Method and device for measuring transmission speed of electroneurographic signals

Abstract

The invention provides an electroneurographic signal transmission speed measurement method and device, relates to the technical field of neural electrophysiology, and solves the technical problem of insufficient transmission speed measurement precision caused by difficulty in accurately acquiring an actual spatial transmission path of an electroneurographic signal of a single nerve fiber in the existing electroneurographic signal speed measurement field. The method specifically comprises the following steps: acquiring a first electroneurographic signal based on an electroneurographic signal acquisition point; using a SpikeSorting algorithm to decompose the first electroneurographic signal of each acquisition point to obtain a second electroneurographic signal, and recording a timestamp; determining an electroneurographic signal acquisition point sequence through which a second electroneurographic signal flows based on a second electroneurographic signal timestamp of the same nerve fiber, and fitting an actual spatial propagation path; determining a timestamp difference according to the first timestamp and the second timestamp; and determining a second electroneurographic signal transmission speed based on the actual spatial propagation path and the timestamp difference. The method is used for measuring the transmission speed of the electroneurographic signals.

Description

Technical Field

[0001] This application relates to the field of neurophysiology, and in particular to a method and apparatus for measuring the transmission speed of neural electrical signals. Background Art

[0002] Neuroelectric velocity (NCV) is a key indicator for quantifying neurological function, localizing neuropathological conditions, and assessing nerve damage, and is of great significance for clinical diagnosis and treatment monitoring. Current NCV measurement techniques mostly use two-point electrodes to record mixed action potential signals from multiple nerve fibers within a nerve trunk, and often approximate the propagation path of the nerve signal with the straight-line distance between the two points, leading to inaccurate measurement results. Therefore, current NCV measurement technology suffers from the difficulty in accurately obtaining the actual spatial propagation path of the nerve signal in a single nerve fiber, resulting in insufficient measurement accuracy. Summary of the Invention

[0003] This application provides a method and apparatus for measuring the transmission speed of neural electrical signals, which solves the technical problem in the existing field of neural electrical signal velocity measurement that it is difficult to accurately obtain the actual spatial propagation path of the neural electrical signal of a single nerve fiber, thus leading to insufficient accuracy in the transmission speed measurement.

[0004] To achieve the above objectives, this application adopts the following technical solution: In a first aspect, a method for measuring the transmission speed of neural electrical signals is provided, comprising: acquiring a first neural electrical signal within a nerve trunk at each acquisition point based on neural electrical signal acquisition points on a neural electrode; decomposing the first neural electrical signal at each acquisition point using the SpikeSorting algorithm to obtain a second neural electrical signal of the same nerve fiber in the nerve trunk at each acquisition point, and recording the timestamp of the second neural electrical signal at each acquisition point; determining the sequence of neural electrical signal acquisition points through which the second neural electrical signal flows based on the timestamp of the second neural electrical signal of the same nerve fiber, and fitting the actual spatial propagation path of the second neural electrical signal of the same nerve fiber; determining the timestamp difference of the actual spatial propagation path based on the first timestamp and the second timestamp; and determining the transmission speed of the second neural electrical signal based on the actual spatial propagation path and the timestamp difference.

[0005] In conjunction with the first aspect mentioned above, in one possible implementation, the SpikeSorting algorithm includes preprocessing, detection, feature extraction, and clustering processes. The SpikeSorting algorithm decomposes the first neural electrical signal at each acquisition point to obtain the second neural electrical signal of the same nerve fiber in the nerve trunk at each acquisition point. This includes: preprocessing the first neural electrical signal at each acquisition point; the preprocessing includes filtering out low-frequency baseline drift and high-frequency noise from the first neural electrical signal using a bandpass filter; detecting the peak value of the action potential in the preprocessed signal using an adaptive thresholding method, and extracting the complete waveform segment of each action potential at each acquisition point; performing feature dimensionality reduction processing on the extracted action potential waveform segments to extract the time-domain and frequency-domain features of the waveform; the time-domain features include peak-to-peak amplitude and rising / falling edge slope, and the frequency-domain features include the dominant frequency component; calculating the waveform feature similarity of the action potentials at each acquisition point based on the time-domain and frequency-domain features; and clustering the action potentials at each acquisition point whose waveform feature similarity is higher than a preset waveform threshold into one category, with each category corresponding to the second neural electrical signal of a nerve fiber.

[0006] In conjunction with the first aspect mentioned above, in one possible implementation, the waveform feature similarity of the action potential at each acquisition point, calculated based on time-domain and frequency-domain features, satisfies the following formula:

[0007] in, For waveform feature similarity, The weighting coefficients for the peak-to-peak amplitude characteristics. This is the amplitude attenuation coefficient. The peak-to-peak amplitude of the action potential at sampling point i. The peak-to-peak amplitude of the action potential at sampling point j. The weighting coefficients for the slope trend characteristics. Let i be the rising slope of the action potential at acquisition point i. Let be the rising slope of the action potential at acquisition point j. Let i be the falling edge slope of the action potential at acquisition point i. Let be the falling edge slope of the action potential at acquisition point j. The weighting coefficients for the main frequency characteristics, The dominant frequency component of the action potential at acquisition point i. Let j be the dominant frequency component of the action potential at acquisition point j, and .

[0008] In conjunction with the first aspect mentioned above, in one possible implementation, the sequence of neural electrical signal acquisition points through which the second neural electrical signal flows is determined based on the timestamps of the second neural electrical signal of the same nerve fiber, and the actual spatial propagation path of the second neural electrical signal of the same nerve fiber is fitted. This includes: sorting the timestamps of the second neural electrical signal of the same nerve fiber at each acquisition point in ascending order to obtain the sequence of neural electrical signal acquisition points through which the second neural electrical signal flows; obtaining the three-dimensional spatial coordinates of each acquisition point in the acquisition point sequence; fitting the three dimensions of the three-dimensional spatial coordinates of the acquisition point sequence using cubic spline interpolation to obtain three-dimensional one-dimensional fitting functions; and calculating the arc length integral based on the three-dimensional one-dimensional fitting functions to obtain the continuous actual spatial propagation path.

[0009] In conjunction with the first aspect mentioned above, one possible implementation involves using cubic spline interpolation to fit the three dimensions of the three-dimensional spatial coordinates of the acquisition point sequence, resulting in a three-dimensional one-dimensional fitting function. This includes: splitting the three-dimensional spatial coordinates of the acquisition point sequence into discrete data points of x, y, and z dimensions according to timestamp order, and determining the interpolation nodes for each dimension; for each discrete data point in each dimension, constructing cubic polynomial functions as spline basis functions for the intervals formed by two adjacent interpolation nodes; and solving for the coefficients of each segment of the cubic polynomial function based on the conditions that the function values ​​at adjacent nodes are equal, the first derivative is continuous, and the second derivative is continuous. Finally, concatenating the cubic polynomial functions of each interval in the three dimensions yields the three-dimensional one-dimensional fitting function.

[0010] In conjunction with the first aspect mentioned above, in one possible implementation, the cubic polynomial function in the x-dimensional dimension satisfies the following formula:

[0011] in, For intervals in the x-dimensional dimension The function is a cubic polynomial, k=1,2,...,n-1, where n is the total number of sampling points in the sampling point sequence, and x is the independent variable in dimension x. Let x be the coordinates of the starting point of the k-th interval in the x-dimensional space. Let be the coefficient of the constant term in the k-th interval of dimension x. Let be the coefficient of the first-order term in the k-th interval of dimension x. Let be the coefficient of the quadratic term in the k-th interval of dimension x. Let be the coefficient of the cubic term in the k-th interval of dimension x.

[0012] In conjunction with the first aspect mentioned above, in one possible implementation, the arc length integral is calculated based on a three-dimensional one-dimensional fitting function, yielding a continuous actual spatial propagation path that satisfies the following formula:

[0013] in, The size of the continuous actual spatial propagation path. As the first timestamp, This is the second timestamp. Let x be a one-dimensional fitting function. Let be a one-dimensional fitting function in the y-dimension. It is a one-dimensional fitting function in the z-dimensional dimension.

[0014] In conjunction with the first aspect mentioned above, in one possible implementation, after determining the transmission speed of the second neural electrical signal based on the actual spatial propagation path and the time stamp difference, the method further includes: verifying the effectiveness of fitting the actual spatial propagation path based on the residuals between the actual coordinates of the acquisition points and the fitted path, and the SpikeSorting clustering results; if the mean residual value does not exceed the minimum axial spacing threshold of the acquisition points and there are no abnormal clustering signals, it is deemed effective; and verifying the consistency of the transmission speed based on the path segment speed and the similarity of the waveform features of the second neural electrical signal at each acquisition point; if the mean waveform similarity value is not lower than the waveform threshold and the coefficient of variation of the segment speed does not exceed the coefficient of variation threshold, it is deemed reliable.

[0015] Secondly, a device for measuring the transmission speed of neural electrical signals is provided, comprising: a communication unit and a processing unit; the communication unit is used to acquire first neural electrical signals within the nerve trunk at each acquisition point based on neural electrical signal acquisition points on neural electrodes; the processing unit is used to decompose the first neural electrical signal at each acquisition point using the SpikeSorting algorithm to obtain second neural electrical signals of the same nerve fiber in the nerve trunk at each acquisition point, and record the timestamps of the second neural electrical signals at each acquisition point; determine the sequence of neural electrical signal acquisition points through which the second neural electrical signal flows based on the timestamps of the second neural electrical signals of the same nerve fiber, and fit the actual spatial propagation path of the second neural electrical signal of the same nerve fiber; determine the timestamp difference of the actual spatial propagation path based on the first timestamp and the second timestamp; and determine the transmission speed of the second neural electrical signal based on the actual spatial propagation path and the timestamp difference.

[0016] Thirdly, a computer-readable storage medium is provided, wherein instructions are stored therein, which, when executed on a neural electrical signal transmission speed measuring device, cause the neural electrical signal transmission speed measuring device to perform the method as described in the first aspect and any possible implementation thereof.

[0017] Fourthly, this application provides a neural electrical signal transmission speed measurement device, comprising: a processor and a storage medium; the storage medium includes instructions, and the processor is configured to execute the instructions to implement the method described in the first aspect and any possible implementation thereof. This neural electrical signal transmission speed measurement device may be an electronic device or a chip within an electronic device.

[0018] Fifthly, this application provides a neural electrical signal transmission velocity measurement system, comprising: a high-density three-dimensional neural electrode and a neural electrical signal transmission velocity measurement device; wherein, the high-density three-dimensional neural electrode is used to acquire a first neural electrical signal within a nerve trunk at each acquisition point; the neural electrical signal transmission velocity measurement device is used to acquire the first neural electrical signal within a nerve trunk at each acquisition point; the SpikeSorting algorithm is used to decompose the first neural electrical signal at each acquisition point to obtain a second neural electrical signal of the same nerve fiber in the nerve trunk at each acquisition point, and the timestamp of the second neural electrical signal at each acquisition point is recorded; the sequence of neural electrical signal acquisition points through which the second neural electrical signal flows is determined based on the timestamp of the second neural electrical signal of the same nerve fiber, and the actual spatial propagation path of the second neural electrical signal of the same nerve fiber is fitted; the timestamp difference of the actual spatial propagation path is determined according to the first timestamp and the second timestamp; and the transmission velocity of the second neural electrical signal is determined based on the actual spatial propagation path and the timestamp difference.

[0019] In a sixth aspect, this application provides a computer program product containing instructions that, when run on a neural electrical signal transmission speed measuring device, cause the neural electrical signal transmission speed measuring device to perform the methods described in the first aspect and any possible implementation thereof.

[0020] This application provides a method and apparatus for measuring the transmission velocity of neural electrical signals. It can comprehensively and accurately acquire the first neural electrical signals at various acquisition points within a nerve trunk using three-dimensional high-density neural electrodes, ensuring spatial coverage and integrity of signal acquisition. Then, the SpikeSorting algorithm is used to effectively decompose the mixed signals, accurately extracting the second neural electrical signals of individual nerve fibers at each acquisition point and recording the corresponding timestamps, solving the problem of traditional techniques struggling to distinguish individual nerve fiber signals. Simultaneously, based on the timestamps of the second neural electrical signals of the same nerve fiber, the sequence of acquisition points through which the signal flows is determined, and a spatial propagation path conforming to the actual direction of the nerve fiber is fitted, avoiding the path calculation deviation caused by approximating a straight path. Finally, by combining the accurate actual spatial propagation path with the timestamp difference, high-precision measurement of the transmission velocity of neural electrical signals of individual nerve fibers is achieved, solving the technical problem in the existing field of neural electrical signal velocimetry where it is difficult to accurately obtain the actual spatial propagation path of neural electrical signals of individual nerve fibers, leading to insufficient accuracy in transmission velocity measurement.

[0021] It should be understood that the descriptions of technical features, technical solutions, beneficial effects, or similar language in this application do not imply that all features and advantages can be achieved in any single embodiment. Rather, it is understood that the description of a feature or beneficial effect means that a specific technical feature, technical solution, or beneficial effect is included in at least one embodiment. Therefore, the descriptions of technical features, technical solutions, or beneficial effects in this specification do not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions, and beneficial effects described in this embodiment can be combined in any suitable manner. Those skilled in the art will understand that embodiments can be implemented without one or more specific technical features, technical solutions, or beneficial effects of a particular embodiment. In other embodiments, additional technical features and beneficial effects may be identified in specific embodiments that do not embody all embodiments. Attached Figure Description

[0022] Figure 1 A system architecture diagram of a neural electrical signal transmission speed measurement system provided in this application embodiment; Figure 2 A flowchart illustrating a method for measuring the transmission velocity of neural electrical signals provided in an embodiment of this application; Figure 3 A flowchart illustrating another method for measuring the transmission speed of neural electrical signals provided in an embodiment of this application; Figure 4 A flowchart illustrating another method for measuring the transmission speed of neural electrical signals provided in an embodiment of this application; Figure 5 A flowchart illustrating another method for measuring the transmission speed of neural electrical signals provided in an embodiment of this application; Figure 6 This is a schematic diagram of the structure of a neural electrical signal transmission speed measuring device provided in an embodiment of this application; Figure 7 This is a schematic diagram of the hardware structure of a neural electrical signal transmission speed measuring device provided in an embodiment of this application. Detailed Implementation

[0023] In the description of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B. The "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. Furthermore, "at least one" means one or more, and "multiple" means two or more. The terms "first," "second," etc., do not limit the quantity or order of execution, and "first," "second," etc., do not necessarily imply differences.

[0024] It should be noted that, in this application, the terms "exemplary" or "for example" are used to indicate that something is being described as an example, illustration, or illustration. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or design solutions. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0025] The neural electrical signal transmission velocity measurement method provided in this application embodiment can be applied to, for example... Figure 1 The neural electrical signal transmission velocity measurement system shown includes: a high-density three-dimensional neural electrode 101 and a neural electrical signal transmission velocity measurement device 102.

[0026] The high-density three-dimensional neural electrode 101 is used to acquire the first neural electrical signal in the nerve trunk at each acquisition point; the neural electrical signal transmission speed measuring device 102 is used to acquire the first neural electrical signal in the nerve trunk at each acquisition point; the SpikeSorting algorithm is used to decompose the first neural electrical signal at each acquisition point to obtain the second neural electrical signal of the same nerve fiber in the nerve trunk at each acquisition point, and the timestamp of the second neural electrical signal at each acquisition point is recorded; the sequence of neural electrical signal acquisition points through which the second neural electrical signal flows is determined based on the timestamp of the second neural electrical signal of the same nerve fiber, and the actual spatial propagation path of the second neural electrical signal of the same nerve fiber is fitted; the timestamp difference of the actual spatial propagation path is determined according to the first timestamp and the second timestamp; and the transmission speed of the second neural electrical signal is determined based on the actual spatial propagation path and the timestamp difference.

[0027] As an example, the high-density three-dimensional neural electrode uses a soft sheet electrode wrapped around the nerve epineurium, and uses the neural electrical signal acquisition points on the soft sheet electrode to acquire the first neural electrical signal in the nerve trunk.

[0028] To address the technical problem in existing neural electrical signal velocimetry fields where it is difficult to accurately obtain the actual spatial propagation path of neural electrical signals from individual nerve fibers, leading to insufficient measurement accuracy, this application provides a method for measuring the transmission speed of neural electrical signals. This method includes: comprehensively and accurately acquiring the first neural electrical signals at each acquisition point within the nerve trunk using three-dimensional high-density neural electrodes, ensuring spatial coverage and integrity of signal acquisition; then effectively decomposing the mixed signals using the SpikeSorting algorithm to accurately extract the second neural electrical signals of individual nerve fibers at each acquisition point and recording the corresponding timestamps, solving the problem of traditional techniques failing to distinguish individual nerve fiber signals; simultaneously determining the sequence of acquisition points through which the signal flows based on the timestamps of the second neural electrical signals of the same nerve fiber, fitting a spatial propagation path that conforms to the actual direction of the nerve fiber, avoiding path calculation errors caused by approximating a straight path; finally, combining the accurate actual spatial propagation path with the timestamp difference to achieve high-precision measurement of the transmission speed of neural electrical signals from individual nerve fibers, thus solving the technical problem in existing neural electrical signal velocimetry fields where it is difficult to accurately obtain the actual spatial propagation path of neural electrical signals from individual nerve fibers, leading to insufficient measurement accuracy.

[0029] Figure 2 This is a flowchart illustrating the neural electrical signal transmission velocity measurement method provided in the embodiments of this application, as shown below. Figure 2 As shown, the method includes: Step 201: The neural electrical signal transmission speed measurement device acquires the first neural electrical signal in the nerve trunk at each acquisition point based on the neural electrical signal acquisition point on the neural electrode.

[0030] Among them, the neural electrode is a high-density three-dimensional electrode with M rows and N columns of acquisition points (such as a soft sheet electrode that can be formed into a tube around the nerve trunk), and the acquisition points are distributed in an equally spaced matrix; the first neural electrical signal is a mixed action potential signal generated by multiple nerve fibers in the nerve trunk, which contains the electrical activity information of multiple fibers.

[0031] In this embodiment, a soft sheet electrode is used to wrap the nerve epidermis, and the spacing between acquisition points can be adjusted according to the diameter of the nerve trunk. The acquisition frequency is set to 1-10kHz to capture action potential details, while being compatible with signal acquisition from different types of nerve trunks (such as motor nerves and sensory nerves), avoiding the limitations of a single deployment method.

[0032] It should be noted that during the acquisition process, it is necessary to ensure that the electrodes are in close contact with the nerve epineurium to reduce signal attenuation, while avoiding areas of nerve damage to prevent the acquisition of abnormal interference signals.

[0033] As an example, for the sciatic nerve trunk of rats, a soft sheet-like three-dimensional electrode with a 4×6 matrix was used to wrap the epineurium, with a sampling point spacing of 1 mm, and a mixed action potential signal containing the electrical activity of 20+ nerve fibers was acquired at a frequency of 5 kHz, which is the first nerve electrical signal.

[0034] Based on the above steps, high-density three-dimensional electrodes are used to achieve simultaneous acquisition of multiple acquisition points, ensuring the spatial coverage integrity of neural electrical signals.

[0035] Step 202: The neural electrical signal transmission speed measurement device uses the SpikeSorting algorithm to decompose the first neural electrical signal at each acquisition point to obtain the second neural electrical signal of the same nerve fiber in the nerve trunk at each acquisition point, and records the timestamp of the second neural electrical signal at each acquisition point.

[0036] Among them, the SpikeSorting algorithm is a signal processing algorithm used to separate mixed neural electrical signals. Its core is to distinguish the action potentials of different nerve fibers through waveform features; the second neural electrical signal is the independent action potential signal of a single nerve fiber; and the timestamp is the precise time when the signal arrives at the corresponding acquisition point.

[0037] In this embodiment, the SpikeSorting algorithm combines the waveform time-domain (e.g., peak value, slope) and frequency-domain (e.g., main frequency) features of the second neural electrical signal to separate the signal and records the timestamp through a synchronous clock. The algorithm supports dynamic adjustment of feature weights to adapt to the signal differences of different nerve fibers and is not limited to a single feature combination.

[0038] It should be noted that the timestamp records must correspond one-to-one with the location of the collection point to avoid deviations in subsequent path fitting due to clock asynchrony.

[0039] Based on the above steps, the mixed signal is accurately decomposed into a single fiber signal, solving the problem that traditional techniques cannot distinguish individual nerve fiber signals. At the same time, the time-stamped records provide time-dimensional data for subsequent path tracing.

[0040] Step 203: The neural electrical signal transmission speed measurement device determines the sequence of neural electrical signal acquisition points through which the second neural electrical signal flows based on the timestamp of the second neural electrical signal of the same nerve fiber, and fits the actual spatial propagation path of the second neural electrical signal of the same nerve fiber.

[0041] Among them, the acquisition point sequence is the order of acquisition points through which the signal of the same nerve fiber flows according to the timestamp; the actual spatial propagation path is a fitted continuous path that conforms to the natural direction of the nerve fiber.

[0042] In this embodiment, the timestamps of the second neural electrical signals of the same fiber flowing through the acquisition points are first sorted from smallest to largest, and the corresponding acquisition points are matched to form a sequence; then, the spatial position of each acquisition point in the sequence is obtained by three-dimensional coordinate acquisition, the path is fitted by cubic spline interpolation, and the size of the actual spatial propagation path is determined by calculating the fused three-dimensional path through curve arc length integral.

[0043] It should be noted that three-dimensional coordinate acquisition only requires clarifying the relative positions of different acquisition points. The origin can be selected as a specified neural electrical signal acquisition point, and the three-dimensional coordinates of all neural electrical signal acquisition points can be determined based on the size and structure of the nerve trunk after the electrode is wrapped and the distribution of neural electrical signal acquisition points.

[0044] Based on the above steps, the traditional method of approximating a straight path is abandoned, and a path that closely matches the actual direction of the nerve fiber is obtained, thus solving the problem of path calculation deviation and providing accurate path data for high-precision speed measurement.

[0045] Step 204: The neural electrical signal transmission speed measuring device determines the timestamp difference of the actual spatial propagation path based on the first timestamp and the second timestamp.

[0046] The first timestamp is the timestamp when the second neural electrical signal arrives at the starting acquisition point, and the second timestamp is the timestamp when the second neural electrical signal of the same nerve fiber arrives at the ending acquisition point; the timestamp difference is the difference between the two, that is, the time it takes for the signal to travel through the actual path.

[0047] It should be noted that the starting point and the ending point must strictly correspond to the beginning and end of the actual propagation path to avoid errors in time difference calculation due to node misalignment.

[0048] As an example, the start time stamp (first time stamp) of a certain fiber path is 10.2ms, the end time stamp (second time stamp) is 18.7ms, and the calculated timestamp difference is 8.5ms.

[0049] Based on the above steps, the precise time intervals of the signal flowing through the actual path are obtained, irrelevant time interference is eliminated, and accurate time data is provided for subsequent speed calculations.

[0050] Step 205: The neural electrical signal transmission speed measurement device determines the second neural electrical signal transmission speed based on the actual spatial propagation path and the time stamp difference.

[0051] The transmission speed is the ratio of the actual spatial propagation path length to the timestamp difference, reflecting how fast the electrical signal is transmitted in a single nerve fiber.

[0052] In this embodiment, the speed calculation adopts the formula of path length ÷ timestamp difference, which supports the calculation of overall speed (full path) and segmented speed (between multiple nodes within the path); the calculation results are filtered for outliers (such as removing speed values ​​that exceed the physiological range) to adapt to the physiological characteristics of different types of nerve fibers.

[0053] It should be noted that velocity calculations must be combined with the physiological velocity range of the nerve fiber type to avoid misjudgment due to abnormal data.

[0054] As an example, if the actual path length of a certain fiber is 8.5mm and the timestamp difference is 8.5ms, the calculated transmission speed is 8.5mm÷8.5ms=1m / s, or 1000mm / s, which is within the physiological speed range of unmyelinated fibers.

[0055] Based on the above steps, and combined with precise actual path and time difference, high-precision measurement of the transmission speed of a single nerve fiber is achieved, solving the problem of insufficient accuracy in traditional techniques.

[0056] Based on the above technical solution, the first neural electrical signals of each acquisition point within the nerve trunk can be comprehensively and accurately acquired using three-dimensional high-density neural electrodes, ensuring the spatial coverage and integrity of signal acquisition. Then, the SpikeSorting algorithm is used to effectively decompose the mixed signals, accurately extracting the second neural electrical signals of individual nerve fibers at each acquisition point and recording the corresponding timestamps, solving the problem of traditional techniques' difficulty in distinguishing individual nerve fiber signals. Simultaneously, based on the timestamps of the second neural electrical signals of the same nerve fiber, the sequence of acquisition points through which the signal flows is determined, fitting a spatial propagation path that conforms to the actual direction of the nerve fiber, avoiding the path calculation deviation caused by approximating a straight path. Finally, by combining the accurate actual spatial propagation path with the timestamp difference, high-precision measurement of the transmission speed of the neural electrical signals of individual nerve fibers is achieved, solving the technical problem in the existing field of neural electrical signal velocimetry where it is difficult to accurately obtain the actual spatial propagation path of the neural electrical signals of individual nerve fibers, leading to insufficient accuracy in transmission speed measurement.

[0057] In one possible implementation, combining the above... Figure 2 ,like Figure 3 As shown, step 202 above uses the SpikeSorting algorithm to decompose the first neural electrical signal at each acquisition point to obtain the second neural electrical signal of the same nerve fiber in the nerve trunk at each acquisition point. This process can be specifically implemented through the following steps 301-305: Step 301: The neural electrical signal transmission speed measurement device preprocesses the first neural electrical signal at each acquisition point.

[0058] The preprocessing includes filtering out low-frequency baseline drift and high-frequency noise from the first neural electrical signal using a bandpass filter.

[0059] In this embodiment, the neural electrical signal transmission speed measurement device filters out low-frequency baseline drift (such as breathing and heartbeat interference) and high-frequency noise (such as electromagnetic radiation) through a 100-3000Hz bandpass filter. At the same time, it can be combined with a signal smoothing algorithm (such as moving average) to further reduce fluctuations and adapt to the first neural electrical signal of different intensities (narrowband filtering for strong signals and broadband filtering for weak signals), avoiding the limitations of a single processing method.

[0060] As an example, the first nerve electrical signal of the rat sciatic nerve was filtered using a 150-2500Hz bandpass filter and then smoothed using a 5-point moving average to eliminate baseline drift below 3Hz and electromagnetic noise above 3500Hz.

[0061] Based on the above steps, signal interference is reduced, providing a clear signal basis for subsequent action potential detection and avoiding signal misjudgment caused by noise.

[0062] Step 302: The neural electrical signal transmission velocity measurement device uses an adaptive threshold method to detect the peak value of the action potential in the preprocessed signal and extracts the complete waveform segment of each action potential at each acquisition point.

[0063] Among them, the adaptive threshold method refers to the threshold being dynamically adjusted according to the real-time amplitude of the signal, and the peak value of the action potential is the local maximum value of the signal exceeding the threshold; the complete waveform segment usually includes the signal before and after the peak (such as 1ms before the peak and 2ms after the peak) to ensure coverage of the entire cycle of the action potential.

[0064] In this embodiment, the threshold can be set to 2-4 times the standard deviation within the signal sliding window (window duration 0.5-1s). When the signal amplitude exceeds the threshold and is a local maximum value, it is determined to be a peak value. The duration of the extracted waveform segment can be adjusted according to the action potential width (e.g., 3ms before and after a wide waveform, and 1.5ms before and after a narrow waveform) to adapt to the differences in action potentials of different nerve fibers.

[0065] Based on the above steps, the peak value of the action potential can be accurately located and the waveform information can be completely extracted.

[0066] Step 303: The neural electrical signal transmission velocity measurement device performs feature dimensionality reduction processing on the extracted action potential waveform segments to extract the time domain features and frequency domain features of the waveform.

[0067] Among them, feature dimensionality reduction refers to reducing the dimensionality of waveform features through algorithms to avoid redundant features increasing the amount of computation; time-domain features reflect the time-domain shape of the waveform, and frequency-domain features reflect the frequency distribution of the waveform, both of which are key bases for distinguishing signals from different nerve fibers.

[0068] In the embodiments of this application, feature dimensionality reduction can be achieved using the principal component analysis (PCA) algorithm, retaining more than 85% of the feature information; time-domain features can be extracted by peak-to-peak amplitude, rising / falling edge slope, and peak delay; frequency-domain features can be extracted by the main frequency components and frequency bandwidth through fast Fourier transform (FFT), and are not limited to a single feature combination. The specific process of using the principal component analysis algorithm is not specifically limited in this application.

[0069] It should be noted that the dimensionality reduction process should retain core distinguishing features (such as dominant frequency and peak-to-peak amplitude) to avoid the loss of key information that could lead to subsequent clustering bias.

[0070] Based on the above steps, the feature extraction operation can simplify the feature dimensions and highlight the key differences in the waveform.

[0071] Step 304: The neural electrical signal transmission speed measurement device calculates the waveform feature similarity of the action potential at each acquisition point based on time domain features and frequency domain features.

[0072] Among them, waveform feature similarity is an indicator that quantifies the similarity of different action potential waveforms (with a value of 0-1, the closer to 1 the higher the similarity). The core is to integrate time domain (amplitude, slope) and frequency domain (main frequency) features to determine whether they come from the same nerve fiber.

[0073] In this embodiment, the neural electrical signal transmission speed measurement device can first normalize the time domain and frequency domain features (e.g., map them to the 0-1 interval), and then perform fusion calculations according to preset weights; the weights can be adjusted according to the type of neural fiber (e.g., thick fibers emphasize amplitude, and thin fibers emphasize main frequency) to ensure that the similarity calculation is adapted to the signal characteristics of different fibers.

[0074] Optionally, the waveform feature similarity of the action potential at each acquisition point, calculated based on time-domain and frequency-domain features, satisfies the following formula:

[0075] in, For waveform feature similarity, The weighting coefficients for the peak-to-peak amplitude characteristics. This is the amplitude attenuation coefficient. The peak-to-peak amplitude of the action potential at sampling point i. The peak-to-peak amplitude of the action potential at sampling point j. The weighting coefficients for the slope trend characteristics. Let i be the rising slope of the action potential at acquisition point i. Let be the rising slope of the action potential at acquisition point j. Let i be the falling edge slope of the action potential at acquisition point i. Let be the falling edge slope of the action potential at acquisition point j. The weighting coefficients for the main frequency characteristics, The dominant frequency component of the action potential at acquisition point i. Let j be the dominant frequency component of the action potential at acquisition point j, and .

[0076] As an example, =0.4, =0.3, =0.3, =2, =4.5mV, =4.3mV, =3mV / ms =2.8mV / ms, =2.2mV / ms, =2.1mV / ms, =1.2kH, =1.1kH, the calculated amplitude term is 0.366, the slope term is 0.297, the frequency term is 0.287, and the total similarity is... 0.95.

[0077] It should be noted that the core function of this formula is to accurately quantify the waveform similarity of action potentials at different acquisition points through multi-dimensional feature weighted fusion, providing a reliable quantitative basis for subsequent clustering and separation of single-fiber signals. From a structural design perspective, the amplitude term uses an exponential function, which can smoothly attenuate the peak-to-peak amplitude difference, avoiding a sharp drop in similarity due to small amplitude fluctuations; the slope term uses a cosine similarity form, which can effectively measure the trend consistency of the rising / falling edge slopes, adapting to the temporal morphological differences of action potentials in different nerve fibers; the frequency term is calculated using the minimum dominant frequency proportion, highlighting the matching degree of the dominant frequency component and compensating for the insufficient differentiation of high-frequency signals by time-domain features. Simultaneously, the design of a weight coefficient sum of 1 allows for flexible adjustment of the contribution of each feature to the similarity, adapting to the signal characteristics of different types of nerve fibers, such as coarse fibers (more significant amplitude features) and fine fibers (more prominent frequency features), ultimately achieving multi-dimensional collaborative discrimination in the "time domain + frequency domain," which is more accurate than single-feature judgment and effectively avoids misclassifying different fiber signals as the same type or the same fiber signal as different types.

[0078] Based on the above steps, the waveform similarity is quantified by multi-feature fusion to accurately distinguish the action potentials of different nerve fibers, providing a quantitative basis for subsequent clustering.

[0079] Step 305: The neural electrical signal transmission speed measuring device clusters action potentials with waveform feature similarity higher than a preset waveform threshold at each acquisition point into one category, and each category corresponds to the second neural electrical signal of a nerve fiber.

[0080] Clustering refers to the algorithm process of grouping action potentials with high similarity; the preset waveform threshold is the similarity threshold for determining "same type"; each type of signal corresponds to an independent action potential (second neural electrical signal) of a single nerve fiber.

[0081] In the embodiments of this application, clustering can be performed using the K-means algorithm. First, the number of clusters (e.g., 5-8) is determined by the elbow rule. Then, action potentials with similarity higher than a threshold (e.g., 0.85) are classified into the same cluster. The threshold can be dynamically adjusted (e.g., reduced to 0.8 when noise is high, and increased to 0.9 when noise is low) to adapt to different acquisition environments. The specific clustering process of the K-means algorithm is not specifically limited in this application.

[0082] Based on the above steps, the precise separation of mixed action potentials into independent signals of single fibers is achieved, solving the problem that traditional techniques cannot distinguish signals from individual nerve fibers.

[0083] Based on the above technical solution, by preprocessing the first neural electrical signal to eliminate low-frequency drift and high-frequency noise, adaptive threshold detection to extract the complete action potential waveform, feature dimensionality reduction to highlight key time-domain and frequency-domain features, multi-feature fusion to calculate waveform similarity, and clustering to separate high-similarity signals, the mixed action potential is accurately decomposed into independent signals of a single nerve fiber, effectively solving the problem that traditional techniques are difficult to distinguish signals of a single nerve fiber.

[0084] In one possible implementation, combining the above... Figure 2 ,like Figure 4 As shown, step 203 above, which determines the sequence of neural electrical signal acquisition points through which the second neural electrical signal flows based on the timestamp of the second neural electrical signal of the same nerve fiber, and fits the actual spatial propagation path of the second neural electrical signal of the same nerve fiber, can be specifically implemented through the following steps 401-406: Step 401: The neural electrical signal transmission speed measuring device sorts the timestamps of the second neural electrical signal of the same nerve fiber at each acquisition point in ascending order to obtain the sequence of neural electrical signal acquisition points through which the second neural electrical signal flows.

[0085] Among them, the acquisition point sequence is the order in which the signal of the same nerve fiber flows through the acquisition points in chronological order, reflecting the direction of signal propagation on the nerve trunk.

[0086] In this embodiment, the correspondence between timestamps and acquisition points needs to be verified before sorting (e.g., removing timestamps of acquisition points that have not recorded signals). It supports filtering valid sequences by "time difference threshold" (e.g., the difference between adjacent timestamps is <5ms to avoid interference from abnormal time points) and adapts to the differences in signal propagation speed of different nerve fibers.

[0087] It should be noted that if multiple timestamps are recorded on the same fiber at the same collection point, the first arrival time should be taken (to avoid duplicate counting) to ensure the uniqueness of the sequence.

[0088] As an example, the timestamps of a certain nerve fiber are acquisition points A (10.2ms), C (12.5ms), and B (11.3ms). After sorting, the acquisition point sequence A→B→C is obtained, which corresponds to the signal propagation direction.

[0089] Based on the above steps, the spatiotemporal order of signal propagation is determined through spatiotemporal relationships, thus avoiding misjudgment of path direction.

[0090] Step 402: The neural electrical signal transmission speed measuring device acquires the three-dimensional spatial coordinates of each acquisition point in the acquisition point sequence.

[0091] Among them, the three-dimensional spatial coordinates are the (x, y, z) position data of the acquisition point in physical space.

[0092] In this embodiment, matrix coordinates can be obtained by preset electrode coordinates (e.g., 4×6 matrix electrodes, with 1mm spacing between adjacent acquisition points in the x / y direction and the z-direction conforming to the arc of the nerve trunk). If the electrode wraps around the nerve trunk, the z-direction coordinate is adjusted according to the arc of the nerve trunk to ensure conformity to the physiological shape, and is not limited to planar coordinates.

[0093] It should be noted that the origin of the coordinate system can be set as the starting point of the sequence to simplify subsequent calculations, and the coordinates of all sampling points must be based on the same coordinate system.

[0094] Based on the above steps, the physical spatial location of the collection points is obtained, providing three-dimensional data support for subsequent path fitting.

[0095] Step 403: The neural electrical signal transmission speed measurement device splits the three-dimensional spatial coordinates of the acquisition point sequence into discrete data points of three dimensions, x, y, and z, according to the timestamp order, and determines the interpolation nodes of each dimension.

[0096] Here, discrete data points are individual coordinate values ​​for each dimension; interpolation nodes are key coordinate points in each dimension used to construct the fitting function, that is, the coordinates of each dimension corresponding to the collection points in the sequence.

[0097] In this embodiment, the splitting is strictly in the order of the timestamps of the collection point sequence to ensure that the data points in the x / y / z dimensions correspond one-to-one (e.g., point A (x1,y1,z1) corresponds to time t1, and point B (x2,y2,z2) corresponds to time t2); the number of interpolation nodes is consistent with the length of the collection point sequence, and the node density can be dynamically adjusted according to the sequence length (e.g., one auxiliary node is added for every 3 points in a long sequence).

[0098] As an example, the collection point sequence A(0,0,0), B(1,0,0.2), C(2,0,0.3) is split into x-dimensional [0,1,2], y-dimensional [0,0,0], and z-dimensional [0,0.2,0.3], with the interpolation nodes for each dimension being the corresponding discrete values.

[0099] Based on the above steps, the three-dimensional coordinates are decomposed into single-dimensional data, providing independent dimension processing data for subsequent multi-dimensional fitting functions and simplifying the complexity of high-dimensional fitting.

[0100] Step 404: The neural electrical signal transmission speed measurement device constructs a cubic polynomial function as a spline basis function for each discrete data point in each dimension, for the interval formed by two adjacent interpolation nodes.

[0101] Among them, the cubic polynomial function is a piecewise function that describes the change of a single-dimensional coordinate over time; the spline basis function refers to the independent cubic polynomial within the interval of adjacent nodes, ensuring the continuity of the interval connection.

[0102] In the embodiments of this application, when constructing a cubic polynomial function, the neural electrical signal transmission speed measurement device first determines the number of intervals in each dimension (e.g., n interpolation nodes correspond to n-1 intervals). The cubic polynomial function of each interval can be initially constructed by "node coordinates + smoothing constraints". It supports adjusting the function order according to the direction of the nerve fiber (e.g., using a linear function for straight segments and a cubic function for curved segments), and is not limited to a single cubic polynomial.

[0103] Optionally, the cubic polynomial function in the x-dimensional dimension satisfies the following formula:

[0104] in, For intervals in the x-dimensional dimension The function is a cubic polynomial, k=1,2,...,n-1, where n is the total number of sampling points in the sampling point sequence, and x is the independent variable in dimension x. Let x be the coordinates of the starting point of the k-th interval in the x-dimensional space. Let be the coefficient of the constant term in the k-th interval of dimension x. Let be the coefficient of the first-order term in the k-th interval of dimension x. Let be the coefficient of the quadratic term in the k-th interval of dimension x. Let be the coefficient of the cubic term in the k-th interval of dimension x.

[0105] As an example: k=1 (the first interval [0,1] in the x dimension). =0 (coordinates of the starting point of the interval), assuming =0, =1, =0.5, --0.1, Constructor function For the x-dimensional interval [0,1], substitute x=0.5 and calculate... , which corresponds to the fitted coordinates at that position in the x-dimensional direction.

[0106] It should be noted that the coefficients in the formula need to be solved in conjunction with the continuity conditions of adjacent intervals. The coefficients are only temporarily set as examples, and subsequent step 405 will be calibrated by constraints.

[0107] Based on the above steps, a piecewise fitting basis is constructed for each dimension. The curvature of the nerve fiber is adapted by a cubic polynomial, avoiding the path deviation of traditional linear fitting.

[0108] Step 405: Based on the conditions that the function values ​​at adjacent nodes are equal, the first derivative is continuous, and the second derivative is continuous, the neural electrical signal transmission speed measurement device solves for the coefficients of each segment of the cubic polynomial function, and splices the cubic polynomial functions of each interval in the three dimensions to obtain a three-dimensional one-dimensional fitting function.

[0109] Among them, the continuity condition is a constraint that ensures the smooth connection of the fitting functions in each interval by solving the coefficients of the cubic polynomial function; the one-dimensional fitting function is a complete function after splicing the dimensions, which can describe the continuous change of the coordinates of that dimension over time.

[0110] In this embodiment of the application, when the neural electrical signal transmission velocity measurement device solves for the coefficients, it first takes the x-dimensional dimension as an example, for interval 1 ([x1,x2]) and interval 2 ([x2,x3]), let... (Equal function values) (The first derivative is continuous) (The second derivative is continuous), establish a system of equations and solve for the coefficients; the same applies to the y / z dimensions, and finally obtain complete one-dimensional fitting functions for each dimension.

[0111] It should be noted that if a certain dimension is a straight line (such as the y-coordinate remains unchanged), it can be simplified to a linear function (second derivative = 0), reducing the amount of computation.

[0112] As an example, the coefficients for interval 1 ([0,1]) in the x dimension are a1=0, b1=1.05, c1=0.4, d1=-0.05, and the coefficients for interval 2 ([1,2]) are a2=1, b2=1.05, c2=0.3, d2=-0.02. After splicing, the x-dimensional fitting function satisfies the continuity of function value and first / second derivative at x=1, forming a complete one-dimensional fitting function.

[0113] Based on the above steps, a smooth and continuous one-dimensional fitting function is obtained, ensuring that there are no breaks in the path of each dimension.

[0114] Step 406: The neural electrical signal transmission velocity measurement device performs arc length integral calculation based on a three-dimensional single-dimensional fitting function to obtain a continuous actual spatial propagation path.

[0115] Among them, arc length integral is a mathematical method for calculating the length of a continuous curve in three-dimensional space; the actual spatial propagation path is a continuous path that follows the direction of nerve fibers, and its length is the result of integration.

[0116] In this embodiment, the integration range is from the first timestamp to the second timestamp. First, the single-dimensional fitting functions of each dimension are combined with the timestamp to transform them into "coordinate-time" functions (such as X(t), Y(t), Z(t)), and then the path length is calculated using the arc length integral formula.

[0117] Optionally, the arc length integral is calculated based on the three-dimensional one-dimensional fitting function to obtain the continuous actual spatial propagation path that satisfies the following formula:

[0118] in, The size of the continuous actual spatial propagation path. As the first timestamp, This is the second timestamp. Let x be a one-dimensional fitting function. Let be a one-dimensional fitting function in the y-dimension. It is a one-dimensional fitting function in the z-dimensional dimension.

[0119] It should be noted that the progress of calculating the actual spatial propagation path will increase with the increase of the number of collection points. That is, the higher the density of neural electrical signal collection points used, the more accurate the calculated path will be.

[0120] Based on the above steps, the length of a continuous path that conforms to the actual direction of the nerve fiber can be obtained by arc length integral calculation, thus solving the problem of path calculation deviation caused by traditional straight line approximation.

[0121] Based on the above technical solution, the sequence of acquisition points is determined by sorting the timestamps of the second nerve electrical signals of the same nerve fiber, thus clarifying the spatiotemporal order of signal propagation. The three-dimensional spatial coordinates of the acquisition points are obtained by combining the characteristics of three-dimensional high-density electrodes, laying the foundation for physical location. The interpolation nodes are determined by splitting the dimensions, cubic polynomial spline basis functions are constructed, and the coefficients are solved by continuous conditions to obtain a smoothly connected one-dimensional fitting function. Finally, the continuous spatial propagation path that fits the actual direction of the nerve fiber is calculated by arc length integral, which solves the problem of path calculation deviation caused by the approximation of the path by a straight line in traditional technology.

[0122] In one possible implementation, combining the above... Figure 2 ,like Figure 5As shown, after determining the transmission speed of the second neural electrical signal based on the actual spatial propagation path and the timestamp difference in step 205 above, the method of this application further includes verification of the reliability and validity of the measurement transmission speed result. The verification process can be specifically implemented through the following steps 501-502: Step 501: The neural electrical signal transmission speed measurement device verifies the effectiveness of fitting the actual spatial propagation path based on the residual between the actual coordinates of the acquisition point and the fitted path, and the SpikeSorting clustering results. If the mean residual value does not exceed the minimum axial spacing threshold of the acquisition point and there are no abnormal clustering signals, it is deemed effective.

[0123] Among them, residual refers to the distance difference between the actual three-dimensional coordinates of the acquisition point and the corresponding position on the fitting path; abnormal clustering signal refers to the grouping in SpikeSorting clustering with a sample number <3 or waveform similarity <0.7.

[0124] In this embodiment, the neural electrical signal transmission velocity measurement device sets the minimum axial spacing threshold of the acquisition points as the minimum spacing between the acquisition points of the three-dimensional high-density electrode in the x / y / z directions. The residual is calculated using Euclidean distance, and the mean residual is the arithmetic mean of the residuals of all acquisition points. The clustering result verification requires checking each group one by one, removing abnormal clusters, and re-evaluating the fitted path. If the mean residual does not exceed the set minimum axial spacing threshold of the acquisition points and there are no abnormal clustering signals, it is considered valid.

[0125] As an example, the residuals of the sampling points of a certain fitted path are 0.12mm, 0.18mm, and 0.15mm (mean 0.15mm), the minimum axial spacing of the sampling points is 1mm (threshold 0.2mm), and SpikeSorting clustering has no groups with a sample size <3 or similarity <0.7, so the path fitting is deemed effective.

[0126] Based on the above steps, path fitting bias and signal clustering anomalies are eliminated through two-dimensional verification to ensure that the actual spatial propagation path conforms to the true direction of the nerve fiber.

[0127] Step 502: The neural electrical signal transmission speed measurement device verifies the consistency of transmission speed based on the path segment speed and the similarity of the waveform features of the second neural electrical signal at each acquisition point. If the mean waveform similarity is not lower than the waveform threshold and the coefficient of variation of the segment speed does not exceed the coefficient of variation threshold, it is considered reliable.

[0128] Among them, the path segmentation speed refers to the speed of each segment when the actual spatial propagation path is divided into multiple segments (such as A→B, B→C) according to the sequence of acquisition points; the mean of waveform feature similarity is the arithmetic mean of the similarity between the second neural electrical signals of each acquisition point; and the coefficient of variation is the ratio of the standard deviation of the segmentation speed to the mean.

[0129] In this embodiment, the number of path segments is consistent with the length of the acquisition point sequence (n acquisition points are divided into n-1 segments), and the similarity mean calculation needs to cover all adjacent acquisition point pairs; the waveform threshold is set to 0.8, and the coefficient of variation threshold is set to 0.1 (to ensure small speed fluctuations), and the threshold can be fine-tuned according to the nerve fiber type (such as myelinated / unmyelinated).

[0130] As an example, the actual path is divided into two segments with segment speeds of 1.2 m / s and 1.18 m / s respectively (mean 1.19 m / s, standard deviation 0.014 m / s, coefficient of variation ≈ 0.012 < 0.1). The similarity of the second signal at each acquisition point is 0.85 and 0.88 respectively (mean 0.86 ≥ 0.8), indicating that the transmission speed is reliable.

[0131] Based on the above steps, speed anomalies caused by accidental factors are eliminated through dual-dimensional verification of speed fluctuations and signal homology, ensuring that the measurement results are stable and reliable.

[0132] Based on the above technical solution, the effectiveness and reliability of the measurement results of neural electrical signal transmission speed are verified to ensure that the results are accurate and consistent with physiological reality.

[0133] The foregoing mainly describes the solutions of the embodiments of this application from the perspective of device implementation. It is understood that each device, such as a neural electrical signal transmission speed measuring device, includes at least one of the hardware structures and software modules corresponding to the execution of each function in order to achieve the above-mentioned functions. Those skilled in the art should readily recognize that, in conjunction with the units and algorithm steps of the various examples described in the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware 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.

[0134] This application embodiment can divide the neural electrical signal transmission velocity measurement device into functional units according to the above method example. For example, each function can be divided into separate functional units, or two or more functions can be integrated into one processing unit. The integrated unit can be implemented in hardware or as a software functional unit. It should be noted that the unit division in this application embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0135] When using integrated units, Figure 6A possible structural schematic diagram of the neural electrical signal transmission speed measuring device (referred to as neural electrical signal transmission speed measuring device 60) involved in the above embodiments is shown. The neural electrical signal transmission speed measuring device 60 includes a processing unit 601 and a communication unit 602, and may also include a storage unit 603. Figure 6 The schematic diagram shown can be used to illustrate the structure of the neural electrical signal transmission speed measuring device involved in the above embodiments.

[0136] when Figure 6 The schematic diagram shown illustrates the structure of the neural electrical signal transmission speed measuring device involved in the above embodiments. The processing unit 601 is used to control and manage the operation of the neural electrical signal transmission speed measuring device, the communication unit 602 is used for the neural electrical signal transmission speed measuring device to communicate with other devices, and the storage unit 603 is used to store the program code and data of the neural electrical signal transmission speed measuring device.

[0137] For example, the communication unit 602 is used to acquire the first neural electrical signal in the nerve trunk at each acquisition point based on the neural electrical signal acquisition point on the neural electrode; The processing unit 601 is used to decompose the first neural electrical signal at each acquisition point using the SpikeSorting algorithm to obtain the second neural electrical signal of the same nerve fiber in the nerve trunk at each acquisition point, and record the timestamp of the second neural electrical signal at each acquisition point; determine the sequence of neural electrical signal acquisition points through which the second neural electrical signal flows based on the timestamp of the second neural electrical signal of the same nerve fiber, and fit the actual spatial propagation path of the second neural electrical signal of the same nerve fiber; determine the timestamp difference of the actual spatial propagation path based on the first timestamp and the second timestamp; and determine the transmission speed of the second neural electrical signal based on the actual spatial propagation path and the timestamp difference.

[0138] In one possible implementation, the processing unit 601 is further configured to decompose the first neural electrical signal at each acquisition point using the SpikeSorting algorithm to obtain the second neural electrical signal of the same nerve fiber in the nerve trunk at each acquisition point, including: preprocessing the first neural electrical signal at each acquisition point; the preprocessing includes filtering out low-frequency baseline drift and high-frequency noise of the first neural electrical signal using a bandpass filter; detecting the peak value of the action potential in the preprocessed signal using an adaptive thresholding method, and extracting the complete waveform segment of each action potential at each acquisition point; performing feature dimensionality reduction processing on the extracted action potential waveform segments to extract the time-domain features and frequency-domain features of the waveform; the time-domain features include peak-to-peak amplitude and rising / falling slope, and the frequency-domain features include the main frequency component; calculating the waveform feature similarity of the action potential at each acquisition point based on the time-domain features and frequency-domain features; and grouping the action potentials at each acquisition point with a waveform feature similarity higher than a preset waveform threshold into one category by clustering, with each category corresponding to the second neural electrical signal of a nerve fiber.

[0139] In one possible implementation, the waveform feature similarity of the action potential at each acquisition point, calculated based on time-domain and frequency-domain features, satisfies the following formula:

[0140] in, For waveform feature similarity, The weighting coefficients for the peak-to-peak amplitude characteristics. This is the amplitude attenuation coefficient. The peak-to-peak amplitude of the action potential at sampling point i. The peak-to-peak amplitude of the action potential at sampling point j. The weighting coefficients for the slope trend characteristics. Let i be the rising slope of the action potential at acquisition point i. Let be the rising slope of the action potential at acquisition point j. Let i be the falling edge slope of the action potential at acquisition point i. Let be the falling edge slope of the action potential at acquisition point j. The weighting coefficients for the main frequency characteristics, The dominant frequency component of the action potential at acquisition point i. Let j be the dominant frequency component of the action potential at acquisition point j, and .

[0141] In one possible implementation, the processing unit 601 is further configured to determine the sequence of neural electrical signal acquisition points through which the second neural electrical signal flows based on the timestamps of the second neural electrical signal of the same nerve fiber, and to fit the actual spatial propagation path of the second neural electrical signal of the same nerve fiber, including: sorting the timestamps of the second neural electrical signal of the same nerve fiber at each acquisition point in ascending order to obtain the sequence of neural electrical signal acquisition points through which the second neural electrical signal flows; obtaining the three-dimensional spatial coordinates of each acquisition point in the acquisition point sequence; fitting the three dimensions of the three-dimensional spatial coordinates of the acquisition point sequence using cubic spline interpolation to obtain a three-dimensional one-dimensional fitting function; and performing arc length integral calculation based on the three-dimensional one-dimensional fitting function to obtain a continuous actual spatial propagation path.

[0142] In one possible implementation, the processing unit 601 is further configured to use cubic spline interpolation to fit the three dimensions of the three-dimensional spatial coordinates of the acquisition point sequence to obtain a three-dimensional one-dimensional fitting function. This includes: splitting the three-dimensional spatial coordinates of the acquisition point sequence into discrete data points of three dimensions (x, y, and z) according to the timestamp order, and determining the interpolation nodes for each dimension; for each discrete data point in each dimension, constructing a cubic polynomial function as a spline basis function for the interval formed by two adjacent interpolation nodes; and solving for the coefficients of each segment of the cubic polynomial function based on the conditions that the function values ​​at adjacent nodes are equal, the first derivative is continuous, and the second derivative is continuous, and concatenating the cubic polynomial functions of each interval in the three dimensions to obtain a three-dimensional one-dimensional fitting function.

[0143] In one possible implementation, the cubic polynomial function in the x-dimensional dimension satisfies the following formula:

[0144] in, For intervals in the x-dimensional dimension The function is a cubic polynomial, k=1,2,...,n-1, where n is the total number of sampling points in the sampling point sequence, and x is the independent variable in dimension x. Let x be the coordinates of the starting point of the k-th interval in the x-dimensional space. Let be the coefficient of the constant term in the k-th interval of dimension x. Let be the coefficient of the first-order term in the k-th interval of dimension x. Let be the coefficient of the quadratic term in the k-th interval of dimension x. Let be the coefficient of the cubic term in the k-th interval of dimension x.

[0145] In one possible implementation, the arc length integral is calculated based on a three-dimensional one-dimensional fitting function, yielding a continuous actual spatial propagation path that satisfies the following formula:

[0146] in, The size of the continuous actual spatial propagation path. As the first timestamp, This is the second timestamp. Let x be a one-dimensional fitting function. Let be a one-dimensional fitting function in the y-dimension. It is a one-dimensional fitting function in the z-dimensional dimension.

[0147] In one possible implementation, the processing unit 601 is further configured to, after determining the transmission speed of the second neural electrical signal based on the difference between the actual spatial propagation path and the timestamp, further include: verifying the effectiveness of fitting the actual spatial propagation path based on the residual between the actual coordinates of the acquisition point and the fitted path, and the SpikeSorting clustering results; if the mean residual value does not exceed the minimum axial spacing threshold of the acquisition point and there are no abnormal clustering signals, it is deemed effective; and verifying the consistency of the transmission speed based on the segmented speed of the path and the similarity of the waveform features of the second neural electrical signal at each acquisition point; if the mean waveform similarity value is not lower than the waveform threshold and the coefficient of variation of the segmented speed does not exceed the coefficient of variation threshold, it is deemed reliable.

[0148] The processing unit 601 can be a processor or a controller, and the communication unit 602 can be a communication interface, transceiver, transceiver circuit, transceiver device, etc. The term "communication interface" is a general term and may include one or more interfaces. The storage unit 603 can be a memory. When the neural electrical signal transmission speed measuring device 60 is a chip, the processing unit 601 can be a processor or a controller, and the communication unit 602 can be an input interface and / or an output interface, pins, or circuits, etc. The storage unit 603 can be a storage unit within the chip (e.g., a register, cache, etc.) or a storage unit located outside the chip (e.g., read-only memory (ROM), random access memory (RAM, etc.)).

[0149] The communication unit can also be called a transceiver unit. The antenna and control circuit with transceiver functions in the neural electrical signal transmission speed measuring device 60 can be considered as the communication unit 602 of the neural electrical signal transmission speed measuring device 60, and the processor with processing functions can be considered as the processing unit 601 of the neural electrical signal transmission speed measuring device 60. Optionally, the device in the communication unit 602 that implements the receiving function can be considered as the communication unit. The communication unit is used to execute the receiving steps in the embodiments of this application, and the communication unit can be a receiver, a receiver circuit, etc. The device in the communication unit 602 that implements the transmitting function can be considered as the transmitting unit. The transmitting unit is used to execute the transmitting steps in the embodiments of this application, and the transmitting unit can be a transmitter, a transmitter, a transmitting circuit, etc.

[0150] Figure 6 If the integrated units in the process are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, in essence, or the parts that contribute to the prior art, or all or part of the technical solutions, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. Storage media for storing computer software products include various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory, random access memory, magnetic disks, or optical disks.

[0151] Figure 6 The units in the process can also be called modules; for example, a processing unit can be called a processing module.

[0152] This application also provides a hardware structure diagram of a neural electrical signal transmission velocity measuring device (denoted as neural electrical signal transmission velocity measuring device 70), see [link to diagram]. Figure 7 The neural electrical signal transmission speed measuring device 70 includes a processor 701, and optionally, a memory 702 connected to the processor 701.

[0153] In the first possible implementation, see Figure 7 The neural electrical signal transmission speed measuring device 70 also includes a transceiver 703. The processor 701, memory 702, and transceiver 703 are connected via a bus. The transceiver 703 is used to communicate with other devices or communication networks. Optionally, the transceiver 703 may include a transmitter and a receiver. The device in the transceiver 703 that implements the receiving function can be considered as a receiver, which is used to perform the receiving steps in the embodiments of this application. The device in the transceiver 703 that implements the transmitting function can be considered as a transmitter, which is used to perform the transmitting steps in the embodiments of this application.

[0154] Based on the first possible implementation method Figure 7 The schematic diagram shown can be used to illustrate the structure of the neural electrical signal transmission speed measuring device involved in the above embodiments.

[0155] in, Figure 7 The diagram can also illustrate the system chip in the neural electrical signal transmission velocity measurement device. In this case, the actions performed by the aforementioned neural electrical signal transmission velocity measurement device can be implemented by this system chip; the specific actions performed are described above and will not be repeated here.

[0156] In implementation, each step of the method provided in this embodiment can be completed by integrated logic circuits in the processor hardware or by instructions in software. The steps of the method disclosed in the embodiments of this application can be directly manifested as being executed by a hardware processor, or being executed by a combination of hardware and software modules in the processor.

[0157] The processor in this application may include, but is not limited to, at least one of the following: a central processing unit (CPU), a microprocessor, a digital signal processor (DSP), a microcontroller unit (MCU), or an artificial intelligence processor, etc., which are various computing devices that run software. Each computing device may include one or more cores for executing software instructions to perform calculations or processing. The processor may be a separate semiconductor chip or integrated with other circuits into a single semiconductor chip. For example, it may be integrated with other circuits (such as encoding / decoding circuits, hardware acceleration circuits, or various bus and interface circuits) to form a System-on-a-Chip (SoC), or it may be integrated as a built-in processor within an ASIC. The ASIC with the integrated processor may be packaged separately or together with other circuits. In addition to the cores for executing software instructions to perform calculations or processing, the processor may further include necessary hardware accelerators, such as field-programmable gate arrays (FPGAs), programmable logic devices (PLDs), or logic circuits that implement dedicated logic operations.

[0158] The memory in the embodiments of this application may include at least one of the following types: read-only memory (ROM) or other types of static storage devices capable of storing static information and instructions; random access memory (RAM) or other types of dynamic storage devices capable of storing information and instructions; or electrically erasable programmable-only memory (EEPROM). In some scenarios, the memory may also be compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto.

[0159] This application also provides a computer-readable storage medium including instructions that, when run on a computer, cause the computer to perform any of the methods described above.

[0160] This application also provides a computer program product containing instructions that, when run on a computer, cause the computer to perform any of the methods described above.

[0161] This application also provides a chip including a processor and an interface circuit. The interface circuit is coupled to the processor. The processor is used to run computer programs or instructions to implement the above-described method. The interface circuit is used to communicate with other modules outside the chip.

[0162] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented using software programs, implementation can be, in whole or in part, in the form of a computer program product. This computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device containing one or more servers, data centers, etc., that can be integrated with the medium. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state drives (SSDs)).

[0163] Although this application has been described herein in conjunction with various embodiments, those skilled in the art, by reviewing the accompanying drawings, disclosure, and appended claims, will understand and implement other variations of the disclosed embodiments in carrying out the claimed application. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude multiple instances. A single processor or other unit can implement several functions listed in the claims. While different dependent claims may recite certain measures, this does not mean that these measures cannot be combined to produce good results.

[0164] Although this application has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made thereto without departing from the spirit and scope of this application. Accordingly, this specification and drawings are merely exemplary illustrations of this application as defined by the appended claims, and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from the spirit and scope of this application. Thus, if such modifications and modifications of this application fall within the scope of the claims of this application and their equivalents, this application is also intended to include such modifications and modifications.

Claims

1. A method for measuring the transmission velocity of neural electrical signals, characterized in that, include: The first neural electrical signal in the nerve trunk at each acquisition point is obtained based on the neural electrical signal acquisition point on the neural electrode. The neural electrode is a high-density three-dimensional electrode with an M-row N-column matrix of equally spaced neural electrical signal acquisition points; the first neural electrical signal is a mixed action potential signal generated by multiple nerve fibers within the nerve trunk. The SpikeSorting algorithm is used to decompose the first neural electrical signal at each acquisition point to obtain the second neural electrical signal of the same nerve fiber in the nerve trunk at each acquisition point, and the timestamp of the second neural electrical signal at each acquisition point is recorded; the second neural electrical signal is an independent action potential signal of a single nerve fiber; the timestamp includes a first timestamp and a second timestamp, the first timestamp is the timestamp when the second neural electrical signal arrives at the starting acquisition point, and the second timestamp is the timestamp when the second neural electrical signal of the same nerve fiber arrives at the ending acquisition point; Based on the timestamp of the second nerve electrical signal of the same nerve fiber, determine the sequence of nerve electrical signal acquisition points through which the second nerve electrical signal flows, and fit the actual spatial propagation path of the second nerve electrical signal of the same nerve fiber; The timestamp difference of the actual spatial propagation path is determined based on the first timestamp and the second timestamp; The transmission speed of the second neural electrical signal is determined based on the difference between the actual spatial propagation path and the timestamp.

2. The method according to claim 1, characterized in that, The SpikeSorting algorithm includes preprocessing, detection, feature extraction, and clustering processes. The SpikeSorting algorithm is used to decompose the first neural electrical signal at each acquisition point to obtain the second neural electrical signal of the same nerve fiber in the nerve trunk at each acquisition point, including: The first neural electrical signal at each acquisition point is preprocessed; the preprocessing includes filtering out low-frequency baseline drift and high-frequency noise from the first neural electrical signal using a bandpass filter; An adaptive thresholding method was used to detect the peak value of the action potential in the preprocessed signal and to extract the complete waveform segment of each action potential at each acquisition point. The extracted action potential waveform segment is subjected to feature dimensionality reduction processing to extract the time domain features and frequency domain features of the waveform; the time domain features include peak-to-peak amplitude and rising / falling edge slope, and the frequency domain features include the main frequency component; The waveform feature similarity of the action potential at each acquisition point is calculated based on the time-domain and frequency-domain features. By clustering, action potentials with waveform feature similarity higher than a preset waveform threshold at each acquisition point are grouped into one category, and each category corresponds to the second neural electrical signal of a nerve fiber.

3. The method according to claim 2, characterized in that, The waveform feature similarity of the action potential at each acquisition point, calculated based on time-domain and frequency-domain features, satisfies the following formula: in, The waveform feature similarity, The weighting coefficients for the peak-to-peak amplitude characteristics. This is the amplitude attenuation coefficient. The peak-to-peak amplitude of the action potential at sampling point i. The peak-to-peak amplitude of the action potential at sampling point j. The weighting coefficients for the slope trend characteristics. Let i be the rising slope of the action potential at acquisition point i. Let be the rising slope of the action potential at acquisition point j. Let i be the falling edge slope of the action potential at acquisition point i. Let be the falling edge slope of the action potential at acquisition point j. The weighting coefficients for the main frequency characteristics, The dominant frequency component of the action potential at acquisition point i. Let j be the dominant frequency component of the action potential at acquisition point j, and .

4. The method according to claim 1, characterized in that, The process of determining the sequence of neural electrical signal acquisition points through which the second neural electrical signal flows based on the timestamp of the second neural electrical signal from the same nerve fiber, and fitting the actual spatial propagation path of the second neural electrical signal from the same nerve fiber, includes: The timestamps of the second nerve electrical signal of the same nerve fiber at each acquisition point are sorted in ascending order to obtain the sequence of nerve electrical signal acquisition points through which the second nerve electrical signal flows. Obtain the three-dimensional spatial coordinates of each acquisition point in the acquisition point sequence; The three dimensions of the three-dimensional spatial coordinates of the collection point sequence are fitted using cubic spline interpolation to obtain a one-dimensional fitting function in three dimensions. Arc length integrals are calculated using a three-dimensional one-dimensional fitting function to obtain a continuous actual spatial propagation path.

5. The method according to claim 4, characterized in that, The cubic spline interpolation method is used to fit the three dimensions of the three-dimensional spatial coordinates of the collection point sequence to obtain a three-dimensional one-dimensional fitting function, including: The three-dimensional spatial coordinates of the collected point sequence are split into discrete data points in three dimensions (x, y, and z) according to the timestamp order, and the interpolation nodes in each dimension are determined. For each dimension of discrete data points, a cubic polynomial function is constructed as a spline basis function for the interval formed by two adjacent interpolation nodes. Based on the conditions that the function values ​​at adjacent nodes are equal, the first derivative is continuous, and the second derivative is continuous, the coefficients of each segment of the cubic polynomial function are solved, and the cubic polynomial functions of each interval in the three dimensions are spliced ​​together to obtain a three-dimensional one-dimensional fitting function.

6. The method according to claim 5, characterized in that, The cubic polynomial function in the x-dimensional dimension satisfies the following formula: in, For intervals in the x-dimensional dimension The function is a cubic polynomial, k=1,2,...,n-1, where n is the total number of sampling points in the sampling point sequence, and x is the independent variable in dimension x. Let x be the coordinates of the starting point of the k-th interval in the x-dimensional space. Let be the coefficient of the constant term in the k-th interval of dimension x. Let be the coefficient of the first-order term in the k-th interval of dimension x. Let be the coefficient of the quadratic term in the k-th interval of dimension x. Let be the coefficient of the cubic term in the k-th interval of dimension x.

7. The method according to claim 4, characterized in that, The arc length integral calculation based on the three-dimensional one-dimensional fitting function yields a continuous actual spatial propagation path that satisfies the following formula: in, The size of the continuous actual spatial propagation path. This is the first timestamp. This is the second timestamp. Let x be a one-dimensional fitting function. Let be a one-dimensional fitting function in the y-dimension. It is a one-dimensional fitting function in the z-dimensional dimension.

8. The method according to claim 1, characterized in that, After determining the transmission speed of the second neural electrical signal based on the actual spatial propagation path and the timestamp difference, the method further includes: Based on the residuals between the actual coordinates of the acquisition points and the fitted path, and the SpikeSorting clustering results, the effectiveness of the fitting of the actual spatial propagation path is verified. If the mean residual does not exceed the minimum axial spacing threshold of the acquisition points and there are no abnormal clustering signals, it is considered valid. Based on the path segment speed and the similarity of the waveform features of the second neural electrical signals at each acquisition point, the consistency of transmission speed is verified. If the mean waveform similarity is not lower than the waveform threshold and the coefficient of variation of the segment speed does not exceed the coefficient of variation threshold, it is considered reliable.

9. A device for measuring the transmission speed of neural electrical signals, characterized in that, The device includes: a communication unit and a processing unit; The communication unit is used to acquire the first neural electrical signal in the nerve trunk at each acquisition point based on the neural electrical signal acquisition point on the neural electrode; the neural electrode is a high-density three-dimensional electrode with an M-row N-column equally spaced neural electrical signal acquisition point matrix; the first neural electrical signal is a mixed action potential signal generated by multiple nerve fibers in the nerve trunk. The processing unit is configured to decompose the first neural electrical signal at each acquisition point using the SpikeSorting algorithm to obtain the second neural electrical signal of the same nerve fiber in the nerve trunk at each acquisition point, and record the timestamp of the second neural electrical signal at each acquisition point; the second neural electrical signal is an independent action potential signal of a single nerve fiber; the timestamp includes a first timestamp and a second timestamp, the first timestamp being the timestamp when the second neural electrical signal arrives at the starting acquisition point, and the second timestamp being the timestamp when the second neural electrical signal of the same nerve fiber arrives at the ending acquisition point; based on the timestamp of the second neural electrical signal of the same nerve fiber, determine the sequence of neural electrical signal acquisition points through which the second neural electrical signal flows, and fit the actual spatial propagation path of the second neural electrical signal of the same nerve fiber; determine the timestamp difference of the actual spatial propagation path according to the first timestamp and the second timestamp; and determine the transmission speed of the second neural electrical signal based on the actual spatial propagation path and the timestamp difference.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed on the neural electrical signal transmission velocity measuring device, cause the neural electrical signal transmission velocity measuring device to perform the method as described in any one of claims 1-8.

Images