A neural signal processing method, apparatus, device, and storage medium

Abstract

The present invention discloses a neural signal processing method, apparatus, electronic device and storage medium. The method includes: acquiring neural signal data; performing spike retrieval on the neural signal data to obtain neuron discharge waveform data; classifying the neuron discharge waveform data to obtain neuron discharge waveform classification data; performing waveform processing on the waveform data of a target channel and a target waveform type in the neuron discharge waveform classification data to obtain a standard discharge waveform of neurons corresponding to the target channel and the target waveform type. By performing visual processing on neural signals, the present invention obtains a standard discharge waveform characterizing the physiological characteristics of neurons, and based on the obtained standard discharge waveform, it is possible to more accurately and conveniently verify the quality of neural signals collected by high-throughput high-density deep neural electrodes.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and particularly to a method, device, equipment and storage medium for processing nerve signals. Background Art

[0002] The technology of nerve electrophysiological signal acquisition and processing is one of the most important technologies in brain surgery and neuroscience. Nerve electrophysiological signals can intuitively reflect the activity state of nerves. There are approximately 50 billion neurons (also known as nerve cells) in normal adult humans, and the brain accounts for about one-fifth. Information is transmitted between neurons through neurotransmitters (chemical substances), and neurons themselves transmit information from the axon along the nerve fiber to the dendrite through the change of action potential (Spike). Spatially, the axons, nerve fibers, and dendrites that can transmit action potentials occupy most of the space; temporally, the fastest speed of action potentials on nerve fibers can reach 120 m / s. Therefore, whether in terms of space or time, brain electrophysiological signals are the physical quantities that most intuitively and quickly reflect nerve activities.

[0003] The recording method of brain electrophysiological signals has evolved over a long time, starting from the most primitive single metal wire recording of neuron discharges in the mouse brain to the current EEG, ECoG, sEEG, etc. In order to pursue better signal quality and more refined description of neuron information, the recording means has developed from non-invasive extracranial recording to invasive recording, and along with the continuous increase in the number of recording channels per unit area, the data volume has become increasingly large, followed by the problems of data processing accuracy and efficiency. In order to serve the preparation iteration of high-throughput and high-density invasive nerve recording electrodes, it is necessary to provide a nerve signal processing method that can accurately and quickly verify the quality of signals collected by nerve electrodes. Summary of the Invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, a first aspect of the present invention proposes a method for processing nerve signals, including:

[0005] Obtaining nerve signal data; wherein, the nerve signal data is multi-channel nerve signal data collected by high-throughput and high-density deep nerve electrodes;

[0006] Performing discharge spike retrieval on the nerve signal data to obtain neuron discharge waveform data; wherein, the neuron discharge waveform data includes a plurality of waveform data representing neuron discharges;

[0007] Classify the neuron discharge waveform data to obtain neuron discharge waveform classification data; wherein, the neuron discharge waveform classification data includes at least one waveform set, the waveform set includes at least one waveform data with the same waveform type, and the at least one waveform data with the same waveform type corresponds to the same type of neuron;

[0008] Perform waveform processing on the waveform data of the target channel and target waveform type in the neuron discharge waveform classification data to obtain the standard discharge waveform of the neuron corresponding to the target channel and the target waveform type.

[0009] Further, the performing waveform processing on the waveform data of the target channel and target waveform type in the neuron discharge waveform classification data to obtain the standard discharge waveform of the neuron corresponding to the target channel and the target waveform type includes:

[0010] Calculate the performance index of the neurons corresponding to each waveform type in the neuron discharge waveform classification data;

[0011] Screen the neuron discharge waveform classification data according to the performance index of the neurons to obtain the screened neuron discharge waveform classification data; wherein, the performance index of the neurons corresponding to the screened neuron discharge waveform classification data meets the preset conditions;

[0012] Perform waveform processing on the waveform data of the target channel and target waveform type in the screened neuron discharge waveform classification data to obtain the standard discharge waveform of the neurons of the target type in the target channel.

[0013] Further, the classifying the neuron discharge waveform data to obtain neuron discharge waveform classification data includes:

[0014] Perform intra-channel clustering on the neuron discharge waveform data to obtain first waveform clustering data;

[0015] Perform inter-channel clustering on the first waveform clustering data to obtain second waveform clustering data, and use the second waveform clustering data as the neuron discharge waveform classification data.

[0016] Further, the performing intra-channel clustering on the neuron discharge waveform data to obtain first waveform clustering data includes:

[0017] Perform dimensionality reduction processing on the neuron discharge waveform data to obtain the dimensionality-reduced neuron discharge waveform data;

[0018] Perform intra-channel clustering based on the dimensionality-reduced neuron discharge waveform data to obtain the first waveform clustering data.

[0019] Further, performing inter-channel clustering on the first waveform clustering data to obtain second waveform clustering data includes:

[0020] Obtaining the distances between channels of high-throughput and high-density deep neural electrodes to obtain a plurality of channel spacing values and the corresponding relationship between the channel numbers and the channel spacing values;

[0021] Comparing the plurality of channel spacing values with a preset channel spacing threshold to obtain a comparison result;

[0022] Obtaining at least one channel combination within the preset channel spacing threshold according to the comparison result and the corresponding relationship; wherein, the channel combination includes at least two channels;

[0023] Calculating the class spacing projection mapping between the neuron discharge waveform data in the channel combination based on the first waveform clustering data;

[0024] Comparing the size of the class spacing projection mapping with a preset class spacing projection mapping threshold to obtain a class spacing projection mapping comparison result;

[0025] According to the class spacing projection mapping comparison result, clustering the neuron discharge waveform data corresponding to the class spacing projection mapping smaller than the preset class spacing projection mapping threshold into one class of waveform data to obtain the second waveform clustering data.

[0026] Further, performing waveform processing on the waveform data of the target channel and the target waveform type in the neuron discharge waveform classification data to obtain the standard discharge waveform of the neurons corresponding to the target channel and the target waveform type includes:

[0027] Obtaining the target channel and the target type for screening neuron discharge waveform data;

[0028] Extracting the discharge waveform data corresponding to the neurons of the target channel and the target type in the neuron discharge waveform classification data;

[0029] Performing waveform processing on the discharge waveform data corresponding to the neurons of the target channel and the target type to obtain the standard discharge waveform of the neurons corresponding to the target channel and the target type.

[0030] Further, performing waveform processing on the discharge waveform data corresponding to the neurons of the target channel and the target type to obtain the standard discharge waveform of the neurons corresponding to the target channel and the target type includes:

[0031] Calculate the average value of the discharge waveform data corresponding to the neurons of the target channel and the target type based on the discharge waveform data corresponding to the neurons of the target channel and the target type;

[0032] Generate the standard discharge waveform of the target channel and the target neuron based on the average value of the discharge waveform data corresponding to the neurons of the target channel and the target type.

[0033] A second aspect of the present invention provides a neural signal processing device, including:

[0034] An acquisition module, configured to acquire neural signal data; wherein, the neural signal data is multi-channel neural signal data collected by high-throughput and high-density deep neural electrodes;

[0035] A retrieval module, configured to perform spike retrieval on the neural signal data to obtain neuron discharge waveform data; wherein, the neuron discharge waveform data includes a plurality of waveform data representing neuron discharges;

[0036] A classification module, configured to classify the neuron discharge waveform data to obtain neuron discharge waveform classification data; wherein, the neuron discharge waveform classification data includes at least one waveform set, the waveform set includes at least one waveform data with the same waveform type, and the at least one waveform data with the same waveform type corresponds to the same type of neurons;

[0037] A waveform processing module, configured to perform waveform processing on the waveform data of the target channel and the target waveform type in the neuron discharge waveform classification data to obtain the standard discharge waveform of the neurons corresponding to the target channel and the target waveform type.

[0038] A third aspect of the present invention provides an electronic device, the electronic device includes a processor and a memory, and at least one instruction, at least one program, a code set or an instruction set is stored in the memory, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the neural signal processing method proposed in the first aspect of the present invention.

[0039] A fourth aspect of the present invention provides a computer-readable storage medium, and at least one instruction, at least one program, a code set or an instruction set is stored in the storage medium, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by a processor to implement the neural signal processing method proposed in the first aspect of the present invention.

[0040] Implementing the present invention has the following beneficial effects:

[0041] The neural signal processing method, apparatus, device, and storage medium provided by the embodiments of the present invention perform visual processing on neural signals collected by high-throughput, high-density flexible deep neural electrodes to obtain a standard discharge waveform that characterizes the physiological characteristics of neurons of a target channel and a target type. The obtained standard discharge waveform can more accurately and conveniently verify the quality of neural signals collected by high-throughput high-density deep neural electrodes, and has potential applications in the field of brain-computer interfaces and other neuroelectrophysiological research based on high-throughput, high-density deep neurons.

[0042] Additional aspects and advantages of the present invention will be given in part in the following description, become apparent in part from the following description, or be learned through the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings without creative efforts.

[0044] Figure 1 is a flowchart of a neural signal processing method provided by an embodiment of the present invention;

[0045] Figure 2 is a flowchart of classifying neuron discharge waveform data provided by an embodiment of the present invention;

[0046] Figure 3 is a flowchart of inter-channel clustering provided by an embodiment of the present invention;

[0047] Figure 4 is a flowchart of waveform processing provided by an embodiment of the present invention;

[0048] Figure 5 is a schematic diagram of the neural signal processing result provided by an embodiment of the present invention;

[0049] Figure 6 is a structural block diagram of a neural signal processing apparatus provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0050] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention. Examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements with the same or similar functions throughout.

[0051] Embodiment

[0052] Figure 1 FIG. is a flowchart of a neural signal processing method provided by an embodiment of the present invention. This specification provides the method operation steps as described in the embodiment or flowchart, but based on routine or non-creative labor, there may be more or fewer operation steps. The step order listed in the embodiment is only one of the execution orders of numerous steps and does not represent the only execution order. When the actual system or server product executes, it can be executed in the order shown in the embodiment or the accompanying drawings or executed in parallel (for example, in an environment of parallel processors or multi-threaded processing). Specifically, as Figure 1 shown, the method may include the following steps:

[0053] S101: Obtain neural signal data;

[0054] Specifically, the neural signal data is multi-channel neural signal data collected by high-throughput, high-density deep neural electrodes; the multi-channel neural signal data may include one or more time series data.

[0055] Theoretically speaking, the neural signal processing method provided by the embodiments of the present invention does not limit the material of the high-throughput, high-density deep neural electrodes to be rigid or flexible. The high-throughput, high-density deep neural electrodes may be rigid neural electrodes, including but not limited to silicon-based electrodes with a hard material itself, or may also be flexible neural electrodes, including but not limited to high-throughput, high-density flexible neural electrodes based on MEMS processing technology. However, invasive rigid electrodes are likely to cut the brain along with movement when recording signals in the human brain, causing damage to the brain. Flexible electrodes have a soft material itself and can adhere well to the brain tissue, causing no damage to the brain when recording signals, which is particularly important both in the scientific research field and in practical application scenarios.

[0056] To remove the inter-channel noise and low-frequency components, preprocessing operations can also be performed on the neural signal data. The preprocessing operations include temporal whitening processing and / or filtering processing. Since the discharge spikes are generally at high frequencies, the filtering processing here can adopt band-pass filtering and / or high-pass filtering. The lower limit of the value range of the band-pass filtering frequency is not less than 600 Hz, and the upper limit is not greater than 6000 Hz. For example, the band-pass filtering frequency range can be 600 Hz - 6000 Hz.

[0057] S102: Perform discharge spike retrieval on the neural signal data to obtain neuron discharge waveform data;

[0058] Specifically, the neuron discharge waveform data includes multiple waveform data representing neuron discharges. The waveform data representing neuron discharges here can specifically include neuron type serial numbers, channel serial numbers, and spike discharge times, and the neuron type serial numbers, channel serial numbers, and spike discharge times are stored in an associated manner.

[0059] Specifically, after band-pass filtering the original time series signal, signal segments are intercepted in the filtered time series signal according to the channel serial number and the spike discharge time, and the discharge conditions of different types of neurons in different channels can be obtained accordingly.

[0060] Discharge spike retrieval is the process of decomposing neural signal data into single neuron activity information. This process requires using microelectrode array technology to record the single-cell action potential pulses of neurons outside brain cells, select the spikes of neurons near the electrode tip, and determine which spike corresponds to which neuron. Therefore, discharge spike retrieval is also called spike sorting. The full English name is Spike sorting. Spike refers to the action potential (i.e., spike potential) of brain neurons, and its generation and transmission are the basis for the brain nervous system to achieve information interaction and processing.

[0061] Specifically, discharge spike retrieval can include sorting on neighborhoods, merging of neighboring blocks, and fitting event. Sorting on neighborhoods subdivides each neighboring block into event detection, block extraction, clustering, etc.

[0062] S103: Classify the neuron discharge waveform data to obtain neuron discharge waveform classification data;

[0063] Among them, the neuron discharge waveform classification data includes at least one waveform set. The waveform set includes at least one waveform data with the same waveform type, and at least one waveform data with the same waveform type corresponds to the same type of neuron.

[0064] Figure 2It is a flowchart for classifying neuron discharge waveform data provided by an embodiment of the present invention. Specifically, as Figure 2 shown, the steps for classifying neuron discharge waveform data to obtain neuron discharge waveform classification data are as follows:

[0065] S201: Perform in-channel clustering on the neuron discharge waveform data to obtain first waveform clustering data;

[0066] In some embodiments, performing in-channel clustering on the neuron discharge waveform data to obtain first waveform clustering data includes:

[0067] Perform dimensionality reduction processing on the neuron discharge waveform data to obtain the dimensionality-reduced neuron discharge waveform data;

[0068] Perform in-channel clustering based on the dimensionality-reduced neuron discharge waveform data to obtain first waveform clustering data.

[0069] Performing dimensionality reduction processing on the neuron discharge waveform data before in-channel clustering and then extracting the data that may be neuron discharge waveform data on each channel based on the dimensionality-reduced neuron discharge waveform data is beneficial to reducing the data processing volume in the subsequent in-channel clustering and inter-channel clustering stages.

[0070] It should be noted that for high-throughput, high-density deep neural electrodes, since the spacing between different channels is small, the discharge of a single neuron is easily collected on multiple channels. Therefore, there will be a problem that a type of neuron discharge is recognized multiple times in neuron discharge spike retrieval. It is necessary to perform inter-channel clustering processing on different types of neurons in multiple channels with channel spacing not exceeding a preset threshold after obtaining the first waveform clustering data, so as to improve the accuracy of the retrieval result.

[0071] S202: Perform inter-channel clustering on the first waveform clustering data to obtain second waveform clustering data, and use the second waveform clustering data as the neuron discharge waveform classification data.

[0072] Figure 3 It is a flowchart for inter-channel clustering provided by an embodiment of the present invention. Specifically, as Figure 3 shown, in some embodiments, performing inter-channel clustering on the first waveform clustering data to obtain second waveform clustering data includes the following steps:

[0073] S301: Obtain the distances between the channels of the high-throughput, high-density deep neural electrodes to obtain multiple channel spacing values and the corresponding relationship between the channel numbers and the channel spacing values;

[0074] Specifically, the channel spacing value refers to the physical distance between channels. In one embodiment, the distance between channels of a high-throughput, high-density deep neural electrode is the distance between all pairs of channels of the deep neural electrode. In one embodiment, the distance between channels of a high-throughput, high-density deep neural electrode is the distance between adjacent channels of the deep neural electrode.

[0075] S302: Compare multiple channel spacing values with a preset channel spacing threshold to obtain a comparison result.

[0076] Among them, the preset channel spacing threshold can be set according to the actual situation. For example, the preset channel spacing threshold can be any value between 90 and 110 microns. For example, it can be 90 microns, 100 microns, or 110 microns, etc. This embodiment is not limited thereto.

[0077] S303: Obtain at least one channel combination with a channel spacing value within the preset channel spacing threshold according to the comparison result and the corresponding relationship.

[0078] Among them, the channel combination includes at least two channels.

[0079] S304: Calculate the class spacing projection mapping between the neuron discharge waveform data in the channel combination based on the first waveform clustering data.

[0080] Among them, the class spacing, that is, the inter-class distance, specifically refers to the distance between different classes in the first waveform clustering data. The calculation method of the class spacing projection mapping is a prior art and is not an improvement point of this embodiment of the present invention.

[0081] S305: Compare the size of the class spacing projection mapping with a preset class spacing projection mapping threshold to obtain a class spacing projection mapping comparison result.

[0082] Among them, the preset threshold is an empirical value; if the class spacing projection mapping is less than the preset threshold, it is considered that the corresponding different types of neurons are similar and can be grouped into one class. By grouping the judged similar neurons into one class, the neurons retrieved from different channels can be merged.

[0083] S306: According to the class spacing projection mapping comparison result, group the neuron discharge waveform data corresponding to the class spacing projection mapping less than the preset class spacing projection mapping threshold into one class of waveform data to obtain the second waveform clustering data.

[0084] For high-density deep neural electrodes, the discharge of a single neuron can easily be collected on multiple channels. By comparing different types of neurons in adjacent neural channels in this embodiment of the present invention, it is possible to avoid multiple identifications of the discharge of one type of neuron, which is beneficial to improving the accuracy of the neuron discharge spike retrieval result.

[0085] In the embodiment of the present invention, different types of neurons in each channel are first clustered to obtain a first clustering result, and then different types of neurons in adjacent channels are clustered based on the first clustering result to obtain a second clustering result. Through the above two neuron clustering operations, the problems of long time consumption and inability to compare between channels in the retrieval of neuron discharges (Spikes) of neural signals collected by high-throughput and high-density flexible deep neural electrodes are solved.

[0086] S104: Perform waveform processing on the waveform data of the target channel and the target waveform type in the neuron discharge waveform classification data to obtain the standard discharge waveform of the neurons corresponding to the target channel and the target waveform type.

[0087] Optionally, step S104 includes the following steps:

[0088] Calculate the performance indicators of the neurons corresponding to each waveform type in the neuron discharge waveform classification data;

[0089] Screen the neuron discharge waveform classification data according to the performance indicators of the neurons to obtain the screened neuron discharge waveform classification data; wherein, the performance indicators of the neurons corresponding to the screened neuron discharge waveform classification data meet the preset conditions;

[0090] Perform waveform processing on the waveform data of the target channel and the target waveform type in the screened neuron discharge waveform classification data to obtain the standard discharge waveform corresponding to the neurons of the target type in the target channel.

[0091] Specifically, the performance indicators of the neurons include some parameters for describing the performance characteristics of isolated neurons and the value ranges of these parameters. For example, signal-to-noise ratio (SNR), noise overlap (the English name is noise overlap), neuron firing rate, and isolation (the English name is isolation, used to describe the size of the overlap).

[0092] Specifically, when the above parameters for describing the neuron performance characteristics are within their respective preset value ranges (i.e., meet the preset conditions), it is considered that the performance of the neurons meets the requirements, or the performance is relatively good.

[0093] Specifically, according to the values of the performance indicators, neurons can be classified into noise, isolated neurons, and bursting clusters. According to such clustering results, determine which types of neurons will be finally retained as the types of neurons detected, that is, screen out the neurons that meet the preset conditions. In the embodiment of the present invention, the neurons that meet the preset conditions refer to isolated neurons.

[0094] In one embodiment, step S104 includes the following steps:

[0095] Obtain the target channels and target types for screening neuron discharge waveform data;

[0096] Extract the discharge waveform data corresponding to the neurons of the target channels and target types from the neuron discharge waveform classification data;

[0097] Perform waveform processing on the discharge waveform data corresponding to the neurons of the target channels and target types to obtain the standard discharge waveforms corresponding to the neurons of the target channels and target types.

[0098] Under normal circumstances, there will be a large number of similar waveforms in the discharge spike retrieval results, and there will always be deviations. However, neuron discharge has a fixed discharge pattern, and taking the average value can remove these deviations.

[0099] Figure 4 It is a flowchart of waveform processing provided by an embodiment of the present invention. Specifically, as Figure 4 shown, in some embodiments, performing waveform processing on the discharge waveform data corresponding to the neurons of the target channels and target types to obtain the standard discharge waveforms corresponding to the neurons of the target channels and target types includes the following steps:

[0100] S401: Calculate the average value of the discharge waveform data corresponding to the neurons of the target channels and target types according to the discharge waveform data corresponding to the neurons of the target channels and target types;

[0101] Specifically, calculating the average value of the waveform parameters of the target channels and target neurons can be to calculate the average value of all waveform data of each type of neuron in each channel by calling the mean function of the numpy software package in python. According to actual needs, other methods can also be used to calculate the average value of all waveform data of the target channels and target neurons. This embodiment is not limited thereto.

[0102] Specifically, the target channels can be one or more channels. Preferably, the target channels are all channels of the neural electrode;

[0103] Specifically, the target types of neurons can be one or more types of neurons. Preferably, the target types of neurons are all types of neurons whose performance indicators meet the preset conditions.

[0104] S402: Generate the standard discharge waveforms of the target channels and target neurons according to the average value of the discharge waveform data corresponding to the neurons of the target channels and target types.

[0105] Specifically, step S1042 may be to visualize the obtained average waveform through the matplotlib software package of python. According to actual needs, other methods may also be used to visualize the average waveform of each type of target neuron in each channel. This embodiment is not limited thereto.

[0106] To serve the preparation iteration of high-throughput and high-density deep neural electrodes, it is necessary to verify the quality of the neural signals collected by high-throughput and high-density deep neural electrodes. Visualizing the processing results of neural electrode signals can facilitate the measurement and evaluation of the quality of neural signals collected by high-throughput and high-density deep neural electrodes.

[0107] Figure 5 is a schematic diagram of the processing results of neural signals provided by an embodiment of the present invention. Please refer to Figure 5 , after processing the neural signals collected by a 128-channel high-density flexible deep neural electrode with a duration of 10 minutes locally collected according to the neural signal processing method provided by the embodiment of the present invention, a very obvious standard neuron waveform can be obtained.

[0108] Figure 6 is a structural block diagram of a neural signal processing device provided by an embodiment of the present invention. Specifically, as Figure 6 shown, an embodiment of the present invention proposes a neural signal processing device, including the following modules:

[0109] An acquisition module 501, configured to acquire neural signal data; wherein, the neural signal data is multi-channel neural signal data collected by a high-throughput and high-density deep neural electrode;

[0110] A retrieval module 502, configured to perform spike retrieval on the neural signal data to obtain neuron discharge waveform data; wherein, the neuron discharge waveform data includes a plurality of waveform data representing neuron discharges;

[0111] A classification module 503, configured to classify the neuron discharge waveform data to obtain neuron discharge waveform classification data; wherein, the neuron discharge waveform classification data includes at least one waveform set, the waveform set includes at least one waveform data with the same waveform type, and at least one waveform data with the same waveform type corresponds to the same type of neuron;

[0112] A waveform processing module 504, configured to perform waveform processing on the waveform data of a target channel and a target waveform type in the neuron discharge waveform classification data to obtain a standard discharge waveform of the neuron corresponding to the target channel and the target waveform type.

[0113] It should be noted that the present invention is not limited by the described order of actions, because according to the present invention, certain steps may be performed in other orders or simultaneously.

[0114] An embodiment of the present invention further provides an electronic device, which includes a processor and a memory. The memory stores at least one instruction, at least one program, a code set or an instruction set. The at least one instruction, at least one program, code set or instruction set is loaded and executed by the processor to implement the neural signal processing method in the method embodiment.

[0115] An embodiment of the present invention further provides a storage medium, which can be set in a server to store at least one instruction, at least one program, a code set or an instruction set related to implementing the neural signal processing method in the method embodiment. The at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the neural signal processing method provided in the above method embodiment.

[0116] Optionally, in this embodiment, the above storage medium may be located in at least one of multiple network servers in a computer network. Optionally, in this embodiment, the above storage medium may include, but is not limited to: various media that can store program codes such as USB flash drives, read-only memories (ROMs), random access memories (RAMs), mobile hard disks, magnetic disks, or optical discs.

[0117] As can be seen from the embodiments of the neural signal processing method, device, electronic device, or storage medium provided by the present invention above, the neural signal processing method, device, equipment, and storage medium provided by the embodiments of the present invention perform visualization processing on the neural signals collected by high-throughput and high-density deep neural electrodes to obtain a standard discharge waveform representing the physiological characteristics of neurons of a target channel and a target type, greatly reducing the processing time and significantly improving the accuracy of the processing results, which is beneficial to saving the working time of scientific researchers and improving research efficiency; the obtained standard discharge waveform can more accurately and conveniently verify the quality of the neural signals collected by high-throughput and high-density deep neural electrodes, and has potential applications in the fields of brain-computer interfaces and other neuroelectrophysiological research based on high-throughput and high-density deep neurons.

[0118] It should be noted that: the above sequence of embodiments of the present invention is only for description and does not represent the advantages or disadvantages of the embodiments. And the above specific embodiments of this specification have been described. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be executed in a different order than in the embodiments and still achieve the desired results. Additionally, the processes depicted in the drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0119] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other, and the differences between each embodiment and other embodiments are emphasized. In particular, for the apparatus and server embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and reference can be made to the relevant parts of the method embodiments for the relevant content.

[0120] Those of ordinary skill in the art can understand that all or part of the steps to implement the above embodiments can be completed by hardware, or can be completed by a program instructing the relevant hardware. The program can be stored in a computer-readable storage medium, and the above-mentioned storage medium can be a read-only memory, a magnetic disk or an optical disk, etc.

[0121] The above are only the preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for processing neural signals, characterized in that, Including: Obtaining neural signal data; wherein, the neural signal data is multi-channel neural signal data collected by high-throughput and high-density deep neural electrodes; Performing spike retrieval on the neural signal data to obtain neuron discharge waveform data; wherein, the neuron discharge waveform data includes a plurality of waveform data representing neuron discharges; Classifying the neuron discharge waveform data to obtain neuron discharge waveform classification data; wherein, the neuron discharge waveform classification data includes at least one waveform set, the waveform set includes at least one waveform data with the same waveform type, and the at least one waveform data with the same waveform type corresponds to the same type of neuron; Performing waveform processing on the waveform data of the target channel and the target waveform type in the neuron discharge waveform classification data to obtain the standard discharge waveform of the neuron corresponding to the target channel and the target waveform type.

2. The method according to claim 1, characterized in that The performing waveform processing on the waveform data of the target channel and the target waveform type in the neuron discharge waveform classification data to obtain the standard discharge waveform of the neuron corresponding to the target channel and the target waveform type includes: Calculating the performance index of the neurons corresponding to each waveform type in the neuron discharge waveform classification data; Screening the neuron discharge waveform classification data according to the performance index of the neurons to obtain the screened neuron discharge waveform classification data; wherein, the performance index of the neurons corresponding to the screened neuron discharge waveform classification data meets a preset condition; Performing waveform processing on the waveform data of the target channel and the target waveform type in the screened neuron discharge waveform classification data to obtain the standard discharge waveform of the neurons corresponding to the target waveform type in the target channel.

3. The method according to claim 1, wherein The classifying the neuron discharge waveform data to obtain neuron discharge waveform classification data includes: Performing in-channel clustering on the neuron discharge waveform data to obtain first waveform clustering data; Performing inter-channel clustering on the first waveform clustering data to obtain second waveform clustering data, and using the second waveform clustering data as the neuron discharge waveform classification data.

4. The method according to claim 3, wherein The performing in-channel clustering on the neuron discharge waveform data to obtain first waveform clustering data includes: Performing dimensionality reduction processing on the neuron discharge waveform data to obtain the dimensionality-reduced neuron discharge waveform data; Performing in-channel clustering based on the dimensionality-reduced neuron discharge waveform data to obtain the first waveform clustering data.

5. The method according to claim 3, characterized in that, The performing inter-channel clustering on the first waveform clustering data to obtain second waveform clustering data includes: Obtaining the distances between the channels of the high-throughput and high-density deep neural electrodes to obtain a plurality of channel spacing values and the corresponding relationship between the channel numbers and the channel spacing values; Comparing the plurality of channel spacing values with a preset channel spacing threshold to obtain a comparison result; Obtaining at least one channel combination with the channel spacing value within the preset channel spacing threshold according to the comparison result and the corresponding relationship; wherein, the channel combination includes at least two channels; Calculate the class interval projection mapping between the neuron discharge waveform data in the channel combination based on the first waveform clustering data; Compare the class interval projection mapping with a preset class interval projection mapping threshold to obtain a class interval projection mapping comparison result; According to the class interval projection mapping comparison result, cluster the neuron discharge waveform data corresponding to the class interval projection mapping that is less than the preset class interval projection mapping threshold into one type of waveform data to obtain the second waveform clustering data.

6. The method according to claim 1, characterized in that, The waveform processing of the waveform data of the target channel and the target waveform type in the neuron discharge waveform classification data to obtain the standard discharge waveform of the neurons corresponding to the target channel and the target waveform type includes: Obtain the target channel and the target waveform type for screening neuron discharge waveform data; Extract the discharge waveform data corresponding to the neurons of the target channel and the target waveform type from the neuron discharge waveform classification data; Perform waveform processing on the discharge waveform data corresponding to the neurons of the target channel and the target waveform type to obtain the standard discharge waveform of the neurons corresponding to the target channel and the target waveform type.

7. The method according to claim 6, wherein The waveform processing of the discharge waveform data corresponding to the neurons of the target channel and the target waveform type to obtain the standard discharge waveform of the neurons corresponding to the target channel and the target waveform type includes: Calculate the average value of the discharge waveform data corresponding to the neurons of the target channel and the target waveform type according to the discharge waveform data corresponding to the neurons of the target channel and the target waveform type; Generate the standard discharge waveform of the neurons corresponding to the target channel and the target waveform type according to the average value of the discharge waveform data corresponding to the neurons of the target channel and the target waveform type.

8. A neural signal processing device, characterized in that, Includes: An acquisition module for acquiring neural signal data; wherein, the neural signal data is multi-channel neural signal data collected by high-throughput, high-density deep neural electrodes; A retrieval module for performing discharge spike retrieval on the neural signal data to obtain neuron discharge waveform data; wherein, the neuron discharge waveform data includes a plurality of waveform data representing neuron discharges; A classification module for classifying the neuron discharge waveform data to obtain neuron discharge waveform classification data; Wherein, the neuron discharge waveform classification data includes at least one waveform set, the waveform set includes at least one waveform data with the same waveform type, and the at least one waveform data with the same waveform type corresponds to the same type of neurons; A waveform processing module for performing waveform processing on the waveform data of the target channel and the target waveform type in the neuron discharge waveform classification data to obtain the standard discharge waveform of the neurons corresponding to the target channel and the target waveform type.

9. An electronic device, characterized in that, The electronic device includes a processor and a memory. At least one instruction, at least one program, a code set, or an instruction set is stored in the memory. The at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by the processor to implement the neural signal processing method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, At least one instruction, at least one program, a code set, or an instruction set is stored in the storage medium. The at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by a processor to implement the neural signal processing method according to any one of claims 1-7.

Images