automated visual stimulation analysis system: Building an automated analysis system platform, taking the neuron data to be analyzed as the object, and according to step 3), analyzing and obtaining the DSI and OSI, preferred direction and position parameters of each neuron.

[0010] The step 1) is specifically as follows:

[0011] Step 1.1) Introducing fluorescent protein into a specific brain region corresponding to the visual cortex of the living biological sample to be observed (the visual cortex of mice is the V1 brain region) by stereotaxic injection technology to achieve cell transfection;

[0012] Step 1.2) After the fluorescent protein is fully expressed, the mouse is subjected to 360-degree all-round visual stimulation, and the sample is imaged and observed using a two-photon microscope, and video data of the fluorescent image is collected through the device.

[0013] The step 2) is specifically as follows:

[0014] Step 2.1) Calculate the corresponding thresholds for all neurons in the two-photon microscopy video using the following formula, thereby obtaining the corresponding threshold curve, and take the minimum value on the curve as the segmentation threshold:

[0015]

[0016] Where F(t) represents the accumulation of all pixel values ​​in the area marked by the current neuron in the current frame; F0 represents the average value of F(t); σ represents the standard deviation;

[0017] Step 2.2) based on the segmentation threshold obtained in step 2.1), all neurons in the two-photon fluorescence microscopy video are automatically segmented and the outline of each neuron is marked;

[0018] Step 2.3) Using the segmentation results, the segmented neurons are calculated one by one using the following formula to obtain the activity curve corresponding to each neuron:

[0019]

[0020] Among them, F(t) represents the accumulation of all pixel values ​​in the area marked by the current neuron in the current frame number, and F0 represents the average value of F(t).

[0021] The step 3) is specifically as follows:

[0022] 3.1) Based on step 2.2), the segmented neuron image is obtained, and the directional tuning curve of each neuron in the segmented image is drawn to obtain the number of pulses of each neuron under stimulation in different directions per unit time, and to clarify the sensitivity of each neuron to different directions;

[0023] 3.2) For the segmented neuron image data obtained in step 2), based on the activity curves corresponding to each neuron obtained in step 2.3), the response duration of each neuron under visual stimulation at different grating angles is measured to reveal the time course characteristics of each neuron's response.

[0024] The step 4) is specifically as follows:

[0025] Step 4.1) Based on step 3), the directional tuning curves and response durations of all neurons in the data are obtained, and the stimulation angle with the longest response duration and the largest number of pulses is screened and determined as the preferred direction of each neuron; further, the selectivity of the current neuron to different spatial directions is measured by calculating the OSI of the neuron, and the DSI of the neuron is calculated to measure the selectivity of the neuron to different movement directions;

[0026] Step 4.2) Based on step 4.1), an automated analysis system is built to input the neuron data to be analyzed into the system platform, thereby obtaining the preferred direction, DSI and OSI, and position parameters of each neuron, revealing the spatial position and tuning characteristics of each neuron to visual stimuli.

[0027] The step 4.1) is specifically as follows:

[0028] The OSI of a neuron is calculated as follows:

[0029]

[0030] Among them, SN θ represents the number of spikes triggered by the neuron during the grating stimulation in the θ direction, θ max The direction in which the neuron exhibits the highest number of pulses is the preferred direction;

[0031] The DSI of a neuron is calculated as follows:

[0032]

[0033] Among them, SN θ represents the number of spikes triggered by the neuron during the grating stimulation in the θ direction, θ max The direction in which the neuron shows the highest number of pulses is the preferred direction, θ null It indicates the non-preferred direction, which is 180 degrees opposite to the preferred direction.

[0034] In the step 4.1):

[0035] When the OSI value is close to 1, it indicates that the neuron is strongly selective to a specific direction; when it is close to 0, it indicates that the neuron is insensitive to direction or has no obvious preference;

[0036] When the DSI is 0, it indicates no directional selectivity, and 1 indicates complete directional selectivity.

[0037] Suppose a neuron has a high OSI value (such as 0.8) for vertical lines, which means that it is highly selective for vertical lines. At the same time, this neuron has a high DSI value (such as 0.9) for the upward movement of these vertical lines, indicating that it particularly prefers vertical lines moving upward. By combining the analysis of OSI and DSI, we can conclude that this neuron has a particularly strong response to vertical lines moving upward, and a weak response to lines in other directions or vertical lines moving downward.

[0038] Beneficial effects of the present invention:

[0039] The present invention performs automated analysis on existing visual stimulation experimental data, overcoming the cumbersomeness and subjectivity of the visual stimulation analysis process in existing biological experiments. It does not require the modification of the optical imaging system hardware, greatly reducing costs and improving scalability. At the same time, it analyzes the activity patterns of neurons under visual stimulation at different angles, revealing the complex mechanism of visual information processing in the brain, aiming to provide an automated and standardized analysis process paradigm for visual stimulation experiments in the field of neuroscience, which is of great significance to the impact of visual information on behavior and cognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a schematic diagram of the process of the automatic analysis system of visual stimulation based on target perception provided by the present invention.

[0041] Figure 2 It is a schematic diagram of the quantitative analysis of specific neuron activity patterns provided by the present invention.

[0042] Figure 3 It is a schematic diagram of the structure of the automated visual stimulation analysis system provided by the present invention. DETAILED DESCRIPTION

[0043] The process of the automatic analysis system of visual stimulation based on target perception is as follows Figure 1 As shown, the specific steps are as follows:

[0044] Step (a), two-photon imaging video data acquisition: for the fluorescent probe-labeled sample / living body stimulated by grating visual stimulation, two-photon microscopic imaging is performed and video data is acquired;

[0045] Step (b), neuron segmentation and activity information extraction: segment all active neurons in the two-photon imaging video and calculate their activity curves (△F / F curves) respectively;

[0046] Step (c), analysis of specific neuron activity patterns: using the neuron activity curve obtained in step (b), quantifying the frequency and duration of responses of individual neurons to visual stimuli in different directions;

[0047] Step (d), constructing an automated visual stimulation analysis system: constructing an automated analysis system, calculating the DSI index and the OSI index, and combining the results in step (c) to further accurately reveal the spatial tuning characteristics of neurons to visual stimulation.

[0048] Further, the two-photon imaging video data acquisition in step (a) is performed according to the following steps:

[0049] Step 1: Into a specific area of ​​the V1 brain region of a living biological sample to be observed (specifically, C57BL / 6INifidc mice in this case), fluorescent protein is introduced by stereotaxic injection to achieve cell transfection.

[0050] Step 2: After the fluorescent protein is fully expressed, the mice are subjected to 360-degree visual stimulation. At the same time, the sample is imaged and observed using a two-photon microscope and the video data of the fluorescent image is collected through the device.

[0051] Further, step (b) is carried out as follows:

[0052] Step 1: Calculate the corresponding thresholds for all neurons in the two-photon microscopy video using the following formula to obtain the corresponding threshold curve, and take the minimum value on the curve as the segmentation threshold:

[0053]

[0054] Where F(t) represents the accumulation of all pixel values ​​in the area marked by the current neuron in the current frame; F0 represents the average value of F(t); σ represents the standard deviation;

[0055] Step 2, based on the segmentation threshold obtained in step 1, all neurons in the two-photon fluorescence microscopy video are automatically segmented and the outline of each neuron is marked;

[0056] Step 3: Using the segmentation results, calculate the segmented neurons one by one using the following formula to obtain the activity curve corresponding to each neuron:

[0057]

[0058] Among them, F(t) represents the accumulation of all pixel values ​​in the area marked by the current neuron in the current frame number, and F0 represents the average value of F(t).

[0059] Further, step (c) is carried out as follows:

[0060] like Figure 2 As shown, for the obtained segmented neuron image data, the number of neuron pulses per unit time is calculated to determine its response frequency to multi-angle visual stimulation.

[0061] Based on the obtained segmented neuronal activity data, the activation duration of single neurons was measured when stimulated by multi-angle visual stimulation, revealing the temporal characteristics of neuronal responses.

[0062] Further, step (d) is carried out as follows:

[0063] Step 1, such as Figure 3 As shown, based on step (c), the direction tuning curves and response durations of all neurons in the data are obtained, and the stimulation angles with longer response durations and more pulse numbers are screened out and determined as the preferred directions of each neuron.

[0064] Step 2: Build an automated analysis system based on step 1 and input the neuron data to be analyzed into the system platform to obtain the preferred direction, DSI, OSI and position parameters of each neuron, revealing the spatial position and tuning characteristics of each neuron to visual stimuli.

[0065] The method of the present invention can be used to explore the preferred direction spatial position of each neuron and its DSI and OSI after applying 360-degree omnidirectional stimulation, thereby revealing the working principle of the brain in encoding and processing visual information in visual perception and motion detection.

Claims

1. An automated visual stimulus analysis method based on target perception, characterized in that: The following steps are involved: 1) Two-photon imaging video data acquisition: For samples or living organisms labeled with fluorescent probes stimulated by grating visual stimulation, two-photon microscopic imaging is performed and video data is acquired; 2) Neuron segmentation and activity information extraction: Segment all active neurons in the two-photon imaging video and calculate their activity curves respectively; 3) Analysis of specific neuronal activity patterns: Using the neuronal activity curves obtained in step 2), quantify the frequency and duration of responses of individual neurons to visual stimuli in different directions; 4) Constructing an automated visual stimulation analysis system: Building an automated analysis system platform, taking the neuron data to be analyzed as the object, and according to step 3), analyzing and obtaining the DSI and OSI, preferred direction and position parameters of each neuron.

2. The method for automatic visual stimulation analysis based on target perception according to claim 1, characterized in that: The step 1) is specifically as follows: Step 1.1) Introducing fluorescent protein into a specific brain region corresponding to the visual cortex of the living biological sample to be observed by stereotaxic injection technology to achieve cell transfection; Step 1.2) After the fluorescent protein is fully expressed, the mouse is subjected to 360-degree all-round visual stimulation, and the sample is imaged and observed using a two-photon microscope, and video data of the fluorescent image is collected through the device.

3. The method for automatic visual stimulation analysis based on target perception according to claim 2, characterized in that: The step 2) is specifically as follows: Step 2.1) Calculate the corresponding thresholds for all neurons in the two-photon microscopy video using the following formula, thereby obtaining the corresponding threshold curve, and take the minimum value on the curve as the segmentation threshold: Where F(t) represents the accumulation of all pixel values ​​in the area marked by the current neuron in the current frame; F0 represents the average value of F(t); σ represents the standard deviation; Step 2.2) based on the segmentation threshold obtained in step 2.1), all neurons in the two-photon fluorescence microscopy video are automatically segmented and the outline of each neuron is marked; Step 2.3) Using the segmentation results, the segmented neurons are calculated one by one using the following formula to obtain the activity curve corresponding to each neuron: Among them, F(t) represents the accumulation of all pixel values ​​in the area marked by the current neuron in the current frame number, and F0 represents the average value of F(t).

4. The method for automatic visual stimulation analysis based on target perception according to claim 3, characterized in that: The step 3) is specifically as follows: 3.1) Based on step 2.2), the segmented neuron image is obtained, and the directional tuning curve of each neuron in the segmented image is drawn to obtain the number of pulses of each neuron under stimulation in different directions per unit time, and to clarify the sensitivity of each neuron to different directions; 3.2) For the segmented neuron image data obtained in step 2), based on the activity curves corresponding to each neuron obtained in step 2.3), the response duration of each neuron under visual stimulation at different grating angles is measured to reveal the time course characteristics of each neuron's response.

5. The method for automatic visual stimulation analysis based on target perception according to claim 1, characterized in that: The step 4) is specifically as follows: Step 4.1) Based on step 3), the directional tuning curves and response durations of all neurons in the data are obtained, and the stimulation angle with the longest response duration and the largest number of pulses is screened and determined as the preferred direction of each neuron; further, the selectivity of the current neuron to different spatial directions is measured by calculating the OSI of the neuron, and the DSI of the neuron is calculated to measure the selectivity of the neuron to different movement directions; Step 4.2) Based on step 4.1), an automated analysis system is built to input the neuron data to be analyzed into the system platform, thereby obtaining the preferred direction, DSI and OSI, and position parameters of each neuron, revealing the spatial position and tuning characteristics of each neuron to visual stimuli.

6. The method for automatic visual stimulation analysis based on target perception according to claim 1, characterized in that: The step 4.1) is specifically as follows: The OSI of a neuron is calculated as follows: Among them, SN θ represents the number of spikes triggered by the neuron during the grating stimulation in the θ direction, θ max The direction in which the neuron exhibits the highest number of pulses is the preferred direction; The DSI of a neuron is calculated as follows: Among them, SN θ represents the number of spikes triggered by the neuron during the grating stimulation in the θ direction, θ max The direction in which the neuron shows the highest number of pulses is the preferred direction, θ null It indicates the non-preferred direction, which is 180 degrees opposite to the preferred direction.

7. The method for automatic visual stimulation analysis based on target perception according to claim 1, characterized in that: In the step 4.1): When the OSI value is close to 1, it indicates that the neuron has a strong selectivity for a specific direction; when it is close to 0, it indicates that the neuron is insensitive to spatial direction or has no obvious preference; When DSI is 0, it means there is no preference for motion direction, and 1 means that the neuron has the highest selectivity for motion direction, that is, it only responds to motion stimuli in a specific direction.

Images