Transcranial photostimulation regulation and control method and system

Abstract

The invention provides a transcranial photostimulation regulation and control method and system, and a brain light field model module in the system is used for constructing a target multi-classifier, and outputting transcranial photostimulation parameter information based on a target patient data set by using the target multi-classifier; the virtual reality module is used for providing a virtual reality rehabilitation training task of the user; the light emission control module is used for adjusting parameters of the light stimulation module according to the output transcranial light stimulation parameters, and the light stimulation module is used for carrying out light stimulation on a patient; the information acquisition and analysis module is used for acquiring and analyzing eye movement and face video information of a user in a virtual reality rehabilitation training task process in a transcranial photostimulation process; the brain function evaluation module is used for evaluating the brain function condition of the patient before and after transcranial light regulation; and the feedback module is used for feeding the eye movement and face video information output by the information acquisition and analysis module and the brain function information output by the brain function evaluation module back to the light emission control module to adjust the light stimulation parameters, so as to improve the transcranial light stimulation regulation efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of medical devices, and in particular, to a transcranial light stimulation regulation method and system. Background Art

[0002] With the aging of the world's population, the prevalence of brain dysfunction such as stroke and cognitive impairment has gradually increased, seriously threatening the health of the elderly and bringing a heavy burden to families and society. Its prevention and treatment is an important challenge. As a new non-invasive cranial nerve stimulation regulation technology, transcranial light irradiates the brain through low-intensity infrared light (wavelength 625 - 740nm) or near-infrared light (wavelength 750 - 1100nm) to penetrate the skull, thereby regulating brain neurons. It has the characteristics of high safety, high adaptability, and non-invasiveness, and has received more and more attention in the field of neuromodulation.

[0003] How to determine the transcranial light parameters used in transcranial light stimulation regulation to achieve better regulation effects is a very important issue. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a transcranial light stimulation regulation method and system, which can timely and accurately adjust the parameters of the light emission unit and improve the efficiency of transcranial light stimulation regulation.

[0005] According to one aspect of the present invention, a transcranial light stimulation regulation method is provided. The method includes:

[0006] Obtaining historical sample data of multiple groups of patients with brain dysfunction in the historical record and sample labels of the historical sample data to obtain a target training set;

[0007] Obtaining multiple groups of target patient data of a target patient to be subjected to transcranial light stimulation to obtain a target patient data set;

[0008] Training a target multi-classifier based on each piece of the historical sample data and sample labels in the target training set, where the target multi-classifier is used to output transcranial light stimulation parameters;

[0009] Inputting the target patient data set into the target multi-classifier so that the target multi-classifier outputs the target transcranial light stimulation parameters of the target patient based on the target patient data set;

[0010] Using the target transcranial light stimulation parameters to perform transcranial light stimulation regulation on the target patient, and collecting target brain function feedback information, target eye movement information, and target facial video information of the target patient during the transcranial light stimulation regulation process;

[0011] Based on the target eye movement information and the corresponding relationship between the preset eye movement information and the visual spatial attention index, obtain the target visual spatial attention index of the target patient; use a pre-trained convolutional neural network to obtain the target emotion index of the target patient based on the target facial video information;

[0012] Adjust the target transcranial light stimulation parameters based on the target brain function feedback information, the target visual spatial attention index, and the target emotion index of the target patient;

[0013] Adjust the light emission parameters of the light emission unit according to the target transcranial light stimulation regulation parameters, where the light emission parameters include one or more of the wavelength, frequency, pulse width, and emission time of the light.

[0014] According to another aspect of the present invention, a transcranial light stimulation regulation system is provided, and the system includes:

[0015] A first acquisition module, configured to acquire historical sample data of multiple groups of patients with brain dysfunction in the history and sample labels of the historical sample data to obtain a target training set;

[0016] A second acquisition module, configured to acquire multiple groups of target patient data of a target patient to be subjected to transcranial light stimulation regulation to obtain a target patient data set;

[0017] A classifier training module, configured to train a target multi-classifier based on each piece of the historical sample data and the sample labels in the target training set, and the target multi-classifier is used to output transcranial light stimulation parameters;

[0018] A parameter output module, configured to input the target patient data set into the target multi-classifier, so that the target multi-classifier outputs the target transcranial light stimulation parameters of the target patient based on the target patient data set;

[0019] An information acquisition module, configured to perform transcranial light stimulation regulation on the target patient by using the target transcranial light stimulation parameters, and collect the target brain function feedback information, the target eye movement information, and the target facial video information of the target patient during the transcranial light stimulation regulation process;

[0020] An index calculation module, configured to obtain the target visual spatial attention index of the target patient based on the target eye movement information and the corresponding relationship between the preset eye movement information and the visual spatial attention index; and obtain the target emotion index of the target patient based on the target facial video information by using a pre-trained convolutional neural network;

[0021] A parameter adjustment module, configured to adjust the target transcranial light stimulation parameters based on the target brain function feedback information, the target visual spatial attention index, and the target emotion index of the target patient;

[0022] A light emitting unit for adjusting light emission parameters according to the target transcranial light stimulation parameters, where the light emission parameters include one or more of the wavelength, frequency, pulse width, and emission time of light.

[0023] According to another aspect of the present invention, there is provided a non-transitory computer-readable storage medium storing computer instructions, where the computer instructions are used to cause a computer to execute the transcranial light stimulation regulation method described in any one of the above.

[0024] In one or more technical solutions provided in the embodiments of the present invention, an existing target multi-classifier is trained using patient data, and the target multi-classifier is used to specifically output target transcranial light stimulation parameters based on multiple sets of data of a target patient. Furthermore, the target transcranial light stimulation parameters are used to perform transcranial light stimulation regulation on the target patient, improving the acquisition efficiency and accuracy of the target transcranial light stimulation parameters and the regulation efficiency. Moreover, by collecting the brain function, eye movement, and facial video information of the target patient during the regulation process and performing feedback adjustment on the target transcranial light stimulation parameters based on this information, the transcranial light can timely respond to the feedback of the patient on the transcranial light stimulation parameters, and thus can timely and accurately adjust the parameters of the light emitter, further improving the transcranial light nerve stimulation regulation effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In the following description of exemplary embodiments in conjunction with the drawings, more details, features, and advantages of the present invention are disclosed. In the drawings:

[0026] Figure 1 It is a schematic flowchart of a transcranial light stimulation regulation method provided by an embodiment of the present invention;

[0027] Figure 2 It is a schematic flowchart of obtaining target transcranial light stimulation parameters in the transcranial light stimulation regulation method provided by an embodiment of the present invention;

[0028] Figure 3 It is a schematic logical structure diagram of a transcranial light stimulation regulation system provided by an embodiment of the present invention;

[0029] Figure 4 It is another schematic logical structure diagram of a transcranial light stimulation regulation system provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0030] Embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although some embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to more thoroughly and completely understand the present invention. It should be understood that the drawings and embodiments of the present invention are only for exemplary purposes and are not used to limit the protection scope of the present invention.

[0031] It should be understood that the various steps recited in the method embodiments of the present invention can be executed in a different order and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this regard.

[0032] As used herein, the term "comprising" and its variations are open-ended, that is, "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description. It should be noted that the concepts such as "first" and "second" mentioned in the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order of the functions executed by these devices, modules or units or their interdependent relationships.

[0033] It should be noted that the modifications of "one" and "plural" mentioned in the present invention are illustrative rather than restrictive. Those skilled in the art should understand that unless otherwise clearly specified in the context, it should be understood as "one or more".

[0034] The names of the messages or information exchanged between multiple devices in the embodiments of the present invention are only for illustrative purposes and are not used to limit the scope of these messages or information.

[0035] In order to improve the efficiency of patient rehabilitation training, embodiments of the present invention provide a transcranial light stimulation regulation method, system and storage medium. The solution of the present invention will be described below with reference to the accompanying drawings:

[0036] Figure 1 It is a schematic flowchart of a transcranial light stimulation regulation method provided by an embodiment of the present invention, and may include the following steps:

[0037] S101. Obtain historical sample data of multiple groups of patients with brain dysfunction in the historical record and the sample labels of the historical sample data to obtain a target training set; wherein, each group of the historical sample data includes near-infrared brain function data, eye movement data, and facial video data; the sample label is the actual transcranial light stimulation parameter corresponding to the historical sample data, and the actual transcranial light stimulation parameter includes at least stimulation wavelength, stimulation frequency, stimulation pulse width, and stimulation time;

[0038] S102. Obtain multiple groups of target patient data of the target patient to be subjected to transcranial light stimulation to obtain a target patient data set, and each group of the target patient data includes the near-infrared brain function data, eye movement data, and facial video data of the target patient;

[0039] S103. Train a target multi-classifier based on each piece of the historical sample data and the sample label in the target training set, and the target multi-classifier is used to output transcranial light stimulation parameters;

[0040] S104. Input the target patient data set into the target multi-classifier so that the target multi-classifier outputs the target transcranial light stimulation parameter of the target patient based on the target patient data set;

[0041] S105. Perform transcranial light stimulation on the target patient using the target transcranial light stimulation parameter, and collect the target brain function feedback information, target eye movement information, and target facial video information of the target patient during the transcranial light stimulation process; wherein, the target brain function feedback information is used to characterize the change in brain function data of the target patient before and after transcranial light stimulation;

[0042] S106. Based on the target eye movement information and the corresponding relationship between the preset eye movement information and the visuospatial attention index, obtain the target visuospatial attention index of the target patient; use the pre-trained convolutional neural network to obtain the target emotion index of the target patient based on the target facial video information;

[0043] S107. Adjust the target transcranial light stimulation parameter based on the target brain function feedback information, target visuospatial attention index, and target emotion index of the target patient.

[0044] S108. Adjust the light emission parameters of the light emission unit according to the target transcranial light stimulation parameter, wherein the light emission parameters include the wavelength, frequency, pulse width, and emission time of the light.

[0045] By applying the embodiment of the present invention, a target multi-classifier is trained using existing patient data, and the target multi-classifier is used to output target transcranial light stimulation parameters based on multiple groups of data of the target patient, and then the target transcranial light stimulation parameters are used to perform light stimulation on the target patient, thereby improving the efficiency and accuracy of obtaining the target transcranial light stimulation parameters, and improving efficiency. Furthermore, by collecting the target patient's brain function, eye movement and facial video information during the stimulation process, and feedback-adjusting the target transcranial light stimulation parameters based on this information, the transcranial light can respond to the patient's feedback on the transcranial light stimulation parameters in a timely manner, thereby improving the transcranial light stimulation effect.

[0046] The above S101-S108 are exemplarily described below:

[0047] In a possible embodiment, with the patient's consent, various preset indicator data and corresponding transcranial light stimulation parameters of patients with brain dysfunction can be recorded during transcranial light stimulation, and the preset indicator data and transcranial light stimulation parameters can be stored in a database in correspondence. Exemplarily, during the stimulation process, various preset indicator data and corresponding transcranial light stimulation parameters of the patient can be collected according to a preset collection cycle, and the collection cycle can be set according to the actual application scenario. Exemplarily, near-infrared brain function parameters can be collected every 15 minutes or 20 minutes, and one collection lasts for 15 or 20 minutes. Eye movement data and facial video data can be continuously collected during the light stimulation regulation process. For example, if the light stimulation lasts for 20 minutes, the patient's eye movement data and facial video data can be continuously collected within the 20 minutes.

[0048] The above-mentioned preset indicators may include the patient's near-infrared brain function data, eye movement data, and facial video data, wherein the near-infrared brain function data refers to the physiological and functional information about brain activity obtained through near-infrared spectroscopy (NIRS). This technology uses the characteristics of near-infrared light to penetrate the scalp and skull to detect hemodynamic changes in the cerebral cortex. Eye movement data refers to visual behavior data recorded and analyzed by eye tracking technology, which may include gaze point, gaze time and number, eye saccade distance, and pupil size.

[0049] In actual application scenarios, virtual reality (VR) rehabilitation training technology is used to regulate transcranial light stimulation for patients. Through the use of electronic device modeling and simulation, the virtual reality rehabilitation training technology enables patients to interact with a 3D visual or other sensory environment. As a possible implementation method, during the virtual reality rehabilitation training process, devices such as a brain functional imaging device, a near-infrared spectroscopy device, a camera, and an eye tracker can be connected to collect the preset index data of the patient. Specifically, near-infrared brain function data of the patient can be collected through the brain functional imaging device and the near-infrared spectroscopy device, eye movement data of the patient can be collected using the eye tracker, and facial video data of the patient can be collected using a depth camera.

[0050] In a possible embodiment, the acquisition timestamp can be used as a reference to distinguish different groups of preset index data of the patient and the corresponding transcranial light stimulation parameters. Exemplarily, when collecting the above-mentioned preset index data and transcranial light stimulation parameters, a timestamp mark can be added to the collected data, and the preset index data and transcranial light stimulation parameters with the same timestamp mark are stored in the database as a group of historical data.

[0051] In S101, all the data in the database can be obtained to construct the target training set, or partial data in the database can be obtained to form the target training set. The present invention does not make specific limitations on this. The target training set includes multiple groups of historical sample data of different patients and the corresponding sample labels. The historical sample data is the above-mentioned preset index data, and the sample label is the transcranial light stimulation parameter. Exemplarily, the transcranial light stimulation parameter can include information such as stimulation wavelength, stimulation frequency, stimulation pulse width, and stimulation time.

[0052] In S102, multiple groups of near-infrared brain function information, eye movement information, and facial video information of the target patient can be collected through the above data collection process to obtain the target patient dataset, which will not be elaborated here.

[0053] In a possible embodiment, after obtaining the above-mentioned target training set, a multi-classifier can be trained based on the target training set. The multi-classifier can be constructed using any feasible structure. Exemplarily, the classifier can include an input layer, a hidden layer, and an output layer. The input layer is used to receive the input of data, the hidden layer is used to extract data features, and the input layer is used to output results based on the extracted data features. As a possible implementation method, the output layer of the multi-classifier can include multiple output units, and each output unit is used to output different types of transcranial light stimulation parameters. Exemplarily, when the transcranial light stimulation parameters include stimulation wavelength, stimulation frequency, stimulation pulse width, and stimulation time, the number of the above output units can be 4, and the 4 output units are respectively used to output the stimulation wavelength, stimulation frequency, stimulation pulse width, and stimulation time.

[0054] In the present invention, the multi-classifier can be trained by any feasible method. Exemplarily, each piece of historical sample data in the target training set can be output to the multi-classifier to be trained, and the transcranial light stimulation prediction parameters output by the multi-classifier to be trained are obtained. The light stimulation prediction parameters include a predicted wavelength, a predicted frequency, a predicted pulse width, and a predicted time. Based on the difference between the preset parameters and the sample labels corresponding to the historical sample data, the parameters of the multi-classifier are adjusted until the difference converges, and a target multi-classifier is obtained.

[0055] After obtaining the target multi-classifier, the target patient data set can be input into the target multi-classifier, and the predicted value of the light stimulation parameter output by the target multi-classifier based on the target patient data set is obtained as the target light stimulation parameter.

[0056] In a possible embodiment, in order to improve the accuracy of the multi-classifier and thus improve the prediction accuracy of the target light stimulation parameter, the target training set and the target patient data set can be obtained through the following steps:

[0057] S111. Obtain a first source data set and a first target data set. The first source data set includes multiple groups of the historical sample data and the sample labels of each group of the historical sample data, and the first target data set includes multiple groups of target patient data of the target patient.

[0058] The historical sample data and the corresponding sample labels contained in the first source data set are obtained from the database. The first target data set includes multiple groups of near-infrared brain function data, eye movement data, and facial video data of the target patient collected according to a preset collection period.

[0059] S112. Calculate the first similarity between multiple groups of the historical sample data and the first target data set respectively, and delete the historical sample data with the first similarity higher than a first preset threshold from the first source data set to obtain a second source data set.

[0060] Both the historical sample data and the data of the target patient include multiple data types, which may include the above-mentioned near-infrared brain function data, eye movement data, and facial video data, etc. As a possible implementation manner, the similarity between the first source data set and the first target data set can be calculated based on the above data types, and the similarity calculation can be performed by any feasible method. Exemplarily, the cosine similarity distance, Mahalanobis similarity distance, etc. between each type of data can be calculated to calculate the similarity between the sample data and the target data.

[0061] Exemplarily, the similarity between each group of data in the first source data set and the first target data set can be calculated by the following formula:

[0062] A(S 1 , D 1 ) = (S 1nty - D 1nty ) 2 + (S 1yty - D 1yty ) 2 + (S 1mty - D 1mty ) 2

[0063] Wherein, S 1 is the first source data set, D 1 is the first target data set, S 1nty， S 1yty， S 1mty are the near-infrared brain function information data, eye movement information data, and facial video information data in the historical sample data of the first source data set respectively, and D 1nty， D 1yty， D 1mty are the near-infrared brain function information data, eye movement information data, and facial video information data in each patient data of the first target data set respectively. A(S 1 , D 1 ) represents the similarity distance between a set of historical sample data in the first source data set and the target patient data in the first target data set.

[0064] Exemplarily, if the first source data set includes historical sample data A, B, and C, and the first target data set includes target patient data a, b, and c, then the statistical value of the similarity between A and a, b, and c can be calculated as the similarity between A and the first target data set. The above statistical value can be an average value, a median value, etc.

[0065] In another possible embodiment, the similarity between each group of historical sample data in the first source data set and each group of target patient data in the first target data set can also be calculated separately.

[0066] When the similarity distance is higher than the first preset threshold, it indicates that there is a large difference between the historical sample data of the first source data set and the target patient data in the first target data set. Therefore, the historical sample data can be removed from the first source data set. Remove each historical sample data with a similarity distance greater than the first preset threshold between the first source data set and the first target data set, so as to obtain the second source data set.

[0067] S113. Classify the second source data set using the first target classifier to obtain the classification results of each group of second sample data included in the second source data set, and delete the second sample data with incorrect classification results from the second source data set to obtain the third source data set.

[0068] The above first target classifier may be an SVM (Support Vector Machine), a DT (Decision Tree), a CNN (Convolutional Neural Network), an LSTM (Long Short-Term Memory networks), etc., and the specific structure may be selected according to the actual application scenario.

[0069] In a possible embodiment, the above first target classifier may be trained based on each second historical sample data and the sample label. Exemplarily, the above first target classifier may be selected through the following steps:

[0070] S1131. Test multiple candidate classifiers based on each second sample data in the second source dataset, and obtain multiple candidate classifiers with classification accuracy higher than a preset accuracy threshold based on the sample labels corresponding to each second sample data in the second source dataset as the to-be-determined classifiers; the number of the to-be-determined classifiers is preset.

[0071] In this step, the second source dataset can be used to train each candidate classifier, and the candidate classifier may include SVM, DT, CNN, and LSTM, etc. As a possible implementation manner, each second sample data included in the second dataset can be input into each candidate classifier, and the classification prediction result output by each candidate classifier for each second sample data can be obtained. Each candidate classifier can be trained based on the difference between the classification prediction result and the sample label of the second sample data until the difference converges. And each second sample data is input into each trained candidate classifier, and based on the proportion of the correct classification prediction results output by the candidate classifier for each second sample data in the number of second sample data, the prediction accuracy of each candidate classifier is determined, and then multiple candidate classifiers with the highest accuracy are selected as the to-be-determined classifiers. The above preset number can be selected according to the actual application scenario, such as it can be 4.

[0072] Exemplarily, after obtaining the prediction accuracy of each candidate classifier, each candidate classifier can be sorted in descending order of prediction accuracy, and the first 4 candidate classifiers in the sorting are selected as the to-be-determined classifiers.

[0073] In a possible embodiment, if there are candidate classifiers with equal prediction accuracy, resulting in the number of classifiers with the highest accuracy being more than the preset number, then the to-be-determined classifiers can be randomly selected from the candidate classifiers with equal prediction accuracy so that the number of to-be-determined classifiers meets the above preset number.

[0074] In a possible embodiment, each candidate classifier can be trained using the historical sample data stored in the database and the sample labels of the historical sample data, and the prediction accuracy of each candidate classifier can be tested using each second sample data in the second source dataset, and then multiple candidate classifiers can be selected as the to-be-determined classifiers; the number of the to-be-determined classifiers is preset.

[0075] After that, the first target classifier can be determined based on the prediction result accuracy of each of the to-be-determined classifiers for each target patient data included in the first target dataset. Specifically, the process of determining the first target classifier may further include the following steps:

[0076] S1132. Classify each group of target patient data in the first target dataset using each of the to-be-determined classifiers to obtain multiple candidate classification labels corresponding to the first target dataset. In the case where there are duplicate candidate classification labels, the duplicate candidate classification labels are used as the target label classification result of the first target dataset; wherein, the target label classification result is the light stimulation parameter predicted for the first target dataset.

[0077] As described above, there are multiple to-be-determined classifiers, and the predicted classification results output by each to-be-determined classifier for each patient data in the first target dataset may be the same or different, and this classification result is the light stimulation parameter. In this step, if the same predicted classification result appears, the predicted classification result can be retained as the target label classification result corresponding to the corresponding patient data. The above "same" may mean that there are two or more identical candidate classification labels among the candidate classification labels. Exemplarily, the number of to-be-determined classifiers is 4, and the candidate classification labels output by the 4 to-be-determined classifiers for the same group of patient data are A, A, B, and C respectively. Therefore, the predicted classification result A can be used as the target label classification result corresponding to this group of patient data.

[0078] S1133. Classify each group of target patient data in the first target dataset using each of the to-be-determined classifiers to obtain the to-be-determined classification results.

[0079] S1134. Determine the classification accuracy of each of the to-be-determined classifiers based on the to-be-determined classification results and the target label classification results, and select the to-be-determined classifier with the highest classification accuracy as the first target classifier.

[0080] In a possible embodiment, each target patient data in the first target data set can be input into each to-be-determined classifier, and the classification result prediction value output by each to-be-determined classifier for the patient data can be obtained as the to-be-determined classification result. Based on the target label classification result corresponding to the patient data, it is determined whether the to-be-determined classification result is correct, and then the accuracy rate of each to-be-determined classifier for the first target data set is obtained, so as to select the to-be-determined classifier with the highest accuracy rate as the first target classifier.

[0081] Through the above technical solution, the candidate classifier is tested multiple times using the second source data set and the first target classifier, and then the candidate classifier with the best performance, that is, the highest prediction accuracy rate, in the two data sets is selected as the first target classifier, which helps to improve the accuracy of the classification result output for the second source data set.

[0082] In a possible embodiment, each second sample data included in the second source data set can be input into the first target classifier, and the classification result output by the first target classifier for each second sample data can be obtained. This classification result is the predicted value of the light stimulation parameter output by the first target classifier for the second sample data. Correspondingly, if the classification result corresponding to the second sample data is inconsistent with the sample label corresponding to the second sample, it indicates that there is an abnormality in the second sample data or its sample label. Therefore, the second sample label can be removed from the second source data set to obtain the third source data set.

[0083] In a possible embodiment, the similarity distance between each third sample data in the third source data set and each target patient data in the first target data set can be calculated, and the third samples and target patient data with a similarity distance greater than the second preset threshold can be deleted from the third source data set and the first target data set to obtain the target training set and the target patient data set. The second preset threshold can be higher than the above first preset threshold.

[0084] As a possible implementation manner, the process of obtaining the target training set and the target patient data set may further include the following steps:

[0085] S114. Input each group of third sample data in the third source data set and each group of target sample data in the first target data set into a plurality of preset binary classifiers respectively, and obtain the loss function values output by each of the preset classifiers; different data types are included in the loss functions of each of the preset binary classifiers, and the data types include: near-infrared brain function data, eye movement data, and facial video data; the number of the preset binary classifiers is the same as the number of data types included in the sample data.

[0086] In this step, a set of third sample data and a set of target patient data can be input into each preset binary classifier simultaneously. The preset binary classifier can output a result indicating whether they are similar based on the similarity between the third sample data and the target patient data. The above-mentioned multiple preset binary classifiers can be selected according to the actual application scenario. For example, they can be the above-mentioned SVM, DT, CNN, LSTM, etc. The above-mentioned multiple preset binary classifiers can be of the same type or different types, and the present invention does not make specific limitations on this.

[0087] The loss functions used by the above-mentioned preset binary classifiers are different. Specifically, the loss functions used by each classifier can be obtained based on different types of historical sample data and patient data. Exemplarily, each loss function can be a cross-entropy loss function obtained based on near-infrared brain function information, eye movement information, and facial video information respectively.

[0088] As a possible implementation manner, the loss functions of the above-mentioned multiple preset binary classifiers can be the following functions respectively:

[0089]

[0090] Among them, G(f nsy ) is the cross-entropy loss function of the near-infrared brain function information data binary classifier f nsy . n and m are the numbers of the third sample data in the third source dataset and the target patient data in the first target dataset respectively. S 3nsyi is the i-th near-infrared brain function sample data in the third source dataset, and D 1nsyj is the j-th near-infrared brain function data in the first target dataset. α is a preset first correction coefficient.

[0091]

[0092] Among them, G(f ysy ) is the cross-entropy loss function of the eye movement information data binary classifier f ysy . n and m are the numbers of the third sample data in the third source dataset and the sample numbers of the target patient data in the first target dataset respectively. S 3ysyi is the i-th eye movement sample data in the third source dataset, and D 1nsyj is the j-th eye movement data in the first target dataset. β is a preset second correction coefficient.

[0093]

[0094] Among them, G(f ysy ) is the facial video information data binary classifier f msyThe cross-entropy loss function, where n and m are the number of third sample data in the third source dataset and the number of target patient data in the first target dataset respectively, S 3msyi is the i-th facial video sample data in the third source dataset, D 1msyj is the j-th facial video data in the first target dataset, and γ is a preset third correction coefficient.

[0095] S115. When the values of all the loss functions reach the minimum, determine the target data type corresponding to the preset binary classifier with the lowest accuracy among all the preset binary classifiers.

[0096] In this step, based on the similarity results between the third sample data output by each preset binary classifier and the target patient data, the preset binary classifier with the largest difference from the similarity results output by other binary classifiers can be determined as the preset binary classifier with the lowest accuracy. In another possible embodiment, the preset binary classifier with the lowest accuracy can also be determined based on the above similarity results and the actual similarity results between the sample labels corresponding to the third sample data and the target classification labels corresponding to the target patient data.

[0097] As described above, different preset binary classifiers set cross-entropy loss functions based on different types of data. Therefore, in this step, the data type included in the cross-entropy loss function used by the preset binary classifier with the lowest accuracy can be used as the target data type.

[0098] S116. Calculate the similarity distance between the target data type in the third source dataset and the target data type in the first target dataset, and delete the third sample data and the target patient data with the similarity distance greater than the second preset threshold from the third source dataset and the first target dataset to obtain a fourth source dataset and a second target dataset. Then, use the fourth source dataset as the target training set and the second target dataset as the target patient dataset.

[0099] Exemplarily, when the values of the above three loss functions all reach the minimum, the accuracy of the binary classifier corresponding to the near-infrared brain function information data is the lowest. Then, the similarity distance between the near-infrared brain function data in the third sample data in the third source dataset and the near-infrared brain function information data in the target patient data can be calculated. If the similarity distance is greater than the preset distance threshold, the third sample data can be deleted, and at the same time, the corresponding target patient data can be deleted from the first target dataset to obtain the second target dataset.

[0100] By means of the above technical means, the original source data set and the target patient data set are screened to obtain a target training set and a target data set, so that the data similarity between the target training set and the target data set is relatively high. Furthermore, when using the multi-classifier trained in the target training set to predict the parameters of the target data set, a better effect can be achieved, and the accuracy of the obtained target transcranial light stimulation parameters can be further improved.

[0101] As Figure 2 shown, Figure 2 FIG. is a schematic flowchart of obtaining light stimulation parameters in an embodiment of the present invention. Specifically, first, a first source data set and a first target data set are obtained. Based on the similarity between the data in the first source data set and the first target data set, sample data in the first source data set with a similarity lower than a first preset threshold to the first target data set is removed to obtain a second data set. The first target classifier is screened using the second source data set and the first target data set, and the classification results of each second sample data in the second source data set are calculated using the first target classifier. Second sample data with incorrect classification results is removed from the second source data set to obtain a third source data set. The similarity results between the third source data set and the first target data set are calculated using multiple preset binary classifiers. Based on the similarity results and the data types included in the cross-entropy loss functions used by the preset binary classifiers, the target data type for which similarity is to be calculated is determined, and the similarity between the data corresponding to the target data type in the third source data set and the first target data set is calculated. Third sample data and target patient data with a similarity lower than a second preset threshold are removed from the third source data set and the first target data set, so as to obtain a fourth source data set and a second target data set, which are used as the target training set and the target patient data set respectively. The multi-classifier is trained using the target training set, and the classifier is used to output the light stimulation parameters predicted for the target patient based on the second target data set.

[0102] After obtaining the light stimulation parameters, the light stimulation parameters can be used to control the transcranial light to perform light stimulation on the target patient in a virtual reality rehabilitation training task. In a possible embodiment, the light emission parameters of the light emission unit can be adjusted according to the light stimulation parameters, so that the light emission unit emits light that conforms to the above-mentioned target transcranial light stimulation parameters. The above light emission parameters may include the wavelength, frequency, pulse width, and emission time of the emitted light. As a possible implementation manner, the wavelength of the emitted light can be adjusted according to the stimulation wavelength included in the target transcranial light stimulation parameters, the frequency of the emitted light can be adjusted according to the stimulation frequency in the target transcranial light stimulation parameters, the pulse width of the emitted light can be adjusted according to the stimulation pulse width in the target transcranial light stimulation parameters, and the emission duration of the emitted light can be adjusted according to the stimulation duration in the target transcranial light stimulation parameters.

[0103] In a possible embodiment, during transcranial light stimulation, brain function feedback information, eye movement information, and facial video information of the target patient can be collected. The collection methods of the above-mentioned eye movement information and facial video information have been described in detail above and will not be elaborated here.

[0104] The target brain function feedback information of the target patient is used to characterize the changes in the brain function of the target patient before and after transcranial light stimulation. In a possible embodiment, the target brain function feedback information includes a target brain activation value index and a target brain function network connection index number. Among them, the target brain activation index is obtained through the following formula:

[0105]

[0106] Among them, GJ rht is the target brain activation value index; m, n, and l are the numbers of cerebral blood oxygen channels in the prefrontal lobe, occipital lobe, and parietal lobe of the brain, respectively; W qei、 W hei is the wavelet amplitude of the cerebral blood oxygen signal of the i-th channel in the prefrontal lobe of the brain before and after light stimulation; W qzi、 W hzi is the wavelet amplitude of the cerebral blood oxygen signal of the i-th channel in the occipital lobe of the brain before and after light stimulation; W qdi、 W hdi is the wavelet amplitude of the cerebral blood oxygen signal of the i-th channel in the parietal lobe of the brain before and after light stimulation; W qxi、 W hxi is the wavelet amplitude of the cerebral blood oxygen signal of the i-th channel in the prefrontal lobe, occipital lobe, and parietal lobe of the brain before and after light stimulation; λ is a preset fourth correction coefficient, which can be any constant set according to the actual application scenario.

[0107] The target brain function network connection index is obtained through the following formula:

[0108]

[0109] Among them, BM is the target brain function network connection index, Le, Lz, and Ld are the numbers of channels with functional connections in the prefrontal lobe, occipital lobe, and parietal lobe of the brain, respectively, φ Lehi , φ Leqi is the connection efficiency of the i-th channel with functional connection in the prefrontal lobe of the brain before and after light stimulation, φ Lzhi , φ Lzqi is the connection efficiency of the i-th channel with functional connection in the occipital lobe of the brain before and after light stimulation, φ Ldhi , φ Ldqi is the connection efficiency of the i-th channel with functional connection in the parietal lobe of the brain before and after light stimulation, α 1 , α 2 , α 3They are a preset fifth correction coefficient, a preset sixth correction coefficient, and a preset seventh correction coefficient respectively.

[0110] In a possible embodiment, after obtaining the target eye movement information, based on the target eye movement information and the corresponding relationship between the preset eye movement information and the visual spatial attention index, the target visual spatial attention of the target patient can be obtained. Specifically, the target visual spatial attention SZ can be calculated based on the target eye movement information through the following formula y :

[0111]

[0112] where HY and LY are the high-frequency power and low-frequency power of the eye movement signal respectively, P k (e jw ) is the power spectrum of the eye movement signal within different sampling periods, σ(e jw ) is the power spectrum of the impulse function, ZX t is the forward saccade delay time of the time-domain feature of the eye movement signal, FX t is the backward saccade delay time of the time-domain feature of the eye movement signal, α, β, and γ are weight coefficients, e refers to the natural constant, w is the central frequency of the eye movement signal, and j is the imaginary part of the complex number.

[0113] Based on the corresponding relationship between the preset visual spatial attention range and the visual spatial attention index, the target visual spatial attention index of the target patient is obtained based on the target visual spatial attention.

[0114] Exemplarily, the value range of SZ y can be set between 0 and 1, where (0 - 0.4) indicates a better visual spatial attention index, [0.4 - 0.7) indicates an average visual spatial attention index, and [0.7 - 1] indicates a relatively poor visual spatial attention index.

[0115] In a possible embodiment, the obtaining of the target emotion index of the target patient based on the target facial video information by using the pre-trained convolutional neural network includes:

[0116] Cropping the target facial video to obtain each target facial video segment; using the convolutional neural network to output multiple target emotion states of the target patient based on each target facial video segment, where the target emotion states include negative, positive, and neutral; determining the target emotion index of the target patient based on the proportion of the preset emotion state in the target emotion state and the corresponding relationship between the preset proportion range and the emotion index.

[0117] In this step, the target facial video can be cropped based on a preset segment length to obtain multiple target video segments. In one possible embodiment, different segment lengths can be set for cognitive tasks with different difficulty coefficients, and the preset segment length can be determined based on the difficulty coefficient corresponding to the cognitive task set for the target patient. Then, based on this segment length, the target facial video can be cropped to obtain multiple target video segments.

[0118] After that, the pre-trained convolutional neural network can be used to output the target emotional states of the target patient based on each target facial video segment. The convolutional neural network can be pre-trained based on a large number of facial video segments and the emotional states corresponding to the facial video segments. The above target emotional states can include negative, positive, and neutral.

[0119] Exemplarily, the acquisition process of the target emotional state can be characterized by the following formula:

[0120] QX i =G CNN (W CNN ,B CNN ;JJ CNN ,CH CNN ;SP i ,HK i )

[0121] Wherein, W CNN is the weight matrix of the convolutional neural network model, B CNN is the bias parameter; JJ CNN is the convolutional layer of the convolutional module, the convolutional kernel is 3*2, and the convolutional stride is 2*2; CH CNN is the pooling layer of the convolutional module, and the pooling layer is a 2*2 maximum pooling kernel; SP i the i-th cropped facial video; SP i is the difficulty coefficient of the virtual reality task corresponding to the i-th facial video; QX i the emotional state corresponding to the i-th facial video, including: negative, positive, neutral.

[0122] After that, the target request index can be calculated based on each target emotional state. Specifically, the proportion of the target emotional state in each emotional state can be calculated, and based on the corresponding relationship between the data range of this proportion and the preset emotional index, the target emotional index can be obtained. Exemplarily, the target emotional index of the target patient can be calculated by the following formula:

[0123]

[0124] Wherein, SL cj is the number of times of positive emotions, SL czThe number of neutral emotions is \(X\), the total number of emotional states is \(N\); \(LQ\) is the emotion index. The value range of the \(LQ\) is between 0 and 1, where \([0.7 - 1)\) indicates excellent emotion index, \([0.5 - 0.7)\) indicates good emotion index, \([0.3 - 0.5]\) indicates poor emotion index, and \((0 - 0.3)\) indicates extremely poor emotion index.

[0125] In a possible embodiment, the transcranial light stimulation parameters can be adjusted based on the target brain function feedback information, the target visuospatial attention index, the target emotion index, and a preset adjustment strategy collected during the transcranial light stimulation process, and the above adjustment strategy can be set according to the actual application scenario.

[0126] In a possible embodiment, the stimulation duration in the target transcranial light stimulation parameters can be adjusted based on the target visuospatial attention index, the stimulation frequency in the target transcranial light stimulation parameters can be adjusted based on the target emotion index, the stimulation wavelength in the target transcranial light stimulation parameters can be adjusted based on the target brain activation index, and the stimulation pulse width in the target transcranial light stimulation parameters can be adjusted based on the target brain network connection index.

[0127] Exemplarily, the adjustment strategy may include that when the target visuospatial attention index is good, the stimulation time is adjusted to be longer; when the target visuospatial attention index is average, the stimulation time remains unchanged; when the target visuospatial attention index is poor, the stimulation time is adjusted to be shorter. When the target emotion index is excellent, the stimulation frequency is adjusted to be higher; when the target emotion index is good, the stimulation frequency remains unchanged; when the target emotion index is poor or extremely poor, the stimulation frequency is adjusted to be lower. When the brain activation value index output by the brain function evaluation module exceeds the threshold \(M1\), the light stimulation wavelength is adjusted to be lower; when the target brain activation value index does not exceed the threshold \(M1\), the light stimulation wavelength is adjusted to be higher; when the target brain network connection index exceeds the threshold \(M2\), the light stimulation pulse width is adjusted to be smaller; when the target brain network connection index does not exceed the threshold \(M2\), the light stimulation pulse width is adjusted to be larger.

[0128] In a possible embodiment, after adjusting the target transcranial light stimulation parameters based on the feedback information of the target patient, the parameters of the light emitting unit can be adjusted based on the adjusted target transcranial light stimulation parameters, so that the light emitting unit can emit transcranial light in a timely and accurate manner according to the adjusted target transcranial light stimulation parameters.

[0129] In a possible embodiment, the regulation parameters used in each transcranial light stimulation of the target patient can be correspondingly stored for the target patient information and used as the initial value of the regulation parameters for the next transcranial light stimulation of the target patient.

[0130] In a possible embodiment, information such as the above-mentioned light stimulation parameters, target brain function feedback information, target visuospatial metrics, etc. can all be displayed through a display device to present to relevant personnel the transcranial light stimulation effect and the adjustment situation of the light stimulation parameters in real time.

[0131] Applying the embodiments of the present invention, the cognitive regulation state of a patient is evaluated through visuospatial attention metrics, emotion metrics, brain activation metrics, and brain network connection metrics, fully considering the collaborative control effects of multi-source information such as the patient's brain and physiological information before, during, and after the virtual training process, and being able to comprehensively evaluate the patient's brain function state.

[0132] Moreover, by making full use of the multi-level information changes of the patient's visuospatial attention metrics, emotion metrics, brain activation metrics, and brain network connection metrics, the transcranial light stimulation parameters of the patient are adjusted in real time, promoting the collaborative optimization and real-time feedback of the brain and light stimulation data. It can provide a personalized and adaptive light stimulation parameter adjustment scheme for the patient, enabling the light stimulation-assisted rehabilitation training to exert the maximum gain effect and improving the efficiency and effect of the patient's rehabilitation training.

[0133] Based on the same inventive concept, the embodiments of the present invention also provide a transcranial light stimulation regulation system, as Figure 3 shown. The system 300 may include:

[0134] A first acquisition module 301, configured to acquire historical sample data of multiple groups of patients with brain dysfunction in the historical record and the sample labels of the historical sample data to obtain a target training set; wherein, each group of the historical sample data includes near-infrared brain function data, eye movement data, and facial video data; the sample label is the actual transcranial light stimulation parameter corresponding to the historical sample data, and the actual transcranial light stimulation parameter at least includes a stimulation wavelength, a stimulation frequency, a stimulation pulse width, and a stimulation time;

[0135] A second acquisition module 302, configured to acquire multiple groups of target patient data of a target patient to be subjected to transcranial light stimulation to obtain a target patient data set, and each group of the target patient data includes the target patient's near-infrared brain function data, eye movement data, and facial video data;

[0136] A classifier training module 303, configured to train a target multi-classifier based on each piece of the historical sample data and the sample label in the target training set, and the target multi-classifier is used to output transcranial light stimulation parameters;

[0137] A parameter output module 304, configured to input the target patient data set into the target multi-classifier, so that the target multi-classifier outputs the target transcranial light stimulation parameter of the target patient based on the target patient data set;

[0138] An information acquisition module 305 is configured to perform transcranial optoneuroregulation on the target patient by using the target transcranial light stimulation parameters, and collect target brain function feedback information, target eye movement information, and target facial video information of the target patient during the transcranial light stimulation process; wherein, the target brain function feedback information is used to characterize the change in brain function data of the target patient before and after transcranial light stimulation.

[0139] An index calculation module 306 is configured to obtain the target visual spatial attention index of the target patient based on the target eye movement information and the corresponding relationship between the preset eye movement information and the visual spatial attention index; and obtain the target emotion index of the target patient by using a pre-trained convolutional neural network based on the target facial video information.

[0140] A parameter adjustment module 307 is configured to adjust the target transcranial light stimulation parameters based on the target brain function feedback information, the target visual spatial attention index, and the target emotion index of the target patient.

[0141] A light emission unit 308 is configured to adjust the light emission parameters of the light emission unit according to the target transcranial light stimulation parameters sent by the parameter adjustment module 307, including the wavelength, frequency, pulse width, and emission time of the light.

[0142] In a possible embodiment, as Figure 4 shown, the transcranial light stimulation regulation system may include a cranial light field model module, a virtual reality task module, an information acquisition and analysis module, a light stimulation module, a light emission control module, a brain function evaluation module, and a feedback module.

[0143] The cranial light field model module is configured to construct a target multi-classifier and output transcranial light stimulation parameter information based on the target patient dataset by using the target multi-classifier. The virtual reality module is configured to provide a virtual reality rehabilitation training task for the user. The light emission control module is configured to adjust the parameters of the light stimulation module according to the transcranial light stimulation parameters output by the cranial light field model module, and the light stimulation module is configured to perform light stimulation on the patient. The information acquisition and analysis module is configured to collect and analyze the eye movement and facial video information during the user's virtual reality rehabilitation training task during the transcranial light stimulation process. The brain function evaluation module is configured to evaluate the brain function of the patient before and after transcranial optoregulation. The feedback module is configured to feedback the eye movement, facial video information output by the information acquisition and analysis module and the brain function information output by the brain function evaluation module to the light emission control module for adjustment of the light stimulation parameters.

[0144] Wherein, the collection, storage, use, processing, transmission, provision, and disclosure of the user's personal information involved in the present invention all comply with the provisions of relevant laws and regulations and do not violate public order and good customs.

[0145] An exemplary embodiment of the present invention also provides a non-transitory computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor of a computer, is configured to cause the computer to execute the method according to an embodiment of the present invention.

[0146] An exemplary embodiment of the present invention also provides a computer program product, including a computer program, wherein the computer program, when executed by a processor of a computer, is configured to cause the computer to execute the method according to an embodiment of the present invention.

[0147] The program code for implementing the method of the present invention may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that the program codes, when executed by the processor or controller, cause the functions / operations specified in the flowchart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, as an independent software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0148] In the context of the present invention, a machine-readable medium may be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0149] As used in the present invention, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, apparatus, and / or device (e.g., a disk, an optical disc, a memory, a programmable logic device (PLD)) for providing machine instructions and / or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term "machine-readable signal" refers to any signal for providing machine instructions and / or data to a programmable processor.

Claims

1. A transcranial light stimulation control method, characterized in that: The method comprises: Acquire historical sample data of multiple groups of patients with brain dysfunction and sample labels of the historical sample data in historical records to obtain a target training set; Acquire multiple groups of target patient data of target patients to be subjected to transcranial light stimulation to obtain a target patient data set; A target multi-classifier is trained based on each piece of the historical sample data and the sample label in the target training set, wherein the target multi-classifier is used to output transcranial light stimulation parameters; Inputting the target patient data set into the target multi-classifier, so that the target multi-classifier outputs the target transcranial photostimulation parameters of the target patient based on the target patient data set; Performing transcranial light stimulation regulation on the target patient using the target transcranial light stimulation parameters, and collecting target brain function feedback information, target eye movement information, and target facial video information of the target patient during the transcranial light stimulation regulation process; Based on the target eye movement information and the correspondence between the preset eye movement information and the visual-spatial attention index, the target visual-spatial attention index of the target patient is obtained; and based on the target facial video information, the target emotion index of the target patient is obtained using a pre-trained convolutional neural network; Adjusting the target transcranial light stimulation parameters based on the target brain function feedback information, target visual-spatial attention index, and target emotion index of the target patient; The light emission parameters of the light emission unit are adjusted according to the target transcranial light stimulation control parameters, and the light emission parameters include one or more of the wavelength, frequency, pulse width and emission time of the light.

2. The method according to claim 1, characterized in that The target training set and the target patient data set are obtained by the following steps: Acquire a first source data set and a first target data set, wherein the first source data set includes multiple groups of the historical sample data and sample labels of each group of the historical sample data, and the first target data set includes multiple groups of target patient data of the target patient; Calculating first similarities between the plurality of groups of historical sample data and the first target data set respectively, and deleting historical sample data whose first similarity is higher than a first preset threshold from the first source data set, to obtain a second source data set; Using the first target classifier to classify the second source data set, obtain classification results of each group of second sample data included in the second source data set, and delete the second sample data with incorrect classification results from the second source data set to obtain a third source data set; Calculate the similarity distance between each third sample data in the third source data set and each target patient data in the first target data set, and delete the third samples and target patient data whose similarity distance is greater than a second preset threshold from the third source data set and the first target data set to obtain a target training set and a target patient data set.

3. The method according to claim 2, characterized in that The method further comprises: Testing a plurality of candidate classifiers based on each second sample data in the second source data set, and acquiring a plurality of candidate classifiers having a classification accuracy higher than a preset accuracy threshold based on a sample label corresponding to each second sample data in the second source data set as pending classifiers; A first target classifier is determined based on the accuracy of the prediction results output by each of the pending classifiers for each target patient data included in the first target data set.

4. The method according to claim 1, characterized in that: Adjusting the target transcranial light stimulation parameters based on the target brain function feedback information, the target visual-spatial attention index, and the target emotion index of the target patient includes: The stimulation duration in the target transcranial light stimulation parameters is adjusted based on the target visual-spatial attention index, the stimulation frequency in the target transcranial light stimulation parameters is adjusted based on the target emotion index, the stimulation wavelength in the target transcranial light stimulation parameters is adjusted based on the target brain activation index, and the stimulation pulse width in the target transcranial light stimulation parameters is adjusted based on the target brain network connection index.

5. A transcranial light stimulation control system, characterized in that: The system comprises: A first acquisition module is used to acquire historical sample data of multiple groups of patients with brain dysfunction in historical records and sample labels of the historical sample data to obtain a target training set; A second acquisition module is used to acquire multiple groups of target patient data of target patients to be subjected to transcranial light stimulation regulation to obtain a target patient data set; A classifier training module, used for training a target multi-classifier based on each of the historical sample data and sample labels in the target training set, wherein the target multi-classifier is used for outputting transcranial light stimulation parameters; A parameter output module, used for inputting the target patient data set into the target multi-classifier, so that the target multi-classifier outputs the target transcranial optical stimulation parameters of the target patient based on the target patient data set; An information acquisition module, used to perform transcranial light stimulation regulation on the target patient using the target transcranial light stimulation parameters, and to collect target brain function feedback information, target eye movement information, and target facial video information of the target patient during the transcranial light stimulation regulation process; An index calculation module, used to obtain a target visual-spatial attention index of the target patient based on the target eye movement information and a preset correspondence between the eye movement information and the visual-spatial attention index; and to obtain a target emotion index of the target patient based on the target facial video information using a pre-trained convolutional neural network; A parameter adjustment module, used to adjust the target transcranial light stimulation parameters based on the target brain function feedback information, target visual-spatial attention index, and target emotion index of the target patient; The light emission unit is used to adjust the light emission parameters according to the target transcranial light stimulation parameters, and the light emission parameters include one or more of the wavelength, frequency, pulse width and emission time of the light.

6. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to make a computer execute the method according to any one of claims 1-4.

Images