Performance evaluation method, device and medium for brain-computer interface electrodes

Abstract

The present application provides a performance evaluation method, device and medium for brain-computer interface electrodes, including constructing a training / testing data set based on the behavioral data of the subject using the brain-computer interface to perform a preset brain control task, and the neural activity data collected by each electrode channel corresponding to the behavioral data; using the training data to train the neural signal decoding algorithm for multiple rounds, and the number of electrode channels selected for each round of training is reduced by a preset ratio compared to the number of electrode channels in the previous round; after each round of training, the average decoding accuracy of the neural signal decoding algorithm in this round is tested, and recorded as electrode feature data in pairs with the number of electrode channels in this round, and the performance of the brain-computer interface electrode is evaluated based on the electrode evaluation feature data composed of the number of electrode channels and the average decoding accuracy of the neural signal decoding algorithm in each round. The present application can make a more reasonable evaluation of the performance of brain-computer interface electrodes from the perspective of brain-computer interface system application.

Description

Technical Field

[0001] The present application belongs to the field of brain-computer interface technology, and in particular, relates to a performance evaluation method, device and medium for brain-computer interface electrodes. Background Art

[0002] The human nervous system can be roughly divided into the central nervous system and the peripheral nervous system. Normally, the activities of the central nervous system of the brain connect the body's sensory, motor and other information-transmitting organs through the peripheral nervous system to achieve information exchange between the body's internal and external environment. The brain-computer interface establishes a new communication and control channel between the brain and the external environment that is independent of peripheral nerves and muscles. It measures and collects brain neural activity and decodes it into signals or commands that can be recognized by external artificial devices, thereby achieving communication and control between the brain and external devices. At present, brain-computer interface technology is expected to be used as a new assistive technology for patients with movement disorders to improve the independence of patients' lives.

[0003] The performance of brain-computer interface is closely related to the ease of use and accuracy of the patient-controlled device. The quality of neural signal acquisition is a key factor affecting the performance of brain-computer interface. Brain-computer interface electrodes are used to collect and record neural information from the brain and create information channels between the brain and external devices. Therefore, how to evaluate the performance of brain-computer interface electrodes has become an important research topic. At present, the evaluation criteria for brain-computer interface electrodes mainly include indicators such as impedance size, cutoff frequency and signal-to-noise ratio. These indicators mainly reflect the electrical properties of the electrode itself, but fail to consider its actual effect when used in brain-computer interface. In addition, with the development of chip technology, the number of channels that can be accommodated on a single electrode is increasing. Generally speaking, the more electrode channels there are, the richer the signals collected, and the more accurate the decoded information and subsequent input information can be regulated. However, in other cases, especially for electrodes with different processing techniques and material types, the number of electrode recording channels is not necessarily sufficient to reflect the performance of the electrode in actual application.

[0004] It can be seen that there is currently no unified and universal evaluation method and indicator system that can be used to reasonably evaluate the performance of brain-computer interface electrodes from the perspective of brain-computer interface system application. Summary of the invention

[0005] In view of the above problems, the present application is proposed to solve the above problems existing in the prior art.

[0006] The purpose of this application is to provide a performance evaluation method, device and medium for brain-computer interface electrodes, which does not need to focus on the electrical performance of the brain-computer interface electrodes themselves, but is based on the relationship between the number of electrode channels of the brain-computer interface electrodes and the decoding accuracy of the neural signal decoding algorithm, to achieve a more reasonable evaluation of the performance of the brain-computer interface electrodes from the perspective of brain-computer interface system application.

[0007] According to the first scheme of the present application, a performance evaluation method for a brain-computer interface electrode is provided, wherein the brain-computer interface electrode is implanted in a specific brain region of a subject and comprises a given number of electrode channels, each electrode channel being used to record neural activity data in the specific brain region, the performance evaluation method comprising: constructing an original data set based on behavioral data of the subject performing a preset brain control task using a brain-computer interface, and neural activity data collected by each electrode channel corresponding to the behavioral data, and dividing the original data set into a training set and a test set; performing multiple rounds of training on a neural signal decoding algorithm using the training data in the training set, wherein the number of electrode channels selected for each round of training gradually decreases compared to the number of electrode channels in the previous round, and each round of training starts from the initial parameters of the neural signal decoding algorithm; testing the neural signal decoding algorithm trained in this round using the test data in the test set to obtain the average decoding accuracy of this round, and evaluating the performance of the brain-computer interface electrode based on the electrode evaluation feature data of each round consisting of the number of electrode channels and the average decoding accuracy of the neural signal decoding algorithm.

[0008] According to the second scheme of the present application, another performance evaluation method for different types of brain-computer interface electrodes is provided, and the performance evaluation method is used to compare and evaluate the performance of different types of brain-computer interface electrodes, wherein each type of brain-computer interface electrode is implanted in the same brain region of the same type of subjects, and each type of brain-computer interface electrode contains the same or different numbers of electrode channels, and each type of electrode channel is used to record neural activity data of the same signal type in the brain region where the electrode is implanted. The performance evaluation method includes, for each type of brain-computer interface electrode: respectively obtaining behavioral data of the respective subjects performing the same preset brain control task using the brain-computer interface, and neural activity data collected by each electrode channel corresponding to the behavioral data, and executing the performance evaluation method for brain-computer interface electrodes as described in each embodiment of the first scheme of the present application to obtain electrode evaluation feature data of various types of brain-computer interface electrodes, and for comparing and evaluating the performance of different types of brain-computer interface electrodes; wherein various types of brain-computer interface electrodes all use the same neural signal decoding algorithm.

[0009] According to the third scheme of the present application, another performance evaluation method for brain-computer interface electrodes is provided, wherein the brain-computer interface electrodes are implanted in a specific brain area of ​​a subject and comprise a given number of electrode channels, each electrode channel being used to record neural activity data in the specific brain area, the performance evaluation method comprising: recording behavioral data of the subject performing the same preset brain control task using the brain-computer interface in different time periods, and neural activity data collected by each electrode channel corresponding to the behavioral data; based on the behavioral data of the subject performing the preset brain control task using the brain-computer interface in each time period, and the neural activity data collected by each electrode channel corresponding to the behavioral data, executing the performance evaluation method as described in each embodiment of the first scheme of the present application to obtain performance evaluation results of the brain-computer interface electrodes in each time period; based on the performance evaluation results of the brain-computer interface electrodes in each time period, evaluating the time domain related performance of the brain-computer interface electrodes.

[0010] According to the fourth scheme of the present application, a performance evaluation device for brain-computer interface electrodes is provided, wherein the brain-computer interface electrodes are implanted in a specific brain area of ​​a subject and comprise a given number of electrode channels, each electrode channel being used to record neural activity data in the specific brain area, and the performance evaluation device comprises at least an interface and a processor, wherein the interface is configured to obtain behavioral data of the subject performing a preset brain control task using the brain-computer interface, and neural activity data collected by each electrode channel corresponding to the behavioral data; and the processor is configured to perform the operations of the performance evaluation method for brain-computer interface electrodes as described in each embodiment of the first scheme of the present application.

[0011] According to the fifth scheme of the present application, another performance evaluation device for brain-computer interface electrodes is provided, and the performance evaluation device is used to compare and evaluate the performance of different types of brain-computer interface electrodes, wherein each type of brain-computer interface electrode is implanted in the same brain area of ​​the same type of subjects, and each type of brain-computer interface electrode contains the same or different numbers of electrode channels, and each type of electrode channel is used to record neural activity data of the same signal type in the brain area where the electrode is implanted, and the performance evaluation device includes at least an interface and a processor, wherein the interface is configured to obtain behavioral data of subjects implanted with different types of brain-computer interface electrodes using the brain-computer interface to perform the same preset brain control task, and neural activity data collected by each electrode channel corresponding to the behavioral data; the processor is configured to perform the operations of the performance evaluation method for different types of brain-computer interface electrodes as described in each embodiment of the second scheme of the present application, and wherein each type of brain-computer interface electrode adopts the same neural signal decoding algorithm.

[0012] According to the sixth scheme of the present application, there is provided another performance evaluation device for brain-computer interface electrodes, wherein the brain-computer interface electrodes are implanted in a specific brain region of a subject and comprise a given number of electrode channels, each electrode channel being used to record neural activity data in the specific brain region, and the performance evaluation device comprises at least an interface and a processor, wherein the interface is configured to obtain behavioral data recorded in different time periods when the subject uses the brain-computer interface to perform the same preset brain control task, and neural activity data collected by each electrode channel corresponding to the behavioral data; and the processor is configured to perform the operations of the performance evaluation method for brain-computer interface electrodes as described in each embodiment of the third scheme of the present application.

[0013] According to the seventh embodiment of the present application, a non-temporary computer-readable storage medium is provided, on which computer-executable instructions are stored, wherein when the computer-executable instructions are executed by a processor, the operations of the performance evaluation method for brain-computer interface electrodes as described in the various embodiments of the first embodiment and the third embodiment of the present application are performed, or the operations of the performance evaluation method for different types of brain-computer interface electrodes as described in the various embodiments of the second embodiment of the present application are performed.

[0014] According to the performance evaluation method, device and medium for brain-computer interface electrodes of various embodiments of the present application, the training data set of the neural signal decoding algorithm is constructed using the behavioral data of the subject and the neural activity data collected by each corresponding electrode channel, and the number of electrode channels is gradually reduced in each round of multiple training rounds, and then the performance of the brain-computer interface electrode is evaluated based on the electrode evaluation feature data composed of the number of electrode channels in each round and the average decoding accuracy of the corresponding round. Compared with some existing technologies that focus on the electrical performance of the brain-computer interface electrode itself in isolation, the present application does not test the electrical performance indicators of the electrode, but starts from the application effect of the electrode in the brain-computer interface, using the neural activity data collected by the brain-computer interface electrode and its corresponding behavioral data as a training set, and training a specific neural signal decoding algorithm, and using the decoding accuracy of the trained neural signal decoding algorithm and the corresponding number of electrode channels as electrode evaluation feature data, so that the actual contribution of the brain-computer interface electrode to the decoding performance can be more reasonably evaluated under the consideration of the number of electrode channels. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In the drawings, which are not necessarily drawn to scale, the same reference numerals may describe similar parts in different views. The same reference numerals with letter suffixes or different letter suffixes may represent different instances of similar parts. The accompanying drawings generally illustrate various embodiments by way of example and not limitation, and together with the specification and claims, are used to illustrate the disclosed embodiments. When appropriate, the same reference numerals are used throughout the drawings to refer to the same or similar parts. Such embodiments are illustrative and are not intended to be exhaustive or exclusive embodiments of the present apparatus or method.

[0016] Figure 1 A schematic diagram showing a brain-computer interface and brain-computer interface electrodes used on a subject according to an embodiment of the present application.

[0017] Figure 2 A schematic flow chart of a method for evaluating the performance of a brain-computer interface electrode according to an embodiment of the present application is shown.

[0018] Figure 3 Another flowchart of a method for evaluating the performance of a brain-computer interface electrode according to an embodiment of the present application is shown.

[0019] Figure 4 A flow chart of a method for evaluating the time-domain correlation performance of brain-computer interface electrodes according to an embodiment of the present application is shown.

[0020] Figure 5 A schematic diagram showing some components of a performance evaluation device for brain-computer interface electrodes according to an embodiment of the present application is shown.

[0021] Figure 6 A schematic diagram showing the performance evaluation of different types of brain-computer interface electrodes according to an embodiment of the present application. DETAILED DESCRIPTION

[0022] In order to enable those skilled in the art to better understand the technical solution of the present disclosure, the present disclosure is described in detail below in conjunction with the drawings and specific implementation methods. The embodiments of the present disclosure are further described in detail below in conjunction with the drawings and specific implementation methods, but are not intended to limit the present disclosure.

[0023] The words "first", "second" and similar terms used in this application do not indicate any order, quantity or importance, but are only used to distinguish different parts. The expressions "first" and "second" are only used to distinguish by numbering for the convenience of expression, and are not intended to imply that the "first component" and the "second component" must have different physical properties. In fact, the "first component" and the "second component" may have the same or different structures, and are not limited here, as long as the "first component" and the "second component" are discrete components. Furthermore, if the context provides sufficient explanation, the "first component" and the "second component" may not even be discrete components, but may be integrated into the same component, or may be interchangeable with each other.

[0024] In the present application, when a specific device is described as being located between a first device and a second device, there may or may not be an intermediate device between the specific device and the first device or the second device. When a specific device is described as being connected to other devices, the specific device may be directly connected to the other device without an intermediate device, or may not be directly connected to the other device but have an intermediate device.

[0025] "Include" or "comprising" and other similar words mean that the elements before the word include the elements listed after the word, and do not exclude the possibility of including other elements. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0026] Figure 1 A schematic diagram showing a brain-computer interface and a brain-computer interface electrode used in a subject according to an embodiment of the present application is shown. Figure 1 As shown, the brain-computer interface electrode 10 is implanted in a specific brain area of ​​the subject 100, for example, it can be a motor area near the precentral gyrus, or it can be implanted in other brain areas such as the frontal lobe, parietal lobe, temporal lobe, occipital lobe, hippocampus, etc. according to the function to be achieved. The brain-computer interface electrode 10 includes a given number of electrode channels 101, wherein each electrode channel 101 is used to record neural activity data in a specific brain area. In some embodiments, the number of electrode channels 101 can be, for example, a dozen, dozens or even more, and this application does not limit this.

[0027] Figure 2 A schematic flow chart of a method for evaluating the performance of a brain-computer interface electrode according to an embodiment of the present application is shown.

[0028] Only Figure 1 Taking the brain-computer interface electrode 10 with a given number of electrode channels 101 shown in the figure as an example, the performance evaluation method for the brain-computer interface electrode according to the embodiment of the present application can be completed through the following steps 201 to 203.

[0029] First, in step 201, an original data set can be constructed based on the behavioral data of the subject 100 performing a preset brain control task using a brain-computer interface, and the neural activity data collected by each electrode channel 101 corresponding to the behavioral data, and the original data set can be divided into a training set and a test set.

[0030] In some embodiments, the above-mentioned preset brain control tasks can be, for example, center-out, grid tasks, and moving cursor typing tasks performed by the subject 100 using a brain-computer interface. The behavioral data may include the direction of the target position of each trial relative to the starting position and the actual movement direction or position obtained by decoding the neural activity. The neural activity data can generally include the high-frequency action potential (spike) firing rate of each electrode channel, etc., and may also include one or more neural activity data such as low-frequency field potential (LFP), ECoG (Electrocorticography, electroencephalogram) signals, and the present application does not impose any restrictions on this.

[0031] In some embodiments, for example, a cross-validation method can be used to divide the original data set into a training set and a test set. As an example only, the data in the original data set can be divided into 5 parts, 4 of which are used as training sets and the remaining 1 is used as a test set.

[0032] Then, in step 202, the training data in the training set can be used to perform multiple rounds of training on the neural signal decoding algorithm, wherein the number of electrode channels selected in each round of training is gradually reduced compared to the number of electrode channels in the previous round, and each round of training starts from the initial parameters of the neural signal decoding algorithm, that is: based on the electrode channels used in the previous round of training, firstly, the data of some channels are discarded, and then new training based on the initialization parameters is started, so that the number of electrode channels used in each round of training shows a decreasing trend.

[0033] In some embodiments, the neural signal decoding algorithm can be, for example, a Kalman filter algorithm, a neural network algorithm, or any other applicable decoding algorithm commonly used for decoding neural activity data. This application does not impose any restrictions on this, as long as it can complete the task of decoding neural activity data through training.

[0034] When training the neural signal decoding algorithm, the neural activity data collected by each electrode channel 101 is used as the input of the neural signal decoding algorithm to be trained, and the behavioral data of the subject 100 using the brain-computer interface to perform the preset brain control task is used as the true value data of the output of the neural signal decoding algorithm to be trained. In the training process, the loss function is used to calculate the loss between the actual behavioral data output by the neural signal decoding algorithm and the corresponding true value data, and the algorithm parameters of the neural signal decoding algorithm are adjusted according to the loss. When the loss is less than or equal to the preset threshold or convergence is reached, it indicates that the neural signal decoding algorithm training converges. Optionally, when calculating the loss, for example, a Dice loss function, a cross entropy loss function or any other applicable loss function can be used, and this application does not impose any restrictions on this. When adjusting the algorithm parameters, a stochastic gradient descent SGD optimizer or other types of optimizers can be used, and this application does not impose any specific restrictions on this.

[0035] Next, in step 203, the neural signal decoding algorithm trained in this round is tested using the test data in the test set to obtain the average decoding accuracy of this round. Furthermore, the average decoding accuracy of the number of electrode channels and the neural signal decoding algorithm in each round is used as electrode evaluation feature data, and based on the electrode evaluation feature data, the performance of the brain-computer interface electrode is evaluated. It can be understood that when the average decoding accuracy of the neural signal decoding algorithm trained in the current round reaches or is lower than a certain preset value, it can be considered that the number of electrode channels has been reduced to a level that cannot meet the normal application of the brain-computer interface. Therefore, the next round of training may not be performed at this time. The above preset values ​​can be specifically set according to the requirements of the electrode performance evaluation, and this application does not impose any restrictions on this.

[0036] According to the performance evaluation method for brain-computer interface electrodes of the embodiment of the present application, the brain control task behavior data and the corresponding neural activity data collected by each electrode channel, which are available in brain-computer interface applications and are very easy to obtain, are used. By gradually reducing the number of electrode channels used for training and testing in multiple rounds of training of the neural signal decoding algorithm, the corresponding relationship between the number of electrode channels actually used and the decoding accuracy that can be achieved by the neural signal decoding algorithm is explored for a specific brain-computer interface electrode, and a construction method and evaluation method of a system-level performance indicator of a brain-computer interface electrode are given. Compared with the prior art of testing the electrical performance of the brain-computer interface, or believing that the more electrode channels, the better the performance, the present application evaluates the performance of the brain-computer interface electrode based on the corresponding relationship as described above, and can focus on the actual neural signal recording capability of the electrode without having to care about the underlying indicators of the brain-computer interface electrode such as electrical performance. In other words, it can also be considered as a consideration of the contribution of a unit channel (e.g., one channel) on the brain-computer interface electrode to the decoding accuracy rate when redundant channels are excluded.

[0037] In the practical process of proposing the present application, it was found that when the number of electrode channels is greater than a specific threshold (the specific value varies with the type of electrode), the decoding accuracy that can be achieved through training the neural signal decoding algorithm is not much different. Only after the number of electrode channels is lower than the threshold, the decoding accuracy of the neural signal decoding algorithm begins to decrease at a larger gradient. Therefore, from another perspective, the results of the performance evaluation of brain-computer interface electrodes using the performance evaluation method of the present application can be used to evaluate the lower limit of the number of electrode channels that meets the needs of the subjects and the corresponding brain control. For example, when the number of channels of a specific type of brain-computer interface electrode is associated with the electrode surface area, the number of electrodes or the electrode surface area implanted in the brain can be controlled according to the evaluation results. In this way, the waste and unnecessary tissue damage caused by implanting too many electrodes or too large an electrode surface area can be reduced to a certain extent.

[0038] In some embodiments, when constructing the original data set based on the behavioral data of the subject using the brain-computer interface to perform the preset brain control task and the neural activity data collected by each electrode channel corresponding to the behavioral data, the behavioral data of the subject's good performance can be selected from all the behavioral data of the subject using the brain-computer interface to perform the preset brain control task. For example, in the case of performing center-out, grid tasks and cursor movement typing, the behavioral data in which the cursor movement is completely determined by the decoding of the neural signal and completed without the help of auxiliary vectors can be selected. In some embodiments, the proportion of positive samples in the original data set can be greater than a first threshold, for example, greater than 80%, that is, the decoding accuracy in the original data set is greater than 80%, thereby at least partially excluding atypical data generated in the process of the subject's adaptation to the brain control task and algorithm, especially when the subject is an animal subject.

[0039] In some embodiments, each round of training for a neural signal decoding algorithm using the same number of electrode channels includes multiple trainings, and the average decoding accuracy of each round can be obtained based on the decoding accuracy of each training in this round, on the basis of testing the neural signal decoding algorithm trained in each round in this round using the test data in the test set. For example, the arithmetic mean of the decoding accuracy of each training can be used as the average decoding accuracy of this round, and so on. The present application does not impose any restrictions on this.

[0040] In addition, when each round includes multiple trainings, for each training of each round (i.e., each round), for example, bootstrap sampling method and other methods can be used to randomly discard at least part of the channel data in the previous round. In this way, while the number of electrode channels selected for each round of training is gradually reduced compared to the number of electrode channels in the previous round, it can also ensure that the channel data used for each training of each round is not exactly the same, and to a certain extent, the distribution of channels discarded / selected for each training is relatively random and uniform, so that the training effect is not biased due to the distribution of electrode channels and training data. In specific implementation, other sampling methods can also be used to select the channel data to be discarded, and this application does not limit this.

[0041] In some embodiments, the number of electrode channels discarded in each round can be a preset proportion of the number of electrode channels in the previous round, that is, the number of electrode channels selected for each round of training is reduced by a preset proportion compared to the number of electrode channels in the previous round (for example, any value between [1 / 4, 1 / 3] can be selected).

[0042] As described above, the ratio of the number of electrode channels discarded in each round can be the same. In other embodiments, the number of channels to be discarded or the ratio of the number of channels can also be different in each round. For example, the ratio of the number of channels / number of channels to be discarded in the next round can be dynamically adjusted according to the average decoding accuracy of the neural signal decoding algorithm in the current round compared with the previous round. For example, when the average decoding accuracy of adjacent rounds changes slowly, the ratio of the number of channels / number of channels to be discarded in the next round is appropriately increased. Conversely, the ratio of the number of channels / number of channels to be discarded in the next round can be appropriately reduced, thereby making the average decoding accuracy change gradient appropriate and the entire process complete faster. In other embodiments, the ratio of the number of channels / number of channels to be discarded in the next round can also be determined in combination with the degree of proximity between the average decoding accuracy value of the current round and the preset value to be achieved by the decoding accuracy, so that the gradient of change is slower when the average decoding accuracy is close to the preset value, so as not to miss the preset value of the average decoding accuracy.

[0043] Figure 3 Another flowchart of a method for evaluating the performance of a brain-computer interface electrode according to an embodiment of the present application is shown.

[0044] like Figure 3 As shown, in step 301, the number of electrode channels selected for each round and the average decoding accuracy of the round may be recorded in pairs as electrode evaluation feature data of the round.

[0045] Then, in step 302, based on the electrode evaluation characteristic data of each round, a characteristic curve of the number of electrode channels-decoding accuracy of the brain-computer interface electrode is drawn, and the characteristic curve of the number of electrode channels-decoding accuracy reflects the corresponding relationship between the decoding accuracy of the trained neural signal decoding algorithm and the number of electrode channels selected for the training neural signal decoding algorithm.

[0046] Next, in step 303, based on the electrode channel number-decoding accuracy characteristic curve, the minimum number of electrode channels that can be selected while meeting the preset decoding accuracy requirement is obtained, and it is determined that the smaller the minimum number of electrode channels that can be selected while meeting the preset decoding accuracy requirement, the better the performance evaluation result of the brain-computer interface electrode.

[0047] In some embodiments, the preset decoding accuracy requirement may be, for example, the minimum decoding accuracy requirement that satisfies the requirements for performing a preset brain control task. In other embodiments, the preset decoding accuracy may also be other preset values, such as a preset value of the decoding accuracy requirement determined by considering the redundancy of future applications and / or comprehensively considering the execution of multiple brain control tasks, and this application does not limit this.

[0048] Therefore, the performance evaluation method according to the embodiment of the present application further proposes a more intuitive and multi-angle evaluation method for the performance of brain-computer interface electrodes based on the characteristic curve of the number of electrode channels and decoding accuracy. Figure 3 The minimum number of electrode channels that can be selected to meet the preset decoding accuracy requirements proposed in the performance evaluation method for brain-computer interface electrodes described above can be used to measure the single-channel signal acquisition capability of electrodes under the premise of meeting the application requirements of specific systems. It can be imagined that electrodes made of different materials and / or manufacturing processes may contain the same number of channels on a single electrode. Figure 3 After the operations of each step, there may be large differences in the above-mentioned indicators. Therefore, compared with judging the performance of brain-computer interface electrodes based on the number of electrode channels, the electrode performance evaluation method according to the embodiment of the present application can more accurately correspond to the neural signal acquisition capability and acquisition efficiency of the electrode channel, and is a more reasonable system-level performance evaluation method, which can provide stronger support for the selection and application of electrodes.

[0049] In other embodiments, based on the electrode channel number-decoding accuracy characteristic curve, the minimum number of electrode channels that can be selected while meeting the preset decoding accuracy requirement is obtained, and multiple system-level electrode performance evaluation indicators such as those associated with the electrode surface area and the electrode signal acquisition efficiency can be further determined. As an example only, for example, at least one of the electrode surface area characteristic parameter of the brain-computer interface electrode and the first electrode signal acquisition efficiency characteristic parameter can be included, wherein the product of the electrode surface area of ​​the brain-computer interface electrode and the minimum number of electrode channels is used as the electrode surface area characteristic parameter, and the inverse of the minimum number of electrode channels is used as the first electrode signal acquisition efficiency characteristic parameter, and when performing the performance evaluation of the brain-computer interface electrode, the evaluation result is inversely proportional to the above-mentioned electrode surface area characteristic parameter and directly proportional to the first electrode signal acquisition efficiency characteristic parameter.

[0050] It can be understood that the smaller the minimum number of electrode channels, the larger the characteristic parameter of the first electrode signal acquisition efficiency, and the better the performance evaluation result of the brain-computer interface electrode. In addition, when the electrode surface area is taken into consideration, the two factors of the electrode surface area characteristic parameter - the electrode surface area of ​​the brain-computer interface electrode and the minimum number of electrode channels, when both or either of them have a large value, will lead to poor electrode performance evaluation results, so the electrode surface area characteristic parameter is inversely proportional to the electrode performance evaluation result.

[0051] In other embodiments, the performance of the brain-computer interface electrodes can be evaluated in other ways by utilizing the electrode evaluation feature data composed of the number of electrode channels and the average decoding accuracy of the neural signal decoding algorithm in each round. As an example only, the highest decoding accuracy and its corresponding number of electrode channels can be obtained based on the electrode evaluation feature data, and the quotient of the highest decoding accuracy and the corresponding number of electrode channels can be used as the second electrode signal acquisition efficiency characteristic parameter, and the performance evaluation result of the brain-computer interface electrode is proportional to the second electrode signal acquisition efficiency characteristic parameter.

[0052] Different from the first electrode signal acquisition efficiency characteristic parameter, the second electrode signal acquisition efficiency characteristic parameter reflects the contribution of the corresponding single electrode channel to the decoding accuracy when the highest decoding accuracy is achieved. Therefore, the higher the second electrode signal acquisition efficiency characteristic parameter, the better the performance evaluation result of the brain-computer interface electrode.

[0053] In other embodiments, the electrode surface area-decoding accuracy characteristic curve can be determined and plotted in a similar manner to the method of determining the characteristic data of each round of electrode evaluation. Alternatively, in other cases, especially when the number of channels of the brain-computer interface electrode is associated with the electrode surface area, the characteristic data of each round of electrode evaluation can be directly used or the electrode channel number-decoding accuracy characteristic curve can be converted into an electrode surface area-decoding accuracy characteristic curve. Thus, based on the electrode surface area-decoding accuracy characteristic curve, the minimum electrode surface area that can be selected to meet the preset decoding accuracy requirement, as well as the highest decoding accuracy and its corresponding electrode surface area and other electrode performance evaluation results can be obtained.

[0054] Furthermore, the various electrode performance evaluation characteristic parameters proposed above can not only be applied individually, but also can be used to construct a composite evaluation index or index system in a certain way (for example, the weights and other numerical relationships between the various characteristic parameters can be determined through experiments and fitting), and the composite index or index system can be used to conduct a comprehensive performance evaluation of the brain-computer interface electrode from multiple angles. The specific construction method of the composite evaluation index or index system is not specifically limited in this application.

[0055] According to an embodiment of the present application, a performance evaluation method for different types of brain-computer interface electrodes is also provided. In this case, each type of brain-computer interface electrode is implanted in the same brain region of the same type of subjects, and each type of brain-computer interface electrode contains the same or different numbers of electrode channels. Each type of brain-computer interface electrode channel is used to record neural activity data of the same signal type in the brain region where the electrode is implanted. For example, each type of brain-computer interface electrode records spike signals or LFP signals, etc. The specific signal type is not limited in this application as long as the electrodes remain consistent.

[0056] On this basis, various types of brain-computer interface electrodes can be used to obtain the behavioral data of the same preset brain control task performed by the respective subjects using the brain-computer interface, as well as the neural activity data collected by each electrode channel corresponding to the behavioral data, and the steps in the performance evaluation method for brain-computer interface electrodes described in the previous embodiments can be performed, so as to obtain the electrode evaluation characteristic data of various types of brain-computer interface electrodes, and the obtained electrode evaluation characteristic data of various types of brain-computer interface electrodes are used for comparative evaluation of the performance of different types of brain-computer interface electrodes, and for various types of brain-computer interface electrodes, the same neural signal decoding algorithm is used. More specifically, when comparing and evaluating, for example, the same characteristic parameters of different types of brain-computer interface electrodes can be compared, including the minimum number of electrode channels that can be selected under the preset decoding accuracy requirements, the electrode surface area characteristic parameter, the first electrode signal acquisition efficiency characteristic parameter, or one of the first electrode signal acquisition efficiency characteristic parameters, or a combination of multiple characteristic parameters in the same manner, and the specific parameters and compounding methods used can be set according to the focus of the comparative evaluation of different types of brain-computer interface electrodes, and this application does not make specific restrictions.

[0057] As described above, since the various types of brain-computer interface electrodes are implanted in the same type of subjects and the same brain area, the signal types recorded by the electrodes and the neural signal decoding algorithms used in training / testing are also exactly the same, it can be ensured that the various types of brain-computer interface electrodes are under the same conditions when executing the performance evaluation method described in the embodiments of the present application, that is, there is a credible and consistent basis for comparative evaluation, which can be used in particular for comparative evaluation of electrodes made of different materials / different processing techniques, and the evaluation results are more meaningful for reference than judging only by the number of electrode channels.

[0058] Figure 6 Schematic diagram showing the performance evaluation of different types of brain-computer interface electrodes according to an embodiment of the present application. Figure 6 As shown, the circle marked with "○" is the characteristic curve of the number of electrode channels of the brain-computer interface electrode 1-decoding accuracy, and the cross marked with "X" is the characteristic curve of the number of electrode channels of the brain-computer interface electrode 2-decoding accuracy. Among them, the brain-computer interface electrode 1 and the brain-computer interface electrode 2 are different types of electrodes. For convenience, the number of channels of the two electrodes used in this embodiment is the same. Figure 6 In the figure, the horizontal axis represents the number of electrode channels selected for each round of training of the neural signal decoding algorithm. Figure 6The R1 shown is the number of electrode channels used in the first round, and R2-R5 is the number of electrode channels used in the second to fifth rounds. R1-R5 are fixed values ​​determined according to the decreasing method of the number of electrode channels proposed in the embodiment of the present application. Therefore, this also means that the number of electrode channels used by the two brain-computer interface electrodes in each round of training is equal; the ordinate represents the average decoding accuracy of the neural signal decoding algorithm after the training is completed (expressed as a percentage), and the P0 value on the ordinate represents the minimum decoding accuracy requirement to meet the requirements of performing the preset brain control task. Figure 6 It can be seen that the average decoding accuracy of brain-computer interface electrode 1 in the first round is P1, which is higher than the average decoding accuracy of brain-computer interface electrode 2 in the first round, P1'. Figure 6 It can also be seen that for the brain-computer interface electrode 1, the number of electrode channels corresponding to the minimum decoding accuracy requirement P0 is about R5, and when the number of electrode channels of the brain-computer interface electrode is reduced to R4, the average decoding accuracy is close to P0. Therefore, according to Figure 6 , according to the electrode performance evaluation method described in this application, it can be determined that the performance of brain-computer interface electrode 1 is better than that of brain-computer interface electrode 1.

[0059] According to an embodiment of the present application, another performance evaluation method for brain-computer interface electrodes is also provided. Figure 4 A flow chart of a method for evaluating the time-domain correlation performance of brain-computer interface electrodes according to an embodiment of the present application is shown.

[0060] exist Figure 4 The performance evaluation method shown in the figure is the same as the above Figure 2 Similar to the embodiment shown, the brain-computer interface electrode is implanted in a specific brain region of the subject and comprises a given number of electrode channels, each electrode channel being used to record neural activity data in the specific brain region. The specific performance evaluation method is shown in steps 401 to 403.

[0061] First, in step 401, it is necessary to record the behavior data of the subject performing the same preset brain control task using the brain-computer interface, and the neural activity data collected by each electrode channel corresponding to the behavior data in different time periods. Just as an example, for example, from the time when the brain-computer interface electrode is implanted in the subject's brain area to the current time, the behavior data of the subject performing the brain control task and the corresponding electrode recorded neural activity data can be collected at regular intervals (e.g., one month, half a year, etc.) during the use of the brain-computer interface electrode.

[0062] Then, in step 402, based on the behavioral data of the subject performing the preset brain control task using the brain-computer interface in each time period, and the neural activity data collected by each electrode channel corresponding to the behavioral data, the following is performed: Figure 2The steps of the performance evaluation method for brain-computer interface electrodes described in each embodiment are used to obtain the performance evaluation results of the brain-computer interface electrodes in each time period.

[0063] In some embodiments, similar to the characteristic parameters used in the performance evaluation of the aforementioned brain-computer interface electrodes, the performance evaluation of the brain-computer interface electrodes in each time period may also use the minimum number of electrode channels that can be selected while meeting the preset decoding accuracy requirements, electrode surface area characteristic parameters, first electrode signal acquisition efficiency characteristic parameters, or one of the first electrode signal acquisition efficiency characteristic parameters, or a combination of multiple characteristic parameters, and obtain the performance evaluation results of the brain-computer interface electrodes corresponding to each time period accordingly.

[0064] Next, in step 403, based on the performance evaluation results of the brain-computer interface electrodes corresponding to each time period, the time domain related performance of the brain-computer interface electrode is evaluated. Specifically, for example, a time domain change characteristic curve of the electrode performance that changes with time can be drawn based on the performance evaluation results of one or more brain-computer interface electrodes corresponding to each time period. The figure may contain one or more time-varying characteristic curves, and the time domain related performance of the brain-computer interface electrode is evaluated by analyzing and calculating each curve in the figure. In some embodiments, the time domain related performance may include, for example, a time stability index, an error index of the brain-computer interface electrode performance in a specific time period, the time length after which the decoding accuracy drops to a specific threshold, etc., and the present application does not limit this. As an example only, for example, the time stability index of the brain-computer interface electrode can be calculated by combining the fluctuation range of the minimum number of electrode channels that can be selected when the brain-computer interface electrode meets the preset decoding accuracy requirements in each time period, etc. The specific algorithm can be set as needed, and the present application does not limit this.

[0065] Combination Figure 4 The performance evaluation method for brain-computer interface electrodes described can effectively evaluate the specific conditions of overall electrode performance degradation after long-term implantation due to reasons such as degradation or failure of signal quality of some channels, and can thereby evaluate the stability indicators of specific types of electrodes that change over time, and provide strong support for electrode selection in long-term application scenarios.

[0066] According to an embodiment of the present application, a performance evaluation device for a brain-computer interface electrode is also provided. Figure 5 A schematic diagram showing some components of a performance evaluation device for brain-computer interface electrodes according to an embodiment of the present application is shown.

[0067] Similar to the above, Figure 5The brain-computer interface electrode associated with the performance evaluation device 500 for brain-computer interface electrodes is implanted in a specific brain region of the subject and includes a given number of electrode channels, each of which is used to record neural activity data in the specific brain region. Figure 5 As shown, the performance evaluation device 500 includes at least an interface 501 and a processor 502, wherein the interface 501 can be configured to obtain behavioral data of the subject performing a preset brain control task using a brain-computer interface, as well as neural activity data collected by each electrode channel corresponding to the behavioral data.

[0068] In some embodiments, the interface 501 may include a network adapter, a cable connector, a serial connector, a USB connector, a parallel connector, a high-speed data transmission adapter (such as optical fiber, USB 3.0, Thunderbolt interface, etc.), a wireless network adapter (such as a WiFi adapter), a telecommunications (3G, 4G / LTE, etc.) adapter, etc., and this application does not limit this. The performance evaluation device 500 can transmit the obtained behavioral data of the subject performing the preset brain control task using the brain-computer interface, and the neural activity data collected by each electrode channel corresponding to the behavioral data, etc. to other parts such as the processor 502 through the interface 501.

[0069] In some embodiments, the processor 502 may be configured to perform Figure 2 Various operations of the performance evaluation method for brain-computer interface electrodes described in various embodiments.

[0070] In some embodiments, the processor 502 may be a processing device including one or more general-purpose processing devices, such as a microprocessor, a central processing unit (CPU), a graphics processing unit (GPU), etc. More specifically, the processor may be a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, a processor running other instruction sets, or a processor running a combination of instruction sets. The processor may also be one or more special-purpose processing devices, such as an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), a digital signal processor (DSP), a system on a chip (SoC), etc.

[0071] According to an embodiment of the present application, another performance evaluation device for brain-computer interface electrodes is also included, and the performance evaluation device is used to compare and evaluate the performance of different types of brain-computer interface electrodes, wherein each type of brain-computer interface electrode is implanted in the same brain region of the same type of subject, and each type of brain-computer interface electrode contains the same or different number of electrode channels, and each type of brain-computer interface electrode channel is used to record the neural activity data of the same signal type in the brain region where the electrode is implanted. The performance evaluation device includes at least an interface and a processor, wherein the interface is configured to obtain the behavioral data of the subjects implanted with different types of brain-computer interface electrodes using the brain-computer interface to perform the same preset brain control task, and the neural activity data collected by each electrode channel corresponding to the behavioral data. The processor is configured to perform various operations of the performance evaluation method for different types of brain-computer interface electrodes as described in the embodiment of the present application, wherein various types of brain-computer interface electrodes use the same neural signal decoding algorithm.

[0072] According to an embodiment of the present application, another performance evaluation device for brain-computer interface electrodes is also provided, wherein the brain-computer interface electrodes are implanted in a specific brain region of a subject and include a given number of electrode channels, each electrode channel being used to record neural activity data in the specific brain region, and the performance evaluation device includes at least an interface and a processor, wherein the interface is configured to: obtain behavioral data recorded in different time periods when the subject uses the brain-computer interface to perform the same preset brain control task, and neural activity data collected by each electrode channel corresponding to the behavioral data; the processor is configured to perform the following steps: Figure 4 Various operations of the performance evaluation method for brain-computer interface electrodes described in various embodiments.

[0073] According to an embodiment of the present application, a non-temporary computer-readable storage medium is also provided, on which computer-executable instructions are stored, wherein when the computer-executable instructions are executed by a processor, the operations of the performance evaluation method for brain-computer interface electrodes as described in the various embodiments of the present application are performed, or the operations of the performance evaluation method for different types of brain-computer interface electrodes as described in the various embodiments of the present application are performed.

[0074] In some embodiments, the above-mentioned non-transitory computer-readable medium can be a read-only memory (ROM), random access memory (RAM), phase change random access memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), electrically erasable programmable read-only memory (EEPROM), other types of random access memory (RAM), flash disk or other forms of flash memory, cache, register, static memory, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, cassette tape or other magnetic storage device, or any other possible non-transitory medium used to store information or instructions that can be accessed by a computer device.

[0075] In addition, although exemplary embodiments have been described herein, the scope includes any and all embodiments based on the present disclosure with equivalent elements, modifications, omissions, combinations (e.g., various embodiments intersecting schemes), adaptations or changes. The elements in the claims are to be interpreted broadly based on the language adopted in the claims, and are not limited to the examples described in this specification or during the enforcement of this application, which examples are to be interpreted as non-exclusive. Therefore, this specification and examples are intended to be considered as examples only, and the true scope and spirit are indicated by the claims and the full scope of their equivalents.

[0076] The above description is intended to be illustrative rather than restrictive. For example, the above examples (or one or more schemes thereof) may be used in combination with each other. For example, a person of ordinary skill in the art may use other embodiments when reading the above description. In addition, in the above-mentioned specific embodiments, various features may be grouped together to simplify the present disclosure. This should not be interpreted as an intention that a disclosed feature that is not claimed for protection is necessary for any claim. On the contrary, the subject matter of the present invention may be less than all the features of a particular disclosed embodiment. Thus, the claims are incorporated herein into the specific embodiments as examples or embodiments, wherein each claim is independently a separate embodiment, and it is considered that these embodiments may be combined with each other in various combinations or arrangements. The scope of the present invention should be determined with reference to the claims and the full scope of equivalent forms granted by these claims.

Claims

1. A performance evaluation method for brain-computer interface electrodes, characterized in that: The brain-computer interface electrode is implanted in a specific brain region of the subject and comprises a given number of electrode channels, each electrode channel being used to record neural activity data in the specific brain region. The performance evaluation method comprises: Constructing an original data set based on the behavioral data of the subject performing a preset brain control task using a brain-computer interface and the neural activity data collected by each electrode channel corresponding to the behavioral data, and dividing the original data set into a training set and a test set; Performing multiple rounds of training on the neural signal decoding algorithm using the training data in the training set, wherein the number of electrode channels selected in each round of training is gradually reduced compared to the number of electrode channels in the previous round, and each round of training starts from the initial parameters of the neural signal decoding algorithm; The neural signal decoding algorithm trained in this round is tested using the test data in the test set to obtain the average decoding accuracy of this round, wherein each round of training includes multiple trainings, and the average decoding accuracy is obtained by testing the neural signal decoding algorithm trained in each round to obtain the average of multiple decoding accuracy rates; and, Based on the electrode evaluation feature data composed of the number of electrode channels and the average decoding accuracy of the neural signal decoding algorithm in each round, the performance of the brain-computer interface electrode is evaluated, specifically including: The number of electrode channels selected for each round and the average decoding accuracy of the round are recorded in pairs as electrode evaluation feature data for the round; based on the electrode evaluation feature data of each round, a characteristic curve of the number of electrode channels-decoding accuracy of the brain-computer interface electrode is drawn; the characteristic curve of the number of electrode channels-decoding accuracy reflects the corresponding relationship between the decoding accuracy of the trained neural signal decoding algorithm and the number of electrode channels selected for the training neural signal decoding algorithm; based on the characteristic curve of the number of electrode channels-decoding accuracy, the minimum number of electrode channels that can be selected while meeting the preset decoding accuracy requirement is obtained, and it is determined that the smaller the minimum number of electrode channels that can be selected while meeting the preset decoding accuracy requirement, the better the performance evaluation result of the brain-computer interface electrode.

2. The performance evaluation method according to claim 1, characterized in that: The performance evaluation method further comprises: Based on the electrode channel number-decoding accuracy characteristic curve, the minimum number of electrode channels that can be selected to meet the preset decoding accuracy requirement is obtained, and at least one of the electrode surface area characteristic parameter of the brain-computer interface electrode and the first electrode signal acquisition efficiency characteristic parameter is further determined, wherein: The product of the electrode surface area of ​​the brain-computer interface electrode and the minimum number of electrode channels is used as the electrode surface area characteristic parameter, and the inverse of the minimum number of electrode channels is used as the first electrode signal acquisition efficiency characteristic parameter, and the performance evaluation result of the brain-computer interface electrode is inversely proportional to the electrode surface area characteristic parameter and directly proportional to the first electrode signal acquisition efficiency characteristic parameter.

3. The performance evaluation method according to claim 1 or 2, characterized in that: The preset decoding accuracy requirement is the minimum decoding accuracy requirement that meets the needs of performing a preset brain control task.

4. The performance evaluation method according to claim 1, characterized in that: Based on each round of electrode evaluation feature data consisting of the number of electrode channels and the average decoding accuracy of the neural signal decoding algorithm, the performance of the brain-computer interface electrode is evaluated further including: Based on the electrode evaluation characteristic data, the highest decoding accuracy and its corresponding number of electrode channels are obtained, and the quotient of the highest decoding accuracy and the corresponding number of electrode channels is used as the second electrode signal acquisition efficiency characteristic parameter, and the performance evaluation result of the brain-computer interface electrode is proportional to the second electrode signal acquisition efficiency characteristic parameter.

5. The performance evaluation method according to claim 1 or 2, characterized in that: Constructing an original data set based on the behavioral data of the subject performing a preset brain control task using a brain-computer interface and the neural activity data collected by each electrode channel corresponding to the behavioral data further includes: An original data set is constructed based on the behavioral data of the subject performing a preset brain control task using a brain-computer interface, and the neural activity data collected by each electrode channel corresponding to the behavioral data, and the proportion of positive samples in the original data set is greater than a first threshold.

6. The performance evaluation method according to claim 1 or 2, characterized in that: Each round of training includes multiple training sessions, and the number of electrode channels used in each round of training gradually decreases compared to the number of electrode channels in the previous round, further including: In each round of training, the bootstrap sampling method is used to randomly discard at least part of the channel data in the previous round, so that the number of electrode channels selected in each round of training is gradually reduced compared with the number of electrode channels in the previous round, and the channel data used in each round of training are not exactly the same.

7. The performance evaluation method according to claim 1 or 2, characterized in that: The number of electrode channels selected in each round of training is gradually reduced compared with the number of electrode channels in the previous round. Specifically, the number of electrode channels selected in each round of training is reduced by a preset ratio compared with the number of electrode channels in the previous round.

8. The performance evaluation method according to claim 1 or 2, characterized in that: The neural signal decoding algorithm is one of Kalman filtering and neural network.

9. A performance evaluation method for different types of brain-computer interface electrodes, characterized in that: The performance evaluation method is used to compare and evaluate the performance of different types of brain-computer interface electrodes, wherein each type of brain-computer interface electrode is implanted in the same brain region of the same type of subject, and each type of brain-computer interface electrode contains the same or different numbers of electrode channels, and each type of brain-computer interface electrode channel is used to record neural activity data of the same signal type in the brain region where the electrode is implanted, and the performance evaluation method includes: For various types of brain-computer interface electrodes: respectively obtain the behavioral data of the respective subjects performing the same preset brain control task using the brain-computer interface, and the neural activity data collected by each electrode channel corresponding to the behavioral data, and execute the performance evaluation method for brain-computer interface electrodes as described in any one of claims 1-8 to obtain electrode evaluation feature data of various types of brain-computer interface electrodes, and use it to compare and evaluate the performance of different types of brain-computer interface electrodes; wherein various types of brain-computer interface electrodes all use the same neural signal decoding algorithm.

10. A performance evaluation method for brain-computer interface electrodes, characterized in that: The brain-computer interface electrode is implanted in a specific brain region of the subject and comprises a given number of electrode channels, each electrode channel being used to record neural activity data in the specific brain region. The performance evaluation method comprises: Recording behavioral data of the subject performing the same preset brain control task using the brain-computer interface in different time periods, as well as neural activity data collected by each electrode channel corresponding to the behavioral data; Based on the behavioral data of the subject using the brain-computer interface to perform the preset brain control task in each time period, and the neural activity data collected by each electrode channel corresponding to the behavioral data, the performance evaluation method for the brain-computer interface electrode according to any one of claims 1 to 8 is executed to obtain the performance evaluation results of the brain-computer interface electrodes corresponding to each time period; Based on the performance evaluation results of the brain-computer interface electrodes corresponding to each time period obtained, the time domain related performance of the brain-computer interface electrodes is evaluated.

11. A performance evaluation device for brain-computer interface electrodes, characterized in that: The brain-computer interface electrode is implanted in a specific brain region of the subject and comprises a given number of electrode channels, each of which is used to record neural activity data in the specific brain region. The performance evaluation device comprises at least an interface and a processor, wherein: The interface is configured to: obtain behavioral data of the subject performing a preset brain control task using the brain-computer interface, and neural activity data collected by each electrode channel corresponding to the behavioral data; The processor is configured to: execute the operations of the performance evaluation method for brain-computer interface electrodes as described in any one of claims 1-8.

12. A performance evaluation device for different types of brain-computer interface electrodes, characterized in that: The performance evaluation device is used to compare and evaluate the performance of different types of brain-computer interface electrodes, wherein each type of brain-computer interface electrode is implanted in the same brain region of the same type of subject, and each type of brain-computer interface electrode contains the same or different numbers of electrode channels, and each type of brain-computer interface electrode channel is used to record neural activity data of the same signal type in the brain region where the electrode is implanted, and the performance evaluation device includes at least an interface and a processor, wherein The interface is configured to: obtain behavioral data of subjects implanted with different types of brain-computer interface electrodes performing the same preset brain control task using the brain-computer interface, and neural activity data collected by each electrode channel corresponding to the behavioral data; The processor is configured to: execute the operations of the performance evaluation method for different types of brain-computer interface electrodes as described in claim 9, and wherein various types of brain-computer interface electrodes all use the same neural signal decoding algorithm.

13. A performance evaluation device for brain-computer interface electrodes, characterized in that: The brain-computer interface electrode is implanted in a specific brain region of the subject and comprises a given number of electrode channels, each of which is used to record neural activity data in the specific brain region. The performance evaluation device comprises at least an interface and a processor, wherein: The interface is configured to: obtain behavioral data recorded in different time periods when the subject uses the brain-computer interface to perform the same preset brain control task, and neural activity data collected by each electrode channel corresponding to the behavioral data; The processor is configured to: execute the operations of the performance evaluation method for brain-computer interface electrodes as described in claim 10.

14. A non-temporary computer-readable storage medium having computer-executable instructions stored thereon, wherein when the computer-executable instructions are executed by a processor, the operations of the performance evaluation method for brain-computer interface electrodes as described in any one of claims 1 to 8 and 10 are performed, or the operations of the performance evaluation method for different types of brain-computer interface electrodes as described in claim 9 are performed.

Images