EEG modeling method for motor imagery in patients with unilateral limbs based on neuron optimization
By performing neuron optimization of deep learning model of EEG data, the problem of model complexity and high computational cost in EEG signal analysis is solved, and the neurons and parameters are reduced while ensuring the accuracy of intention recognition, which is suitable for brain computers Interface real-time system.
Patent Information
- Application Number
- CN202210436298.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-04-25
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2042-04-25
AI Technical Summary
In the analysis of EEG signal, due to the complexity of the model and high computing cost, it is difficult to reduce training parameters and neurons while ensuring the accuracy of intention recognition, and adapt to the real-time online system requirements of the brain-computer interface.
Neuron optimization of deep learning models through the motor imaginary EEG data of unilateral limb patients, including preprocessing, feature extraction, average pooling, full-connection layer adjustment and parameter optimization, reduce redundant neurons, and use fast recursion algorithms Fitting with polynomial functions, reducing model size and parameters.
It realizes a real-time system suitable for brain-computer interfaces and is suitable for mobile and personalized rehabilitation strategies without affecting classification accuracy.
Smart Images

Figure CN114818803B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to an electroencephalogram (EEG) modeling method, and in particular to an EEG modeling method for motor imagery in patients with unilateral limbs based on neuron optimization. Background Art
[0002] Stroke is the leading cause of death and disability among adults in my country. It is characterized by high morbidity, disability, mortality, and recurrence rates, placing a heavy burden on individuals, families, and society[1]. Motor imagery brain-computer interface technology can determine the patient's movement intention by collecting and decoding the patient's EEG signals. The movement intention triggers the rehabilitation peripherals to perform treatment, achieving closed-loop training and active rehabilitation. Compared with traditional rehabilitation training technology, motor imagery brain-computer interface has a better effect on the rehabilitation training of stroke patients.
[0003] EEG signals are inherently non-stationary, nonlinear, and highly noisy, making it difficult to analyze and extract signal features relevant to a patient's motor intentions. Excessive redundant information interferes with the classification of patient intentions. Deep learning, a branch of traditional machine learning algorithms, has demonstrated excellent signal analysis capabilities in fields such as natural language processing and computer vision analysis, and has gradually become a mainstream approach for EEG signal analysis. However, to extract useful information from EEG signals, deep learning often requires building relatively complex models, which in turn requires more neurons for training. This increased number of model neurons also requires more training parameters and features for each learning iteration, significantly increasing training and testing time and computational costs. Given that brain-computer interface technology often requires real-time online systems to facilitate timely interaction between patients and computers or other rehabilitation devices, deep learning models need to reduce training and testing time. How to significantly reduce training parameters and eliminate redundant neurons to lower computational costs while maintaining accuracy in intention recognition remains a pressing technical challenge. Summary of the Invention
[0004] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides an EEG modeling method for motor imagery in patients with unilateral limbs based on neuron optimization.
[0005] The present invention solves the above technical problems through the following technical solutions: a method for EEG modeling of motor imagery for unilateral limb patients based on neuron optimization, characterized in that it comprises the following steps:
[0006] Step 1: Professionals help patients wear and use EEG acquisition equipment;
[0007] Step 2: The patient performs motor imagery according to the prompts, and the EEG acquisition device transmits the motor imagery EEG signals to the computer through wired or wireless transmission for personalized training modeling;
[0008] Step 3: After receiving the EEG data, the computer system first pre-processes the data, including filtering, baseline removal, and data slice integration;
[0009] Step 4: The processed data is fed into the deep learning model for feature extraction;
[0010] Step 5: After extracting the spatiotemporal features of the EEG data, average pooling is performed, summing every four neurons, and then calculating the average value as the output;
[0011] Step 6: The processed features are finally sent to the fully connected layer for integration and output to build the model;
[0012] Step 7: Reduce the number of neurons in the fully connected layer of the model and recalculate and adjust the parameters in the new layer;
[0013] Step 8: Calculate the difference between the output of the new connection layer and the original output;
[0014] Step nine: The final model after adjusting the parameters is used for the patient's subsequent movement intention recognition and judgment.
[0015] Preferably, the step six uses a fully connected layer to integrate the features and finally outputs the classification result through a softmax activation function.
[0016] Preferably, in step 4, a one-dimensional convolutional layer is first used to extract the temporal features of each channel of the EEG data, and then a deep convolutional layer is used to learn the spatial features of the EEG data. The obtained spatiotemporal features will be sent to the next pooling layer.
[0017] Preferably, the EEG acquisition device adopts the Quick30 series lead dry electrode cap of CGX company.
[0018] Preferably, the step seven uses a fast recursive algorithm to optimize the parameters of the fully connected layer and remove redundant neurons.
[0019] Preferably, in steps seven to nine, each input neuron on the fully connected layer is sequentially superimposed from one to all as a new input, and a polynomial function is used to fit the output, and the fitted output is then compared with the output of the original model.
[0020] The positive progress effect of the present invention is:
[0021] First, the present invention uses deep learning to decode the EEG data of motor imagery in patients with unilateral limbs. After modeling, the neurons in the fully connected layer are optimized to achieve similar output results using fewer neurons without compromising classification accuracy. The reduced number of neurons in the final model means fewer parameters, lower computational costs, and shorter response times, which facilitates the rapid and accurate recognition and classification of motor imagery intentions by real-time brain-computer interfaces (BCIs).
[0022] Second, this method also reduces model size, reducing the amount of computer memory used. As EEG devices become more mobile and miniaturized, models will likely be stored on mobile chips in the future. This reduced model footprint allows more models to be stored on the same chip, providing personalized rehabilitation strategies for more patients and facilitating the development of additional features for researchers. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] Figure 1 This is a flow chart of the EEG modeling method for motor imagery in unilateral limb patients based on neuron optimization of the present invention. DETAILED DESCRIPTION
[0024] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.
[0025] In the description of the present invention, it should be understood that the terms "upper", "lower", "front", "back", "left", "right", "top", "bottom", "inside", "outside", etc., indicating directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as limiting the present invention.
[0026] like Figure 1 As shown, the present invention's method for modeling motor imagery EEG for unilateral limb patients based on neuron optimization includes the following steps:
[0027] Step 1: Professionals help patients wear and use EEG acquisition equipment;
[0028] Step 2: The patient performs motor imagery according to the prompts, and the EEG acquisition device transmits the motor imagery EEG signals to the computer through wired or wireless transmission for personalized training modeling;
[0029] Step three: After receiving the EEG data, the computer system first preprocesses the data, including filtering, baseline removal, and data slicing and integration; this can effectively filter out useless information in the EEG signal, reduce data redundancy, and improve the signal-to-noise ratio, so that the deep learning model can better learn the effective features in the EEG signal.
[0030] Step 4: The processed data is fed into a deep learning model for feature extraction. First, a one-dimensional convolutional layer is used to extract the temporal features of each EEG data pass, and then a deep convolutional layer is used to learn the spatial features of the EEG data. The obtained spatiotemporal features are fed into the next pooling layer.
[0031] In step 5, after extracting the spatiotemporal features of the EEG data, average pooling is performed, summing every four neurons and calculating the average as the output. The step size of average pooling is also four to ensure that the pooled content is not repeated.
[0032] Step 6: The processed features are finally sent to the fully connected layer for integration and output to build the model;
[0033] Step 7: Reduce the number of neurons in the fully connected layer of the model and recalculate and adjust the parameters in the new layer;
[0034] Step 8: Calculate the difference between the output of the new connection layer and the original output. If the difference is too large, it means that the number of neurons used is too small, and the classification curve is not well fitted, resulting in an inability to properly classify data with different labels. If the difference is too small, it means that the number of neurons is large enough, the fitting effect is good enough, and there is room for further improvement to reduce the number of neurons. After continuous attempts, set a threshold so that the difference between the new output and the original output is not large, while greatly reducing the number of neurons used;
[0035] Step nine: The final model after adjusting the parameters is used for the patient's subsequent movement intention recognition and judgment.
[0036] In step six, a fully connected layer is used to integrate the features and finally output the classification results through the softmax activation function. In this way, all features will be integrated into one layer, and these features will affect the final classification and better judge the patient's movement intention.
[0037] In step seven, a fast recursive algorithm is used to optimize the parameters of the fully connected layer, remove redundant neurons, and reduce the model size. The final model can save a lot of time and memory space and quickly provide classification results of the patient's EEG intentions.
[0038] In steps 7 to 9, each input neuron on the fully connected layer is superimposed one to all in sequence as a new input, and the output is fitted with a polynomial function. The fitted output is then compared with the output of the original model. If the difference between the two is lower than a preset threshold (i.e., the difference is very small), then it can be considered that even if some neurons are deleted, a similar output can still be obtained. During the fitting process, new weights and biases are calculated and saved under the condition of reducing neurons. The final model will use these new parameters. This method can reduce the size of the final model, reduce parameters, and reduce computational costs while ensuring that the recognition of the patient's intentions is not affected.
[0039] The neurons in the fully connected layer have a large number of redundant parameters. When using the established model to discriminate patient intent, even if the weights and biases of some neuron parameters are deleted, the final classification result will be largely unaffected. This method can significantly reduce the final parameters used by the model, shrink the model's memory usage, and lower computational costs.
[0040] The EEG acquisition device can use CGX's Quick30 series lead dry electrode cap, which is connected to the computer via wired serial communication or wireless Bluetooth. Hospital professionals use the computer to help patients collect motor imagery sample data. It has a simple structure and is easy to connect and use.
[0041] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention. Therefore, from all points of view, the embodiments should be regarded as illustrative and non-restrictive. The scope of the present invention is defined by the appended claims rather than the above description, and it is intended that all changes that fall within the meaning and range of equivalents of the claims are included in the present invention. Any reference signs in the claims should not be construed as limiting the claim to which they relate.
[0042] In addition, it should be understood that although this specification is described in terms of implementation methods, not every implementation method contains only one independent technical solution. This narrative method of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.
Claims
1. A method for EEG modeling of motor imagery in unilateral limb patients based on neuron optimization, characterized in that: It includes the following steps: Step 1: Professionals help patients wear and use EEG acquisition equipment; Step 2: The patient performs motor imagery according to the prompts, and the EEG acquisition device transmits the motor imagery EEG signals to the computer through wired or wireless transmission for personalized training modeling; Step 3: After receiving the EEG data, the computer system first pre-processes the data, including filtering, baseline removal, and data slice integration; Step 4: The processed data is fed into the deep learning model for feature extraction; Step 5: After extracting the spatiotemporal features of the EEG data, average pooling is performed, summing every four neurons, and then calculating the average value as the output; Step 6: The processed features are finally sent to the fully connected layer for integration and output to build the model; Step 7: Reduce the number of neurons in the fully connected layer of the model and recalculate and adjust the parameters in the new layer; Step 8: Calculate the difference between the output of the new connection layer and the original output; Step nine, the final model after adjusting the parameters is used for the patient's subsequent movement intention recognition and judgment; The step 7 uses a fast recursive algorithm to optimize the parameters of the fully connected layer and remove redundant neurons. In steps 7 to 9, each input neuron on the fully connected layer is stacked from one to all in sequence as a new input, and the output is fitted with a polynomial function. The fitted output is then compared with the output of the original model. If the difference between the two is lower than a preset threshold, it is considered that even if some neurons are deleted, an output of similar size can be obtained. At the same time, new weights and biases are calculated and saved under the condition of reducing neurons, and the final model uses these new parameters.
2. The method for EEG modeling of motor imagery for unilateral limb patients based on neuron optimization as claimed in claim 1, characterized in that: The step six uses a fully connected layer to integrate the features and finally outputs the classification result through a softmax activation function.
3. The method for EEG modeling of motor imagery for unilateral limb patients based on neuron optimization as claimed in claim 1, characterized in that: In step 4, a one-dimensional convolutional layer is first used to extract the temporal features of each channel of the EEG data, and then a deep convolutional layer is used to learn the spatial features of the EEG data. The obtained spatiotemporal features are sent to the next pooling layer.
4. The method for EEG modeling of motor imagery for unilateral limb patients based on neuron optimization as claimed in claim 1, characterized in that: The EEG acquisition device adopts the Quick30 series lead dry electrode cap of CGX Company.
Citation Information
Cited By
Limb nerve rehabilitation state evaluation method based on multi-modal sensor data fusion
CN119318470A
Method for evaluating limb nerve rehabilitation status based on multi-modal sensor data fusion
CN119318470B
