WO2022267385A1

WO2022267385A1 - Neuronal signal processing method and processing apparatus, and readable storage medium

Info

Publication number: WO2022267385A1
Application number: PCT/CN2021/138077
Authority: WO
Inventors: 岳斌; 李骁健
Original assignee: 中国科学院深圳先进技术研究院
Priority date: 2021-06-22
Filing date: 2021-12-14
Publication date: 2022-12-29
Also published as: CN113627599B; CN113627599A

Abstract

A neuronal signal processing method and processing apparatus, and a readable storage medium. The method comprises: acquiring a neuronal signal, the neuronal signal comprising membrane potential data (11); performing derivation processing on the membrane potential data to obtain derivative data (12); and determining at least one of a spike position, a single spike timing range, and a continuous spike timing range of the neuronal signal according to the derivative data (13). In this way, the extraction speed and accuracy of extracting the spike position, the single spike timing range, and the continuous spike timing range from the neuronal signal can be improved.

Description

Neuron signal processing method, processing device and readable storage medium

technical field

The present application relates to the technical field of neuron signal processing, and in particular to a neuron signal processing method, processing device and readable storage medium.

Background technique

Electrophysiology is the field of study that studies changes in electrical current or voltage across a cell membrane. Electrophysiological techniques are used in a wide variety of neuroscience and physiology applications, from understanding the behavior of individual ion channels in a cell membrane, to whole-cell changes in membrane potential data in a cell, to field potentials in brain slices in vitro or in brain regions in vivo large-scale changes. Ion channels are prime targets for researchers due to their key roles and physiology in many neurological and cardiovascular diseases, and patch clamping, one of the most widely used electrophysiological techniques, is the most effective way to study ion channel activity. best tool.

There are defects in the processing of neuron signals collected by the patch clamp in the related art, which will lead to inaccurate extracted data.

Contents of the invention

The technical problem mainly solved by this application is to provide a neuron signal processing method, a processing device and a readable storage medium, which can improve the extraction speed of the pulse position, the single pulse firing time range and the continuous pulse firing time range from the neuron signal, and accuracy.

In order to solve the above problems, a technical solution adopted by the present application is to provide a neuron signal processing method, the method comprising: acquiring the neuron signal, the neuron signal includes membrane potential data; deriving the membrane potential data, to obtain derivative data; according to the derivative data, determine at least one of the pulse position, single pulse firing time range, and continuous pulse firing time range of the neuron signal.

Wherein, derivation processing is performed on the membrane potential data to obtain derivative data, including: obtaining adjacent first membrane potential data and second membrane potential data in the neuron signal; the second membrane potential data is after the first membrane potential data ; Using the difference between the first membrane potential data and the second membrane potential data as reference membrane potential data; obtaining derivative data based on multiple reference membrane potential data.

Wherein, obtaining the derivative data based on a plurality of reference membrane potential data includes: performing a binarization operation on the plurality of reference membrane potential data; converting the value of the reference membrane potential data satisfying the first preset condition into a first preset value; Taking the value of the reference membrane potential data satisfying the second preset condition as a second preset value; and obtaining derivative data based on a plurality of first preset values and a plurality of second preset values.

Wherein, determining the time range of continuous pulse firing of neuron signals according to the derivative data includes: obtaining the target reference membrane potential data in the derivative data and the first number of reference membrane potential data and the right neighbor in the left neighborhood of the target reference membrane potential data The second number of reference membrane potential data in the domain to obtain a plurality of first membrane potential data segments; determine the average value of all reference membrane potential data in each first membrane potential data segment in order from left to right, as each The first value of the target reference membrane potential data in a first membrane potential data segment; wherein, the first value of the target reference membrane potential data within the first number of ranges in the left neighborhood will participate in the current first membrane potential data Segment calculation: determine the average value of all reference membrane potential data in each first membrane potential data segment in order from right to left, as the second value of the target reference membrane potential data in each first membrane potential data segment; Wherein, the second value of the target reference membrane potential data within the second number of ranges in the right neighborhood will participate in the calculation of the current first membrane potential data segment; determine the continuous Pulse emission time range.

Wherein, determining the continuous pulse emission time range according to the plurality of first values and the plurality of second values includes: determining the time range corresponding to the first value or the second value that continuously satisfies the preset condition as the continuous pulse emission time range.

Wherein, determining the pulse position of the neuron signal according to the derivative data includes: obtaining adjacent first reference membrane potential data and second reference membrane potential data in the derivative data; the second reference membrane potential data is after the first reference membrane potential data ; If the first reference membrane potential data is greater than the second reference membrane potential data and greater than 0, and the second reference membrane potential data is less than 0, then use the time corresponding to the first reference membrane potential data as the pulse position.

Wherein, before determining the single pulse firing time range of the neuron signal according to the derivative data, it includes: converting the value of the reference membrane potential data corresponding to the pulse position into a first preset value; converting the value of the reference membrane potential data not corresponding to the pulse position converting to a second preset value; obtaining derivative data based on a plurality of first preset values and a plurality of second preset values.

Wherein, according to the derivative data, determining the pulse position of the neuron signal, the single pulse firing time range, and the single pulse firing time range of at least one of the continuous pulse firing time ranges include: obtaining the target reference membrane potential data in the derivative data and The first number of reference membrane potential data in the left neighborhood of the target reference membrane potential data and the second number of reference membrane potential data in the right neighborhood to obtain a plurality of second membrane potential data fragments; determine each second membrane potential data The average value of all the reference membrane potential data of the segment is used as the value of the target reference membrane potential data; the single pulse firing time range is determined according to the values of multiple target reference membrane potential data.

Wherein, determining the single-pulse delivery time range according to the values of multiple target reference membrane potential data includes: determining the time range corresponding to multiple target reference membrane potential data that continuously meet preset conditions as the single-pulse delivery time range.

In order to solve the above problems, another technical solution adopted by this application is to provide a neuron signal processing device, the processing device includes a processor and a memory coupled to the processor, a computer program is stored in the memory, and the processor is used for Execute the computer program to realize the processing method provided by the above technical solution.

In order to solve the above problems, another technical solution adopted by the present application is to provide a computer-readable storage medium, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it realizes the above-mentioned technical solution. processing method.

The beneficial effects of the present application are: different from the prior art, the present application provides a neuron signal processing method, a processing device and a readable storage medium. This method can correlate the discrete membrane potential data by deriving the membrane potential data, and then can determine the pulse position, single pulse firing time range, and continuous pulse firing time range of the neuron signal from the associated derivative data. At least one of them can improve the extraction speed and accuracy of extracting the pulse position, the single pulse firing time range and the continuous pulse firing time range from the neuron signal.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings that need to be used in the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application. For those skilled in the art, other drawings can also be obtained based on these drawings without creative effort. in:

Fig. 1 is a schematic flow chart of an embodiment of a neuron signal processing method provided by the present application;

Fig. 2 is a schematic diagram of an embodiment of a neuron signal provided by the present application;

Fig. 3 is a schematic flow chart of an embodiment of step 12 in Fig. 1 provided by the present application;

Fig. 4 is a schematic flowchart of another embodiment of the neuron signal processing method provided by the present application;

Fig. 5 is a schematic diagram of an embodiment of the pulse position provided by the application;

Fig. 6 is a partial schematic diagram after comparison between Fig. 2 and Fig. 5 provided by the present application;

Fig. 7 is a schematic flowchart of another embodiment of the neuron signal processing method provided by the present application;

Fig. 8 is a schematic flowchart of another embodiment of the neuron signal processing method provided by the present application;

Fig. 9 is a schematic diagram of an embodiment of the single pulse delivery time range provided by the present application;

Figure 10 is a partial schematic diagram after comparison between Figure 2 and Figure 9 provided by the present application;

Fig. 11 is a schematic flowchart of another embodiment of the neuron signal processing method provided by the present application;

Fig. 12 is a schematic diagram of an embodiment of the continuous pulse delivery time range provided by the present application;

Fig. 13 is a partial schematic diagram after comparison between Fig. 2 and Fig. 12 provided by the present application;

Fig. 14 is a schematic structural diagram of an embodiment of a neuron signal processing device provided by the present application;

Fig. 15 is a schematic structural diagram of an embodiment of a computer-readable storage medium provided by the present application.

detailed description

The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the specific embodiments described here are only used to explain the present application, but not to limit the present application. In addition, it should be noted that, for the convenience of description, only some structures related to the present application are shown in the drawings but not all structures. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the scope of protection of this application.

The terms "first", "second", etc. in this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "include" and "have", as well as any variations thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but optionally also includes unlisted steps or units, or optionally further includes For other steps or units inherent in these processes, methods, products or apparatuses.

Reference herein to an "embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the present application. The occurrences of this phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is understood explicitly and implicitly by those skilled in the art that the embodiments described herein can be combined with other embodiments.

The patch clamp technique is a versatile electrophysiological tool for understanding ion channel behavior. Every cell expresses ion channels, but the most common cells studied by patch clamp techniques include nerve cells, muscle fibers, cardiomyocytes, and oocytes that highly express a single ion channel. To assess the conductivity of individual ion channels, the microelectrode forms a high-resistance seal with the cell membrane and removes the cell membrane sheet containing the ion channel of interest. Alternatively, when the microelectrode is sealed to the cell membrane, the cell membrane is ruptured, allowing the electrode to be in electrical communication with the entire cell. A voltage is then applied, a voltage clamp is formed, and the membrane current is measured. Current clamp can also be used to measure changes in voltage across the cell membrane (known as membrane potential data). Voltage or current changes within the cell membrane can be altered by adding compounds to block or activate channels. These techniques allow researchers to understand how ion channels behave in normal and disease states, and how different drugs, ions or other analytes alter these states.

Membrane potential data usually refer to the potential difference generated between two solutions separated by a membrane. Generally, it refers to the electrical phenomenon accompanying the process of cell life activities, and the potential difference between the two sides of the cell membrane. Membrane potential data play an important role in the process of nerve cell communication.

Neurons are the basic unit of computation in the biological brain, exchanging and transmitting information through pulses, and memory and learning through synapses. Biological brains contain networks of billions of neurons connected to each other by trillions of synapses. In this network, the pulse-based temporal processing mechanism enables sparse and efficient information transfer in the human brain.

However, the membrane potential data collected by patch clamp is often disordered, how to find effective information from these disordered membrane potential data is a problem that needs to be solved at present.

Based on this, the present application adopts the following methods for signal processing.

Referring to FIG. 1 , FIG. 1 is a schematic flowchart of an embodiment of a neuron signal processing method provided in the present application. The method includes:

Step 11: Obtain neuron signals, which include membrane potential data.

Wherein, the neuron signal may be the membrane potential data collected within a preset time by the aforementioned patch clamp. It can be understood that each membrane potential data corresponds to a collection time.

At this point, the neuron signal can be expressed in the form of FIG. 2 . In Fig. 2, the abscissa is the acquisition time, and the ordinate is the membrane potential data of the neuron signal.

Step 12: Derivative processing is performed on the membrane potential data to obtain derivative data.

Combined with Figure 2 for illustration:

Because the membrane potential data are chaotic and disorderly, the neuron signal needs to be derived to correlate the discrete membrane potential data.

In some embodiments, a first order derivative can be performed on the membrane potential data to obtain derivative data. Wherein, each derivative data corresponds to the membrane potential data.

For example, referring to FIG. 3, step 12 may be as follows:

Step 121: Obtain adjacent first membrane potential data and second membrane potential data in neuron signals.

Wherein, the second membrane potential data follows the first membrane potential data.

Step 122: Use the difference between the first membrane potential data and the second membrane potential data as the reference membrane potential data.

Wherein, the second membrane potential data can be subtracted from the first membrane potential data, and the obtained first difference can be used as the reference membrane potential data, and the reference membrane potential data can correspond to represent the first membrane potential data. The second membrane potential data can also be subtracted from the second membrane potential data to obtain a second difference as the reference membrane potential data, and the reference membrane potential data can correspond to the second membrane potential data.

Step 123: Obtain derivative data based on multiple reference membrane potential data.

Since the neuron signal has multiple membrane potential data, multiple reference membrane potential data can be correspondingly generated, and the reference membrane potential data can be used to form derivative data in the manner of the neuron signal. Wherein, each time of the reference membrane potential data corresponds to a time of the membrane potential data in the neuron signal.

At this time, discrete neuron signals can be represented by derivative data.

Step 13: Determine at least one of the pulse position, single pulse firing time range, and continuous pulse firing time range of the neuron signal according to the derivative data.

For example, judging the adjacent reference membrane potential data in the derivative data can determine whether there is a pulse position in the adjacent reference membrane potential data.

As another example, neighborhood points are acquired for each reference membrane potential data in the derivative data, and it is judged according to the neighborhood points whether the reference membrane potential data is within the time range of single pulse firing.

As another example, neighborhood points are acquired for each reference membrane potential data in the derivative data, and it is judged according to the neighborhood points whether the reference membrane potential data is within the continuous pulse emission time range.

In this embodiment, by deriving the membrane potential data, the discrete membrane potential data can be correlated, and then the pulse position, single pulse firing time range, continuous At least one of the pulse firing time ranges can improve the extraction speed and accuracy of extracting the pulse position, the single pulse firing time range and the continuous pulse firing time range from the neuron signal.

The applicant has found in long-term research that when a neuron fires a pulse, the membrane potential data at the corresponding moment is the maximum value within a certain period of time before and after the moment. And when neurons emit pulses, the time range of single pulse emission can also be collected, and the emission of continuous pulses can also be collected. These pulse data are of great significance for further modeling of neuron models, simulation of neuron electrophysiological properties, and research of spiking neural networks.

Therefore, the accuracy of the pulse position, single pulse firing time range, and continuous pulse firing time range of neuron signals is also very important. Based on this, the application provides the following examples for illustration:

Referring to FIG. 4 , FIG. 4 is a schematic flowchart of another embodiment of the neuron signal processing method provided in the present application. The method includes:

Step 41: Acquire neuron signals.

Step 42: Obtain adjacent first membrane potential data and second membrane potential data in the neuron signal.

Step 43: Using the difference between the first membrane potential data and the second membrane potential data as reference membrane potential data.

Step 44: Obtain derivative data based on multiple reference membrane potential data.

Steps 41-44 have the same or similar technical solutions as those of the above-mentioned embodiments, which will not be repeated here.

In this embodiment, step 43 may be to subtract the first membrane potential data from the second membrane potential data to obtain a first difference, and use the first difference as the reference membrane potential data corresponding to the second membrane potential data.

Step 45: Acquiring adjacent first reference membrane potential data and second reference membrane potential data in the derivative data.

Wherein, the second reference membrane potential data follows the first reference membrane potential data.

Since the derivative data can approximate the neuronal signal, the derivative data can be used to determine the pulse position.

Step 46: If the first reference membrane potential data is greater than the second reference membrane potential data and greater than 0, and the second reference membrane potential data is less than 0, then use the time corresponding to the first reference membrane potential data as the pulse position.

It can be understood that if the first reference membrane potential data is greater than the second reference membrane potential data, it means that the membrane potential data on the neuron signal corresponding to the first reference membrane potential data is greater than the membrane potential data corresponding to the second reference membrane potential data.

Moreover, the reference membrane potential data in the derivative data is the difference between adjacent membrane potential data in the neuron signal, indicating that the reference membrane potential data on the derivative data will appear in two situations, greater than 0 or less than 0.

If the reference membrane potential data is greater than 0, it means that the membrane potential data on the neuron signal corresponding to the reference membrane potential data is greater than the membrane potential data on the neuron signal at the moment before this moment.

The first reference membrane potential data is greater than the second reference membrane potential data and greater than 0, and the second reference membrane potential data is less than 0, indicating that the membrane potential data on the neuron signal corresponding to the first reference membrane potential data is greater than the neuron signal at this time Membrane potential data at the next moment. It can be determined that the pulse is delivered at this moment, and the moment corresponding to the first reference membrane potential data can be determined as the pulse position.

In other embodiments, the following conditions can be used to determine:

The reference membrane potential data meeting the first preset condition is converted into a first preset value, and the reference membrane potential data meeting the second preset condition is converted into a second preset value. For example, the conversion is performed by the following formula.

Among them, S(j) is the function expression after derivative data conversion,

Represents the functional expression of the derivative data, that is, the reference membrane potential data at the jth moment, and j represents the acquisition time of the corresponding reference membrane potential data, that is, the acquisition time of the membrane potential data in neurons.

It can be known from the above formula that the time corresponding to S(j)=1 is the pulse position, that is, the neuron fires a pulse at this time.

In an application scenario, refer to Figure 2, Figure 5 and Figure 6 for illustration:

FIG. 5 is a schematic diagram of the pulse position determined using the method of this embodiment. The pulse position in FIG. 5 can be compared with the neuron signal in FIG. 2 to obtain the image shown in FIG. 6 . Therefore, it can be determined that according to the error between the two, if the error meets the threshold, the pulse position obtained in Figure 5 can be used to model the neuron model, simulate the electrophysiological properties of the neuron, and study the pulse neural network.

In this embodiment, by comparing the magnitude of the reference membrane potentials in the derivative data and the positive and negative of the reference membrane potentials, the pulse position in the neuron signal can be accurately determined, thereby improving the accuracy and comprehensiveness of the pulse position acquisition , the pulse position collected by this method can more accurately reflect the activity of neurons.

In some embodiments, after the pulse position is obtained in the above embodiments, the time range of single pulse delivery can be determined according to the pulse position. Specifically, referring to Figure 7, the method includes:

Step 71: converting the value of the reference membrane potential data corresponding to the pulse position into a first preset value.

Step 72: Convert the value of the reference membrane potential data not corresponding to the pulse position into a second preset value.

Step 73: Obtain derivative data based on a plurality of first preset values and a plurality of second preset values.

In some embodiments, conversion can be carried out according to the formula, the formula is as follows:

At this point, the derivative data becomes a value of 0 or 1, which is convenient for subsequent calculations.

Step 74: Obtain the target reference membrane potential data in the derivative data and the first number of reference membrane potential data in the left neighborhood of the target reference membrane potential data and the second number of reference membrane potential data in the right neighborhood to obtain multiple Second membrane potential data segment.

Wherein, the first quantity is the same as the second quantity, so that the balance of data around the target reference membrane potential data can be ensured. Then the number of reference membrane potential data in the second membrane potential data segment is an odd number.

Wherein, the first quantity may be 10 to 20. For example, you can choose 12, 15, 17 or 19.

It can be understood that if there are multiple reference membrane potential data in the derivative data, the selection of the target reference membrane potential data is also performed sequentially, that is, the next target reference membrane potential data is adjacent to the current target reference membrane potential data.

Step 75: Determine the average value of all reference membrane potential data in each second membrane potential data segment as the value of the target reference membrane potential data.

The target reference membrane potential data, the first number of reference membrane potential data and the second number of reference membrane potential data in each second membrane potential data segment are summed, and then averaged. The average value can represent the value of any reference membrane potential data in the second membrane potential data segment, that is, it can be used as the value of the target reference membrane potential data.

Step 76: Determine the single pulse delivery time range according to the values of multiple target reference membrane potential data.

In some embodiments, the time range corresponding to a plurality of target reference membrane potential data that continuously satisfy the preset condition may be determined as the time range of single pulse delivery.

If the calculation is performed according to the value of the derivative data being 0 or 1, the value of the target reference membrane potential data in step 55 will be equal to 0 or not equal to 0.

When it is equal to 0, it means that the reference membrane potential data in the second membrane potential data segment does not have a pulse position corresponding to it, and when it is not equal to 0, it means that there is a pulse position corresponding to the reference membrane potential data in the second membrane potential data segment.

for example:

For example, there are 100 reference membrane potential data in the conduction data, the seventh reference membrane potential data is 1, and the rest are all 0. The first quantity is 10, then the following results will occur.

When step 55 is executed, the value of the target reference membrane potential data is as follows:

The value of the first target reference membrane potential data is not equal to 0, the value of the second target reference membrane potential data is not equal to 0, the value of the third target reference membrane potential data is not equal to 0, and the fourth target reference membrane potential data The value of the target reference membrane potential data is not equal to 0, the value of the fifth target reference membrane potential data is not equal to 0, the value of the sixth target reference membrane potential data is not equal to 0, the value of the seventh target reference membrane potential data is not equal to 0, and the eighth target reference membrane potential data value is not equal to 0. The values of the 1st target reference membrane potential data to the 90th target reference membrane potential data are equal to 0.

Then the time range corresponding to the first target reference membrane potential data to the seventh target reference membrane potential data is determined as the single pulse firing time range.

It can be understood that the single-pulse firing time range can represent the whole process from the preparation before the pulse firing to the decay after the pulse firing.

In this embodiment, the left and right neighbors are used to determine the membrane potential data segment for the target reference membrane potential data, and the average value of the reference membrane potential data in this segment is used as the value of the target reference membrane potential. Therefore, each reference membrane potential data There will be an average value, and the range of the average value that satisfies the preset value can be determined as the single pulse delivery range. Because the target reference membrane potential data is based on the pulse position, the determined single pulse delivery range can be more accurate, thereby improving the accuracy of the pulse. Accuracy and comprehensiveness of single-pulse firing time range collection, the single-pulse firing time range collected by this method can more accurately reflect neuron activity.

Referring to FIG. 8 , FIG. 8 is a schematic flowchart of another embodiment of a neuron signal processing method provided in the present application. The method includes:

Step 801: Acquire neuron signals, where the neuron signals include membrane potential data.

Step 802: Obtain adjacent first membrane potential data and second membrane potential data in neuron signals.

Step 803: Using the difference between the first membrane potential data and the second membrane potential data as reference membrane potential data.

Steps 801-803 have the same or similar technical solutions as those in the foregoing embodiments, which will not be repeated here.

Step 804: Binarize multiple reference membrane potential data.

Step 805: Convert the value of the reference membrane potential data satisfying the first preset condition into a first preset value.

Step 806: Use the value of the reference membrane potential data satisfying the second preset condition as the second preset value.

Through the binarization operation, the reference membrane potential data can be denoised to improve the accuracy of subsequent calculations.

In some embodiments, step 804-step 806 can be represented by the following formula:

Wherein, the first preset value is 0, the second preset value is the reference membrane potential data itself, and A and B are preset conditions. Optionally, A can be -0.5, and B can be 0.5.

Step 807: Obtain derivative data based on a plurality of first preset values and a plurality of second preset values.

According to the above formula, the derivative data becomes a set of 0 and a value corresponding to the reference membrane potential data satisfying the second preset condition.

Step 808: Obtain the target reference membrane potential data in the derivative data, the first quantity of reference membrane potential data in the left neighborhood of the target reference membrane potential data and the second quantity of reference membrane potential data in the right neighborhood of the target reference membrane potential data, so as to obtain multiple Second membrane potential data segment.

Wherein, the first quantity may be 5 to 20. For example, you can choose 7, 8, 9, 10, 12, 15, 17 or 19.

Step 809: Determine the average value of all reference membrane potential data in each second membrane potential data segment as the value of the target reference membrane potential data.

Step 810: Determine the single pulse delivery time range according to the values of multiple target reference membrane potential data.

Wherein, steps 807-810 are the same as or similar to the technical solutions of the foregoing embodiments, and details are not repeated here.

In an application scenario, refer to Fig. 2, Fig. 9 and Fig. 10 for description:

FIG. 9 shows the time range of single pulse firing determined by the method of this embodiment. The pulse position in FIG. 9 can be compared with the neuron signal in FIG. 2 to obtain the image shown in FIG. 10 . Therefore, it can be found from FIG. 10 that the single pulse firing time range obtained by this method is consistent with the data in the neuron signal in FIG. 2 . Then, the single pulse firing time range obtained in Fig. 9 can be used for modeling of neuron model, simulation of neuron electrophysiological properties, and research of spiking neural network.

In this embodiment, the left and right neighbors are used to determine the membrane potential data segment for the target reference membrane potential data, and the average value of the reference membrane potential data in this segment is used as the value of the target reference membrane potential. Therefore, each reference membrane potential data There will be an average value, and the range of the average value that satisfies the preset value can be determined as the single-pulse release range, which can accurately determine the single-pulse release time range, thereby improving the accuracy and comprehensiveness of the collection of single-pulse release time ranges. The time range of single pulse firing collected by this method can more accurately reflect the activity of neurons.

Referring to FIG. 11 , FIG. 11 is a schematic flowchart of another embodiment of the neuron signal processing method provided in the present application. The method includes:

Step 101: Obtain the target reference membrane potential data in the derivative data and the first quantity of reference membrane potential data in the left neighborhood of the target reference membrane potential data and the second quantity of reference membrane potential data in the right neighborhood of the target reference membrane potential data, so as to obtain multiple First slice of membrane potential data.

Wherein, the derivative data in this embodiment is obtained according to the technical solutions in any of the above-mentioned embodiments, and details are not described here.

For example, the derivative data can be obtained according to the following formula:

In some embodiments, the derivative data can also be obtained according to the following formula:

According to the above formula, the derivative data becomes a set of 0 and 1.

Step 102: Determine the average value of all reference membrane potential data in each first membrane potential data segment in order from left to right, as the first value of the target reference membrane potential data in each first membrane potential data segment; where , the first value of the target reference membrane potential data will be involved in the calculation of the next first membrane potential data segment.

In some embodiments, after the average value is obtained, the average value is judged, and if the average value is less than the set threshold, the average value is changed to 0, and then used as the target reference membrane potential data in each first membrane potential data segment the first value of . For example, set the threshold to 0.0001, 0.0002.

In an application scenario, due to the use of electronic equipment for calculation, due to hardware limitations, the smallest decimal stored by the hardware limitation can be used as the set threshold.

For example, in the case where there is no value in the reference membrane potential data other than the first value in the continuous first membrane potential data segment, the first value participating in the calculation from the beginning will rapidly follow the reference value in the first membrane potential data segment. The amount of membrane potential data is reduced by a factor. At the same time, in the continuous first membrane potential data segment, except for the first value, the situation that other reference membrane potential data has no value corresponds to the situation of no pulse.

In other embodiments, the second value of the target reference membrane potential data within the first number of ranges in the left neighborhood will participate in the calculation of the current first membrane potential data segment.

Steps

101 and 102 are illustrated with examples:

After calculating the average value of the first first membrane potential data segment, the first value of the target reference membrane potential data is obtained, and then performing the average value calculation on the second first membrane potential data segment, at this time the second The first membrane potential data segment includes the target reference membrane potential data in the first first membrane potential data segment, then when the second first membrane potential data segment performs average calculation, the target reference value uses the first Numerical values are involved in calculations.

Therefore, when calculating from left to right, the target reference value in each first membrane potential data segment that has been calculated on the left will get the first value, and when the next first membrane potential data segment is calculated, the target reference value in the left adjacent The first value of the target reference membrane potential data within the first number of ranges of the field will participate in the calculation of the current first membrane potential data segment.

In this way, the time range can be emitted as successive pulses in the left-to-right direction as acquired.

It can be understood that, because the value of the reference membrane potential data is 0 or a positive number or a negative number, when calculating the average value of the first membrane potential data segment, the average value will be 0. When it is 0, it means that no pulse is generated at this moment, and it is also within the pulse emission time range.

However, there is a problem of omission in this way, because the reference membrane potential data of the left neighborhood is involved in the calculation of the next first membrane potential data segment, therefore, the reference membrane potential data of the right neighborhood is omitted. Therefore, step 103 is performed.

Step 103: Determine the average value of all reference membrane potential data in each first membrane potential data segment in order from right to left, as the second value of the target reference membrane potential data in each first membrane potential data segment; where , the second value of the target reference membrane potential data will be involved in the calculation of the next first membrane potential data segment.

In some embodiments, after the average value is obtained, the average value is judged, and if the average value is less than the set threshold, the average value is changed to 0, and then used as the target reference membrane potential data in each first membrane potential data segment the second value of .

Step 103 is similar to step 102, except that the calculation sequence at this time is from right to left. In this way, it can be determined whether the reference membrane potential data in the right neighborhood is the membrane potential data in the continuous pulse emission time range.

In other embodiments, the second value of the target reference membrane potential data within the second number of ranges in the right neighborhood will participate in the calculation of the current first membrane potential data segment.

After calculating the average value of the first first membrane potential data segment, the second value of the target reference membrane potential data is obtained, and then performing the average value calculation on the second first membrane potential data segment, at this time the second The first membrane potential data segment includes the target reference membrane potential data in the first first membrane potential data segment, then when the second first membrane potential data segment performs average calculation, the target reference value uses the second Numerical values are involved in calculations.

When calculating from right to left, the target reference value in each first membrane potential data segment that has been calculated on the right will get the second value. When calculating the next first membrane potential data segment, the target reference value in the left neighbor The first value of the target reference membrane potential data within the second number of ranges will participate in the calculation of the current first membrane potential data segment.

In this way, the time range can be emitted as successive pulses in the right-to-left direction as acquired.

Step 104: Determine the time range of continuous pulse delivery according to multiple first values and multiple second values.

The time range corresponding to the first numerical value or the second numerical value that satisfies the preset condition is determined as the continuous pulse emission time range.

In some embodiments, the determination of the continuous pulse emission time range can be performed according to the following formula.

First, use the following formula to obtain the first numerical value:

in,

Represents the function corresponding to the first value, H represents the first quantity, L represents the length or quantity of the reference membrane potential data, P represents the total number of reference membrane potential data in the first membrane potential data segment, P=2H+1,

solving

, values are taken in the manner of k ₁ =0, 1...N, that is, from left to right. Among them, in

is less than the set threshold, the

The corresponding value is changed to 0, and then used as

The value of , which is the first value.

Then, a plurality of second values are obtained by using the following formula:

in,

denotes a function corresponding to the second value, H denotes the second quantity, and L denotes the length or quantity of the reference membrane potential data. P represents the total number of reference membrane potential data in the first membrane potential data segment, P=2H+1,

solving

, values are taken in the manner of k ₂ =N,N-1,...1,0, that is, from right to left. Among them, in

is less than the set threshold, the

The corresponding value is changed to 0, and then used as

The value of , the second value.

then after getting

with

Finally, because the value order of the two is opposite, it is necessary to unify the order of the two, for example, unify from right to left, or from left to right, then you can get

with

Then calculate according to the following formula:

At this time, the time corresponding to D(k)=1 is the continuous pulse emission time range.

In an application scenario, refer to Figure 2, Figure 12 and Figure 13 for description:

FIG. 12 shows the time range of continuous pulse firing determined by the method of this embodiment. The pulse position in FIG. 12 can be compared with the neuron signal in FIG. 2 to obtain the image shown in FIG. 13 . Therefore, it can be found from FIG. 13 that the single pulse firing time range obtained by this method is consistent with the data in the neuron signal in FIG. 2 . Then, the single pulse firing time range obtained in Fig. 13 can be used to model neuron models, simulate neuron electrophysiological properties, and study spiking neural networks.

In this embodiment, the left and right neighbors are used to determine the membrane potential data segment for the target reference membrane potential data, and the average value of the reference membrane potential data in the segment is used as the value of the target reference membrane potential, and the target reference membrane potential The value of is involved in the calculation of the next membrane potential data segment, so that the time corresponding to the pulse release can be correlated, and then the continuous pulse release time range can be determined, which can improve the accuracy of the acquisition of the continuous pulse release time range, and Determining the continuous pulse firing time range from left to right and from right to left can make the collected continuous pulse firing time range more comprehensive and can better reflect neuron activity.

Referring to FIG. 14 , FIG. 14 is a schematic structural diagram of an embodiment of a neuron signal processing device provided in the present application. The processing device 140 includes a processor 141 and a memory 142 coupled to the processor 141, a computer program is stored in the memory 142, and the processor 141 is used to execute the computer program to realize the following method:

Acquire neuron signals, which include membrane potential data; perform derivation processing on the membrane potential data to obtain derivative data; determine the pulse position, single pulse firing time range, and continuous pulse firing time range of the neuron signal according to the derivative data at least one of .

It can be understood that the processor 141 in this embodiment may also implement the method in any of the foregoing embodiments, and reference may be made to the foregoing embodiments for specific implementation steps thereof, which will not be repeated here.

It can be understood that the processing device 140 may further have a communication interface (not shown in the figure), and the communication interface is connected with the processor 141 and used for connecting the patch clamp. Wherein, the patch clamp may be an in vivo patch clamp, which is used to collect signals from neurons of a target organism. For example, the target organism can be a mouse, a rabbit, or the like.

Referring to FIG. 15 , FIG. 15 is a schematic structural diagram of an embodiment of a computer-readable storage medium provided by the present application. The computer-readable storage medium 150 stores a computer program 151, and when the computer program 151 is executed by a processor, the following method is realized:

Acquire neuron signals, which include membrane potential data; perform derivation processing on the membrane potential data to obtain derivative data; determine the pulse position, single pulse emission time range, and continuous pulse emission time range of neuron signals according to the derivative data at least one of .

It can be understood that the computer-readable storage medium 150 in this embodiment is applied to the processing device 140, and its specific implementation steps can refer to the above-mentioned embodiments, and details are not repeated here.

In addition, the original data in Fig. 2, Fig. 5, Fig. 6, Fig. 9, Fig. 10 and Fig. 13 are neuron signals in this application.

The present application provides a neuron signal processing method, a processing device, and a readable storage medium. This method obtains the derivative data by deriving the membrane potential data. On the one hand, by comparing the magnitude of the reference membrane potential in the derivative data and the positive and negative of the reference membrane potential, the pulse position in the neuron signal can be accurately determined. On the one hand, for the target reference membrane potential data in the derivative data, the left and right neighbors are used to determine the membrane potential data segment, and the average value of the reference membrane potential data in this segment is used as the value of the target reference membrane potential, so each reference membrane potential There will be an average value in the data, and the range of the average value that satisfies the preset value can be determined as the single pulse emission range. On the other hand, the left and right neighbors are used to determine the membrane potential data segment for the target reference membrane potential data in the derivative data , and take the average value of the reference membrane potential data in this segment as the value of the target reference membrane potential, and make the value of the target reference membrane potential participate in the calculation of the next membrane potential data segment, so that Correlate between them, and then determine the time range of continuous pulse release.

To sum up, the neuron signal processing method, processing device and readable storage medium provided by the present application can improve the accuracy of collecting the pulse position, single pulse distribution range and continuous pulse distribution time range, and can make the collected pulse position, The range of single-pulse firing and the time range of continuous pulse firing are more comprehensive and can better reflect the activity of neurons, which is convenient for further modeling of neuron models, simulation of neuron electrophysiological properties, and research on spiking neural networks.

In the several implementation manners provided in this application, it should be understood that the disclosed methods and devices may be implemented in other ways. For example, the device implementation described above is only illustrative. For example, the division of the modules or units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be Incorporation may either be integrated into another system, or some features may be omitted, or not implemented.

The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

If the integrated units in the above other embodiments are realized in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or part of the contribution to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , including several instructions to make a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor (processor) execute all or part of the steps of the methods described in various embodiments of the present application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disc, etc., which can store program codes. .

The above is only the implementation of the application, and does not limit the patent scope of the application. Any equivalent structure or equivalent process conversion made by using the specification and drawings of the application, or directly or indirectly used in other related technologies fields, are all included in the scope of patent protection of this application in the same way.

Claims

A method for processing neuron signals, characterized in that the method comprises:

acquiring neuron signals, the neuron signals comprising membrane potential data;

performing derivation processing on the membrane potential data to obtain derivative data;

At least one of a pulse position, a single pulse firing time range, and a continuous pulse firing time range of the neuron signal is determined according to the derivative data.
The method according to claim 1, characterized in that,

The derivation processing of the membrane potential data to obtain derivative data includes:

Acquiring adjacent first membrane potential data and second membrane potential data in the neuron signal; the second membrane potential data is after the first membrane potential data;

using the difference between the first membrane potential data and the second membrane potential data as reference membrane potential data;

The derivative data is obtained based on a plurality of the reference membrane potential data.
The method according to claim 2, characterized in that,

The obtaining the derivative data based on a plurality of the reference membrane potential data includes:

performing a binarization operation on a plurality of reference membrane potential data;

converting the value of the reference membrane potential data satisfying the first preset condition into a first preset value;

Taking the value of the reference membrane potential data satisfying the second preset condition as the second preset value;

The derivative data is obtained based on a plurality of the first preset values and a plurality of the second preset values.
The method according to claim 3, characterized in that,

The determining the continuous pulse firing time range of the neuron signal according to the derivative data includes:

Acquiring the target reference membrane potential data in the derivative data, the first quantity of reference membrane potential data in the left neighborhood of the target reference membrane potential data and the second quantity of reference membrane potential data in the right neighborhood of the target reference membrane potential data, so as to obtain multiple a first membrane potential data segment;

The average value of all reference membrane potential data in each of the first membrane potential data segments is determined in order from left to right as the first value of the target reference membrane potential data in each of the first membrane potential data segments. Value; wherein, the first value of the target reference membrane potential data will participate in the calculation of the next first membrane potential data segment;

The average value of all reference membrane potential data in each of the first membrane potential data segments is determined in order from right to left as the second value of the target reference membrane potential data in each of the first membrane potential data segments. Value; wherein, the second value of the target reference membrane potential data will participate in the calculation of the next first membrane potential data segment;

The continuous pulse emission time range is determined according to a plurality of the first values and a plurality of the second values.
The method according to claim 4, characterized in that,

The determining the continuous pulse delivery time range according to a plurality of the first values and a plurality of the second values includes:

A time range corresponding to the first numerical value or the second numerical value that satisfies a preset condition is determined as the continuous pulse emission time range.
The method according to claim 2, characterized in that,

The determining the pulse position of the neuron signal according to the derivative data includes:

Acquiring adjacent first reference membrane potential data and second reference membrane potential data in the derivative data; the second reference membrane potential data is after the first reference membrane potential data;

If the first reference membrane potential data is greater than the second reference membrane potential data and greater than 0, and the second reference membrane potential data is less than 0, then use the time corresponding to the first reference membrane potential data as the pulse Location.
The method according to claim 6, characterized in that,

Before the determination of the single pulse firing time range of the neuron signal according to the derivative data, it includes:

converting the value of the reference membrane potential data corresponding to the pulse position into a first preset value;

converting the value of the reference membrane potential data not corresponding to the pulse position into a second preset value;

The derivative data is obtained based on a plurality of the first preset values and a plurality of the second preset values.
The method according to claim 3 or 7, characterized in that,

The determining the single pulse firing time range of at least one of the pulse position, single pulse firing time range, and continuous pulse firing time range of the neuron signal according to the derivative data includes:

Acquiring the target reference membrane potential data in the derivative data, the first quantity of reference membrane potential data in the left neighborhood of the target reference membrane potential data and the second quantity of reference membrane potential data in the right neighborhood of the target reference membrane potential data, so as to obtain multiple a second membrane potential data segment;

determining the average value of all reference membrane potential data for each of the second membrane potential data segments as the value of the target reference membrane potential data;

The single pulse firing time range is determined according to a plurality of values of the target reference membrane potential data.
The method according to claim 8, characterized in that,

The determining the single pulse delivery time range according to a plurality of values of the target reference membrane potential data includes:

The time range corresponding to the plurality of target reference membrane potential data that continuously meet the preset condition is determined as the single pulse delivery time range.
A neuron signal processing device, characterized in that the processing device includes a processor and a memory coupled to the processor, a computer program is stored in the memory, and the processor is used to execute the computer program To realize the processing method as described in any one of claims 1-9.
A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the processing method according to any one of claims 1-9 .