Behavior neural signal optimal weighted integration algorithm applied to language brain diseases

Abstract

The invention relates to the technical field of neural signal processing, in particular to a behavior neural signal optimal weighted integration algorithm applied to language brain diseases, which comprises the following steps of: 1, acquiring a basic activity mode of a brain of a user through electroencephalogram data, and analyzing characteristics of low-frequency oscillation and a functional connection network; 2, in task state electroencephalogram data collection, a user completes a language processing task; step 3, identifying two sets of electroencephalogram indexes closely related to language brain diseases through feature extraction and screening; 4, through an iterative optimization method, different weights are given to different indexes, and the reasonability of weight distribution is verified through a machine learning technology; and 5, generating a comprehensive language brain health score for clinical evaluation. According to the method, multi-dimensional data are integrated into index weight distribution of neural signals, and brain multi-dimensional electrical data and behavioral data are combined, so that a set of efficient language brain function evaluation system is constructed.

Description

Technical Field

[0001] The present invention relates to the technical field of neural signal processing, and particularly relates to a weighted integration algorithm for behavioral neural signals applied to language brain diseases. Background Art

[0002] Due to the irreversibility of language function defects, traditional medical means (such as language rehabilitation training and neuropharmacological intervention) have limited therapeutic effects on aphasia, dysarthria, etc. caused by severe nerve injuries (such as stroke, amyotrophic lateral sclerosis, traumatic brain injury), and patients have long faced communication barriers. In addition, language interaction is physiologically limited. In the late stage of amyotrophic lateral sclerosis (ALS), patients cannot achieve language expression through muscle movement due to the degeneration of motor neurons. Existing assistive devices rely on eye tracking or residual muscle signals, with low efficiency and limited applicability. Summary of the Invention

[0003] The present invention provides a weighted integration algorithm for behavioral neural signals applied to language brain diseases, which realizes the identification and evaluation of language brain diseases by allocating index weights for neural signals through multi-dimensional data integration, and constructs an efficient language brain function evaluation system in combination with multi-dimensional brain electrical data and behavioral data.

[0004] To achieve the above object, the present invention provides the following technical solution: A weighted integration algorithm for behavioral neural signals applied to language brain diseases, which includes:

[0005] Step 1, obtaining the basic activity pattern of the user's brain through electroencephalogram data, and analyzing the characteristics of low-frequency oscillation and functional connection network;

[0006] Step 2, in the acquisition of task-state electroencephalogram data, the user needs to complete a series of language processing tasks designed based on the theory of text cognition;

[0007] Step 3, identifying two sets of electroencephalogram indicators closely related to language brain diseases through feature extraction and screening;

[0008] Step 4, assigning different weights to different indicators through an iterative optimization method, and verifying the rationality of the weight assignment through machine learning technology;

[0009] Step 5, generating a comprehensive language brain health score, presented in a percentage system, to facilitate clinicians to intuitively evaluate the language brain function status of the user.

[0010] Preferably, in the second step, specific neural activities related to language processing are stimulated through the task, and electroencephalogram features related to text visual processing, semantic integration, and motor execution are captured.

[0011] Preferably, in the third step, behavioral data is also obtained through indicators such as task completion time, accuracy rate, and reaction speed, providing objective behavioral evidence for the user's language ability.

[0012] Preferably, in the fourth step, a scaling coefficient and an offset coefficient are introduced to optimize the sensitivity and specificity of the index formula and maximize the classification effect of distinguishing between language brain disease patients and healthy people.

[0013] The beneficial effects of the present invention are as follows: Through the multimodal signal fusion algorithm, the EEG of language-related brain regions and multi-source signals of the whole brain network are integrated to improve the decoding accuracy of complex semantic intentions. At the same time, through the personalized adaptive model and the dynamic transfer learning framework based on the lesion location of the patient, the influence of individual differences in neural signals on the generalization of the system is reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0015] Figure 1 It is a flowchart of the algorithm of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] The following will clearly and completely describe the technical solutions of the present invention in conjunction with the drawings of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0017] According to Figure 1 As shown, a behavior neural signal optimal weighted integration algorithm applied to language brain diseases includes:

[0018] Step 1: Obtain the basic activity pattern of the user's brain through EEG data and analyze the characteristics of low-frequency oscillations and functional connection networks;

[0019] Step 2: In the collection of task-state EEG data, the user needs to complete a series of language processing tasks designed based on the theory of text cognition; and specific neural activities related to language processing are also stimulated through the tasks, and EEG characteristics related to text visual processing, semantic integration, and motor execution are captured.

[0020] Step 3: Through feature extraction and screening, two sets of EEG indicators closely related to language brain diseases are identified; behavioral data is also obtained through indicators such as task completion time, accuracy rate, and reaction speed, providing objective behavioral evidence for the user's language ability.

[0021] Step 4: Through the iterative optimization method, different weights are assigned to different indicators, and the rationality of the weight assignment is verified through machine learning techniques; in addition, a scaling coefficient and an offset coefficient are introduced to optimize the sensitivity and specificity of the indicator formula and maximize the classification effect of distinguishing between language brain disease patients and healthy people.

[0022] Step 5: Generate a comprehensive language brain health score, presented in a percentage system, to facilitate clinicians' intuitive assessment of the user's language brain function status.

[0023] Through the above steps, multi-dimensional data is integrated into the weight assignment of neural signal indicators to achieve the identification and assessment of language brain diseases. The core of this algorithm lies in the language processing tasks designed based on the theory of text cognition. Combining multi-dimensional EEG data and behavioral data, a set of efficient language brain function assessment systems is constructed. First, through EEG data, the basic activity pattern of the user's brain is obtained, and the characteristics of low-frequency oscillation and functional connection network are mainly analyzed. These indicators can reflect the overall neural activity state of the brain. Secondly, in the collection of task-state EEG data, the user needs to complete a series of language processing tasks designed based on the theory of text cognition. These tasks can not only stimulate specific neural activities related to language processing but also capture EEG characteristics related to text visual processing, semantic integration, and motor execution. Behavioral data provides objective behavioral evidence for the user's language ability through indicators such as task completion time, accuracy rate, and reaction speed.

[0024] After the data collection is completed, the algorithm identifies two sets of EEG indicators closely related to language brain diseases through feature extraction and screening. Subsequently, the algorithm uses the iterative optimization method to assign different weights to different indicators and verifies the rationality of the weight assignment through machine learning techniques. This process aims to maximize the classification effect of distinguishing between language brain disease patients and healthy people. In addition, the algorithm also introduces a scaling coefficient and an offset coefficient to further optimize the sensitivity and specificity of the indicator formula. Finally, the algorithm generates a comprehensive language brain health score, presented in a percentage system, to facilitate clinicians' intuitive assessment of the user's language brain function status.

[0025] The innovation of this algorithm lies in its multi-dimensional data integration and dynamic weight optimization mechanism. By combining resting-state and task-state data, the algorithm can comprehensively evaluate the static and dynamic characteristics of language brain function; and the task design based on the theory of text cognition makes the algorithm more in line with the language processing characteristics of text users, improving cultural adaptability and diagnostic accuracy. In addition, the dynamic weight optimization mechanism of the algorithm can automatically adjust the evaluation model according to the characteristics of different individuals, thus realizing personalized evaluation. This algorithm is not only applicable to the early screening and diagnosis of language brain diseases, but also can be used in the fields of language cognitive aging, neurodegenerative diseases (such as Alzheimer's disease), and language function recovery after brain injury, etc., with broad application prospects.

[0026] Among them, through the language processing tasks designed by combining the theory of text cognition, the resting-state EEG data, task-state EEG data, and behavioral data are successfully integrated, and two sets of EEG indicators closely related to language brain diseases are identified. Through iterative optimization, the algorithm assigns different weights to different indicators, and combines the scaling coefficient and offset coefficient to construct an indicator formula that can maximize the distinction between language brain disease patients and healthy people. This formula can not only effectively identify language cognitive aging, but also generate language brain health scores in the form of a percentage system, providing a quantitative basis for clinical diagnosis and intervention. In addition, this algorithm has high generalization ability and can adapt to individuals of different ages and disease stages, significantly improving the accuracy and efficiency of early screening and evaluation of language brain diseases. Through this technology, doctors and researchers can more accurately identify the potential risks of language dysfunction and provide scientific support for the formulation of personalized treatment plans.

[0027] The specific advantages of this algorithm are as follows:

[0028] (1) Multi-dimensional data integration to improve diagnostic accuracy.

[0029] This algorithm innovatively combines resting-state EEG data, task-state EEG data, and behavioral data, making full use of the advantages of different data sources. Resting-state data reflects the basic activity patterns of the brain, task-state data captures the dynamic changes during language processing, and behavioral data provides external evidence of actual performance. Through the integration of multi-dimensional data, the algorithm can more comprehensively evaluate language brain function, significantly improving the accuracy and reliability of diagnosis. In addition, through weighted integration and optimization, the algorithm avoids the limitations of a single data source, further enhancing the scientific nature and practicality of the results. For example, resting-state data can reveal the overall connection state of the brain network, while task-state data can reflect the neural activity patterns under specific language tasks. The combination of the two can more accurately locate the neural mechanisms of language dysfunction. Behavioral data provides an objective manifestation of language ability, compensating for the deficiencies of EEG data at the functional output level. This multi-dimensional integration not only improves the diagnostic sensitivity but also provides a basis for the subtype classification of diseases, such as distinguishing whether language disorders are caused by neurodegenerative diseases or brain injuries. In addition, the algorithm can be combined with other neuroimaging techniques (such as fMRI or MEG) to further expand its application scope and provide a new perspective for the mechanism research of language brain diseases.

[0030] (2) Task design based on the theory of character cognition to improve diagnostic sensitivity.

[0031] This algorithm adopts a language processing task designed based on the theory of character cognition, making the experimental task more in line with the language processing characteristics of Chinese users. As a logographic language, Chinese has unique characteristics of combining form, sound, and meaning, and its cognitive processing process is significantly different from that of alphabetic languages. For example, the visual complexity of Chinese characters, the radical structure, and the existence of polyphonic and polysemous characters require Chinese users to have stronger visual analysis and semantic integration abilities in language processing. Through this task design, the algorithm can identify unique EEG indicators related to character cognition, such as alpha wave activity related to visual-spatial processing and theta wave synchrony related to semantic integration. In addition, character cognition tasks can also stimulate more neural activity patterns, providing richer feature information for the algorithm and thus enhancing the discrimination ability of the model. This task design is not only applicable to the evaluation of language brain diseases but also can be used in other fields related to character cognition, such as the diagnosis of Chinese learning disorders and the evaluation of character writing ability. At the same time, the task design based on the theory of character cognition also provides a new tool for cross-cultural language research, such as comparing the neural mechanism differences between Chinese users and alphabetic language users in language processing, so as to provide a scientific basis for cross-cultural diagnosis of language brain diseases.

[0032] (3) Dynamic weight optimization to achieve personalized evaluation.

[0033] This algorithm realizes the dynamic optimization evaluation of language brain function by iteratively attempting to assign differentiated weights to different indicators and combining the scaling coefficient and the offset coefficient. This dynamic weight assignment method can automatically adjust the evaluation model according to the characteristics of different individuals, thus better adapting to individual differences. For example, for the elderly population, the algorithm may pay more attention to indicators related to cognitive aging (such as the increase in slow-wave activity), while for patients with brain injuries, it may pay more attention to indicators related to neuroplasticity (such as power changes in specific frequency bands). In addition, the algorithm can further optimize the weight assignment according to factors such as an individual's medical history and lifestyle, thereby achieving truly personalized evaluation. This dynamic optimization not only improves the accuracy of the algorithm but also provides clinicians with more detailed diagnostic information, such as the disease development stage, potential risk factors, etc. At the same time, the percentile language brain health score generated by the algorithm is intuitive and easy to understand, facilitating understanding by clinicians and patients, and providing an important reference for the formulation of personalized treatment plans. For example, individuals with scores below a certain threshold may require further neurorehabilitation training, while those with higher scores can prevent further degradation of language function through cognitive training.

[0034] (4) Effectively distinguish between disease and healthy populations and assist in early screening.

[0035] By optimizing the indicator weights and formula parameters, this algorithm can maximize the distinction between language brain disease populations and healthy populations. This high-efficiency discrimination ability makes the algorithm have important application value in the early screening of language brain diseases. For example, the algorithm can identify early-stage patients who have not shown obvious symptoms in traditional clinical evaluations, thus providing opportunities for early intervention. By early identifying potential language function disorders, doctors can take timely intervention measures to delay the disease progression and improve the quality of life of patients. At the same time, the high efficiency of the algorithm also reduces the screening cost and provides technical support for large-scale population screening. For example, the algorithm can be integrated into portable electroencephalogram devices for routine screening in community hospitals or nursing homes, thus expanding the screening coverage. In addition, the algorithm can be combined with other health data (such as genetic data, lifestyle data) to construct a more comprehensive risk assessment model, providing a scientific basis for the prevention of language brain diseases. For example, the algorithm can identify high-risk populations with specific gene mutations or lifestyles and provide them with personalized health management suggestions.

[0036] (5) Wide applicability, supporting multi-scenario applications.

[0037] This algorithm is not only applicable to the diagnosis and evaluation of language brain diseases, but also can be widely used in fields such as language cognitive aging, neurodegenerative diseases (such as Alzheimer's disease), and language function recovery after brain injury. Its flexible design and strong generalization ability enable the algorithm to adapt to individuals of different ages and different disease stages, providing a general solution for language brain function evaluation in various scenarios. For example, in the early diagnosis of Alzheimer's disease, the algorithm can identify potential signs of cognitive decline by analyzing the semantic integration ability in language tasks. In brain injury rehabilitation, the algorithm can evaluate the language function recovery of patients, providing a basis for adjusting the rehabilitation training plan. In addition, the output results of the algorithm are easy to integrate with other medical data, facilitating multidisciplinary collaboration. For example, the algorithm can be combined with neuroimaging data, gene data, and clinical scale data to construct a more comprehensive disease model to support precision medicine. At the same time, the algorithm can also be applied to the education field, such as evaluating the language development level of children or for personalized evaluation of language learning ability. This wide applicability makes the algorithm not only of great value in clinical medicine, but also shows broad application prospects in fields such as psychology and education.

[0038] As described above, it is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A behavioral neural signal optimal weighted integration algorithm for language brain diseases, characterized in that: include: Step 1: Obtain the basic activity pattern of the user's brain through EEG data, and analyze the characteristics of low-frequency oscillations and functional connection networks; Step 2: In the task-based EEG data collection, the user needs to complete a series of language processing tasks designed based on text cognition theory; Step 3: Through feature extraction and screening, two sets of EEG indicators closely related to language brain diseases are identified; Step 4: Assign differentiated weights to different indicators through iterative optimization methods, and verify the rationality of weight allocation through machine learning technology; Step 5: Generate a comprehensive language brain health score in percentage form to facilitate clinicians to intuitively evaluate the user's language brain function status.

2. According to claim 1, a behavioral neural signal optimal weighted integration algorithm for language brain diseases is characterized by: In the second step, the task is used to stimulate specific neural activities related to language processing, and capture EEG features related to text visual processing, semantic integration, and motor execution.

3. The optimal weighted integration algorithm of behavioral neural signals for language brain diseases according to claim 2 is characterized in that: In step three, behavioral data is also obtained through indicators such as task completion time, accuracy, and reaction speed, providing objective behavioral performance evidence for the user's language ability.

4. The optimal weighted integration algorithm of behavioral neural signals for language brain diseases according to claim 1 is characterized in that: In the step 4, a scaling factor and an offset factor are introduced to optimize the sensitivity and specificity of the indicator formula and maximize the classification effect of distinguishing people with language brain diseases from healthy people.

Images