Method and system for reconstructing brain spatial neural signal field based on non-simultaneous local observation

Abstract

The application discloses a brain spatial neural signal field reconstruction method and system based on non-simultaneous local observation, relates to the technical field of neural electrophysiological signal processing, and comprises the following steps: controlling a neural probe with multiple columns of neural signal recording sites to collect neural signals of a first scanning section at a first spatial position; controlling the neural probe to move a preset distance along a main scanning axis to a second spatial position and collect neural signals of a second scanning section; generating a two-dimensional detection event rate field and a two-dimensional amplitude field respectively; calculating an event rate difference and an amplitude difference based on spatially overlapping neural signal recording sites; when the event rate difference and the amplitude difference meet a preset difference condition, performing coordinate splicing and overlapping region fusion to obtain a target expanded spatial neural signal field; otherwise, outputting an unsplicable reconstruction result. The activity scale calibration relationship between two non-simultaneous local spatial fields can be established by using coincident probe sites, and the reliability of the non-simultaneous local observation neural signal field reconstruction is improved.

Description

Technical Field

[0001] This invention relates to the field of neurophysiological signal processing technology, and in particular to a method and system for reconstructing the brain spatial neural signal field based on non-simultaneous local observation. Background Technology

[0002] Implantable brain-computer interfaces typically acquire neurophysiological signals via neural probes for brain function research and neural signal analysis. However, due to limitations in the size, number of channels, and implantable range of these probes, a single fixed-location recording usually only captures neural signals from a localized brain region. Expanding the recording range by increasing the number of probes, their size, or the number of channels may lead to problems such as more complex wiring, increased data throughput, larger implant volume, increased risk of tissue damage, and higher costs.

[0003] To expand the cumulative spatial coverage of limited-channel neural probes, the same probe can be actively scanned and recorded at different spatial locations. However, neural signals from different scan segments are usually acquired at different times, and neural activity states, probe-tissue interfaces, background noise, and detection thresholds may all change, thus preventing direct stitching like image data. Existing active scanning recordings are mostly used for probe navigation, target brain region identification, insertion trajectory evaluation, or brain region boundary detection, and still struggle to address whether non-simultaneous local observations of neural signals can be stitched together and how to reconstruct an extended spatial neural signal field.

[0004] Therefore, how to determine whether the neural signals of different scan segments meet the splicing conditions when the neural probes with limited channels actively scan and obtain non-simultaneous, locally overlapping neural signals, and how to reconstruct the extended spatial neural signal field when the conditions are met, is a technical problem to be solved in this field. Summary of the Invention

[0005] In view of this, the purpose of this invention is to provide a method and system for reconstructing the brain spatial neural signal field based on non-simultaneous local observation. This method can determine whether neural signals from different scan segments meet splicing conditions when non-simultaneous, locally overlapping neural signals are obtained through active scanning with a limited-channel neural probe, and reconstruct the extended spatial neural signal field when the conditions are met. The specific solution is as follows: In a first aspect, this application discloses a method for reconstructing the spatial neural signal field of the brain based on non-simultaneous local observations, including: A neural probe with multiple neural signal recording sites is controlled to acquire neural signals in the first scan segment at a first spatial location in the target brain region; The neural probe is controlled to move a preset distance along the active scanning axis to a second spatial position, and neural signals of a second scanning segment are acquired at the second spatial position; wherein, the preset distance is less than the span of the recording site of the neural probe along the active scanning axis. A first two-dimensional detection event rate field and a first two-dimensional amplitude field are generated based on the neural signals of the first scan segment, and a second two-dimensional detection event rate field and a second two-dimensional amplitude field are generated based on the neural signals of the second scan segment; wherein, the two-dimensional detection event rate field is a data field of detection event rates arranged according to the row and column positions of the neural signal recording sites in the neural probe; Based on the spatial overlap of the neural signals of the first scan segment and the second scan segment, the event rate difference between the first two-dimensional detection event rate field and the second two-dimensional detection event rate field is calculated, and the amplitude difference between the first two-dimensional amplitude field and the second two-dimensional amplitude field is also calculated. When the event rate difference and the amplitude difference meet the preset difference conditions, the first two-dimensional detection event rate field and the second two-dimensional detection event rate field are spliced ​​and the overlapping region is fused along the active scanning axis to obtain the target extended space neural signal field. If the event rate difference and / or the magnitude difference do not meet the preset difference conditions, an unspantable reconstruction result will be output.

[0006] Optionally, the neural probe includes at least two columns of neural signal recording sites, with adjacent columns of neural signal recording sites arranged alternately along the active scanning axis. Accordingly, controlling the neural probe to move a preset distance along the active scanning axis to the second spatial position includes: The preset distance is determined based on the row spacing and span of the recording sites along the active scanning axis of the neural probe; wherein the preset distance is an integer multiple of the row spacing of the recording sites and is less than the span of the recording sites.

[0007] Optionally, controlling a neural probe with multiple neural signal recording sites to acquire neural signals of a first scan segment at a first spatial location in the target brain region includes: The neural probe is controlled to reach the first spatial position, and after maintaining a first preset stable time, neural signals are collected within a first preset recording time to obtain the neural signal of the first scan segment; Accordingly, the neural probe is controlled to move a preset distance along the active scanning axis to a second spatial position, and neural signals of the second scanning segment are acquired at the second spatial position, including: The neural probe is controlled to move a preset distance along the active scanning axis to the second spatial position, and after maintaining a second preset stable duration, neural signals are collected within a second preset recording duration to obtain the neural signals of the second scanning segment.

[0008] Optionally, before generating the first two-dimensional detection event rate field and the first two-dimensional amplitude field based on the neural signals of the first scan segment, and generating the second two-dimensional detection event rate field and the second two-dimensional amplitude field based on the neural signals of the second scan segment, the method further includes: The neural signals of the first scan segment and the neural signals of the second scan segment are respectively subjected to bandpass filtering to obtain a first bandpass signal and a second bandpass signal; A first event detection threshold is determined based on the background noise of the channel corresponding to each neural signal recording site in the first bandpass signal, and a second event detection threshold is determined based on the background noise of the channel corresponding to each neural signal recording site in the second bandpass signal. Based on the first event detection threshold, multiple first neural event detection results are obtained from the first bandpass signal, and based on the second event detection threshold, multiple second neural event detection results are obtained from the second bandpass signal; If the detection times corresponding to multiple first neural event detection results are within the same preset time window, and the corresponding neural signal recording sites are spatially adjacent neural signal recording sites, then the multiple first neural event detection results are merged into a first multi-channel event, and the neural signal recording site corresponding to the first neural event detection result with the largest absolute peak-to-peak amplitude among the multiple first neural event detection results is determined as the first representative recording site of the first multi-channel event. If the detection times corresponding to multiple second neural event detection results are within the same preset time window, and the corresponding neural signal recording sites are spatially adjacent neural signal recording sites, then the multiple second neural event detection results are merged into a second multi-channel event, and the neural signal recording site corresponding to the second neural event detection result with the largest absolute peak-to-peak amplitude among the multiple second neural event detection results is determined as the second representative recording site of the second multi-channel event.

[0009] Optionally, generating a first two-dimensional detection event rate field based on the neural signals of the first scan segment, and generating a second two-dimensional detection event rate field based on the neural signals of the second scan segment, includes: For each neural signal recording site, the number of first multi-channel events with the neural signal recording site as the first representative recording site is counted within the effective recording time, and the ratio of the number of first multi-channel events to the effective recording time is determined as the detection event rate of the neural signal recording site in the first scan segment; For each neural signal recording site, the number of second multi-channel events using the neural signal recording site as the second representative recording site within the effective recording time is counted, and the ratio of the number of second multi-channel events to the effective recording time is determined as the detection event rate of the neural signal recording site in the second scan segment; Based on the row and column positions of each neural signal recording site in the neural probe, the detection event rates of each neural signal recording site in the first scan segment are arranged into a first two-dimensional detection event rate field, and the detection event rates of each neural signal recording site in the second scan segment are arranged into a second two-dimensional detection event rate field.

[0010] Optionally, generating a first two-dimensional amplitude field based on the neural signals of the first scan segment and generating a second two-dimensional amplitude field based on the neural signals of the second scan segment includes: For each first multi-channel event, the peak-to-peak amplitude of the detection results of each first neural event included in the first multi-channel event is obtained, and the maximum absolute peak-to-peak amplitude among the peak-to-peak amplitudes is determined as the first event amplitude value of the first multi-channel event. For each second multi-channel event, the peak-to-peak amplitude of the detection results of each second neural event included in the second multi-channel event is obtained, and the maximum absolute peak-to-peak amplitude among the peak-to-peak amplitudes is determined as the second event amplitude value of the second multi-channel event; Based on the background noise scale of the channel corresponding to the first representative recording site, the amplitude values ​​of the first events of each first multi-channel event with the first representative recording site as the representative recording site are standardized, and the median of the standardized first event amplitude values ​​is determined as the amplitude value of the first representative recording site in the first scan segment. Based on the background noise scale of the channel corresponding to the second representative recording site, the amplitude values ​​of the second events of each second multi-channel event with the second representative recording site as the representative recording site are standardized, and the median of the standardized second event amplitude values ​​is determined as the amplitude value of the second representative recording site in the second scan segment. According to the row and column positions of each neural signal recording site in the neural probe, the amplitude values ​​of each neural signal recording site in the first scan segment are arranged into a first two-dimensional amplitude field, and the amplitude values ​​of each neural signal recording site in the second scan segment are arranged into a second two-dimensional amplitude field.

[0011] Optionally, the step of calculating the event rate difference between the first two-dimensional detection event rate field and the second two-dimensional detection event rate field based on the spatially overlapping neural signal recording sites of the first scan segment neural signal and the second scan segment neural signal, and calculating the amplitude difference between the first two-dimensional amplitude field and the second two-dimensional amplitude field, includes: Based on the preset distance and the geometric information of each neural signal recording site in the neural probe, determine the spatially overlapping neural signal recording sites between the first scan segment neural signal and the second scan segment neural signal; The event rates of the first two-dimensional detection event rate field and the second two-dimensional detection event rate field at the spatially overlapping neural signal recording sites are extracted respectively, and the difference in event rates is calculated by normalization; The amplitude values ​​of the first two-dimensional amplitude field and the second two-dimensional amplitude field at the spatially overlapping neural signal recording sites are extracted respectively, and the amplitude difference is calculated by normalization.

[0012] Optionally, the step of performing coordinate stitching and overlapping region fusion of the first two-dimensional detection event rate field and the second two-dimensional detection event rate field along the active scanning axis to obtain the target extended spatial neural signal field includes: Align the row positions of the corresponding spatially overlapping neural signal recording sites in the first two-dimensional detection event rate field and the second two-dimensional detection event rate field along the active scanning axis to obtain the splicing coordinate relationship; Based on the splicing coordinate relationship, the first non-overlapping region of the first two-dimensional detection event rate field, the second non-overlapping region of the second two-dimensional detection event rate field, and the overlapping region between the first two-dimensional detection event rate field and the second two-dimensional detection event rate field are determined. The detection event rate in the first non-overlapping region is written into the corresponding first extended region in the initial extended spatial neural signal field, and the detection event rate in the second non-overlapping region is written into the corresponding second extended region in the initial extended spatial neural signal field. The detection event rates of the same tissue spatial location and the same lateral recording site column in the overlapping region are fused, and the fused detection event rates are written into the corresponding overlapping extended region in the initial extended spatial neural signal field to obtain the target extended spatial neural signal field.

[0013] Optionally, the preset difference conditions include: The event rate difference is less than or equal to a preset event rate difference threshold, and the magnitude difference is less than or equal to a preset magnitude difference threshold; Accordingly, when the event rate difference and / or the amplitude difference do not meet the preset difference condition, an unjoinable reconstruction result is output, including: If the event rate difference is greater than the preset event rate difference threshold, and / or the amplitude difference is greater than the preset amplitude difference threshold, then an unjoinable reconstruction result will be output.

[0014] Secondly, this application discloses a brain spatial neural signal field reconstruction system based on non-simultaneous local observation, comprising: a neural probe, a mechanical displacement control component, a neural signal acquisition component, and a processor; wherein, The neural probe has multiple rows of neural signal recording sites; The mechanical displacement control component is used to control the neural probe to move to a first spatial position in the target brain region, and to control the neural probe to move a preset distance along the active scanning axis to a second spatial position; wherein, the preset distance is less than the span of the recording site of the neural probe along the active scanning axis. The neural signal acquisition component is used to acquire a first scan segment neural signal at the first spatial location in the target brain region and acquire a second scan segment neural signal at the second spatial location, and send the first scan segment neural signal and the second scan segment neural signal to the processor. The processor is configured to generate a first two-dimensional detection event rate field and a first two-dimensional amplitude field based on the neural signals of the first scan segment, and to generate a second two-dimensional detection event rate field and a second two-dimensional amplitude field based on the neural signals of the second scan segment; wherein, the two-dimensional detection event rate field is a data field of detection event rates arranged according to the row and column positions of the neural signal recording sites in the neural probe; The processor is further configured to calculate the event rate difference between the first two-dimensional detection event rate field and the second two-dimensional detection event rate field based on the spatially overlapping neural signal recording sites of the first scan segment neural signal and the second scan segment neural signal, and to calculate the amplitude difference between the first two-dimensional amplitude field and the second two-dimensional amplitude field. The processor is further configured to, when the event rate difference and the amplitude difference satisfy a preset difference condition, perform coordinate splicing and overlapping region fusion of the first two-dimensional detection event rate field and the second two-dimensional detection event rate field along the active scanning axis to obtain a target extended spatial neural signal field; The processor is further configured to output an unjoinable reconstruction result when the event rate difference and / or the amplitude difference do not meet a preset difference condition.

[0015] Thirdly, this application discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the steps of the aforementioned disclosed method for reconstructing the brain spatial neural signal field based on non-simultaneous local observations.

[0016] As can be seen, this application discloses a method for controlling a neural probe with multiple neural signal recording sites to acquire neural signals in a first scan segment at a first spatial position in a target brain region; controlling the neural probe to move a preset distance along an active scanning axis to a second spatial position, and acquiring neural signals in a second scan segment at the second spatial position; wherein the preset distance is less than the span of the recording sites of the neural probe along the active scanning axis; generating a first two-dimensional event rate field and a first two-dimensional amplitude field based on the neural signals in the first scan segment, and generating a second two-dimensional event rate field and a second two-dimensional amplitude field based on the neural signals in the second scan segment; wherein the two-dimensional event rate field is arranged according to the row and column positions of the neural signal recording sites in the neural probe. The event rate data field is obtained; based on the spatial overlap of the neural signals of the first scan segment and the second scan segment, the event rate difference between the first two-dimensional event rate field and the second two-dimensional event rate field is calculated, and the amplitude difference between the first two-dimensional amplitude field and the second two-dimensional amplitude field is calculated; when the event rate difference and the amplitude difference meet the preset difference conditions, the first two-dimensional event rate field and the second two-dimensional event rate field are spliced ​​along the active scanning axis and the overlapping area is fused to obtain the target extended spatial neural signal field; when the event rate difference and / or the amplitude difference do not meet the preset difference conditions, the non-spongeable reconstruction result is output.

[0017] Therefore, by controlling a neural probe with multiple neural signal recording sites to acquire neural signals from the first and second scan segments at the first and second spatial positions respectively, and ensuring that the preset distance is less than the span of the recording sites along the active scanning axis of the neural probe, spatially overlapping neural signal recording sites can be formed between two non-simultaneous scan segments, providing a physical spatial reference for determining whether two neural signal segments can be spliced. Furthermore, by generating a two-dimensional event rate field and a two-dimensional amplitude field based on the neural signals from the first and second scan segments respectively, and calculating the event rate difference and amplitude difference at the spatially overlapping neural signal recording sites, the consistency between two non-simultaneous scan segments can be determined from two dimensions: the number of neural events and the event amplitude characteristics. This avoids misjudgments caused by relying solely on a single feature for splicing judgment. When the event rate difference and amplitude difference meet the preset difference conditions, this application performs coordinate stitching and overlapping region fusion on the two two-dimensional detection event rate fields along the active scanning axis to obtain the target extended spatial neural signal field; when the event rate difference and / or amplitude difference do not meet the preset difference conditions, the output is an unstitchable reconstruction result, which can expand the cumulative spatial recording range under the condition of limited channel neural probe, while avoiding forced stitching when the signals in the overlapping region are inconsistent, thus improving the reliability of the reconstruction of the neural signal field of non-simultaneous local observation. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0019] Figure 1 This is a flowchart of a brain spatial neural signal field reconstruction method based on non-simultaneous local observation disclosed in this application; Figure 2 This is a flowchart of a specific brain spatial neural signal field reconstruction method disclosed in this application; Figure 3 This is a schematic diagram of a neural probe scheme disclosed in this application; Figure 4 This is a schematic diagram of splicing and reconstructing a two-dimensional neural signal field as disclosed in this application; Figure 5 This is a schematic diagram of the brain spatial neural signal field reconstruction system based on non-simultaneous local observation disclosed in this application. Detailed Implementation

[0020] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0021] Implantable brain-computer interfaces typically acquire neurophysiological signals via neural probes for brain function research and neural signal analysis. However, due to limitations in the size, number of channels, and implantable range of these probes, a single fixed-location recording usually only captures neural signals from a localized brain region. Expanding the recording range by increasing the number of probes, their size, or the number of channels may lead to problems such as more complex wiring, increased data throughput, larger implant volume, increased risk of tissue damage, and higher costs.

[0022] To expand the cumulative spatial coverage of limited-channel neural probes, the same probe can be actively scanned and recorded at different spatial locations. However, neural signals from different scan segments are usually acquired at different times, and neural activity states, probe-tissue interfaces, background noise, and detection thresholds may all change, thus preventing direct stitching like image data. Existing active scanning recordings are mostly used for probe navigation, target brain region identification, insertion trajectory evaluation, or brain region boundary detection, and still struggle to address whether non-simultaneous local observations of neural signals can be stitched together and how to reconstruct an extended spatial neural signal field.

[0023] To this end, the present invention provides a brain spatial neural signal field reconstruction scheme based on non-simultaneous local observation. It can determine whether the neural signals of different scan segments meet the splicing conditions when the non-simultaneous and locally spatially overlapping neural signals are obtained by active scanning with a limited-channel neural probe, and reconstruct the extended spatial neural signal field when the conditions are met.

[0024] like Figure 1 As shown, this invention provides a method for reconstructing the spatial neural signal field of the brain based on non-simultaneous local observation, including: Step S11: Control the neural probe with multiple neural signal recording sites to collect the first scan segment neural signal at the first spatial position of the target brain region.

[0025] In this embodiment, the neural probe includes at least two columns of neural signal recording sites. Adjacent columns of neural signal recording sites are staggered along the active scanning axis. Along the probe's length, the spacing between neural signal recording sites in the same column is p. z In micrometers, along the width direction of the probe, the spacing between neural signal recording sites in the same row is p. y micrometers, total r OK, c The neural probe has multiple recording sites; the active scanning axis is the direction in which the neural probe actively moves within the target brain region; for example, the active scanning axis can be the implantation direction of the neural probe; the scanned segment neural signal is a segment of neurophysiological signal recorded when the neural probe stops at a spatial position; the neural probe has multiple neural signal recording sites, which are arranged along the length and width of the neural probe, so that the acquired neural signals have a two-dimensional spatial correspondence; the mechanical displacement control component drives the neural probe to move under the control of the processor; the neural signal acquisition component is connected to the neural probe and is used to acquire the neural signals of the corresponding channels of each neural signal recording site and send the neural signals to the processor.

[0026] The neural signal acquisition component can be a 32-channel neural signal acquisition device with a 30 kSa / s sampling rate and 16-bit analog-to-digital conversion accuracy. The neural probe can include 32 neural signal recording sites arranged in 16 rows and 4 columns, with adjacent columns of recording sites staggered. The geometric dimensions of the neural signal recording sites can be 10 micrometers by 10 micrometers. The spacing between neural signal recording sites in the same column along the probe length direction can be 30 micrometers, and the spacing between neural signal recording sites in the same row along the probe width direction can be 30 micrometers. It should be noted that the above values ​​are merely examples and are not intended to limit the scope of protection of this application.

[0027] Specifically, the neural probe is controlled to reach the first spatial position, and after maintaining a first preset stable duration, neural signals are collected within a first preset recording duration to obtain the neural signal of the first scan segment. It can be understood that the mechanical displacement control component first controls the neural probe to enter the target brain region and reach the first spatial position; after the neural probe reaches the first spatial position, it can maintain a first preset stable duration, and then the neural signal acquisition component collects neural signals within the first preset recording duration to obtain the neural signal of the first scan segment.

[0028] Step S12: Control the neural probe to move a preset distance along the active scanning axis to the second spatial position, and collect the neural signal of the second scanning segment at the second spatial position; wherein, the preset distance is less than the recording site span of the neural probe along the active scanning axis.

[0029] In this embodiment, the preset distance is determined based on the row spacing and span of the recording sites along the active scanning axis of the neural probe; wherein, the preset distance is an integer multiple of the row spacing of the recording sites and less than the span of the recording sites. It can be understood that the row spacing of the recording sites in the same column is p. z The distance is micrometers, and the recording site span is p micrometers; therefore, the preset distance is n times p. z The distance is set to micrometers, and the preset distance is less than the span of the recording sites on the neural probe along the probe length direction by p micrometers. This ensures that after recording neural signals at one spatial location, when moving to the next spatial location, some recording sites of the neural probe always overlap.

[0030] In this embodiment, the neural probe is controlled to move a preset distance along the active scanning axis to the second spatial position, and after maintaining a second preset stable duration, neural signals are acquired within a second preset recording duration to obtain the second scan segment neural signal. It can be understood that after the first scan segment neural signal acquisition is completed, the mechanical displacement control component controls the neural probe to move a preset distance along the active scanning axis to the second spatial position, and acquires the second scan segment neural signal at the second spatial position. The preset distance is less than the span of the recording sites along the active scanning axis of the neural probe, so that the first and second scan segment neural signals have spatially overlapping neural signal recording sites.

[0031] like Figure 2 As shown, the system controls the neural probe at the tip of the brain-computer interface device to be implanted into the target brain region at a preset rate of a micrometers per second, and waits for t1 seconds. When the brain tissue tends to stabilize, it begins recording neural signals at time t2 seconds. The recorded neural signals are first packaged by the neural signal acquisition component and then synchronously transmitted to the computer for corresponding processing. After the neural signal recording is completed, the computer sends a command to the mechanical displacement control component to the brain-computer interface system, displacing it n times p in the direction of the neural probe implantation at a speed of b micrometers per second. z The distance is in micrometers. It's important to note that n times p z The distance of a micrometer is less than the span of the recording sites on the neural probe along the probe's length direction (p micrometers). This ensures that after recording a neural signal at one spatial location, when moving to the next spatial location, some recording sites on the neural probe always overlap. In other words, these two non-simultaneous recordings at different spatial locations have a local overlap area. This overlap of some recording sites on the neural probe also provides a reliable physical anchor point for the reconstruction of the subsequent two neural signal segments. Similarly, the neural probe located at the tip of the brain-computer interface device records neural signals at a time t2 seconds at the second spatial location. The neural signals are packaged by the neural signal acquisition component and then synchronously transmitted to the computer for processing. Next, the signal processing module in the computer performs corresponding feature comparisons on the two received and processed neural signal segments, and decides whether to splice and reconstruct the two neural signals from different times and locations based on the comparison results. The probe's end is connected to a 32-channel neural probe, such as... Figure 3 As shown, Figure 3Each white square represents a recording site. The neural signal recording sites on the probe are distributed in multiple rows, with adjacent rows staggered. The geometric dimensions of each recording site are 10 micrometers × 10 micrometers. The spacing between the recording sites along the probe's length is 30 micrometers, and along the probe's width, the spacing is 10 micrometers, resulting in 16 rows and 4 columns of recording sites. After identifying the target brain region, the computer sends instructions to the mechanical displacement control component. This system controls the neural probe at the tip of the brain-computer interface device to be implanted into the target brain region at a preset rate of 5 micrometers per second. After a 300-second waiting period, when the brain tissue stabilizes, 120 seconds of neural signal recording begins. The recorded neural signals are first packaged by the neural signal acquisition component and then synchronously transmitted to the computer for processing. After neural signal recording is complete, the computer sends instructions to the mechanical displacement control component to the brain-computer interface system, displacing the neural probe 150 micrometers in the implantation direction at a speed of 1 micrometer per second (the overlapping area corresponds to the implantation site). Figure 3 (12 neural signal recording sites). Similarly, a neural probe located at the tip of the brain-computer interface device records neural signals for 120 seconds at a second spatial location. The neural signals are packaged by the neural signal acquisition component and then synchronously transmitted to the computer for corresponding processing.

[0032] By controlling the neural probe to move along the active scanning axis by a preset distance less than the span of the recording site, a local spatial overlap region is formed between two non-simultaneous scanning segments. The local spatial overlap region can serve as a physical spatial reference for consistency judgment.

[0033] Step S13: Generate a first two-dimensional detection event rate field and a first two-dimensional amplitude field based on the neural signal of the first scan segment, and generate a second two-dimensional detection event rate field and a second two-dimensional amplitude field based on the neural signal of the second scan segment; wherein, the two-dimensional detection event rate field is a data field of detection event rates arranged according to the row and column positions of the neural signal recording sites in the neural probe.

[0034] In this embodiment, before generating the first two-dimensional detection event rate field and the first two-dimensional amplitude field based on the neural signals of the first scan segment, and before generating the second two-dimensional detection event rate field and the second two-dimensional amplitude field based on the neural signals of the second scan segment, the method further includes: Bandpass filtering is performed on the neural signals of the first and second scan segments respectively to obtain a first bandpass signal and a second bandpass signal. A first event detection threshold is determined based on the background noise of the corresponding channel of each neural signal recording site in the first bandpass signal, and a second event detection threshold is determined based on the background noise of the corresponding channel of each neural signal recording site in the second bandpass signal. Multiple first neural event detection results are obtained from the first bandpass signal based on the first event detection threshold, and multiple second neural event detection results are obtained from the second bandpass signal based on the second event detection threshold. If the detection times corresponding to multiple first neural event detection results are within the same preset time window, and the corresponding neural signal recording sites are spatially adjacent... If the detection results of multiple first neural events are merged into a first multi-channel event, and the neural signal recording site corresponding to the first neural event detection result with the largest absolute peak-to-peak amplitude among the multiple first neural event detection results is determined as the first representative recording site of the first multi-channel event; if the detection times corresponding to multiple second neural event detection results are within the same preset time window, and the corresponding neural signal recording sites are spatially adjacent neural signal recording sites, then the multiple second neural event detection results are merged into a second multi-channel event, and the neural signal recording site corresponding to the second neural event detection result with the largest absolute peak-to-peak amplitude among the multiple second neural event detection results is determined as the second representative recording site of the second multi-channel event.

[0035] Understandably, the processor performs bandpass filtering on the first and second scanned neural signals. Bandpass filtering can be used to preserve the relevant frequency bands of neural events and suppress low-frequency drift or high-frequency noise. For example, the bandpass filtering frequency band can be from 300 Hz to 5 kHz, or other frequency bands can be set according to the neural signal acquisition components and signal characteristics. Further, after obtaining the first and second bandpass signals through bandpass filtering, a first event detection threshold is determined based on the background noise of the corresponding channel of each neural signal recording site in the first bandpass signal, and a second event detection threshold is determined based on the background noise of the corresponding channel of each neural signal recording site in the second bandpass signal. The background noise can be determined by the statistical characteristics of the corresponding channel bandpass signal, for example, the noise scale can be determined based on the median absolute deviation. The event detection threshold can be 60 microvolts, or it can be set according to the background noise of each channel and the acquisition conditions. Then, multiple first neural event detection results are obtained from the first bandpass signal based on the first event detection threshold, and multiple second neural event detection results are obtained from the second bandpass signal based on the second event detection threshold. Each neural event detection result can include the detection time, the corresponding neural signal recording site, and the peak-to-peak amplitude. If the detection times corresponding to multiple first neural event detection results are within the same preset time window, and the corresponding neural signal recording sites are spatially adjacent, the processor merges the multiple first neural event detection results into a first multi-channel event, and determines the neural signal recording site corresponding to the first neural event detection result with the largest absolute peak-to-peak amplitude among the multiple first neural event detection results as the first representative recording site of the first multi-channel event. If the detection times corresponding to multiple second neural event detection results are within the same preset time window, and the corresponding neural signal recording sites are spatially adjacent, the processor merges the multiple second neural event detection results into a second multi-channel event, and determines the neural signal recording site corresponding to the second neural event detection result with the largest absolute peak-to-peak amplitude among the multiple second neural event detection results as the second representative recording site of the second multi-channel event.

[0036] Instead of repeatedly counting the same neural event detected by multiple adjacent channels, the processor merges multiple detection results into a single multi-channel event based on temporal and spatial adjacency and identifies a representative recording site. This reduces the likelihood of the same extracellular neural event being repeatedly counted by multiple adjacent neural signal recording sites, improving the reliability of the event rate field and amplitude field.

[0037] In this embodiment, for each neural signal recording site, the number of first multi-channel events using the neural signal recording site as the first representative recording site within the effective recording duration is counted, and the ratio of the number of first multi-channel events to the effective recording duration is determined as the detection event rate of the neural signal recording site in the first scan segment; for each neural signal recording site, the number of second multi-channel events using the neural signal recording site as the second representative recording site within the effective recording duration is counted, and the ratio of the number of second multi-channel events to the effective recording duration is determined as the detection event rate of the neural signal recording site in the second scan segment; according to the row and column positions of each neural signal recording site in the neural probe, the detection event rates of each neural signal recording site in the first scan segment are arranged into a first two-dimensional detection event rate field, and the detection event rates of each neural signal recording site in the second scan segment are arranged into a second two-dimensional detection event rate field.

[0038] Understandably, for each neural signal recording site, the number of unsorted neural events belonging to that site within its effective recording time is counted and divided by the effective recording duration. The resulting two-dimensional event rate field can be a data field of event rates arranged according to the row and column positions of the neural signal recording sites in the neural probe, where each site is a recording site and the events belonging to that site are multi-channel events. For the scanning segment... k Middle r line, number c The event rate for recording neural signals at a given location is defined as follows: ; in, k This represents two different spaces with locally overlapping regions, namely the first scan segment and the second scan segment; This indicates the number of detection events attributed to that neural signal recording site. This indicates the same effective recording duration (t2 seconds) for both scan segments. r Indicates the probe row. c This indicates a probe array.

[0039] This forms an interleaved four-column two-dimensional detection event rate field for the first scan segment (the first spatial location for recording neural signals). Each row of the matrix corresponds to a tissue spatial location along the probe implantation direction (scanning axis), and each column corresponds to a transverse recording site column of the probe. The final output spatial electrophysiological field uses the unsorted detection event rate as the field value, which is the first two-dimensional detection event rate field.

[0040] The neural signals of the second scan segment, which are synchronously transmitted by the neural signal acquisition component, are preprocessed in the same way as those of the first scan segment. This forms an interleaved four-column two-dimensional detection event rate field for the second scan segment, which is used as one of the criteria for reconstructing and judging the neural signals.

[0041] In this embodiment, for each first multi-channel event, the peak-to-peak amplitude of the detection results of each first neural event included in the first multi-channel event is obtained, and the maximum absolute peak-to-peak amplitude among the peak-to-peak amplitudes is determined as the first event amplitude value of the first multi-channel event; for each second multi-channel event, the peak-to-peak amplitude of the detection results of each second neural event included in the second multi-channel event is obtained, and the maximum absolute peak-to-peak amplitude among the peak-to-peak amplitudes is determined as the second event amplitude value of the second multi-channel event; based on the background noise scale of the channel corresponding to the first representative recording site, the first event amplitude values ​​of each first multi-channel event with the first representative recording site as the representative recording site are standardized. The median of the standardized first event amplitude value is determined as the amplitude value of the first representative recording site in the first scan segment. Based on the background noise scale of the channel corresponding to the second representative recording site, the amplitude values ​​of the second events of each second multi-channel event with the second representative recording site as the representative recording site are standardized, and the median of the standardized second event amplitude values ​​is determined as the amplitude value of the second representative recording site in the second scan segment. According to the row and column positions of each neural signal recording site in the neural probe, the amplitude values ​​of each neural signal recording site in the first scan segment are arranged into a first two-dimensional amplitude field, and the amplitude values ​​of each neural signal recording site in the second scan segment are arranged into a second two-dimensional amplitude field.

[0042] Understandably, the two-dimensional amplitude field can be an amplitude data field arranged according to the row and column positions of the neural signal recording sites in the neural probe. For each unsorted event belonging to a certain neural signal recording site, its peak-to-peak amplitude is calculated. To reduce the influence of background noise differences between different channels on amplitude comparison, the peak-to-peak amplitude of each event is divided by the background noise scale of that recording channel to obtain the standardized peak-to-peak amplitude. The background noise scale adopts a fixed robust noise estimation method: it is calculated based on the median absolute deviation of the bandpass signal of that channel. Specifically, for each recording channel, the median absolute deviation is calculated for the effective recording signal after the same high-frequency bandpass processing, and the median absolute deviation is divided by 0.6745 to obtain the robust background noise scale of that channel. For the same neural signal recording site, the median of the standardized peak-to-peak amplitudes of all events belonging to that site is taken to obtain the standardized median peak-to-peak amplitude of that site. Thus, the first scan segment forms a double-column two-dimensional amplitude field corresponding to the geometry of the neural probe, i.e., the first two-dimensional amplitude field. Similarly, the second scan segment is processed in the same way to obtain the corresponding second two-dimensional amplitude field. The purpose of using the median is to reduce the impact of occasional large-amplitude events, anomalous events, and residual artifacts on the eigenvalues.

[0043] Step S14: Calculate the event rate difference between the first two-dimensional detection event rate field and the second two-dimensional detection event rate field based on the spatial overlap of the neural signal recording sites of the first scan segment neural signal and the second scan segment neural signal, and calculate the amplitude difference between the first two-dimensional amplitude field and the second two-dimensional amplitude field.

[0044] In this embodiment, spatially overlapping neural signal recording sites between the first scan segment neural signal and the second scan segment neural signal are determined based on the preset distance and the geometric information of each neural signal recording site in the neural probe; the detection event rates of the first two-dimensional detection event rate field and the second two-dimensional detection event rate field at the spatially overlapping neural signal recording sites are extracted respectively, and the event rate difference is calculated by normalization; the amplitude values ​​of the first two-dimensional amplitude field and the second two-dimensional amplitude field at the spatially overlapping neural signal recording sites are extracted respectively, and the amplitude difference is calculated by normalization. It can be understood that spatially overlapping neural signal recording sites are neural signal recording sites in the first scan segment neural signal and the second scan segment neural signal that correspond to the same or adjacent tissue spatial locations. Based on the preset distance (n times p... z The spatially overlapping neural signal recording site rows at the end of the first scan segment and the beginning of the second scan segment were determined using the neural probe geometry (p micrometers) and the recording site span along the probe length direction. For each overlapping neural signal recording site row: the leftmost column of the first scan segment corresponds to the leftmost column of the second scan segment; the rightmost column of the first scan segment corresponds to the rightmost column of the second scan segment. The detection event rate and median peak-to-peak amplitude at all overlapping neural signal recording sites of the two scan segments were extracted respectively, and the same normalized difference calculation method was used. The features to be compared are... X The values ​​at the overlapping sites of the first and second scan segments are respectively and Its characteristic differences are defined as: ; in, This indicates all spatially overlapping neural signal recording sites across multiple columns; It is a constant, a preset small quantity to prevent the denominator from being zero; when the characteristic X To detect the event rate, the difference in event rates is obtained. When features X When the median peak-to-peak amplitude is used, the amplitude difference is obtained. .

[0045] It should be noted that in the overlapping region, the two neural signal recording sites are only subjected to the aforementioned dual-feature judgment to determine whether they can be spliced ​​and reconstructed when there are clearly detected unsorted events. Otherwise, the output is a result indicating that splicing and reconstruction is not possible.

[0046] Therefore, before calculating event rate and amplitude differences, the second two-dimensional event rate field or amplitude field is not scaled or baseline-corrected based on data from spatially overlapping neural signal recording sites. Data from spatially overlapping neural signal recording sites is used to directly verify the consistency between the neural signals in the first and second scan segments. This avoids altering the scale of the second scan segment neural signal before consistency verification, ensuring that the spatially overlapping neural signal recording sites maintain their independent verification function.

[0047] This embodiment calculates the difference in event rate and amplitude at spatially overlapping neural signal recording sites to determine whether the activity quantity and amplitude characteristics of two non-simultaneous scan segments are consistent in a common spatial region, providing a basis for subsequent splicing.

[0048] Step S15: When the event rate difference and the amplitude difference meet the preset difference conditions, the first two-dimensional detection event rate field and the second two-dimensional detection event rate field are spliced ​​together and the overlapping region is fused along the active scanning axis to obtain the target extended spatial neural signal field.

[0049] In this embodiment, the preset difference condition includes: the event rate difference is less than or equal to a preset event rate difference threshold, and the magnitude difference is less than or equal to a preset magnitude difference threshold; it can be understood that when the event rate difference... Less than or equal to the unsorted event rate threshold And the difference in magnitude Less than or equal to the amplitude difference threshold If two neural signals recorded at different times and in different spaces are not recorded simultaneously, spatial splicing and reconstruction are performed; otherwise, splicing and reconstruction are not performed.

[0050] In this embodiment, under the condition of satisfying preset differences, the process proceeds to one-dimensional coordinate stitching and two-dimensional field fusion, specifically: Align the row positions of the corresponding spatially overlapping neural signal recording sites in the first two-dimensional detection event rate field and the second two-dimensional detection event rate field along the active scanning axis to obtain a splicing coordinate relationship; determine the first non-overlapping region of the first two-dimensional detection event rate field, the second non-overlapping region of the second two-dimensional detection event rate field, and the overlapping region between the first two-dimensional detection event rate field and the second two-dimensional detection event rate field based on the splicing coordinate relationship; write the detection event rates in the first non-overlapping region into the corresponding first extended region of the initial extended spatial neural signal field, and write the detection event rates in the second non-overlapping region into the corresponding second extended region of the initial extended spatial neural signal field; perform fusion processing on the detection event rates corresponding to the same tissue spatial position and the same transverse recording site column in the overlapping region, and write the fused detection event rates into the corresponding overlapping extended region of the initial extended spatial neural signal field to obtain the target extended spatial neural signal field.

[0051] Understandably, the processor aligns the row positions of corresponding spatially overlapping neural signal recording sites in the first and second two-dimensional event rate fields along the active scanning axis to obtain a splicing coordinate relationship. This splicing coordinate relationship is used to determine the row position correspondence between the first and second two-dimensional event rate fields in the extended space. Although each local spatial field retains two dimensions—the vertical row and the horizontal column of the neural signal recording sites—the neural probe only moves along the vertical scanning axis (probe implantation direction). Therefore, only a one-dimensional coordinate transformation is performed between the two local fields, that is, the overlapping neural signal recording sites in the first and second scan segments are aligned to obtain the splicing coordinate relationship. In this splicing coordinate relationship, the number of vertical columns in the two-dimensional field is equal to that of a single scan segment. c Columns, expanded to c + n The number of columns and rows remains unchanged.

[0052] Based on the splicing coordinate relationship, the first non-overlapping region of the first two-dimensional detection event rate field, the second non-overlapping region of the second two-dimensional detection event rate field, and the overlapping region between the first and second two-dimensional detection event rate fields are determined. The first non-overlapping region can be a region in the first two-dimensional detection event rate field that does not coincide with the second two-dimensional detection event rate field; the second non-overlapping region can be a region in the second two-dimensional detection event rate field that does not coincide with the first two-dimensional detection event rate field; the overlapping region can be a region that corresponds to the same tissue spatial location and the same lateral recording point column.

[0053] Furthermore, when the preset difference conditions are met, the extended spatial electrophysiological field is formed according to the following fixed rules: For the first non-overlapping region, the detection event rate of the neural signal recording site corresponding to the first scan segment is directly retained. Similarly, for the second non-overlapping region, the detection event rate of the neural signal recording site corresponding to the second scan segment is directly retained. For the overlapping region of the two scan segments, the two observations of the detection event rates corresponding to the same tissue spatial location and the same transverse recording site column are weighted using an equal-weighted arithmetic mean (since the two scan segments have the same effective recording duration, the same acquisition parameters, the same event detection rules, and the same experimental conditions, and both overlapping features have passed consistency judgment). The two-dimensional fusion field can be represented as: ; in, This represents the detection event rate at the corresponding position in the first two-dimensional detection event rate field. This represents the detection event rate at the corresponding position in the second two-dimensional detection event rate field. Finally, after splicing and reconstruction, the output is a multi-column two-dimensional discrete unsorted detection event rate spatial neural signal field extended along the active scanning axis.

[0054] like Figure 4 The image shows the multi-column, discrete, unsorted event rate spatial neural signal field extended along the active scanning axis after stitching and reconstruction. The color represents the magnitude of the unsorted event rate; darker red indicates a higher event rate, and darker blue indicates a lower event rate. After stitching and reconstruction, the neural signal recording is equivalent to a 52-channel recording from the original 32-channel recording, effectively increasing the spatial recording range of the limited-channel neural probe.

[0055] Therefore, by identifying the first non-overlapping region, the second non-overlapping region, and the overlapping region through splicing coordinate relationships, and by writing and fusing different regions respectively, a target extended spatial neural signal field extending along the active scanning axis can be obtained.

[0056] Step S16: If the event rate difference and / or the amplitude difference do not meet the preset difference conditions, then output the non-splitterable reconstruction result.

[0057] In this embodiment, when the event rate difference is greater than the preset event rate difference threshold, and / or the amplitude difference is greater than the preset amplitude difference threshold, an unjoinable reconstruction result is output.

[0058] Therefore, by controlling a neural probe with multiple neural signal recording sites to acquire neural signals from the first and second scan segments at the first and second spatial positions respectively, and ensuring that the preset distance is less than the span of the recording sites along the active scanning axis of the neural probe, spatially overlapping neural signal recording sites can be formed between two non-simultaneous scan segments, providing a physical spatial reference for determining whether two neural signal segments can be spliced. Furthermore, by generating a two-dimensional event rate field and a two-dimensional amplitude field based on the neural signals from the first and second scan segments respectively, and calculating the event rate difference and amplitude difference at the spatially overlapping neural signal recording sites, the consistency between two non-simultaneous scan segments can be determined from two dimensions: the number of neural events and the event amplitude characteristics. This avoids misjudgments caused by relying solely on a single feature for splicing judgment. When the event rate difference and amplitude difference meet the preset difference conditions, this application performs coordinate stitching and overlapping region fusion on the two two-dimensional detection event rate fields along the active scanning axis to obtain the target extended spatial neural signal field; when the event rate difference and / or amplitude difference do not meet the preset difference conditions, the output is an unstitchable reconstruction result, which can expand the cumulative spatial recording range under the condition of limited channel neural probe, while avoiding forced stitching when the signals in the overlapping region are inconsistent, thus improving the reliability of the reconstruction of the neural signal field of non-simultaneous local observation.

[0059] like Figure 5 As shown, the present invention also provides a brain spatial neural signal field reconstruction system based on non-simultaneous local observation, comprising: a neural probe 11, a mechanical displacement control component 12, a neural signal acquisition component 13, and a processor 14; wherein, The neural probe 11 has multiple rows of neural signal recording sites; The mechanical displacement control component 12 is used to control the neural probe 11 to move to a first spatial position in the target brain region, and to control the neural probe 11 to move a preset distance along the active scanning axis to a second spatial position; wherein, the preset distance is less than the span of the recording site of the neural probe 11 along the active scanning axis. The neural signal acquisition component 13 is used to acquire a first scan segment neural signal at the first spatial location of the target brain region and acquire a second scan segment neural signal at the second spatial location, and send the first scan segment neural signal and the second scan segment neural signal to the processor 14. The processor 14 is configured to generate a first two-dimensional detection event rate field and a first two-dimensional amplitude field based on the neural signals of the first scan segment, and to generate a second two-dimensional detection event rate field and a second two-dimensional amplitude field based on the neural signals of the second scan segment; wherein, the two-dimensional detection event rate field is a data field of detection event rates arranged according to the row and column positions of the neural signal recording sites in the neural probe; The processor 14 is further configured to calculate the event rate difference between the first two-dimensional detection event rate field and the second two-dimensional detection event rate field based on the spatial overlap of the neural signal recording sites of the first scan segment neural signal and the second scan segment neural signal, and to calculate the amplitude difference between the first two-dimensional amplitude field and the second two-dimensional amplitude field. The processor 14 is further configured to, when the event rate difference and the amplitude difference meet a preset difference condition, perform coordinate splicing and overlapping region fusion of the first two-dimensional detection event rate field and the second two-dimensional detection event rate field along the active scanning axis to obtain a target extended spatial neural signal field. The processor 14 is further configured to output an unjoinable reconstruction result when the event rate difference and / or the amplitude difference do not meet a preset difference condition.

[0060] In one specific embodiment, the neural probe 11 specifically includes at least two columns of neural signal recording sites, with adjacent columns of neural signal recording sites arranged alternately along the active scanning axis. Accordingly, the mechanical displacement control component 12 may specifically include: The distance determination unit is used to determine the preset distance based on the row spacing and span of the recording sites along the active scanning axis of the neural probe 11; wherein the preset distance is an integer multiple of the row spacing of the recording sites and is less than the span of the recording sites.

[0061] In one specific embodiment, the neural signal acquisition component 13 may include: The first signal acquisition unit is used to acquire neural signals within a first preset recording time after the mechanical displacement control component 12 controls the neural probe 11 to reach the first spatial position and maintains a first preset stable time, so as to obtain the neural signal of the first scan segment. The second signal acquisition unit is used to acquire neural signals within a second preset recording time after the mechanical displacement control component 12 controls the neural probe 11 to move a preset distance along the active scanning axis to the second spatial position and maintains a second preset stable time, so as to obtain the neural signal of the second scanning segment.

[0062] In one specific embodiment, the system further includes: The filtering processing unit is used to perform bandpass filtering processing on the first scan segment neural signal and the second scan segment neural signal respectively to obtain a first bandpass signal and a second bandpass signal; The threshold determination unit is used to determine a first event detection threshold based on the background noise of the channel corresponding to each neural signal recording site in the first bandpass signal, and to determine a second event detection threshold based on the background noise of the channel corresponding to each neural signal recording site in the second bandpass signal. The detection result acquisition unit is used to acquire multiple first neural event detection results from the first bandpass signal based on the first event detection threshold, and to acquire multiple second neural event detection results from the second bandpass signal based on the second event detection threshold; The first event merging unit is used to merge the multiple first neural event detection results into a first multi-channel event if the detection times corresponding to multiple first neural event detection results are within the same preset time window and the corresponding neural signal recording sites are spatially adjacent neural signal recording sites, and to determine the neural signal recording site corresponding to the first neural event detection result with the largest absolute peak-to-peak amplitude among the multiple first neural event detection results as the first representative recording site of the first multi-channel event. The second event merging unit is used to merge the multiple second neural event detection results into a second multi-channel event if the detection times corresponding to multiple second neural event detection results are within the same preset time window and the corresponding neural signal recording sites are spatially adjacent neural signal recording sites, and to determine the neural signal recording site corresponding to the second neural event detection result with the largest absolute peak-to-peak amplitude among the multiple second neural event detection results as the second representative recording site of the second multi-channel event.

[0063] In one specific embodiment, the processor 14 may include: The first processing unit is used to count the number of first multi-channel events with the neural signal recording site as the first representative recording site within the effective recording time for each neural signal recording site, and to determine the ratio of the number of first multi-channel events to the effective recording time as the detection event rate of the neural signal recording site in the first scan segment. The second processing unit is used to count the number of second multi-channel events with the neural signal recording site as the second representative recording site within the effective recording time for each neural signal recording site, and to determine the ratio of the number of second multi-channel events to the effective recording time as the detection event rate of the neural signal recording site in the second scanning segment. The third processing unit is used to arrange the detection event rates of each neural signal recording site in the first scan segment into a first two-dimensional detection event rate field according to the row and column positions of each neural signal recording site in the neural probe, and to arrange the detection event rates of each neural signal recording site in the second scan segment into a second two-dimensional detection event rate field.

[0064] In one specific embodiment, the processor 14 may include: The fourth processing unit is used to obtain the peak-to-peak amplitude of the detection results of each first neural event included in the first multi-channel event for each first multi-channel event, and determine the maximum absolute peak-to-peak amplitude among the peak-to-peak amplitudes as the first event amplitude value of the first multi-channel event. The fifth processing unit is used to obtain the peak-to-peak amplitude of the detection results of each second neural event included in the second multi-channel event for each second multi-channel event, and determine the maximum absolute peak-to-peak amplitude among the peak-to-peak amplitudes as the second event amplitude value of the second multi-channel event; The sixth processing unit is used to standardize the first event amplitude value of each first multi-channel event with the first representative recording site as the representative recording site based on the background noise scale of the channel corresponding to the first representative recording site, and to determine the median of the standardized first event amplitude value as the amplitude value of the first representative recording site in the first scan segment. The seventh processing unit is used to standardize the second event amplitude values ​​of each second multi-channel event with the second representative recording site as the representative recording site based on the background noise scale of the channel corresponding to the second representative recording site, and to determine the median of the standardized second event amplitude values ​​as the amplitude value of the second representative recording site in the second scan segment. The eighth processing unit is used to arrange the amplitude values ​​of each neural signal recording site in the first scanning segment into a first two-dimensional amplitude field according to the row and column positions of each neural signal recording site in the neural probe, and to arrange the amplitude values ​​of each neural signal recording site in the second scanning segment into a second two-dimensional amplitude field.

[0065] In one specific embodiment, the processor 14 may include: The overlapping site determination unit is used to determine the spatially overlapping neural signal recording sites between the first scan segment neural signal and the second scan segment neural signal based on the preset distance and the geometric information of each neural signal recording site in the neural probe; The first difference calculation unit is used to extract the detection event rates of the first two-dimensional detection event rate field and the second two-dimensional detection event rate field at the spatially overlapping neural signal recording sites, and normalize and calculate the event rate difference. The second difference calculation unit is used to extract the amplitude values ​​of the first two-dimensional amplitude field and the second two-dimensional amplitude field at the spatially overlapping neural signal recording sites, and normalize and calculate the amplitude difference.

[0066] In one specific embodiment, the processor 14 may include: The alignment processing unit is used to align the row positions of the corresponding spatially overlapping neural signal recording sites in the first two-dimensional detection event rate field and the second two-dimensional detection event rate field along the active scanning axis to obtain the splicing coordinate relationship; The overlapping region determination unit is used to determine, based on the splicing coordinate relationship, the first non-overlapping region of the first two-dimensional detection event rate field, the second non-overlapping region of the second two-dimensional detection event rate field, and the overlapping region between the first two-dimensional detection event rate field and the second two-dimensional detection event rate field; An event writing unit is used to write the detection event rate in the first non-overlapping region into the corresponding first extended region in the initial extended spatial neural signal field, and to write the detection event rate in the second non-overlapping region into the corresponding second extended region in the initial extended spatial neural signal field. The fusion processing unit is used to fuse the detection event rates of the same tissue spatial location and the same lateral recording site column in the overlapping region, and write the fused detection event rates into the corresponding overlapping extended region in the initial extended spatial neural signal field to obtain the target extended spatial neural signal field.

[0067] In one specific implementation, the preset difference condition may specifically include the event rate difference being less than or equal to a preset event rate difference threshold, and the magnitude difference being less than or equal to a preset magnitude difference threshold. Accordingly, the processor 14 may specifically include: The result output unit is used to output an unjoinable reconstruction result when the event rate difference is greater than the preset event rate difference threshold and / or the amplitude difference is greater than the preset amplitude difference threshold.

[0068] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.

[0069] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application. The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly in hardware, software modules executed by a processor, or a combination of both. The software module may be located in random access memory (RAM), memory, read-only memory (ROM), electrically programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), registers, hard disks, removable disks, CD-ROMs (Compact Disc-Read Only Memory), or any other form of storage medium known in the art.

[0070] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0071] The solution provided by the present invention has been described in detail above. Specific examples have been used to illustrate the principle and implementation of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation and application scope based on the idea of ​​the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for reconstructing the spatial neural signal field of the brain based on non-simultaneous local observation, characterized in that, include: A neural probe with multiple neural signal recording sites is controlled to acquire neural signals in the first scan segment at a first spatial location in the target brain region; The neural probe is controlled to move a preset distance along the active scanning axis to a second spatial position, and neural signals of a second scanning segment are acquired at the second spatial position; wherein, the preset distance is less than the span of the recording site of the neural probe along the active scanning axis. A first two-dimensional detection event rate field and a first two-dimensional amplitude field are generated based on the neural signals of the first scan segment, and a second two-dimensional detection event rate field and a second two-dimensional amplitude field are generated based on the neural signals of the second scan segment; wherein, the two-dimensional detection event rate field is a data field of detection event rates arranged according to the row and column positions of the neural signal recording sites in the neural probe; Based on the spatial overlap of the neural signals of the first scan segment and the second scan segment, the event rate difference between the first two-dimensional detection event rate field and the second two-dimensional detection event rate field is calculated, and the amplitude difference between the first two-dimensional amplitude field and the second two-dimensional amplitude field is also calculated. When the event rate difference and the amplitude difference meet the preset difference conditions, the first two-dimensional detection event rate field and the second two-dimensional detection event rate field are spliced ​​and the overlapping region is fused along the active scanning axis to obtain the target extended space neural signal field. If the event rate difference and / or the magnitude difference do not meet the preset difference conditions, an unspantable reconstruction result will be output.

2. The brain spatial neural signal field reconstruction method based on non-simultaneous local observation according to claim 1, characterized in that, The neural probe includes at least two columns of neural signal recording sites, with adjacent columns of neural signal recording sites arranged alternately along the active scanning axis. Accordingly, controlling the neural probe to move a preset distance along the active scanning axis to the second spatial position includes: The preset distance is determined based on the row spacing and span of the recording sites along the active scanning axis of the neural probe; wherein the preset distance is an integer multiple of the row spacing of the recording sites and is less than the span of the recording sites.

3. The brain spatial neural signal field reconstruction method based on non-simultaneous local observation according to claim 1, characterized in that, The control of a neural probe with multiple neural signal recording sites to acquire neural signals in a first scan segment at a first spatial location in the target brain region includes: The neural probe is controlled to reach the first spatial position, and after maintaining a first preset stable time, neural signals are collected within a first preset recording time to obtain the neural signal of the first scan segment; Accordingly, the neural probe is controlled to move a preset distance along the active scanning axis to a second spatial position, and neural signals of the second scanning segment are acquired at the second spatial position, including: The neural probe is controlled to move a preset distance along the active scanning axis to the second spatial position, and after maintaining a second preset stable duration, neural signals are collected within a second preset recording duration to obtain the neural signals of the second scanning segment.

4. The brain spatial neural signal field reconstruction method based on non-simultaneous local observation according to claim 1, characterized in that, Before generating the first two-dimensional detection event rate field and the first two-dimensional amplitude field based on the neural signals of the first scan segment, and before generating the second two-dimensional detection event rate field and the second two-dimensional amplitude field based on the neural signals of the second scan segment, the method further includes: The neural signals of the first scan segment and the neural signals of the second scan segment are respectively subjected to bandpass filtering to obtain a first bandpass signal and a second bandpass signal; A first event detection threshold is determined based on the background noise of the channel corresponding to each neural signal recording site in the first bandpass signal, and a second event detection threshold is determined based on the background noise of the channel corresponding to each neural signal recording site in the second bandpass signal. Based on the first event detection threshold, multiple first neural event detection results are obtained from the first bandpass signal, and based on the second event detection threshold, multiple second neural event detection results are obtained from the second bandpass signal; If the detection times corresponding to multiple first neural event detection results are within the same preset time window, and the corresponding neural signal recording sites are spatially adjacent neural signal recording sites, then the multiple first neural event detection results are merged into a first multi-channel event, and the neural signal recording site corresponding to the first neural event detection result with the largest absolute peak-to-peak amplitude among the multiple first neural event detection results is determined as the first representative recording site of the first multi-channel event. If the detection times corresponding to multiple second neural event detection results are within the same preset time window, and the corresponding neural signal recording sites are spatially adjacent neural signal recording sites, then the multiple second neural event detection results are merged into a second multi-channel event, and the neural signal recording site corresponding to the second neural event detection result with the largest absolute peak-to-peak amplitude among the multiple second neural event detection results is determined as the second representative recording site of the second multi-channel event.

5. The brain spatial neural signal field reconstruction method based on non-simultaneous local observation according to claim 4, characterized in that, Generating a first two-dimensional detection event rate field based on the neural signals of the first scan segment, and generating a second two-dimensional detection event rate field based on the neural signals of the second scan segment, including: For each neural signal recording site, the number of first multi-channel events with the neural signal recording site as the first representative recording site is counted within the effective recording time, and the ratio of the number of first multi-channel events to the effective recording time is determined as the detection event rate of the neural signal recording site in the first scan segment. For each neural signal recording site, the number of second multi-channel events using the neural signal recording site as the second representative recording site within the effective recording time is counted, and the ratio of the number of second multi-channel events to the effective recording time is determined as the detection event rate of the neural signal recording site in the second scan segment; Based on the row and column positions of each neural signal recording site in the neural probe, the detection event rates of each neural signal recording site in the first scan segment are arranged into a first two-dimensional detection event rate field, and the detection event rates of each neural signal recording site in the second scan segment are arranged into a second two-dimensional detection event rate field.

6. The brain spatial neural signal field reconstruction method based on non-simultaneous local observation according to claim 4, characterized in that, Generating a first two-dimensional amplitude field based on the neural signals of the first scan segment, and generating a second two-dimensional amplitude field based on the neural signals of the second scan segment, including: For each first multi-channel event, the peak-to-peak amplitude of the detection results of each first neural event included in the first multi-channel event is obtained, and the maximum absolute peak-to-peak amplitude among the peak-to-peak amplitudes is determined as the first event amplitude value of the first multi-channel event. For each second multi-channel event, the peak-to-peak amplitude of the detection results of each second neural event included in the second multi-channel event is obtained, and the maximum absolute peak-to-peak amplitude among the peak-to-peak amplitudes is determined as the second event amplitude value of the second multi-channel event; Based on the background noise scale of the channel corresponding to the first representative recording site, the amplitude values ​​of the first events of each first multi-channel event with the first representative recording site as the representative recording site are standardized, and the median of the standardized first event amplitude values ​​is determined as the amplitude value of the first representative recording site in the first scan segment. Based on the background noise scale of the channel corresponding to the second representative recording site, the amplitude values ​​of the second events of each second multi-channel event with the second representative recording site as the representative recording site are standardized, and the median of the standardized second event amplitude values ​​is determined as the amplitude value of the second representative recording site in the second scan segment. According to the row and column positions of each neural signal recording site in the neural probe, the amplitude values ​​of each neural signal recording site in the first scan segment are arranged into a first two-dimensional amplitude field, and the amplitude values ​​of each neural signal recording site in the second scan segment are arranged into a second two-dimensional amplitude field.

7. The brain spatial neural signal field reconstruction method based on non-simultaneous local observation according to claim 1, characterized in that, The calculation of the event rate difference between the first two-dimensional detection event rate field and the second two-dimensional detection event rate field based on the spatially overlapping neural signal recording sites of the first scan segment neural signal and the second scan segment neural signal, and the calculation of the amplitude difference between the first two-dimensional amplitude field and the second two-dimensional amplitude field, includes: Based on the preset distance and the geometric information of each neural signal recording site in the neural probe, determine the spatially overlapping neural signal recording sites between the first scan segment neural signal and the second scan segment neural signal; The event rates of the first two-dimensional detection event rate field and the second two-dimensional detection event rate field at the spatially overlapping neural signal recording sites are extracted respectively, and the difference in event rates is calculated by normalization; The amplitude values ​​of the first two-dimensional amplitude field and the second two-dimensional amplitude field at the spatially overlapping neural signal recording sites are extracted respectively, and the amplitude difference is calculated by normalization.

8. The brain spatial neural signal field reconstruction method based on non-simultaneous local observation according to claim 1, characterized in that, The step of stitching together the first two-dimensional detection event rate field and the second two-dimensional detection event rate field along the active scanning axis and fusing the overlapping regions to obtain the target extended spatial neural signal field includes: Align the row positions of the corresponding spatially overlapping neural signal recording sites in the first two-dimensional detection event rate field and the second two-dimensional detection event rate field along the active scanning axis to obtain the splicing coordinate relationship; Based on the splicing coordinate relationship, the first non-overlapping region of the first two-dimensional detection event rate field, the second non-overlapping region of the second two-dimensional detection event rate field, and the overlapping region between the first two-dimensional detection event rate field and the second two-dimensional detection event rate field are determined. The detection event rate in the first non-overlapping region is written into the corresponding first extended region in the initial extended spatial neural signal field, and the detection event rate in the second non-overlapping region is written into the corresponding second extended region in the initial extended spatial neural signal field. The detection event rates of the same tissue spatial location and the same lateral recording site column in the overlapping region are fused, and the fused detection event rates are written into the corresponding overlapping extended region in the initial extended spatial neural signal field to obtain the target extended spatial neural signal field.

9. The brain spatial neural signal field reconstruction method based on non-simultaneous local observation according to claim 1, characterized in that, The preset difference conditions include: The event rate difference is less than or equal to a preset event rate difference threshold, and the magnitude difference is less than or equal to a preset magnitude difference threshold; Accordingly, when the event rate difference and / or the amplitude difference do not meet the preset difference condition, the output of an unjoinable reconstruction result includes: If the event rate difference is greater than the preset event rate difference threshold, and / or the amplitude difference is greater than the preset amplitude difference threshold, then an unjoinable reconstruction result will be output.

10. A brain spatial neural signal field reconstruction system based on non-simultaneous local observation, characterized in that, include: The system comprises a neural probe, a mechanical displacement control component, a neural signal acquisition component, and a processor; among which, The neural probe has multiple rows of neural signal recording sites; The mechanical displacement control component is used to control the neural probe to move to a first spatial position in the target brain region, and to control the neural probe to move a preset distance along the active scanning axis to a second spatial position; wherein, the preset distance is less than the span of the recording site of the neural probe along the active scanning axis. The neural signal acquisition component is used to acquire a first scan segment neural signal at the first spatial location in the target brain region and acquire a second scan segment neural signal at the second spatial location, and send the first scan segment neural signal and the second scan segment neural signal to the processor. The processor is configured to generate a first two-dimensional detection event rate field and a first two-dimensional amplitude field based on the neural signals of the first scan segment, and to generate a second two-dimensional detection event rate field and a second two-dimensional amplitude field based on the neural signals of the second scan segment; wherein, the two-dimensional detection event rate field is a data field of detection event rates arranged according to the row and column positions of the neural signal recording sites in the neural probe; The processor is further configured to calculate the event rate difference between the first two-dimensional detection event rate field and the second two-dimensional detection event rate field based on the spatially overlapping neural signal recording sites of the first scan segment neural signal and the second scan segment neural signal, and to calculate the amplitude difference between the first two-dimensional amplitude field and the second two-dimensional amplitude field. The processor is further configured to, when the event rate difference and the amplitude difference satisfy a preset difference condition, perform coordinate splicing and overlapping region fusion of the first two-dimensional detection event rate field and the second two-dimensional detection event rate field along the active scanning axis to obtain a target extended spatial neural signal field; The processor is further configured to output an unjoinable reconstruction result when the event rate difference and / or the amplitude difference do not meet a preset difference condition.

Images