A synergistic control method based on consciousness disorder assessment and monitoring and neural synchronous stimulation

Abstract

This invention provides a collaborative control method based on consciousness impairment assessment and monitoring and synchronous neural stimulation. The method includes: an eye-tracking monitoring and analysis system extracting eye-tracking feature data from eye-tracking signals using image preprocessing and eye-tracking feature extraction algorithms; a server-side management system combining scale assessments and eye-tracking feature data, utilizing a multi-source data fusion assessment algorithm to output the patient's consciousness state assessment results and stimulation parameter information; a neural correlation electrical stimulator receiving the stimulation parameter information and using a timing synchronization control algorithm to synchronously initiate neural electrical stimulation with the eye-tracking monitoring and analysis system; and the server-side management system dynamically adjusting the consciousness state assessment results and stimulation parameter information based on post-stimulation eye-tracking response characteristics and stimulation parameter feedback, reusing the multi-source data fusion assessment algorithm to form a closed-loop control. This invention achieves timing synchronization and command closed-loop through a collaborative linkage algorithm, ensuring efficient collaborative operation of multiple modules and adapting to complex clinical scenarios such as ICUs and rehabilitation departments.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a collaborative control method based on consciousness disorder assessment and monitoring and neural synchronous stimulation. Background Technology

[0002] Disorders of consciousness are common and serious complications following central nervous system injury, including vegetative states and minimally conscious states. Their assessment, monitoring, and intervention have always been challenging issues in clinical medicine. Current clinical methods for assessing disorders of consciousness primarily rely on subjective scales, which suffer from drawbacks such as susceptibility to assessor experience and a lack of objective data support. Furthermore, existing neurostimulation devices are mostly single-function devices, failing to achieve synergistic integration of assessment, monitoring, and stimulation therapy. This makes it difficult to dynamically adjust stimulation parameters based on the patient's real-time state of consciousness, resulting in insufficient precision and effectiveness of stimulation therapy.

[0003] Furthermore, current technologies lack the ability to integrate and manage multi-dimensional data from patients with disorders of consciousness, failing to achieve systematic archiving and analysis of medical records, assessment data, stimulation parameters, and monitoring results. This hinders clinicians from fully understanding changes in patients' conditions and developing personalized treatment plans. Simultaneously, eye movement signals, as an important objective indicator reflecting the state of consciousness, are not fully utilized in existing devices to leverage the correlation and synergy between eye movement monitoring data and neural stimulation, making it difficult to achieve closed-loop stimulation modulation based on the patient's objective physiological signals. Therefore, developing an integrated device capable of accurately assessing and monitoring disorders of consciousness, integrating and managing multi-dimensional data, and providing personalized synchronous neural stimulation has become an urgent technical challenge. Summary of the Invention

[0004] Therefore, the purpose of this invention is to provide a collaborative control method based on consciousness disorder assessment and monitoring and neural synchronous stimulation, so as to at least solve the shortcomings of the above-mentioned technology.

[0005] This invention proposes a collaborative control method based on consciousness disorder assessment and monitoring and neural synchronous stimulation, comprising: Step 1: The server-side management system sends start commands to the assessment system, eye-tracking monitoring and analysis system, and neural correlation electrical stimulator. Step 2: The eye movement monitoring and analysis system collects the patient's eye movement signals in real time, extracts the corresponding eye movement feature data from the eye movement signals through image preprocessing and eye movement feature extraction algorithms, and transmits the eye movement feature data to the server management system in real time; Step 3: The server-side management system combines the scale assessment with the eye movement feature data, and uses a multi-source data fusion assessment algorithm to output the patient's state of consciousness assessment results and stimulation parameter information; Step 4: The neural correlation electrical stimulator receives stimulation parameter information from the server management system, uses a timing synchronization control algorithm to synchronously start neural electrical stimulation with the eye movement monitoring and analysis system, and applies electrical stimulation to the patient's central and / or peripheral nerve regions through the application terminal electrode; Step 5: The server-side management system, based on the eye movement response characteristics and stimulation parameter feedback after stimulation, reuses the multi-source data fusion evaluation algorithm to dynamically adjust the consciousness state evaluation results and stimulation parameter information, and feeds back the adjusted parameters to the neural correlation electrical stimulator in real time to form a closed-loop control. Step Six: The server-side management system stores, analyzes, and visualizes the data throughout the entire process, generates trend analysis reports, and supports multi-terminal collaborative control and data sharing.

[0006] Furthermore, step two includes: The Canny edge detection and ellipse fitting algorithm is used to extract the pupil center and corneal reflection point from the eye movement signal, and the gaze vector is calculated. The basic eye movement features and electrically induced eye movement features are extracted based on the Kalman filter algorithm, and the effective features are screened by the random forest algorithm to obtain the corresponding eye movement feature data. The basic eye movement features include fixation duration, peak saccadic velocity, and pupil diameter change.

[0007] Furthermore, step three includes: The eye movement feature data, electrical stimulation response data, and clinical scale scores are input into an improved multimodal attention fusion model, and the comprehensive evaluation score is calculated through adaptive weighting. A prototype network optimized by transfer learning is used to extract deep features, and the decision results of support vector machine and random forest are fused together. Ensemble learning is achieved through meta-classifier to output the patient's state of consciousness, which includes vegetative state, minimal state of consciousness, out of minimal state of consciousness, and conscious state. Based on the changes in eye movement response characteristics after electrical stimulation, combined with the difference between the comprehensive assessment score and the clinical scale score, it is determined whether latent consciousness exists, so as to generate the patient's consciousness status assessment result.

[0008] Furthermore, step three also includes: Based on a pre-set clinical case database, initial stimulation parameters are recommended through a similar case matching algorithm; The patient's real-time eye movement response characteristics are output through the eye movement feature data, and the current and frequency of the initial stimulation parameters are optimized by the gradient descent algorithm based on the eye movement response characteristics and the consciousness state assessment results to obtain the corresponding stimulation parameter information. The system monitors the stimulation parameters and preset safety thresholds in real time. If the parameters exceed the allowable range, it automatically adjusts to the standard parameters and triggers a prompt.

[0009] Furthermore, step four includes: A timestamp synchronization mechanism is adopted to ensure that the time error between eye-tracking monitoring, evaluation and electrical stimulation initiation is controlled within a preset error threshold; It supports the coordinated initiation of electrical stimulation and visual stimulation to form a corresponding closed-loop induction mechanism.

[0010] This invention also proposes a collaborative control system based on consciousness disorder assessment and monitoring and neural synchronous stimulation, comprising: The data acquisition module is used to control the server management system to send start commands to the evaluation system, eye-tracking monitoring and analysis system, and neural correlation electrical stimulator. The feature extraction module is used to control the eye movement monitoring and analysis system to collect the patient's eye movement signals in real time, extract the corresponding eye movement feature data from the eye movement signals through image preprocessing and eye movement feature extraction algorithms, and transmit the eye movement feature data to the server management system in real time. The data evaluation module is used to control the server management system to combine the scale evaluation with the eye movement feature data, and use a multi-source data fusion evaluation algorithm to output the patient's consciousness state evaluation results and stimulation parameter information. The synchronization control module is used to control the neural correlation electrical stimulator to receive stimulation parameter information sent by the server management system, use a time-series synchronization control algorithm to start neural electrical stimulation synchronously with the eye movement monitoring and analysis system, and apply electrical stimulation to the central and / or peripheral nerve areas of the patient through the application terminal electrode; The closed-loop control module is used to control the server management system to dynamically adjust the consciousness state assessment results and stimulation parameter information by reusing the multi-source data fusion evaluation algorithm based on the eye movement response characteristics and stimulation parameter feedback after stimulation, and to feed back the adjusted parameters to the neural correlation electrical stimulator in real time to form a closed-loop regulation. The data storage module is used to control the server-side management system to store, analyze, and visualize the data throughout the entire process, generate trend analysis reports, and support multi-terminal collaborative control and data sharing.

[0011] Furthermore, the feature extraction module is specifically used for: The Canny edge detection and ellipse fitting algorithm is used to extract the pupil center and corneal reflection point from the eye movement signal, and the gaze vector is calculated. The basic eye movement features and electrically induced eye movement features are extracted based on the Kalman filter algorithm, and the effective features are screened by the random forest algorithm to obtain the corresponding eye movement feature data. The basic eye movement features include fixation duration, peak saccadic velocity, and pupil diameter change.

[0012] Furthermore, the data evaluation module is specifically used for: The eye movement feature data, electrical stimulation response data, and clinical scale scores are input into an improved multimodal attention fusion model, and the comprehensive evaluation score is calculated through adaptive weighting. A prototype network optimized by transfer learning is used to extract deep features, and the decision results of support vector machine and random forest are fused together. Ensemble learning is achieved through meta-classifier to output the patient's state of consciousness, which includes vegetative state, minimal state of consciousness, out of minimal state of consciousness, and conscious state. Based on the changes in eye movement response characteristics after electrical stimulation, combined with the difference between the comprehensive assessment score and the clinical scale score, it is determined whether latent consciousness exists, so as to generate the patient's consciousness status assessment result.

[0013] Furthermore, the data evaluation module is also specifically used for: Based on a pre-set clinical case database, initial stimulation parameters are recommended through a similar case matching algorithm; The patient's real-time eye movement response characteristics are output through the eye movement feature data, and the current and frequency of the initial stimulation parameters are optimized by the gradient descent algorithm based on the eye movement response characteristics and the consciousness state assessment results to obtain the corresponding stimulation parameter information. The system monitors the stimulation parameters and preset safety thresholds in real time. If the parameters exceed the allowable range, it automatically adjusts to the standard parameters and triggers a prompt.

[0014] Furthermore, the synchronization control module is specifically used for: A timestamp synchronization mechanism is adopted to ensure that the time error between eye-tracking monitoring, evaluation and electrical stimulation initiation is controlled within a preset error threshold; It supports the coordinated initiation of electrical stimulation and visual stimulation to form a corresponding closed-loop induction mechanism.

[0015] The present invention also proposes a storage medium on which a computer program is stored, which, when executed by a processor, implements the above-described collaborative control method based on consciousness disorder assessment and monitoring and neural synchronous stimulation.

[0016] The present invention also proposes a computer, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described collaborative control method based on consciousness impairment assessment and monitoring and neural synchronous stimulation.

[0017] The collaborative control method based on consciousness impairment assessment and monitoring and synchronous neural stimulation in this invention combines an assessment system with an eye movement monitoring and analysis system to achieve a dual assessment mechanism of subjective scale assessment and objective eye movement signal monitoring. Based on real-time monitoring data, neural stimulation parameters can be dynamically adjusted to form a closed-loop control, significantly improving the accuracy and therapeutic effect of synchronous neural stimulation. It introduces synchronous stimulation of central and peripheral nerves, simultaneously stimulating the median nerve and the dorsolateral area of ​​the left prefrontal cortex to achieve bidirectional activation and synergistic regulation of neural pathways. This more effectively promotes the reshaping of the consciousness network and increases the probability of awakening and cognitive recovery compared to traditional single stimulation methods. High-precision image acquisition and intelligent algorithms are used to achieve continuous monitoring of patients' spontaneous eye movements without the need for wearable devices, avoiding secondary injury. Multiple anti-interference algorithms are integrated into eye movement monitoring and signal processing to effectively separate and record signals such as electrical stimulation, blinking, and head movements, ensuring the accuracy and stability of data acquisition. Through collaborative linkage algorithms, temporal synchronization and command closed-loop are achieved, ensuring efficient collaborative operation of multiple modules and adapting to complex clinical scenarios such as ICUs and rehabilitation departments. Attached Figure Description

[0018] Figure 1 This is a flowchart of the collaborative control method based on consciousness disorder assessment and monitoring and neural synchronous stimulation in the first embodiment of the present invention; Figure 2 This is a block diagram showing the distribution of each system in the collaborative control method based on consciousness disorder assessment and monitoring and neural synchronous stimulation in the first embodiment of the present invention. Figure 3 This is a structural block diagram of the collaborative control system based on consciousness disorder assessment and monitoring and neural synchronous stimulation in the second embodiment of the present invention; Figure 4 This is a structural block diagram of the computer in the third embodiment of the present invention.

[0019] The following detailed description, in conjunction with the accompanying drawings, will further illustrate the present invention. Detailed Implementation

[0020] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Several embodiments of the invention are illustrated in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0022] Example 1 Please see Figure 1 The figure shows a collaborative control method based on consciousness disorder assessment and monitoring and neural synchronous stimulation in the first embodiment of the present invention. The method specifically includes steps S101 to S106: S101, The server management system sends a start command to the assessment system, the eye-tracking monitoring and analysis system and the neural correlation electrical stimulator; Please see Figure 2 In this embodiment, the collaborative control method includes four core modules: a server-side management system, an assessment system, an eye-tracking monitoring and analysis system, and a neural correlation electrical stimulator. These modules work together to achieve the assessment and monitoring of consciousness disorders and synchronized neural stimulation. Data interaction between the modules is achieved through Wi-Fi wireless communication. The specific structure is as follows: 1. Server-side Management System: As the core data management and control hub of the entire electrical stimulation device, it is used to integrate and manage data from multiple modules, issue commands, and monitor the entire process. The server-side management system uses an industrial-grade server with a Windows operating system, and the data center uses a MySQL database to store various types of patient data, including the following sub-modules: Data Center: Used to store all types of data generated by all modules, including patient medical record data, assessment data, eye movement monitoring data, stimulation parameter data, monitoring data, etc., and provides data classification, retrieval, backup and export functions to provide data support for clinical analysis and subsequent research; The medical record management module is used to input, edit, and query patients' basic information (such as patient name, age, cause of illness, medical history, etc.), medical history information, diagnosis information, and other medical record data, so as to realize the standardized management and long-term archiving of patient medical records. Specifically, medical records can be retrieved and edited through the patient's unique identifier. Prescription Management Module: This module is used by doctors to create personalized stimulation therapy prescriptions based on patient assessment results and conditions. It includes the input, review, and distribution of stimulation parameters (e.g., stimulation frequency, pulse width, intensity), stimulation duration, stimulation cycle, and other information to the stimulation management module. The module can also adjust the prescription content in real time according to changes in the patient's condition. Stimulation Management Module: This module receives stimulation prescriptions from the prescription management module, sends stimulation control commands to the neural correlative electrical stimulator, and simultaneously receives feedback from the neural correlative electrical stimulator on the stimulation execution status (such as whether stimulation has started normally, current stimulation parameters, etc.) to ensure that stimulation therapy is implemented accurately according to the prescription. The monitoring and management module is used to collect and monitor various monitoring data (such as eye movement signals, stimulation signals, and impedance signals) uploaded by the assessment system, eye movement monitoring and analysis system, and nerve-related electrical stimulator in real time. When the data exceeds the preset threshold, an alarm is triggered to ensure the safety of the treatment process. Specifically, when the patient's eye movement amplitude exceeds 5mm, an audible and visual alarm is triggered, and the alarm information is pushed to the mobile terminal of medical staff. Report Management Module: This module is used to automatically generate patient assessment reports and treatment reports based on stored assessment data, monitoring data, and stimulus data. It supports printing and exporting reports, providing a basis for clinical diagnosis and treatment effect evaluation. Specifically, this module can automatically generate weekly / monthly assessment and treatment reports based on patient assessment data, monitoring data, and stimulus data, including data trend charts and text analysis. The system management module divides users into three permission levels: administrator, doctor, and nurse. Administrators can configure system parameters and manage user accounts, doctors can perform operations such as assessment and prescription formulation, and nurses can only perform data entry and equipment operation. The Help Center module includes built-in video tutorials and frequently asked questions documents, and supports online searching, making it easy for users to quickly familiarize themselves with system operations and resolve problems encountered during use.

[0023] S102, the eye movement monitoring and analysis system collects the patient's eye movement signals in real time, extracts corresponding eye movement feature data from the eye movement signals through image preprocessing and eye movement feature extraction algorithms, and transmits the eye movement feature data to the server management system in real time; Furthermore, step S102 specifically includes steps S1021 to S1022: S1021, The Canny edge detection and ellipse fitting algorithm is used to extract the pupil center and corneal reflection point in the eye movement signal, and the gaze vector is calculated; S1022, Based on the Kalman filter algorithm, basic eye movement features and electrically induced eye movement features are extracted, and effective features are screened through the random forest algorithm to obtain the corresponding eye movement feature data. The basic eye movement features include fixation duration, peak saccadic velocity, and pupil diameter change.

[0024] In practical implementation, the eye movement signal preprocessing algorithm in the eye movement monitoring and analysis system uses the Canny edge detection algorithm with a threshold set to 0.1-0.3, the ellipse fitting algorithm uses the least squares method with a fitting error ≤0.05mm, and the Kalman filter's state equation is... In the formula, Indicates the first The state vector, Indicates the first Process noise at time step; observation equation is , Indicates the first The observation vector at time t, Indicates the first The observation noise at any given moment is automatically adjusted by the A and H coefficient matrices based on eye movement states (fixation / saccade). The eye movement feature extraction algorithm of this eye movement monitoring and analysis system extracts 28 basic eye movement features (8-dimensional fixation features, 8-dimensional saccade features, 4-dimensional smooth tracking features, and 8-dimensional pupillary response features) and 4-dimensional electrically induced eye movement features. It then uses a random forest algorithm to filter features, remove redundant features, and retain effective features. The real-time monitoring and anomaly warning algorithm sets the sliding window size, and the warning signal uses sound and light prompts, which are synchronized to the evaluation end and the server end.

[0025] In this embodiment, eye-tracking monitoring, through image preprocessing and eye-tracking feature extraction, can analyze voluntary or involuntary eye movements, effectively improving the analysis of the degree of consciousness impairment.

[0026] S103, the server-side management system combines the scale assessment with the eye movement feature data, and uses a multi-source data fusion assessment algorithm to output the patient's state of consciousness assessment results and stimulation parameter information; Furthermore, step S103 specifically includes steps S1031 to S1033: S1031, The eye movement feature data, electrical stimulation response data and clinical scale scores are input into the improved multimodal attention fusion model, and the comprehensive evaluation score is calculated through adaptive weighting. S1032, a prototype network optimized by transfer learning is used to extract deep features, and the decision results of support vector machine and random forest are fused together. Ensemble learning is achieved through meta-classifier to output the patient's state of consciousness. The patient's state of consciousness includes vegetative state, minimal state of consciousness, out of minimal state of consciousness, and awake state. S1033, based on the change in eye movement response characteristics after electrical stimulation, combined with the difference between the comprehensive assessment score and the clinical scale score, determine whether latent consciousness exists, so as to generate the patient's consciousness status assessment result.

[0027] Furthermore, step S103 also includes steps S1131 to S1133: S1131, based on a pre-set clinical case database, recommends initial stimulus parameters through a similar case matching algorithm; S1132, output the patient's real-time eye movement response characteristics through the eye movement feature data, and optimize the current and frequency of the initial stimulation parameters using a gradient descent algorithm based on the eye movement response characteristics and the consciousness state assessment results to obtain the corresponding stimulation parameter information; S1133, monitor the stimulation parameter information and preset safety threshold in real time. If the value exceeds the allowable range, automatically adjust to the standard parameter and trigger a prompt.

[0028] S104, the neural correlation electrical stimulator receives stimulation parameter information sent by the server management system, uses a timing synchronization control algorithm to synchronously start neural electrical stimulation with the eye movement monitoring and analysis system, and applies electrical stimulation to the central and / or peripheral nerve areas of the patient through the application terminal electrode; Furthermore, step S104 specifically includes steps S1041 to S1042: S1041 employs a timestamp synchronization mechanism to ensure that the time error between eye-tracking monitoring, assessment and electrical stimulation activation is controlled within a preset error threshold. S1042 supports the coordinated initiation of electrical and visual stimulation to form a corresponding closed-loop induction mechanism.

[0029] S105, the server management system, based on the eye movement response characteristics and stimulation parameter feedback after stimulation, reuses the multi-source data fusion evaluation algorithm to dynamically adjust the consciousness state evaluation results and stimulation parameter information, and feeds the adjusted parameters back to the neural correlation electrical stimulator in real time to form a closed-loop control. S106, the server-side management system stores, analyzes, and visualizes the data throughout the entire process, generates trend analysis reports, and supports multi-terminal collaborative control and data sharing.

[0030] In this embodiment, the server-side collaborative management algorithm is implemented as follows: 1. Data Synchronization Algorithm: Each terminal relies on a high-precision clock synchronization (such as NTP) to generate UTC timestamps, with a sampling period of 10ms. After receiving data, the server eliminates network transmission delays based on the timestamp difference to achieve data time alignment; the CRC-32 algorithm is used to verify the transmitted data blocks, and if the verification fails, automatic retransmission is triggered to ensure data integrity during transmission; incremental backup is adopted, backing up only the data added or modified since the last backup each day to save storage space and backup time; at the same time, a full backup is performed once every morning to form a complete data recovery point and ensure data security. 2. Instruction Distribution and Cooperative Control Algorithm: Scheduling is based on an instruction priority queue, with priorities from high to low as follows: stimulus parameter adjustment instructions, evaluation start instructions, data query instructions, and data backup instructions. The system ensures that high-priority instructions are processed first, and the instruction response time (from issuance to terminal confirmation) is ≤50ms. Reliable communication between the terminal and the server is achieved based on a custom TCP / IP protocol stack. The design ensures a data transmission rate of no less than 10Mbps to meet the requirements of real-time instruction issuance and status feedback. At the same time, for instructions with high real-time requirements, optimization on top of TCP or the use of UDP + application layer confirmation mechanism can be considered to reduce latency. 3. Data Management and Analysis Algorithms: The MySQL database adopts a table partitioning strategy to store patient basic information, assessment data, eye movement data, and stimulation parameter data separately, and indexes are built to optimize query speed, with a single data query time of ≤100ms; the trend analysis algorithm uses a multinomial fitting algorithm to fit the continuous treatment monitoring data and generate a trend curve of changes in consciousness state; the multi-patient comparison algorithm uses a clustering analysis algorithm to classify and statistically analyze the assessment results of different patients and output a comparison report.

[0031] The assessment system uses a tablet computer as the hardware platform, running the Android operating system. The built-in scale assessment software algorithm implements the following steps: 1. Patient information matching and scale recommendation: After medical staff enter the patient's unique identifier, the algorithm automatically retrieves the patient's basic medical condition data from the server-side medical record management module. Based on the scale item adaptation algorithm, it recommends suitable assessment scales (e.g., the CRS-R scale for patients with traumatic brain injury, and the GCS scale for stroke patients), while simultaneously filtering core assessment items (removing items irrelevant to the patient's condition); 2. Assessment process guidance and data entry: The algorithm guides assessors to perform assessment operations for each item using a combination of text and graphics. It supports touch input of raw scores and automatically verifies the validity of the entered data (e.g., whether the score exceeds the preset range of the item), providing prompts and guidance for correction of invalid entries; 3. Score calibration and deviation correction: The algorithm calls the score calibration model to calibrate the entered data... 4. **Comparison of Original Scores with Clinical Scores of Patients with Similar Conditions and Age Groups:** The algorithm calculates the score deviation and corrects the original score (correction formula: calibrated score = original score - deviation value × calibration coefficient, where the calibration coefficient is set based on clinical data validation). 5. **Multi-Scale Result Fusion (if multiple scales are used for assessment):** The algorithm weights and fuses the calibrated scores of different scales. For example, if the CRS-R scale weight is set to 0.6 and the GCS scale weight is set to 0.4, the fused score = CRS-R calibration score × 0.6 + GCS calibration score × 0.4, outputting a standardized assessment score. 6. **Assessment Result Generation and Upload:** The algorithm automatically integrates information such as assessment scale type, original score, calibration score, fused score, assessment time, and assessor to generate a standardized assessment report. The assessment results and report are uploaded to the server-side management system's data center via Wi-Fi, while simultaneously receiving assessment task instructions from the server. Furthermore, the intelligent evaluation algorithm at the evaluation end is implemented as follows: 1. An improved multimodal fusion algorithm with an attention gating module is adopted (in this embodiment, the algorithm adopts the MMA algorithm) to dynamically allocate fusion weights for electrical stimulation-induced eye movement features, basic eye movement features, CRS-R scores, and electrical stimulation response data, and output a comprehensive evaluation score. 2. Consciousness state classification algorithm: Transfer learning is used to train a CNN base model on healthy control data, which is then transferred to patient data (VS patients, MCS patients, eMCS patients) for training. A prototype network is combined to solve the small sample problem. The weights of the SVM and RF integrated algorithm are used. The test dataset is used to verify and show that the model's evaluation accuracy is set. The implicit consciousness judgment threshold is set according to the change of eye movement response characteristics after electrical stimulation and the difference between the comprehensive evaluation score and the CRS-R score. 3. Visualization and Feedback Algorithm for Evaluation Results: The evaluation results are visualized using the Matplotlib library, generating eye-tracking heatmaps, feature importance bar charts, and comprehensive evaluation score curves; the evaluation report is generated using a PDF template, supporting manual editing and PDF export.

[0032] The eye-tracking monitoring and analysis system includes a high-definition camera, an image acquisition card, and a processor. The processor runs on a software algorithm environment for assessing and monitoring consciousness disorders. The specific algorithm implementation steps are as follows: 1. Image acquisition and preprocessing: The high-definition camera acquires images of the patient's eyes. The image acquisition card converts the analog image signal into a digital signal. The algorithm sequentially performs grayscale processing (converting RGB images to grayscale), histogram equalization (enhancing image contrast and improving the distinction between the pupil and iris), Gaussian filtering, and edge enhancement (using the Sobel operator to enhance the edge of the eyeball). 2. Pupil localization and tracking: Based on the improved Hough circle transform algorithm, the pupil is located. A grayscale threshold is set to initially screen candidate pupil regions. Then, the Hough circle transform is used to detect the center coordinates and radius, eliminating interference areas such as eyelids and eyelashes. The KLT optical flow method is used to track the pupil movement trajectory. The number of feature point pyramid layers, iterations, and optical flow thresholds are set to achieve real-time stable tracking of eye movements. 3. Eye movement feature extraction: Based on the tracked pupil trajectory data, eye movement amplitude (distance of pupil center displacement), eye movement frequency (number of eye movements per unit time), and saccade speed are calculated. (Average displacement velocity during saccades), gaze duration (time the pupil center remains in a certain area), and other characteristic parameters are all standardized to dimensionless data in the [0,1] interval; 4. Consciousness state assisted judgment: The standardized eye movement characteristic parameters are input into a preset judgment model (constructed using a support vector machine algorithm), and the model outputs the consciousness state judgment result (no response: all characteristic parameters are below the threshold; weak response: 1-2 characteristic parameters are above the threshold; clear response: 3 or more characteristic parameters are above the threshold), and the judgment confidence is calculated at the same time. When the confidence is ≥0.8, the result is directly output. When the confidence is <0.8, it is marked as pending review and triggers manual intervention by medical staff; 5. Data output and upload: The algorithm integrates the original eye movement data, extracted characteristic parameters, judgment results and confidence into a standardized data packet and transmits it to the server management system via Wi-Fi; The call control module is preset to trigger a call prompt when the patient is detected to have saccades that are more than a set number of times and have an amplitude greater than a set value (clear response judgment result), and at the same time transmits the eye movement signal to the server management system as the core feedback basis for adjusting the stimulation parameters; In this embodiment, the power management module of the neural correlation electrical stimulator uses a 10000mAh rechargeable lithium battery, supports 15V / 2A charging, and has overcharge, over-discharge, and short-circuit protection functions. The remaining battery power is displayed on the device's screen in real time. The central processing unit uses an embedded microcontroller with a built-in neural synchronous stimulation algorithm, which can adjust the stimulation frequency (e.g., when the eye movement frequency is 5Hz, the stimulation frequency is adjusted to 10Hz) and stimulation intensity according to the eye movement frequency uploaded by the eye movement monitoring and analysis system. The output control unit uses a signal generator and a power amplifier, which can generate both DC and pulse wave stimulation waveforms, and outputs... The device has two channels and supports independent parameter settings for different channels. The signal acquisition and processing unit divides and amplifies the voltage at the application end of the stimulation output. The acquired analog voltage signal is converted into a 16-bit digital signal by an AD conversion chip and then transmitted to the central processing unit. At the same time, the acquired voltage signal is converted into current and impedance values ​​for processing. When the impedance is too high, the output is cut off to ensure the safety of the output. The stimulation output at the application end is mainly through two different specifications of physiotherapy electrode pads: saline electrode and silicone electrode. NAES mode current is output at both ends of each specification of physiotherapy electrode pad.

[0033] Specifically, the implementation of neural stimulation synchronization control and parameter modulation algorithms: 1. Timing synchronization control algorithm: The time stamp synchronization mechanism is adopted. The time stamp error of the eye movement monitoring and analysis system, the evaluation terminal, and the neural correlation electrical stimulator is ≤10ms, and the time difference between the start of electrical stimulation and the start of eye movement monitoring is ≤5ms. 2. Stimulation parameter adaptive adjustment algorithm: The parameter adjustment rule base has 1000+ clinical case data built in. It uses the nearest neighbor algorithm to match similar cases and recommend initial stimulation parameters. The gradient descent algorithm adjusts the stimulation parameters every 30 seconds based on the assessment score and eye movement response characteristics. For example, if the initial parameters for a VS patient are 20mA and 40Hz, and the eye movement response is not obvious after stimulation, it will be automatically adjusted to 21mA and 40Hz. 3. Stimulation safety monitoring algorithm: Set the allowable range for current fluctuation and frequency fluctuation. When the range is exceeded, automatically adjust to the nearest standard parameter and trigger a prompt; set an overload protection threshold. When the current reaches the threshold, automatically cut off the stimulation circuit, report to the server and display the fault code.

[0034] To demonstrate the synergistic effect between algorithms, the collaborative workflow in this embodiment is as follows. 1. Device startup: The doctor sends a startup command through the server, which simultaneously activates the assessment terminal, eye movement monitoring and analysis system, and neural correlation electrical stimulation device; the assessment terminal records the patient's basic information and initial CRS-R score, and synchronizes them to the server; 2. Real-time monitoring and assessment: The eye movement monitoring and analysis system collects the patient's eye movement signals in real time, and outputs 35 effective features through preprocessing and feature extraction algorithms, which are transmitted to the assessment end every 100ms; the assessment end runs a multi-source data fusion assessment algorithm and a consciousness state classification algorithm, and combines eye movement features and CRS-R scores to output a comprehensive assessment score and consciousness state, which are synchronized to the server and the neural association electrical stimulation device; 3. Synchronous Neural Stimulation: The neural-associated electrical stimulator receives stimulation parameter suggestions from the assessment end, runs a sequence synchronization control algorithm, and initiates median nerve electrical stimulation synchronously with eye movement monitoring, selecting the right median nerve as the stimulation target; stimulation parameters (current, frequency) are collected in real time and fed back to the server and assessment end; 4. Feedback and parameter adjustment: Based on the eye movement response characteristics and stimulation parameter feedback after stimulation, the assessment end re-runs the assessment algorithm and adjusts the comprehensive assessment score and consciousness state determination results; if the eye movement response is not obvious, stimulation parameter adjustment suggestions are automatically generated and sent to the neural correlation electrical stimulator to adjust the stimulation parameters. 5. Results Output and Data Management: After the stimulation ends, the assessment end generates a standardized assessment report, which supports PDF export; the server runs data management and analysis algorithms to store data throughout the process (eye-tracking data, assessment data, stimulation parameters) and generate a trend curve of changes in the patient's state of consciousness; doctors can query data and compare results from multiple patients through the server to optimize treatment plans.

[0035] In summary, the collaborative control method based on consciousness impairment assessment and monitoring and synchronous neural stimulation in the above embodiments of the present invention combines the assessment terminal system with the eye movement monitoring and analysis system to achieve a dual assessment mechanism of subjective scale assessment and objective eye movement signal monitoring. Based on real-time monitoring data, neural stimulation parameters can be dynamically adjusted to form a closed-loop control, significantly improving the accuracy and therapeutic effect of synchronous neural stimulation. The method introduces synchronous stimulation of central and peripheral nerves, simultaneously stimulating the median nerve and the dorsolateral area of ​​the left prefrontal cortex to achieve bidirectional activation and synergistic regulation of neural pathways. This is more effective than traditional single stimulation methods in promoting the reshaping of the consciousness network and increasing the probability of awakening and cognitive recovery. High-precision image acquisition and intelligent algorithms are used to achieve continuous monitoring of patients' spontaneous eye movements without the need for wearable devices, avoiding secondary injury. Multiple anti-interference algorithms are integrated into eye movement monitoring and signal processing to effectively separate and record signals such as electrical stimulation, blinking, and head movements, ensuring the accuracy and stability of data acquisition. Synergistic linkage algorithms achieve temporal synchronization and command closed-loop, ensuring efficient collaborative operation of multiple modules and adapting to complex clinical scenarios such as ICUs and rehabilitation departments.

[0036] Example 2 In another aspect, this invention proposes a collaborative control system based on the assessment and monitoring of consciousness disorders and synchronous neural stimulation. Please refer to [link / reference needed]. Figure 3The figure shows a collaborative control system based on consciousness disorder assessment and monitoring and neural synchronous stimulation according to a second embodiment of the present invention. The system includes: The data acquisition module 101 is used to control the server management system to send start commands to the evaluation system, the eye movement monitoring and analysis system and the neural correlation electrical stimulator; The feature extraction module 102 is used to control the eye movement monitoring and analysis system to collect the patient's eye movement signals in real time, extract the corresponding eye movement feature data from the eye movement signals through image preprocessing and eye movement feature extraction algorithms, and transmit the eye movement feature data to the server management system in real time. The data evaluation module 103 is used to control the server management system to combine the scale evaluation and the eye movement feature data, and use a multi-source data fusion evaluation algorithm to output the patient's consciousness state evaluation results and stimulation parameter information. The synchronization control module 104 is used to control the neural correlation electrical stimulator to receive stimulation parameter information sent by the server management system, use a time-series synchronization control algorithm to synchronously start neural electrical stimulation with the eye movement monitoring and analysis system, and apply electrical stimulation to the central and / or peripheral nerve areas of the patient through the application terminal electrode; The closed-loop control module 105 is used to control the server management system to dynamically adjust the consciousness state assessment results and stimulation parameter information by reusing the multi-source data fusion evaluation algorithm based on the eye movement response characteristics and stimulation parameter feedback after stimulation, and to feed back the adjusted parameters to the neural correlation electrical stimulator in real time to form a closed-loop regulation. The data storage module 106 is used to control the server-side management system to store, analyze and visualize the data throughout the entire process, generate trend analysis reports, and support multi-terminal collaborative control and data sharing.

[0037] Furthermore, the feature extraction module 102 is specifically used for: The Canny edge detection and ellipse fitting algorithm is used to extract the pupil center and corneal reflection point from the eye movement signal, and the gaze vector is calculated. The basic eye movement features and electrically induced eye movement features are extracted based on the Kalman filter algorithm, and the effective features are screened by the random forest algorithm to obtain the corresponding eye movement feature data. The basic eye movement features include fixation duration, peak saccadic velocity, and pupil diameter change.

[0038] Furthermore, the data evaluation module 103 is specifically used for: The eye movement feature data, electrical stimulation response data, and clinical scale scores are input into an improved multimodal attention fusion model, and the comprehensive evaluation score is calculated through adaptive weighting. A prototype network optimized by transfer learning is used to extract deep features, and the decision results of support vector machine and random forest are fused together. Ensemble learning is achieved through meta-classifier to output the patient's state of consciousness, which includes vegetative state, minimal state of consciousness, out of minimal state of consciousness, and conscious state. Based on the changes in eye movement response characteristics after electrical stimulation, combined with the difference between the comprehensive assessment score and the clinical scale score, it is determined whether latent consciousness exists, so as to generate the patient's consciousness status assessment result.

[0039] Furthermore, the data evaluation module 103 is also specifically used for: Based on a pre-set clinical case database, initial stimulation parameters are recommended through a similar case matching algorithm; The patient's real-time eye movement response characteristics are output through the eye movement feature data, and the current and frequency of the initial stimulation parameters are optimized by the gradient descent algorithm based on the eye movement response characteristics and the consciousness state assessment results to obtain the corresponding stimulation parameter information. The system monitors the stimulation parameters and preset safety thresholds in real time. If the parameters exceed the allowable range, it automatically adjusts to the standard parameters and triggers a prompt.

[0040] Furthermore, the synchronization control module 104 is specifically used for: A timestamp synchronization mechanism is adopted to ensure that the time error between eye-tracking monitoring, evaluation and electrical stimulation initiation is controlled within a preset error threshold; It supports the coordinated initiation of electrical stimulation and visual stimulation to form a corresponding closed-loop induction mechanism.

[0041] The functions or operation steps implemented by the above modules and units are largely the same as those in the above method embodiments, and will not be repeated here.

[0042] The collaborative control system based on consciousness disorder assessment and monitoring and neural synchronous stimulation provided in this embodiment of the invention has the same implementation principle and technical effects as the aforementioned method embodiment. For the sake of brevity, any parts not mentioned in the system embodiment can be referred to the corresponding content in the aforementioned method embodiment.

[0043] Example 3 This invention also proposes a computer, please refer to [link / reference]. Figure 4 The computer shown in the third embodiment of the present invention includes a memory 10, a processor 20, and a computer program 30 stored in the memory 10 and executable on the processor 20. When the processor 20 executes the computer program 30, it implements the above-described collaborative control method based on consciousness disorder assessment and monitoring and neural synchronous stimulation.

[0044] The memory 10 includes at least one type of storage medium, such as flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 10 can be an internal storage unit of a computer, such as the computer's hard disk. In other embodiments, the memory 10 can be an external storage device, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Furthermore, the memory 10 can include both internal and external storage units of the computer. The memory 10 can be used not only to store application software and various types of data installed on the computer, but also to temporarily store data that has been output or will be output.

[0045] In some embodiments, the processor 20 may be an electronic control unit (ECU), a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip, used to run program code stored in the memory 10 or process data, such as executing access restriction programs.

[0046] It should be pointed out that, Figure 4 The structure shown does not constitute a limitation on the computer. In other embodiments, the computer may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0047] This invention also proposes a storage medium storing a computer program that, when executed by a processor, implements the aforementioned collaborative control method based on consciousness impairment assessment and monitoring and neural synchronous stimulation.

[0048] Those skilled in the art will understand that the logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can mean any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0049] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0050] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0051] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0052] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A collaborative control method based on consciousness disorder assessment and monitoring and neural synchronous stimulation, characterized in that, include: Step 1: The server-side management system sends start commands to the assessment system, eye-tracking monitoring and analysis system, and neural sympathetic electrical stimulator. Step 2: The eye movement monitoring and analysis system collects the patient's eye movement signals in real time, extracts the corresponding eye movement feature data from the eye movement signals through image preprocessing and eye movement feature extraction algorithms, and transmits the eye movement feature data to the server management system in real time; Step 3: The server-side management system combines the scale assessment with the eye movement feature data, and uses a multi-source data fusion assessment algorithm to output the patient's state of consciousness assessment results and stimulation parameter information; Step 4: The neural correlation electrical stimulator receives stimulation parameter information from the server management system, uses a timing synchronization control algorithm to synchronously start neural electrical stimulation with the eye movement monitoring and analysis system, and applies electrical stimulation to the patient's central and / or peripheral nerve regions through the application terminal electrode; Step 5: The server-side management system, based on the eye movement response characteristics and stimulation parameter feedback after stimulation, reuses the multi-source data fusion evaluation algorithm to dynamically adjust the consciousness state evaluation results and stimulation parameter information, and feeds back the adjusted parameters to the neural correlation electrical stimulator in real time to form a closed-loop control. Step Six: The server-side management system stores, analyzes, and visualizes the data throughout the entire process, generates trend analysis reports, and supports multi-terminal collaborative control and data sharing.

2. The collaborative control method based on consciousness disorder assessment and monitoring and neural synchronous stimulation according to claim 1, characterized in that, Step two includes: The Canny edge detection and ellipse fitting algorithm is used to extract the pupil center and corneal reflection point from the eye movement signal, and the gaze vector is calculated. The basic eye movement features and electrically induced eye movement features are extracted based on the Kalman filter algorithm, and the effective features are screened by the random forest algorithm to obtain the corresponding eye movement feature data. The basic eye movement features include fixation duration, peak saccadic velocity, and pupil diameter change.

3. The synergistic control method based on consciousness disorder assessment and monitoring and neural synchronous stimulation according to claim 1, characterized in that, Step three includes: The eye movement feature data, electrical stimulation response data, and clinical scale scores are input into an improved multimodal attention fusion model, and the comprehensive evaluation score is calculated through adaptive weighting. A prototype network optimized by transfer learning is used to extract deep features, and the decision results of support vector machine and random forest are fused together. Ensemble learning is achieved through meta-classifier to output the patient's state of consciousness, which includes vegetative state, minimal state of consciousness, out of minimal state of consciousness, and conscious state. Based on the changes in eye movement response characteristics after electrical stimulation, combined with the difference between the comprehensive assessment score and the clinical scale score, it is determined whether latent consciousness exists, so as to generate the patient's consciousness status assessment result.

4. The collaborative control method based on consciousness disorder assessment and monitoring and neural synchronous stimulation according to claim 3, characterized in that, Step three also includes: Based on a pre-set clinical case database, initial stimulation parameters are recommended through a similar case matching algorithm; The patient's real-time eye movement response characteristics are output through the eye movement feature data, and the current and frequency of the initial stimulation parameters are optimized by the gradient descent algorithm based on the eye movement response characteristics and the consciousness state assessment results to obtain the corresponding stimulation parameter information. The system monitors the stimulation parameters and preset safety thresholds in real time. If the parameters exceed the allowable range, it automatically adjusts to the standard parameters and triggers a prompt.

5. The collaborative control method based on consciousness disorder assessment and monitoring and neural synchronous stimulation according to claim 1, characterized in that, Step four includes: A timestamp synchronization mechanism is adopted to ensure that the time error between eye-tracking monitoring, evaluation and electrical stimulation initiation is controlled within a preset error threshold; It supports the coordinated initiation of electrical stimulation and visual stimulation to form a corresponding closed-loop induction mechanism.

6. A collaborative control system based on consciousness disorder assessment and monitoring and neural synchronous stimulation, characterized in that, include: The data acquisition module is used to control the server management system to send start commands to the evaluation system, eye-tracking monitoring and analysis system, and neural correlation electrical stimulator. The feature extraction module is used to control the eye movement monitoring and analysis system to collect the patient's eye movement signals in real time, extract the corresponding eye movement feature data from the eye movement signals through image preprocessing and eye movement feature extraction algorithms, and transmit the eye movement feature data to the server management system in real time. The data evaluation module is used to control the server management system to combine the scale evaluation with the eye movement feature data, and use a multi-source data fusion evaluation algorithm to output the patient's consciousness state evaluation results and stimulation parameter information. The synchronization control module is used to control the neural correlation electrical stimulator to receive stimulation parameter information sent by the server management system, use a time-series synchronization control algorithm to start neural electrical stimulation synchronously with the eye movement monitoring and analysis system, and apply electrical stimulation to the central and / or peripheral nerve areas of the patient through the application terminal electrode; The closed-loop control module is used to control the server management system to dynamically adjust the consciousness state assessment results and stimulation parameter information by reusing the multi-source data fusion evaluation algorithm based on the eye movement response characteristics and stimulation parameter feedback after stimulation, and to feed back the adjusted parameters to the neural correlation electrical stimulator in real time to form a closed-loop regulation. The data storage module is used to control the server-side management system to store, analyze, and visualize the data throughout the entire process, generate trend analysis reports, and support multi-terminal collaborative control and data sharing.

7. The collaborative control system based on consciousness disorder assessment and monitoring and neural synchronous stimulation according to claim 6, characterized in that, The feature extraction module is specifically used for: The Canny edge detection and ellipse fitting algorithm is used to extract the pupil center and corneal reflection point from the eye movement signal, and the gaze vector is calculated. The basic eye movement features and electrically induced eye movement features are extracted based on the Kalman filter algorithm, and the effective features are screened by the random forest algorithm to obtain the corresponding eye movement feature data. The basic eye movement features include fixation duration, peak saccadic velocity, and pupil diameter change.

8. The collaborative control system based on consciousness disorder assessment and monitoring and neural synchronous stimulation according to claim 6, characterized in that, The data evaluation module is specifically used for: The eye movement feature data, electrical stimulation response data, and clinical scale scores are input into an improved multimodal attention fusion model, and the comprehensive evaluation score is calculated through adaptive weighting. A prototype network optimized by transfer learning is used to extract deep features, and the decision results of support vector machine and random forest are fused together. Ensemble learning is achieved through meta-classifier to output the patient's state of consciousness, which includes vegetative state, minimal state of consciousness, out of minimal state of consciousness, and conscious state. Based on the changes in eye movement response characteristics after electrical stimulation, combined with the difference between the comprehensive assessment score and the clinical scale score, it is determined whether latent consciousness exists, so as to generate the patient's consciousness status assessment result.

9. A readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the collaborative control method based on consciousness disorder assessment and monitoring and neural synchronous stimulation as described in any one of claims 1 to 5.

10. A computer comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the collaborative control method based on consciousness disorder assessment and monitoring and neural synchronous stimulation as described in any one of claims 1 to 5.

Images