Real-time data processing method and device for neural signals and computer storage medium

Abstract

The invention provides a real-time data processing method and device for neural signals and a computer storage medium. Based on the method, after a continuous neural signal about a target neural region is acquired as a target neural signal, the target neural signal can be converted into a corresponding discrete target digital signal; carrying out peak classification processing according to a preset peak classification algorithm to obtain a corresponding peak classification result; and obtaining a plurality of pieces of neural signal information corresponding to different types of neurons according to a peak classification result. Therefore, on one hand, the data volume of the finally obtained to-be-transmitted neural signal data can be effectively reduced, the data transmission burden during subsequent transmission of the neural signal data is reduced, and the transmission quality during subsequent transmission of the neural signal data is improved; on the other hand, different types of neural signals can be finely separated from the continuous local field potential, and multiple pieces of neural signal information corresponding to different types of neurons are obtained.

Description

Technical Field

[0001] This specification belongs to the technical field of neural signal data processing, and particularly relates to a real-time data processing method, apparatus, and computer storage medium for neural signals. Background Art

[0002] When performing real-time data processing on neural signals collected in real time by an implanted device inside an implanted subject, the data volume of the neural signal data to be transmitted finally processed based on the existing method is often relatively large, resulting in a heavy data transmission burden. At the same time, when transmitting the above neural signal data, there are also problems such as easy signal distortion and attenuation, and the transmission quality is relatively poor, resulting in a large error in the neural signal data finally received by the external device. In addition, the neural signal data finally processed based on the existing method often can only reflect the sum of the neural signal data corresponding to different neurons in a local area, and cannot finely identify and separate different types of neural signals corresponding to different neurons.

[0003] In view of the above problems, no effective solution has been proposed yet. Summary of the Invention

[0004] This specification provides a real-time data processing method, apparatus, and computer storage medium for neural signals. On the one hand, it can effectively reduce the data volume of the neural signal data to be transmitted finally obtained, reduce the data volume during subsequent transmission of neural signal data, and improve the signal-to-noise ratio of the signal during subsequent communication of neural signal data; on the other hand, it can also separate different types of neural signals from continuous neural signals to obtain multiple neural signal information corresponding to different types of neurons respectively, meeting the complex and diverse scenario requirements.

[0005] This specification provides a real-time data processing method for neural signals, including:

[0006] Obtain a target neural signal;

[0007] Perform analog-to-digital conversion processing on the target neural signal to obtain a corresponding target digital signal;

[0008] Perform spike classification processing on the target digital signal according to a preset spike classification algorithm to obtain a corresponding spike classification result;

[0009] Obtain multiple neural signal information according to the spike classification result; where one neural signal information corresponds to one type of neural signal; the multiple neural signal information includes the peak times of the action potentials of different neurons.

[0010] In one embodiment, obtaining a target neural signal includes:

[0011] Collect the current electrical signals in the target nerve region;

[0012] Perform spike detection on the current electrical signals to obtain corresponding spike detection results;

[0013] According to the spike detection results, determine whether the current electrical signals belong to valid nerve signals;

[0014] When it is determined that the current electrical signals belong to valid nerve signals, collect the nerve signals in the target nerve region as the target nerve signals.

[0015] In one embodiment, perform spike classification processing on the target digital signals according to a preset spike classification algorithm to obtain corresponding spike classification results, including:

[0016] Perform spike classification processing on the current digital signals in the target digital signals according to the following method and the preset spike classification algorithm:

[0017] Detect and intercept the current spike signal from the current digital signals;

[0018] According to the preset spike classification algorithm, calculate the feature distances between the current spike signal and the cluster centers of each preset spike cluster; wherein, one preset spike cluster corresponds to one type of nerve signal;

[0019] Detect whether there is a cluster center of a target spike cluster in the preset spike cluster whose feature distance from the current spike signal is less than or equal to a preset distance threshold;

[0020] When it is determined that there is a cluster center of a target spike cluster in the preset spike cluster whose feature distance from the current spike signal is less than or equal to the preset distance threshold, determine the nerve signal type corresponding to the target spike cluster as the spike classification result of the current digital signals.

[0021] In one embodiment, detecting and intercepting the current spike signal from the current digital signals includes:

[0022] Detect the current digital signals and determine the position points of the peak potentials;

[0023] According to the position points of the peak potentials, intercept the signals of a preset number of sampling points from the current digital signals as the current spike signal.

[0024] In one embodiment, after determining the nerve signal type corresponding to the target spike cluster as the spike classification result of the current digital signals, the method further includes:

[0025] Use the current spike signal to update the cluster center of the target spike cluster.

[0026] In one embodiment, after detecting whether the feature distance between the clustering center of a target spike cluster and the current spike signal in a preset spike cluster is less than or equal to a preset distance threshold, the method further includes:

[0027] In the case of determining that there is no feature distance between the clustering center of a target spike cluster and the current spike signal in a preset spike cluster that is less than or equal to a preset distance threshold, a new preset spike cluster is created; and according to the current spike signal, the clustering center of the preset spike cluster is determined.

[0028] In one embodiment, the preset spike cluster is determined in the following manner:

[0029] Obtain a test electrical signal regarding a target nerve region;

[0030] According to the test electrical signal, obtain a corresponding test digital signal;

[0031] Intercept a plurality of test spike signals from the test digital signal;

[0032] Process the test spike signals to extract a corresponding plurality of test spike features; wherein, the data dimension of the test spike features is smaller than the data dimension of the test spike signals;

[0033] By performing clustering processing on the plurality of test spike features, a plurality of preset spike clusters and the clustering centers of the plurality of preset spike clusters are determined.

[0034] In one embodiment, after obtaining a plurality of nerve signal information, the method further includes:

[0035] According to different nerve signal types, respectively add the plurality of nerve signal information to an echo signal in a matching manner; and externally transmit the echo signal.

[0036] This specification also provides a real-time data processing device for nerve signals, including:

[0037] A first acquisition module, configured to acquire a target nerve signal;

[0038] An analog-to-digital conversion module, configured to perform analog-to-digital conversion processing on the target nerve signal to obtain a corresponding target digital signal;

[0039] A spike classification module, configured to perform spike classification processing on the target digital signal according to a preset spike classification algorithm to obtain a corresponding spike classification result;

[0040] A second acquisition module, configured to acquire a plurality of neural signal information according to the spike classification result; wherein, one neural signal information corresponds to one type of neural signal; the plurality of neural signal information includes the peak times of action potentials of different neurons.

[0041] This specification also provides a computer storage medium, on which computer instructions are stored, and when the instructions are executed by a processor, the related steps of the real-time data processing method of neural signals are implemented.

[0042] Based on the real-time data processing method, device and computer storage medium of neural signals provided in this specification, after collecting continuous neural signals about the target neural region as the target neural signals, the above target neural signals can be first converted from spectral data into corresponding discrete target digital signals; then, according to a preset spike classification algorithm, spike classification processing is performed on the target digital signal to obtain a corresponding spike classification result; and according to the spike classification result, a plurality of neural signal information corresponding to different types of neurons are identified and separated from the target digital signal. In this way, on the one hand, it can effectively reduce the data volume of the finally obtained neural signal data to be transmitted, reduce the data transmission burden when transmitting neural signal data subsequently, improve the signal-to-noise ratio of the signal during subsequent communication of neural signal data, and thus improve the transmission quality of neural signal data during subsequent transmission; on the other hand, it can also separate different types of neural signals from continuous neural signals to obtain a plurality of neural signal information corresponding to different types of neurons, meeting the complex and diverse scenario requirements.

[0043] Moreover, when converting the target neural signal from the original analog signal into the corresponding digital signal, by combining specific scenario requirements, only the peak times of action potentials of different types of neurons required are extracted as the plurality of neural signal information, thereby further reducing the data transmission burden when transmitting neural signal data subsequently.

[0044] In addition, spike detection is performed on the electrical signals in the target neural region collected in real time to identify effective neural signals; and then only the effective neural signals are collected as the target neural signals for subsequent data processing, thereby avoiding wasting processing resources and processing time on invalid neural signals or noise signals, improving the utilization rate of processing resources, and reducing the overall working energy consumption of the implanted device. Description of the Drawings

[0045] In order to more clearly illustrate the embodiments of this specification, the drawings required for the embodiments will be briefly introduced below. The drawings in the following description are only some embodiments recorded in this specification. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0046] Figure 1 It is a schematic flowchart of a real-time data processing method for nerve signals provided by an embodiment of this specification;

[0047] Figure 2 It is a schematic diagram of an embodiment of applying the real-time data processing method for nerve signals provided by the embodiment of this specification in a scenario example;

[0048] Figure 3 It is a schematic diagram of an embodiment of applying the real-time data processing method for nerve signals provided by the embodiment of this specification in a scenario example;

[0049] Figure 4 It is a schematic diagram of an embodiment of applying the real-time data processing method for nerve signals provided by the embodiment of this specification in a scenario example;

[0050] Figure 5 It is a schematic diagram of an embodiment of applying the real-time data processing method for nerve signals provided by the embodiment of this specification in a scenario example;

[0051] Figure 6 It is a schematic diagram of an embodiment of applying the real-time data processing method for nerve signals provided by the embodiment of this specification in a scenario example;

[0052] Figure 7 It is a schematic diagram of the structural composition of an electronic device applying the real-time data processing method for nerve signals provided by the embodiment of this specification;

[0053] Figure 8 It is a schematic diagram of the structural composition of a real-time data processing device for nerve signals provided by an embodiment of this specification. Detailed implementation manners

[0054] In order to enable those skilled in the art to better understand the technical solutions in this specification, the following will clearly and completely describe the technical solutions in the embodiments of this specification in conjunction with the accompanying drawings in the embodiments of this specification. Obviously, the described embodiments are only a part of the embodiments of this specification, rather than all the embodiments. Based on the embodiments in this specification, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of this specification.

[0055] Considering the existing data processing methods for neural signals, most of them directly transmit the neural signals in the form of analog signals collected by the implanted device after performing conventional processing such as denoising. In this way, on the one hand, it leads to a large amount of data of the neural signals to be transmitted, increasing the working energy consumption of the implanted device when transmitting neural signal data, and then resulting in the need to introduce a power supply module or an energy storage module in the implanted device, increasing the size of the implanted device. On the other hand, since the neural signal data is transmitted in the form of analog signals, due to the influence of the spectral data characteristics, the neural signal data is prone to be interfered and affected by external factors during transmission, and problems such as distortion and attenuation are likely to occur, and the transmission quality is poor, making it difficult for the external device to receive neural signal data with small errors and good quality. In addition, based on the existing data processing methods for neural signals, it is also impossible to accurately identify and separate the neural signal information corresponding to different types of neurons in the directly collected neural signals, and thus it cannot meet the requirements of relatively fine and high-demand scenarios.

[0056] In view of the above problems existing in the existing data processing methods for neural signals, this application considers: First, combined with specific scenario requirements, although neural data in the form of analog signals, as a kind of spectral data, has a large amount of data and contains a lot of data information, in fact, most of the information that really needs to be concerned about is the time information when different neurons generate action potentials. Second, when transmitting neural signal data in the form of analog signals, it is easily affected and interfered by external factors, but if the neural signal data is transmitted in the form of digital signals, based on the data characteristics of digital signals and the internal environmental characteristics of the target object, it will show better resistance and is not easily distorted or attenuated significantly. In addition, due to factors such as size and energy consumption, the data processing ability of the implanted device disposed inside the target object is often relatively limited and is not suitable for performing relatively complex data processing in most cases. However, through a large number of experimental tests, it is found that the data processing volume involved in the data processing process based on spike sorting is often relatively small and is more suitable for processing complex neural signals, and can obtain better processing effects.

[0057] Based on the above situation, the present application further considers: First, the analog neural signals directly collected can be converted into digital signals with relatively small data volume and good resistance, so as to reduce the transmission burden during subsequent neural signal data transmission and reduce the working energy consumption of the implant device when transmitting neural signal data; at the same time, the subsequent transmission quality can also be improved. Then, a spike sorting algorithm can be introduced to perform corresponding spike sorting processing on the neural signals in digital signal form. Based on the spike features with smaller data dimensions, multiple different neural signals corresponding to different neurons can be accurately identified and separated from the neural signals with a relatively small amount of data processing. Finally, combined with the specific scenario requirements, only the peak times of the action potentials of different neurons among the multiple different neural signals corresponding to different neurons are further extracted as the multiple neural signal information to be finally transmitted, so as to further reduce the transmission burden during subsequent neural signal data transmission and further reduce the working energy consumption of the implant device when transmitting neural signal data.

[0058] Based on the above considerations, referring to Figure 1 as shown, the embodiments of this specification provide a real-time data processing method for neural signals. Specifically, when implemented, it may include the following content:

[0059] S101: Obtain target neural signals;

[0060] S102: Perform analog-to-digital conversion processing on the target neural signals to obtain corresponding target digital signals;

[0061] S103: Perform spike sorting processing on the target digital signals according to a preset spike sorting algorithm to obtain corresponding spike sorting results;

[0062] S104: According to the spike sorting results, obtain multiple neural signal information; where one neural signal information corresponds to one type of neural signal; the multiple neural signal information includes the peak times of the action potentials of different neurons.

[0063] Among them, the above real-time data processing method for neural signals can be specifically applied to one side of the implant device. The above implant device can also be specifically arranged inside the target object. Correspondingly, the above target neural signals can specifically be the real-time neural signals detected and collected by the implant device from the target neural region inside the target object. Based on the method provided in this specification, real-time data processing can be performed on the above target neural signals.

[0064] Among them, the above target object can specifically be an animal or a human, etc.

[0065] The above-mentioned target nerve region can specifically be understood as the region within the target object where neuron cells to be detected are distributed. Specifically, the above-mentioned target nerve region may include the brain tissue region of the target object, such as the prefrontal lobe, the epidermal layer, the hippocampus, etc. In addition, the above-mentioned target nerve region may also include the sciatic nerve region of the target object, etc. Of course, it should be noted that the above-listed target nerve regions are only illustrative. In specific implementation, according to specific circumstances and processing requirements, the above-mentioned target nerve region may also include other regions within the target object. This specification does not make any limitations in this regard.

[0066] Specifically, referring to Figure 2 As shown, an external device corresponding to the implant device is also arranged outside the target object. The external device may at least include devices such as an ultrasonic probe.

[0067] Among them, the above-mentioned ultrasonic probe can specifically be understood as a device used to transmit and receive ultrasonic waves during ultrasonic detection.

[0068] Correspondingly, the above-mentioned external device can transmit and receive ultrasonic signals through the above-mentioned ultrasonic probe.

[0069] The implant device may at least include an ultrasonic transducer and a neural signal processing chip device.

[0070] Among them, the above-mentioned ultrasonic transducer usually consists of a housing, a matching layer, a piezoelectric ceramic disk transducer, a backing, a lead cable, and a Cymbal array receiver, and can receive and transmit ultrasonic signals, and can also achieve the mutual conversion of mechanical energy and electrical energy.

[0071] The above-mentioned implant device can receive the ultrasonic signals transmitted by the external device through the ultrasonic transducer, and convert the mechanical energy of the ultrasonic waves into electrical energy for the operation of the implant device.

[0072] In this way, it is not necessary to arrange a power module such as a battery or other energy storage modules in the implant device, thereby effectively reducing the size of the implant device, lowering the cost of the implant device, making it relatively easier and more convenient to arrange the implant device inside the target object, and avoiding biocompatibility problems.

[0073] Furthermore, the above-mentioned implant device can also externally reflect echo signals through the ultrasonic transducer.

[0074] Among them, the above-mentioned neural signal processing chip may at least include a neural signal processing circuit; the neural signal processing circuit is connected with a first electrode and a second electrode; the first electrode and the second electrode are used to detect and collect electrical signals in the target nerve region of the target object.

[0075] Specifically, referring to Figure 2 as shown, the above-mentioned first electrode can be inserted into a position in the target nerve area that is relatively far from the neuron cells and used as the negative electrode (for example, V Ref , reference potential). The above-mentioned second electrode can be inserted into a position in the target nerve area that is relatively close to the neuron cells and used as the positive electrode (for example, Neuron, nerve cell). In specific implementation, the potential difference between the two electrodes (that is, the voltage of the electrical signal in the target nerve area) can be monitored through the above-mentioned first electrode and second electrode to detect whether there is an effective nerve signal that needs to be collected.

[0076] Specifically, the above-mentioned nerve signal processing circuit is used to perform corresponding data processing on the nerve signals collected by the electrodes according to the real-time data processing method of the nerve signals provided in this specification, and convert the initial analog signals that are mixed together, have a lot of interference, and a large amount of data (that is, the directly collected target nerve signals) into relatively corresponding digital signals that are finely corresponding to multiple different types of nerve signals, have a high precision, and a small amount of data (that is, the required multiple nerve signal information). Among them, the above-mentioned nerve signal information can specifically include the peak time of the action potential of the corresponding type of neuron.

[0077] Specifically, referring to Figure 2 as shown, the above-mentioned nerve signal processing circuit can at least include: an analog front end (for example, AFE), a clock generation circuit (or a clock signal generation circuit), an analog-to-digital converter (for example, LC-ADC), a spike sorting algorithm module and other devices; among them, the analog front end is electrically connected to the first electrode, the second electrode, and the analog-to-digital converter respectively; the analog-to-digital converter is also electrically connected to the spike sorting algorithm module.

[0078] Specifically, the analog front end can be a filter amplifier.

[0079] Specifically, the above-mentioned analog front end can be electrically connected to the first electrode, the second electrode, and the analog-to-digital converter respectively; the analog-to-digital converter can also be electrically connected to the spike sorting algorithm module and the clock generation circuit respectively; the spike sorting algorithm module can also be electrically connected to the clock generation circuit respectively. Among them, the spike sorting algorithm module is configured with a preset spike sorting algorithm.

[0080] Further, the nerve signal processing chip can also include a spike detector; among them, the spike detector can be electrically connected to the first electrode, the second electrode, the spike sorting algorithm module, the clock generation circuit, and the analog-to-digital converter respectively. Among them, the spike detector is configured with a preset spike threshold.

[0081] The above-mentioned external device may specifically further include devices such as a computer device and a data acquisition card; among them, the data acquisition card is electrically connected to the computer device and the ultrasonic probe respectively.

[0082] Among them, the above-mentioned computer device may specifically include a desktop computer, a server, a laptop computer, etc., or other electronic devices that support functions such as data operation and data storage.

[0083] Based on the above-mentioned implant device and external device, in specific implementation, ultrasonic signals can be first emitted to the implant device arranged inside the target object by using the external device. The implant device uses an ultrasonic transducer to receive the ultrasonic signals and converts the received ultrasonic signals into electrical energy for the operation of the implant device. After the implant device obtains electrical energy and starts to operate, it can detect and collect target nerve signals through electrodes; and then use a nerve signal processing chip to specifically process the collected nerve signals through a corresponding nerve signal processing circuit to obtain multiple nerve signal information that meets the requirements.

[0084] In some embodiments, referring to Figure 3 as shown, the above-mentioned obtaining of the target nerve signal, in specific implementation, may include the following contents:

[0085] S1: Collect the current electrical signal in the target nerve area;

[0086] S2: Perform spike detection on the current electrical signal to obtain the corresponding spike detection result;

[0087] S3: According to the spike detection result, judge whether the current electrical signal belongs to a valid nerve signal;

[0088] S4: When it is determined that the current electrical signal belongs to a valid nerve signal, collect the nerve signal in the target nerve area as the target nerve signal.

[0089] Based on the above embodiments, valid nerve signals can be accurately identified, avoiding errors caused by incorrectly collecting and processing invalid nerve signals or environmental noise.

[0090] In some embodiments, the above-mentioned judging whether the current electrical signal belongs to a valid nerve signal according to the spike detection result, in specific implementation, may include: according to the spike detection result, detecting whether the voltage of the current electrical signal is greater than or equal to a preset spike threshold; when it is determined that the current electrical signal is greater than or equal to the preset spike threshold, it can be judged that the current electrical signal belongs to a valid nerve signal; on the contrary, when it is determined that the current electrical signal is less than the preset spike threshold, it can be judged that the current electrical signal belongs to an invalid nerve signal or environmental noise.

[0091] Among them, the preset spike threshold can be specifically determined in the following manner: A large number of historical valid neural electrical signals in the target area or an approximate area similar to the target area are collected as valid samples; then, the neural electrical signals in the above valid samples are statistically analyzed and sorted to obtain the corresponding preset spike threshold.

[0092] In some embodiments, referring to Figure 4 as shown, the above-mentioned spike classification process is performed on the target digital signal according to the preset spike classification algorithm to obtain the corresponding spike classification result, including: The following method is used to perform spike classification on the current digital signal in the target digital signal according to the preset spike classification algorithm:

[0093] S1: Detect and intercept the current spike signal from the current digital signal;

[0094] S2: According to the preset spike classification algorithm, calculate the characteristic distance between the current spike signal and the cluster centers of each preset spike cluster; among them, one preset spike cluster corresponds to one type of neural signal;

[0095] S3: Detect whether there is a cluster center of the target spike cluster in the preset spike cluster whose characteristic distance from the current spike signal is less than or equal to the preset distance threshold;

[0096] S4: When it is determined that there is a cluster center of the target spike cluster in the preset spike cluster whose characteristic distance from the current spike signal is less than or equal to the preset distance threshold, determine the neural signal type corresponding to the target spike cluster as the spike classification result of the current digital signal.

[0097] In the above manner, the spike classification process of the current spike signal in the current digital signal can be completed. Furthermore, the above manner can be repeated to sequentially complete the spike classification process for the next spike signal and the next digital signal to obtain the corresponding spike classification result. In addition, during the spike classification process, corresponding sorting processing can also be performed according to the corresponding sorting rules to obtain the sorted spike classification result.

[0098] Based on the above embodiments, with a relatively small amount of data processing, effective spike signals can be efficiently and accurately identified from the target digital signal, and the neural signal type corresponding to the spike signal can be determined to complete the spike classification process of the target digital signal.

[0099] Among them, the above-mentioned nerve signal type can specifically be the neuron cell type that emits nerve signals. Specifically, the neuron cell types include: multipolar neurons, bipolar neurons, pseudounipolar neurons, and so on. Of course, it should be noted that the above-listed nerve signal types are only illustrative. In specific implementation, the above-mentioned nerve signal type can also include other types. This specification does not limit this.

[0100] The above-mentioned characteristic distance can specifically be the Euclidean distance.

[0101] In specific implementation, before detecting and intercepting the current spike signal from the current digital signal, the digital signal can also be filtered first.

[0102] In some embodiments, the above-mentioned detecting and intercepting the current spike signal from the current digital signal can specifically include the following when implemented:

[0103] S1: Detect the current digital signal and determine the position points of the peak potentials;

[0104] S2: According to the position points of the peak potentials, intercept the signals of a preset number of sampling points from the current digital signal as the current spike signal.

[0105] Based on the above embodiments, it is not necessary to obtain and use all the signals in the current digital signal for spike classification processing. Only the spike signals with good representativeness in the current digital signal need to be intercepted and utilized separately to participate in the specific spike classification processing, so that while ensuring the reliability of the spike classification result, the data processing amount can be effectively reduced, and the requirement for the data processing ability of the implanted device can be lowered.

[0106] In some embodiments, after determining the nerve signal type corresponding to the target spike clustering as the spike classification result of the current digital signal, when the method is specifically implemented, it can further include: using the current spike signal to update the clustering center of the target spike clustering.

[0107] When specifically implemented, when it is determined that the current spike signal belongs to the target spike clustering, the spike features of the current spike signal can be extracted; and then using the spike features, the average waveform of the target spike clustering can be updated to realize the update of the clustering center of the target spike clustering.

[0108] Furthermore, after updating the clustering center of the target spike clustering, it can be detected whether there is a characteristic distance between the updated target spike clustering and other preset spike clusterings except the target spike clustering that is less than the preset merging distance threshold; in the case of determination, the preset spike clustering with a characteristic distance less than the preset merging distance threshold from the updated target spike clustering can be merged with the updated target spike clustering.

[0109] In some embodiments, after detecting that the characteristic distance between the cluster center of the target spike cluster and the current spike signal in the preset spike clusters is less than or equal to a preset distance threshold, when the method is specifically implemented, it may further include:

[0110] In the case where it is determined that there is no characteristic distance between the cluster center of the target spike cluster and the current spike signal in the preset spike clusters that is less than or equal to the preset distance threshold, a new preset spike cluster is created; and according to the current spike signal, the cluster center of the preset spike cluster is determined.

[0111] Based on the above embodiments, during the process of spike classification processing, the cluster centers of the currently existing preset spike clusters can be continuously updated and optimized, so as to perform spike classification processing more accurately based on the updated and optimized cluster centers of the preset spike clusters.

[0112] In some embodiments, referring to Figure 5 as shown, the preset spike clusters can be specifically determined in the following manner:

[0113] S1: Obtain the test electrical signal regarding the target nerve region;

[0114] S2: Obtain the corresponding test digital signal according to the test electrical signal;

[0115] S3: Intercept multiple test spike signals from the test digital signal;

[0116] S4: Process the test spike signals to extract a corresponding plurality of test spike features; wherein, the data dimension of the test spike features is smaller than the data dimension of the test spike signals;

[0117] S5: Determine a plurality of preset spike clusters and the cluster centers of the plurality of preset spike clusters by performing clustering processing on the plurality of test spike features.

[0118] Based on the above embodiments, a plurality of preset spike clusters with high reliability and good effects can be determined and constructed in advance, and then based on the above plurality of preset spike clusters, spike classification processing can be more accurately realized.

[0119] Among them, the above test electrical signal can specifically be the nerve electrical signal collected by the first electrode and the second electrode in the target nerve region during the test stage before specific implementation.

[0120] In some cases, if it is not possible to directly test the target nerve region, other nerve regions similar to the target nerve region can be found as the test nerve region, and then the test nerve region is tested, and the corresponding nerve electrical signals are collected through the first electrode and the second electrode as the above-mentioned test electrical signals.

[0121] When specifically intercepting the test spike signal, the effective signal segment can be first divided from the digital signal according to the preset spike threshold; then the position points of the peak potentials in the effective signal segment are determined, and based on the position points of the peak potentials, a segment signal containing 100 sampling points is intercepted from the effective signal segment as the test spike signal.

[0122] When specifically processing the test spike signal to extract the test spike features, the features that can best represent the segment characteristics in the segment signal of the test spike signal can be further screened out from the spike signal as the test spike features. Specifically, for example, the feature extraction network pre-trained can be used to process the test spike signal to obtain the test spike features. By extracting the test spike features through the above extraction method, the originally 100-dimensional data can be reduced to 3 dimensions, realizing the dimensionality reduction processing of the signal data and effectively reducing the relevant data processing volume.

[0123] Among them, the above-mentioned feature extraction network can specifically be a neural network structure that is pre-trained with a large number of sample spike signals and can automatically extract the spike features that can effectively characterize the spike waveform characteristics in the input signal segment.

[0124] Finally, the above-mentioned spike features are used to perform clustering processing with a small amount of data processing to obtain multiple preset spike clusters. Specifically, for example, reference can be made to Figure 6 As shown, by performing clustering processing on the 3-dimensional spike features (the three components corresponding to the horizontal axis, vertical axis, and vertical axis), 5 different preset spike clusters can be obtained, namely: spike cluster 1, spike cluster 2, spike cluster 3, spike cluster 4, spike cluster 5.

[0125] Specifically, the Spike Sorting algorithm can be specifically modified based on the waveform characteristics of the digital signal obtained from the nerve signal. For example, a preset spike classification algorithm can be obtained. Based on the preset spike classification algorithm obtained after the modification, on the one hand, when extracting the spike features from the spike signal, dimensionality reduction processing is specifically performed on the premise of retaining the key features, thereby effectively reducing the data processing volume of the spike classification algorithm module; on the other hand, only the cluster centers of the spike clusters are stored during the processing process, rather than the complete spike data, thereby effectively reducing the data storage volume of the spike classification algorithm module. By configuring and using the above-mentioned preset spike classification algorithm, it can help reduce the working energy consumption when the spike classification algorithm module runs.

[0126] In some embodiments, the neural signal information may specifically include the peak time of the neural signal.

[0127] Based on the above embodiments, in combination with specific application scenarios, the peak time of the neural signal can be specifically selected as the required key information for acquisition, so as to obtain the neural signal information that meets the scenario requirements.

[0128] In some embodiments, after obtaining multiple pieces of neural signal information, when the method is specifically implemented, the following may further be included: according to different types of neural signals, adding the multiple pieces of neural signal information to the echo signal in a matching manner respectively; and externally transmitting the echo signal.

[0129] Based on the above embodiments, the implant device can add the required multiple pieces of neural signal information to the echo signal, and then transmit the echo signal carrying the multiple pieces of neural signal information through the ultrasonic transducer. The external device can receive the above echo signal through the ultrasonic probe, and then process the echo signal through the data acquisition card to demodulate the above multiple pieces of neural signal information.

[0130] In some embodiments, during the process of spike classification processing, based on a preset spike classification algorithm, corresponding sorting processing may further be performed according to the corresponding sorting rule to obtain the sorted spike classification result; then, based on the sorted spike classification result, different types of neural signals are distinguished according to the corresponding sorting, and the corresponding multiple pieces of neural signal information are added to the echo signal in sequence in a matching manner. At the same time, the above sorting rule may also be added to the echo signal. Then, the echo signal is externally transmitted.

[0131] Correspondingly, the external device can, according to the corresponding sorting rule through a computer device, finely identify and distinguish the neural signal information corresponding to different types of neurons by processing the multiple pieces of demodulated neural signal information. Further, the external device can also analyze and determine the mapping relationship between the activities of different types of neurons and the specific behavioral manifestations of the target object according to the above neural signal information corresponding to different types of neurons, so as to finely study the behavioral rules of the target object.

[0132] In some embodiments, the implant device transmits an echo signal carrying multiple neural signal information to an external device through an ultrasonic transducer. Specifically, the following steps may be included: The implant device adds the multiple neural signal information to the echo signal in a matching manner based on different amplitude information and / or pulse width information according to the neural signal information, thereby obtaining an echo signal carrying multiple neural signal information; The implant device externally transmits the echo signal through the ultrasonic transducer; Correspondingly, the external device receives the echo signal through an ultrasonic probe.

[0133] Based on the above embodiments, different neural signal types can be distinguished, and different neural signal information can be added to the echo signal in a matching manner; Then, the echo signal carrying multiple neural signal information is transmitted to the external device. In this way, the external device can extract multiple neural signal information corresponding to different neural signal types by detecting and demodulating the received echo signal.

[0134] Specifically, for example, in this implant device, an adjustment switch is also arranged in parallel with the ultrasonic transducer. Specifically, the implant device can distinguish different neural signal types according to the neural signal information, and adjust the different pulse width information of different neural signal information by controlling the closing time of the adjustment switch, so as to be able to add multiple different neural signal information corresponding to different neural signal types to the echo signal in a matching manner (for example, based on the PWM adjustment method). Correspondingly, the external device can identify multiple neural signal information corresponding to different neural signal types by distinguishing different pulse width information.

[0135] Specifically, for example, multiple different adjustment circuits can also be arranged in this implant device; Among them, each adjustment circuit includes an adjustment switch and / or a combination of an adjustment switch and a corresponding impedance resistor. Correspondingly, specifically, the implant device can also distinguish different neural signal types according to the neural signal information, and adjust the amplitude information of different neural signal information corresponding to different neural signal types by controlling the conduction of different adjustment circuits, so as to be able to add multiple different neural signal information corresponding to different neural signal types to the echo signal in a matching manner (for example, based on the ASK adjustment method). Correspondingly, the external device can identify multiple neural signal information corresponding to different neural signal types by distinguishing different amplitude information.

[0136] In some embodiments, specifically, the external device can first use the ultrasonic probe to perform ultrasonic positioning on the implant device inside the target object to determine the position information of the implant device; Then, according to the position information of the implant device, the external device controls the ultrasonic probe to emit ultrasonic signals directionally to the implant device.

[0137] The implant device can use an ultrasonic transducer to convert the received ultrasonic signal into electrical energy for the operation of the implant device; and control the implant device to enter the first mode state; wherein, the neural signal processing circuit is in a standby state in the first mode state, and at this time, the neural signal processing circuit does not execute the real-time data processing method of neural signals provided in this specification.

[0138] In the first mode state, the implant device uses a spike detector to perform spike detection on the electrical signals collected by the first electrode and the second electrode to determine whether the electrical signals collected by the first electrode and the second electrode are valid neural signals; in the case where it is determined that the electrical signals collected by the first electrode and the second electrode are neural signals, it triggers to enter the second mode state; wherein, the neural signal processing circuit is in a working state in the second mode state, and at this time, the neural signal processing circuit starts to execute the real-time data processing method of neural signals provided in this specification.

[0139] In the second mode state, the implant device uses the neural signal processing circuit to perform corresponding data processing on the neural signals collected by the first electrode and the second electrode by executing the real-time data processing method of neural signals to obtain multiple neural signal information that meets the requirements; and then uses the ultrasonic transducer to transmit the echo signal carrying the multiple neural signal information to an external device.

[0140] The external device uses an ultrasonic probe to receive the echo signal, and extracts multiple neural signal information from the echo signal through a data acquisition card; furthermore, a computer device can be used to further process the above multiple neural signal information.

[0141] In some embodiments, after the implant device receives the ultrasonic signal transmitted by the external device through the ultrasonic transducer, it can use the ultrasonic transducer to convert the received ultrasonic signal into electrical energy for the operation of the implant device.

[0142] In the first mode state, in order to reduce the working energy consumption of the implant device, the signal processing circuit is in a standby state and will not operate. However, there are still some components in the implant device (for example, the spike detector) in a working state. The implant device can detect whether the electrical signals collected based on the first electrode and the second electrode are valid neural signals through the above-mentioned components in the working state. When it is detected that the collected electrical signals are valid neural signals, the implant device will trigger to enter the second mode state.

[0143] In the second mode state, the signal processing circuit will enter a working state. Correspondingly, the implant device can use the signal processing circuit to convert the collected electrical signals into corresponding multiple neural signal information.

[0144] In some embodiments, during specific implementation, when the external device performs ultrasonic positioning on the implanted device inside the target object through the ultrasonic probe, a first indication signal for characterizing the positioning can be added to the emitted ultrasonic signal. Correspondingly, when the implanted device receives the above ultrasonic signal through the ultrasonic transducer and detects the above first indication signal, it can determine that it is currently in the positioning stage, and at this time, it may not enter the first mode state.

[0145] When the external device directionally emits an ultrasonic signal to the implanted device through the ultrasonic probe, a second indication signal for characterizing the user's work can be added to the emitted ultrasonic signal. Correspondingly, when the implanted device receives the above ultrasonic signal through the ultrasonic transducer and detects the above second indication signal, it can determine that it is currently in the working stage, and only then will it enter the first mode state.

[0146] In addition, when the implanted device receives an ultrasonic signal emitted by other devices except the external device through the ultrasonic transducer, since the first indication signal and the second indication signal are not detected, it can not make a response. In this way, it can effectively avoid the situation that the implanted device disposed inside the target object is accidentally triggered due to receiving the ultrasonic wave emitted by other devices.

[0147] In some embodiments, the neural signal processing chip further includes a stimulation function circuit. Specifically, the above stimulation function circuit and the neural signal processing circuit are arranged in parallel in the neural signal processing chip. Among them, the stimulation function circuit is connected to the target neural region.

[0148] During specific implementation, the implanted device can apply a stimulation signal (an electrical signal) to the target neural region through the stimulation function circuit to actively stimulate the neuron cells in the target neural region to generate neural signals.

[0149] At the same time, the implanted device can also detect whether there are effective neural signals in the target neural region through the spike detector. When effective neural signals appear in the target neural region, the neural signals are collected through the first electrode and the second electrode; then corresponding data processing is performed through the neural signal processing circuit to obtain corresponding multiple neural signal information; and then the echo signal carrying the multiple neural signal information is provided to the external device for further processing.

[0150] Specifically, the external device can also add a third indication signal for characterizing the stimulation instruction to the ultrasonic signal directionally emitted to the implanted device. Correspondingly, when the implanted device detects the existence of the third indication signal in the ultrasonic signal, it triggers the application of a stimulation signal to the target neural region through the stimulation function circuit.

[0151] In this way, the above-mentioned implantable neural signal acquisition system can be used to simultaneously apply stimulation and acquire neural signals, thereby meeting the requirements of relatively more complex and diverse scenarios.

[0152] In some embodiments, the external device performs ultrasonic positioning on the implant device inside the target object through an ultrasonic probe to determine the position information of the implant device; and according to the position information, emits an ultrasonic signal to the implant device. The specific implementation may include: the external device controls the ultrasonic probe to perform ultrasonic scanning on the target object through a data acquisition card to generate a target ultrasonic image of the target object; the external device determines the position information of the implant device through a computer device according to the target ultrasonic image; the external device determines a matching target emission angle through the computer device according to the position information of the implant device; and the external device controls the ultrasonic probe to emit an ultrasonic signal to the implant device according to the target emission angle through the data acquisition card.

[0153] Based on the above embodiments, the external device can aim at the implant device according to the target emission angle and accurately emit an ultrasonic signal to the implant device in a directional manner to better supply energy to the implant device.

[0154] In some embodiments, the above external device controls the ultrasonic probe to perform ultrasonic scanning on the target object through a data acquisition card to generate a target ultrasonic image of the target object. The specific implementation may include the following steps:

[0155] S1: Determine multiple focal points for the target object and the focal point coordinates;

[0156] S2: Control the ultrasonic probe to sequentially emit focused ultrasound to each of the multiple focal points and collect the corresponding ultrasonic signals;

[0157] S3: Perform ultrasonic imaging processing on the corresponding ultrasonic signals to obtain a qualified target ultrasonic image of the target object.

[0158] When specifically executing step S1, considering that the element spacing in the ultrasonic phased array transducer probe specifically used by the external device is 0.5 mm, the focal length can be set to 10 cm, 121 focal points can be set, and the focal point angle step size is 0.5 degrees; correspondingly, the ultrasonic scanning angle can reach 60 degrees.

[0159] When specifically implementing step S2, the respective vibration elements in the ultrasonic probe can be controlled according to the focal coordinates to send ultrasonic signals with two cycles (the frequency can be 5 MHz) to each focus, serving as focused ultrasound. In this way, the signal-to-noise ratio of the subsequently collected ultrasonic echo signals can be improved. Then, for each focus, after transmitting the ultrasonic signal, the respective vibration elements in the ultrasonic probe can be controlled to collect the corresponding ultrasonic echo signals. By combining the ultrasonic echo signals collected for each focus, an 8-channel signal can be obtained as the corresponding ultrasonic signal.

[0160] When specifically implementing step S3, the following steps can be included: S3-1, performing beamforming on the ultrasonic signal; S3-2, performing time gain control processing on the initial brightness matrix; S3-3, extracting the signal envelope from the first processed signal data; S3-4, performing downsampling processing on the second processed signal data; S3-5, obtaining the required target ultrasonic image through image interpolation based on the third processed signal data.

[0161] Among them, when specifically performing S3-1, the coordinates of each pixel point between the two-point connection can be determined first according to the focal coordinates and the coordinates of the center point of the ultrasonic probe; according to the distance between the pixel point and the vibration element in the ultrasonic probe and the ultrasonic beam, the sampling time of each vibration element for each pixel point can be calculated; the phases of the respective vibration elements in the ultrasonic probe can be aligned according to the sampling time; and then, based on the vibration elements after phase alignment, the 8-channel signal can be synthesized to obtain the initial brightness matrix for ultrasonic imaging.

[0162] When specifically performing S3-2, the trend of energy attenuation during signal propagation can be eliminated for the initial brightness matrix, and only the echo intensity generated by the acoustic impedance can be retained to obtain the first processed signal data.

[0163] When specifically performing S3-3, the ultrasonic signal with a central frequency of 5 MHz in the first processed signal data can be demodulated, and the echo intensity of the scatterer can be reflected using the signal envelope information to obtain the second processed signal data.

[0164] When specifically performing S3-4, according to the demodulation frequency, the signal capacity of the second processed signal data can be reduced to reduce the burden on the hardware caused by large-capacity data transmission, obtaining the third processed signal data (i.e., the downsampled brightness matrix).

[0165] When specifically performing S3-5, according to the third processed signal data, through image interpolation processing, an ultrasonic sector image can be obtained as the target ultrasonic image.

[0166] In some embodiments, after obtaining the target ultrasound image, the external device may perform image recognition on the target ultrasound image through a computer device to identify the implanted device in the image; and then the position information of the implanted device based on the target object can be determined.

[0167] Furthermore, the external device may obtain the position information of the ultrasound probe based on the target object; then, according to the position information of the ultrasound probe based on the target object and the position information of the implanted device based on the target object, a matching target emission angle is determined; and then the data acquisition card is used to control the ultrasound probe to aim at the implanted device according to the target emission angle and emit ultrasonic signals directionally.

[0168] In some embodiments, when the external device is specifically implemented, it may further include the following:

[0169] S1: Monitor the intensity change data of the received echo signal; and detect whether the position information of the implanted device has changed according to the intensity change data of the echo signal.

[0170] S2: When the external device detects that the position information of the implanted device has changed, according to the intensity change data of the echo signal, adjust the target emission angle to obtain the adjusted target emission angle.

[0171] S3: The external device controls the ultrasound probe to emit ultrasonic signals to the implanted device according to the adjusted target emission angle through the data acquisition card.

[0172] Specifically, considering that during the process of directionally emitting ultrasonic signals to the implanted device, the implanted device disposed inside the target object may displace, resulting in the inability to continuously and accurately aim at the implanted device to emit ultrasonic signals.

[0173] Therefore, during the process of emitting ultrasonic signals to the implanted device, the external device can also detect in real time through the data acquisition card and the ultrasound probe whether the intensity change data of the received echo signal returned by the implanted device is greater than a preset change data threshold. If it is detected that the intensity change data of the echo signal is less than or equal to the preset change data threshold, it can be determined that the position information of the current implanted device has not changed, or has only changed relatively slightly, and will not have an obvious impact on the directional emission effect of the ultrasonic wave signal, and there is no need to adjust the emission angle. If it is detected that the intensity change data of the echo signal is greater than the preset change data threshold, it can be determined that the position information of the current implanted device has changed, and this change is relatively large, which will have an obvious impact on the directional emission effect of the ultrasonic wave signal, and the emission angle needs to be adjusted.

[0174] After determining that the emission angle needs to be adjusted, the position information change data of the implanted device and the change trend of the position information can be deduced based on the intensity change data of the echo signal within a current time period. Furthermore, based on the position information change data of the implanted device and the change trend of the position information, the target emission angle can be adjusted specifically to obtain a matching adjusted target emission angle. Then, based on the adjusted target emission angle, the implanted device can be continuously targeted and ultrasonic signals can be accurately emitted to the implanted device.

[0175] In this way, the external device can track the actual position information of the implanted device in real time and adjust the emission angle in a timely manner according to the change of the position information of the implanted device, so as to continuously target the implanted device and emit ultrasonic signals directionally.

[0176] In some embodiments, when the external device emits ultrasonic signals and / or receives echo signals through the ultrasonic probe, one or more of the following listed operation modes can be further used to improve the signal accuracy during emission and / or reception and reduce the errors introduced by factors such as side lobes: performing beam steering, beam focusing, and beam synthesis operations on the emitted ultrasonic signals and / or the received echo signals; performing segmented dynamic apodization operations on the received echo signals; performing dynamic aperture adjustment operations when receiving echo signals, etc.

[0177] In some embodiments, when the external device is specifically implemented, the data acquisition card can also be used to control the ultrasonic probe to receive the echo signals emitted by the implanted device; and extract multiple nerve signal information from the echo signals;

[0178] Furthermore, the external device can also perform corresponding data processing on the multiple nerve signal information through a computer device.

[0179] Specifically, the external device can control the ultrasonic probe to receive the echo signals emitted by the implanted device through the data acquisition card; and perform corresponding detection and demodulation processing (such as digital quadrature demodulation) on the echo signals through the data acquisition card to extract the required multiple nerve signal information.

[0180] Furthermore, the external device can use a corresponding algorithm model to process the multiple nerve signal information through a computer device to analyze and determine the mapping relationship between the behavior of the target object and the nerve signals. Then, subsequently, based on this mapping relationship, the behavior of the target object can be predicted using the collected nerve signal information.

[0181] As can be seen from the above, in the real-time data processing method of neural signals provided by the embodiments of this specification, after continuously collecting neural signals of a target neural region as target neural signals, the target neural signals can be first converted into corresponding discrete target digital signals; then, spike classification processing can be performed on the target digital signals according to a preset spike classification algorithm to obtain corresponding spike classification results; and according to the spike classification results, multiple neural signal information corresponding to different types of neurons can be obtained. In this way, on the one hand, it can effectively reduce the data volume of the finally obtained neural signal data to be transmitted, reduce the data transmission burden when transmitting neural signal data subsequently, and improve the transmission quality of neural signal data during subsequent transmission; on the other hand, it can also finely separate different types of neural signals from continuous neural signals to obtain multiple neural signal information corresponding to different types of neurons, meeting complex and diverse scenario requirements.

[0182] Moreover, when converting the target neural signals from the original analog signals into corresponding digital signals, by combining specific scenario requirements, only the peak times of the action potentials of different types of neurons required are extracted as the multiple neural signal information, thereby further reducing the data transmission burden when transmitting neural signal data subsequently.

[0183] In addition, spike detection is performed on the electrical signals in the target neural region collected in real time to identify effective neural signals; and then only the effective neural signals are collected as target neural signals for subsequent data processing, thereby avoiding wasting processing resources on invalid neural signals or noise signals, improving the utilization rate of processing resources, and reducing the overall working energy consumption of the implanted device.

[0184] The embodiments of this specification also provide an electronic device. Refer to Figure 7 As shown, it includes at least a network communication port 701, a processor 702, and a memory 703. The above structures are connected by internal cables so that each structure can perform specific data interactions.

[0185] Among them, the network communication port 701 can specifically be used to obtain target neural signals.

[0186] The processor 702 can specifically be used to perform analog-to-digital conversion processing on the target neural signals to obtain corresponding target digital signals; perform spike classification processing on the target digital signals according to a preset spike classification algorithm to obtain corresponding spike classification results; obtain multiple neural signal information according to the spike classification results; where one neural signal information corresponds to one type of neural signal; the multiple neural signal information includes the peak times of the action potentials of different neurons.

[0187] The memory 703 can be specifically used to store corresponding instruction programs.

[0188] In this embodiment, the network communication port 701 can be bound to different communication protocols, so as to send or receive different data. For example, the network communication port can be a port responsible for web data communication, can also be a port responsible for FTP data communication, and can also be a port responsible for mail data communication. In addition, the network communication port can also be a physical communication interface or communication chip. For example, it can be a wireless mobile network communication chip, such as GSM, CDMA, etc.; it can also be a Wifi chip; it can also be a Bluetooth chip.

[0189] In this embodiment, the processor 702 can be implemented in any suitable manner. For example, the processor can take the form of, for example, a microprocessor or a processor and a computer-readable medium storing computer-readable program code (such as software or firmware) executable by the (micro)processor, logic gates, switches, application specific integrated circuit (ASIC), programmable logic controller, and embedded microcontroller, etc. This specification does not make a limitation.

[0190] In this embodiment, the memory 703 can include multiple levels. In a digital system, anything that can store binary data can be a memory; in an integrated circuit, a circuit with a storage function without a physical form is also called a memory, such as RAM, FIFO, etc.; in a system, a storage device with a physical form is also called a memory, such as a memory stick, TF card, etc.

[0191] An embodiment of this specification also provides a computer storage medium based on the above neural signal acquisition method. The computer storage medium stores computer program instructions, and when the computer program instructions are executed, the following steps are implemented: obtaining a target neural signal; performing analog-to-digital conversion processing on the target neural signal to obtain a corresponding target digital signal; performing spike classification processing on the target digital signal according to a preset spike classification algorithm to obtain a corresponding spike classification result; obtaining multiple neural signal information according to the spike classification result; where one neural signal information corresponds to one type of neural signal; the multiple neural signal information includes the peak times of action potentials of different neurons.

[0192] In this embodiment, the storage medium includes, but is not limited to, a Random Access Memory (RAM), a Read-Only Memory (ROM), a Cache, a Hard Disk Drive (HDD), or a Memory Card. The memory can be used to store computer program instructions. The network communication unit can be set according to the standards specified by the communication protocol and is used as an interface for network connection communication.

[0193] In this embodiment, the functions and effects specifically implemented by the program instructions stored in the computer storage medium can be explained by comparison with other embodiments and will not be elaborated here.

[0194] Refer to Figure 8 As shown, at the software level, the embodiments of this specification further provide a real-time data processing device for nerve signals. The device may specifically include the following structural modules:

[0195] The first acquisition module 801 can specifically be used to acquire target nerve signals;

[0196] The analog-to-digital conversion module 802 can specifically be used to perform analog-to-digital conversion processing on the target nerve signals to obtain corresponding target digital signals;

[0197] The spike sorting module 803 can specifically be used to perform spike sorting processing on the target digital signals according to a preset spike sorting algorithm to obtain corresponding spike sorting results;

[0198] The second acquisition module 804 can specifically be used to obtain a plurality of nerve signal information according to the spike sorting results; wherein, one nerve signal information corresponds to one type of nerve signal; the plurality of nerve signal information includes the peak times of action potentials of different neurons.

[0199] In some embodiments, when the first acquisition module 801 is specifically implemented, the target nerve signals can be acquired in the following manner: collecting the current electrical signals in the target nerve area; performing spike detection on the current electrical signals to obtain corresponding spike detection results; judging whether the current electrical signals belong to valid nerve signals according to the spike detection results; and when it is determined that the current electrical signals belong to valid nerve signals, collecting the nerve signals in the target nerve area as the target nerve signals.

[0200] In some embodiments, when the spike classification module 803 is specifically implemented, the target digital signal may be subjected to spike classification processing according to a preset spike classification algorithm in the following manner to obtain a corresponding spike classification result: The current digital signal in the target digital signal is subjected to spike classification processing according to the preset spike classification algorithm in the following manner: Detect and intercept the current spike signal from the current digital signal; Calculate the feature distance between the current spike signal and the cluster centers of each preset spike cluster according to the preset spike classification algorithm; wherein, a preset spike cluster corresponds to a type of neural signal; Detect whether there is a cluster center of a target spike cluster in the preset spike clusters whose feature distance from the current spike signal is less than or equal to a preset distance threshold; When it is determined that there is a cluster center of a target spike cluster in the preset spike clusters whose feature distance from the current spike signal is less than or equal to the preset distance threshold, determine the type of neural signal corresponding to the target spike cluster as the spike classification result of the current digital signal.

[0201] In some embodiments, when the spike classification module 803 is specifically implemented, the current spike signal may be detected and intercepted from the current digital signal in the following manner: Detect the current digital signal to determine the position points of the peak potentials; According to the position points of the peak potentials, intercept the signals of a preset number of sampling points from the current digital signal as the current spike signal.

[0202] In some embodiments, after determining the type of neural signal corresponding to the target spike cluster as the spike classification result of the current digital signal, when the device is specifically implemented, it may further be used to: Update the cluster center of the target spike cluster by using the current spike signal.

[0203] In some embodiments, after detecting whether there is a cluster center of a target spike cluster in the preset spike clusters whose feature distance from the current spike signal is less than or equal to the preset distance threshold, when the device is specifically implemented, it may further be used to: When it is determined that there is no cluster center of a target spike cluster in the preset spike clusters whose feature distance from the current spike signal is less than or equal to the preset distance threshold, create a new preset spike cluster; And determine the cluster center of the preset spike cluster according to the current spike signal.

[0204] In some embodiments, before the device is specifically implemented, the preset spike clusters can be determined in the following manner: obtaining test electrical signals regarding a target nerve region; obtaining corresponding test digital signals according to the test electrical signals; intercepting a plurality of test spike signals from the test digital signals; processing the test spike signals to extract a corresponding plurality of test spike features, wherein the data dimension of the test spike features is smaller than the data dimension of the test spike signals; and determining a plurality of preset spike clusters and the cluster centers of the plurality of preset spike clusters by performing clustering processing on the plurality of test spike features.

[0205] In some embodiments, after obtaining a plurality of nerve signal information, when the device is specifically implemented, it can also be used to: add the plurality of nerve signal information to echo signals in a matching manner according to different nerve signal types; and externally transmit the echo signals.

[0206] It should be noted that the units, devices, or modules described in the above embodiments can be specifically implemented by computer chips or entities, or by products with certain functions. For the convenience of description, the above devices are described by dividing them into various modules according to functions. Of course, when implementing this specification, the functions of each module can be implemented in the same or multiple software and / or hardware, or the modules implementing the same function can be realized by a combination of multiple sub-modules or sub-units. The device embodiments described above are only illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be in electrical, mechanical or other forms.

[0207] As can be seen from the above, based on the real-time data processing device for nerve signals provided in the embodiments of this specification, after collecting continuous nerve signals regarding a target nerve region as target nerve signals, the above target nerve signals can be first converted into corresponding discrete target digital signals; then, spike classification processing is performed on the target digital signals according to a preset spike classification algorithm to obtain corresponding spike classification results; and according to the spike classification results, a plurality of nerve signal information corresponding to different types of neurons are obtained. In this way, on the one hand, it can effectively reduce the data volume of the finally obtained nerve signal data to be transmitted, reduce the data transmission burden when transmitting nerve signal data subsequently, and improve the transmission quality when transmitting nerve signal data subsequently; on the other hand, it can also finely separate different types of nerve signals from continuous nerve signals to obtain a plurality of nerve signal information corresponding to different types of neurons, meeting the complex and diverse scenario requirements.

[0208] Although this specification provides method operation steps as described in the embodiments or flowcharts, more or fewer operation steps may be included based on conventional or non-creative means. The order of steps listed in the embodiments is only one way among many execution orders of steps and does not represent the only execution order. When the actual device or client product is executed, it may be executed in the order of the method shown in the embodiments or the drawings or executed in parallel (for example, in an environment of parallel processors or multi-threaded processing, or even in a distributed data processing environment). The terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, such that a process, method, product or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements that are inherent to such process, method, product or device. Without further limitation, there is no exclusion of additional identical or equivalent elements in the process, method, product or device comprising the said elements. The terms such as first, second, etc. are used to denote names and do not denote any particular order.

[0209] Those skilled in the art also know that, in addition to implementing the controller in the form of pure computer-readable program code, the method steps can be logically programmed to enable the controller to implement the same functions in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, embedded microcontrollers, etc. Therefore, such a controller can be regarded as a hardware component, and the devices included therein for implementing various functions can also be regarded as the structures within the hardware component. Or even, the devices for implementing various functions can be regarded as either software modules for implementing the method or the structures within the hardware component.

[0210] This specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, classes, etc. that perform specific tasks or implement specific abstract data types. This specification may also be practiced in a distributed computing environment where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media including storage devices.

[0211] As can be seen from the description of the above embodiments, those skilled in the art can clearly understand that this specification can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of this specification can essentially be embodied in the form of a software product, which can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions for causing a computer device (which can be a personal computer, a mobile terminal, a server, or a network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of this specification.

[0212] The various embodiments in this specification are described in a progressive manner. For the same or similar parts between the various embodiments, reference can be made to each other. The key point of each embodiment is to illustrate the differences from other embodiments. This specification can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multi-processor systems, microprocessor-based systems, set-top boxes, programmable electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and so on.

[0213] Although this specification is depicted through embodiments, those of ordinary skill in the art know that this specification has many variations and changes without departing from the spirit of this specification. It is hoped that the appended claims will cover these variations and changes without departing from the spirit of this specification.

Claims

1. A real-time data processing method for neural signals, characterized in that, Including: Obtain a target nerve signal; Perform analog-to-digital conversion processing on the target nerve signal to obtain a corresponding target digital signal; Perform spike classification processing on the target digital signal according to a preset spike classification algorithm to obtain a corresponding spike classification result; According to the spike classification result, obtain multiple nerve signal information; wherein, one nerve signal information corresponds to one type of nerve signal; the multiple nerve signal information includes the peak times of action potentials of different neurons.

2. The method according to claim 1, wherein Obtaining a target nerve signal includes: Collect the current electrical signal in the target nerve area; Perform spike detection on the current electrical signal to obtain a corresponding spike detection result; According to the spike detection result, determine whether the current electrical signal belongs to a valid nerve signal; When it is determined that the current electrical signal belongs to a valid nerve signal, collect the nerve signal in the target nerve area as the target nerve signal.

3. The method according to claim 1, wherein Performing spike classification processing on the target digital signal according to a preset spike classification algorithm to obtain a corresponding spike classification result includes: According to the following method, perform spike classification processing on the current digital signal in the target digital signal according to a preset spike classification algorithm: Detect and intercept the current spike signal from the current digital signal; According to the preset spike classification algorithm, calculate the characteristic distance between the current spike signal and the cluster centers of each preset spike cluster; wherein, one preset spike cluster corresponds to one type of nerve signal; Detect whether there is a characteristic distance between the cluster center of the target spike cluster and the current spike signal in the preset spike cluster that is less than or equal to a preset distance threshold; When it is determined that there is a characteristic distance between the cluster center of the target spike cluster and the current spike signal in the preset spike cluster that is less than or equal to the preset distance threshold, determine the nerve signal type corresponding to the target spike cluster as the spike classification result of the current digital signal.

4. The method according to claim 3, wherein Detecting and intercepting the current spike signal from the current digital signal includes: Detect the current digital signal and determine the position points of the peak potentials; According to the position points of the peak potentials, intercept the signals of a preset number of sampling points from the current digital signal as the current spike signal.

5. The method according to claim 3, wherein After determining the nerve signal type corresponding to the target spike cluster as the spike classification result of the current digital signal, the method further includes: Use the current spike signal to update the cluster center of the target spike cluster.

6. The method according to claim 3, characterized in that, After detecting whether there is a characteristic distance between the cluster center of the target spike cluster and the current spike signal in the preset spike cluster that is less than or equal to the preset distance threshold, the method further includes: When it is determined that there is no characteristic distance between the cluster center of the target spike cluster and the current spike signal in the preset spike cluster that is less than or equal to the preset distance threshold, create a new preset spike cluster; and according to the current spike signal, determine the cluster center of the preset spike cluster.

7. The method according to claim 3, wherein The preset spike cluster is determined according to the following method: Obtain the test electrical signal regarding the target nerve area; According to the test electrical signal, obtain a corresponding test digital signal; Intercept a plurality of test spike signals from the test digital signal; Process the test spike signals to extract a corresponding plurality of test spike features; wherein, the data dimension of the test spike features is smaller than the data dimension of the test spike signals; Determine a plurality of preset spike clusters and the cluster centers of the plurality of preset spike clusters by performing clustering processing on the plurality of test spike features.

8. The method according to claim 1, characterized in that After obtaining a plurality of neural signal information, the method further includes: Add the plurality of neural signal information to the echo signal in a matching manner according to different neural signal types; and externally transmit the echo signal.

9. A real-time data processing device for nerve signals, characterized in that, Including: A first acquisition module for acquiring a target neural signal; An analog-to-digital conversion module for performing analog-to-digital conversion processing on the target neural signal to obtain a corresponding target digital signal; A spike classification module for performing spike classification processing on the target digital signal according to a preset spike classification algorithm to obtain a corresponding spike classification result; A second acquisition module for acquiring a plurality of neural signal information according to the spike classification result; wherein, one neural signal information corresponds to one neural signal type; the plurality of neural signal information includes the peak times of action potentials of different neurons.

10. A computer storage medium, characterized in that, Stored thereon are computer instructions, and when the instructions are executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images