State data monitoring method and system for electronic medical device

Abstract

According to the electronic medical device-oriented state data monitoring method and system provided by the embodiment of the invention, the physiological signal and the electrical stimulation response data are monitored and integrated in real time, so that the neural activity of the patient is comprehensively and dynamically monitored, the accuracy and timeliness of the monitored data are improved, and the monitoring accuracy is improved. And a detailed integrated state data sequence is generated through intelligent integration, so that rich data support is provided for subsequent medical decisions. Through an index mechanism of the regulation configuration file, a reference regulation scheme sequence is quickly matched and generated, and the response time of medical intervention is effectively shortened. Finally, the target regulation and control scheme is determined based on the flexible regulation and control selection strategy, personalized and precise medical regulation and control are ensured, the performance of the electronic medical device is remarkably improved, and therefore safer and more effective nerve regulation and control services are provided.

Description

Technical Field

[0001] This invention relates to the field of deep learning technology, and more specifically, to a method and system for monitoring the status data of electronic medical devices. Background Technology

[0002] In the field of modern medicine, the widespread application of electronic medical devices has greatly promoted the advancement of medical technology, especially in continuous health monitoring and disease prevention. These devices, such as implantable sensors and wearable devices, can monitor patients' physiological parameters in real time, providing doctors with important diagnostic information. However, with the further development of medical technology, higher demands are being placed on the data monitoring capabilities of electronic medical devices.

[0003] Traditional approaches primarily focus on collecting single state parameters. While these data are crucial for assessing a patient's basic health status, they fall short in some complex medical scenarios. For example, in neuroscience research, single state parameter data struggles to provide a comprehensive understanding of the dynamic changes in the nervous system. Furthermore, with the increasing demand for personalized healthcare, traditional state data monitoring models are no longer sufficient to meet clinical needs. Physiological differences among patients, varying stages of disease progression, and changes in treatment response all necessitate electronic medical devices that can provide more personalized and dynamic data monitoring services. Summary of the Invention

[0004] In order to at least overcome the above-mentioned deficiencies in the prior art, the purpose of this application is to provide a method and system for monitoring the status data of electronic medical devices.

[0005] According to one aspect of this application, a method for monitoring status data of electronic medical devices is provided, the method comprising:

[0006] At the target implantable neural monitoring trigger node, physiological signal data sequences and electrical stimulation response data sequences corresponding to the target electronic medical device are acquired. The physiological signal data sequences are obtained by collecting bioelectrical signals at the target implantable neural monitoring trigger node. The electrical stimulation response data sequences are obtained by collecting neural stimulation feedback data at the target implantable neural monitoring trigger node. The bioelectrical signals represent neural activity signals monitored by the implanted device in the patient's body. The neural stimulation feedback represents the response data generated after the implanted device stimulates the patient's nerves.

[0007] The physiological signal data sequence and the electrical stimulation response data sequence are integrated to generate an integrated state data sequence under the target implantable neural monitoring trigger node;

[0008] In the control configuration file, the reference control scheme corresponding to each integrated state data in the integrated state data sequence is obtained, and a reference control scheme sequence is generated; the control configuration file includes the index relationship between the template integrated state data sequence and the template control scheme sequence; there is an index relationship between a template integrated state data in the template integrated state data sequence and a template control scheme in the template control scheme sequence.

[0009] Based on the regulation selection strategy, the target regulation scheme under the target implanted neural monitoring trigger node is determined in the reference regulation scheme sequence.

[0010] According to one aspect of the embodiments of this application, a deep learning system is provided, the deep learning system including a processor and a machine-readable storage medium, the machine-readable storage medium storing machine-executable instructions, the machine-executable instructions being loaded and executed by the processor to implement the state data monitoring method for electronic medical devices in any of the foregoing possible embodiments.

[0011] According to one aspect of the embodiments of this application, a computer program product or computer program is provided, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods provided in the various optional implementations of the above three aspects.

[0012] In some embodiments of this application, the technical solutions achieve comprehensive and dynamic monitoring of the patient's neural activity through real-time monitoring and integration of physiological signals and electrical stimulation response data. This not only improves the accuracy and timeliness of the monitoring data but also generates a detailed integrated state data sequence through intelligent integration, providing rich data support for subsequent medical decisions. By using an indexing mechanism for the control configuration file, a reference control scheme sequence is quickly matched and generated, effectively shortening the response time of medical intervention. Finally, a target control scheme is determined based on a flexible control selection strategy, ensuring personalized and precise medical control, significantly improving the performance of electronic medical devices, and thus providing safer and more effective neuromodulation services. Attached Figure Description

[0013] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings required in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be extracted in conjunction with these drawings without creative effort.

[0014] Figure 1 A flowchart illustrating the status data monitoring method for electronic medical devices provided in this application embodiment;

[0015] Figure 2 This is a schematic block diagram of the architecture of a deep learning system for implementing the above-described method for monitoring state data of electronic medical devices, provided as an embodiment of this application. Detailed Implementation

[0016] The following description is provided to enable those skilled in the art to implement and combine this application, and is given in a specific application scenario and its requirements. It will be apparent to those skilled in the art that various modifications can be made to the disclosed embodiments, and the general principles defined herein can be applied to other embodiments and application scenarios without departing from the principles and scope of this application. Therefore, this application is not limited to the described embodiments, but should be given the broadest scope consistent with the claims.

[0017] Figure 1 This is a flowchart illustrating a status data monitoring method for electronic medical devices according to an embodiment of this application. The following is a detailed description of this status data monitoring method for electronic medical devices.

[0018] Step S110: Under the target implantable neural monitoring trigger node, acquire the physiological signal data sequence and the electrical stimulation response data sequence corresponding to the target electronic medical device. The physiological signal data sequence is obtained by collecting data on the bioelectrical signals under the target implantable neural monitoring trigger node. The electrical stimulation response data sequence is obtained by collecting data on the neural stimulation feedback data under the target implantable neural monitoring trigger node. The bioelectrical signals represent the neural activity signals monitored by the implanted device in the patient's body. The neural stimulation feedback represents the response data generated after the implanted device stimulates the patient's nerves.

[0019] In detail, the target implantable neural monitoring trigger node refers to a monitoring time point or monitoring cycle that is automatically triggered in an implantable neural monitoring system when specific conditions are met. This target implantable neural monitoring trigger node is the basis for the system's data acquisition and analysis. For example, suppose a patient has an implantable device for monitoring neural activity, and this device is set to automatically start monitoring and record data at a specific time each day (such as 10 pm). This "10 pm" time point is a target implantable neural monitoring trigger node.

[0020] The targeted electronic medical device refers to an electronic device implanted in a patient's body for monitoring and / or modulating neural activity. It typically includes tiny electrodes that connect to the patient's nervous system to acquire data and provide neural stimulation. For example, a typical targeted electronic medical device is a deep brain stimulator (DBS), which uses microelectrodes implanted in a specific area of ​​the patient's brain to monitor neural activity in that area and provide electrical stimulation as needed to modulate neural activity.

[0021] The physiological signal data sequence refers to a continuous data set obtained by collecting bioelectrical signals from a targeted implantable neural monitoring trigger node. This data reflects the real-time state of neural activity within the patient's body. For example, at the targeted implantable neural monitoring trigger node, the device collects neural activity signals from the motor control areas of the patient's brain. These neural activity signals include information such as the firing frequency and amplitude of neurons. Arranging this information in chronological order forms the physiological signal data sequence. The bioelectrical signals are electrical signals generated by cells or tissues within a living organism during life activities. In the nervous system, the excitation and inhibition states of neurons generate electrical signals, which can be monitored externally using implantable devices. For example, electroencephalography (EEG) is a typical bioelectrical signal that reflects the electrical activity of the cerebral cortex. These EEG signals can be monitored in real-time externally using implanted electrodes to analyze the functional state of the brain.

[0022] The electrical stimulation response data sequence refers to a continuous set of data obtained by collecting neural stimulation feedback data at the target implantable neural monitoring trigger node. This data reflects the nervous system's response to electrical stimulation. For example, at the target implantable neural monitoring trigger node, the device electrically stimulates the patient's nervous system and monitors the resulting responses, such as muscle contraction and neurotransmitter release. Arranging these response data in chronological order forms the electrical stimulation response data sequence. The neural stimulation feedback refers to the response data generated by the nervous system after the implantable device electrically stimulates the patient's nervous system. This response data can be used to evaluate the effectiveness of electrical stimulation and adjust stimulation parameters accordingly. For example, when the implantable device electrically stimulates a specific area in the patient's brain, it stimulates neurons in that area to generate action potentials, thereby triggering a series of physiological responses, such as muscle contraction and neurotransmitter release. This response data constitutes the neural stimulation feedback.

[0023] In other words, in this embodiment, it is assumed that a patient with Parkinson's disease needs to improve their symptoms through an implantable neuromonitoring and modulation system. At the target implantable neuromonitoring trigger node, the server begins to acquire the physiological signal data sequence and electrical stimulation response data sequence corresponding to the target electronic medical device.

[0024] A target electronic medical device is an implantable device specifically designed to monitor and stimulate neural activity. This implantable device connects to the patient's nervous system via tiny electrodes, enabling real-time monitoring of bioelectrical signals generated by neural activity and providing electrical stimulation to the nerves when needed.

[0025] At this specific target implantable neural monitoring trigger node, the server first receives bioelectrical signals from the implanted device. These bioelectrical signals reflect neural activity in specific regions of the patient's brain. For example, when the device detects abnormal high-frequency discharges in areas of the patient's brain that control movement, this may indicate that Parkinson's disease symptoms are occurring. The server generates physiological signal data sequences by acquiring and processing these bioelectrical signals.

[0026] Simultaneously, the server also acquired the electrical stimulation response data sequence. When the device electrically stimulates the patient's nerves, it triggers a series of responses. For example, electrical stimulation may lead to changes in the degree of muscle contraction, improvements in the patient's motor abilities, or other physiological responses. The server generates the electrical stimulation response data sequence by collecting and analyzing this neural stimulation feedback data.

[0027] Step S120: Integrate the physiological signal data sequence with the electrical stimulation response data sequence to generate an integrated state data sequence under the target implantable neural monitoring trigger node.

[0028] In this embodiment, after acquiring the physiological signal data sequence and the electrical stimulation response data sequence, the server begins to integrate them.

[0029] Suppose that the physiological signal data sequence shows the frequency and amplitude changes of a patient's neural activity over a specific time period, while the electrical stimulation response data sequence shows the patient's muscle contraction response and motor improvement in response to electrical stimulation over the same time period. The server matches and integrates the data from these two sequences in chronological order.

[0030] For example, at a specific moment, the server combines the physiological signal data of that moment (e.g., neural activity frequency of 100Hz and amplitude of 5mV) with the electrical stimulation response data of that moment (e.g., muscle contraction force increased by 20%) to form integrated state data. The server repeats this process, processing the data throughout the entire time period, and finally generates the integrated state data sequence under the target implanted neural monitoring trigger node.

[0031] Step S130: In the control configuration file, obtain the reference control scheme corresponding to each integrated state data in the integrated state data sequence, and generate a reference control scheme sequence. The control configuration file includes an index relationship between the template integrated state data sequence and the template control scheme sequence. An index relationship exists between a template integrated state data in the template integrated state data sequence and a template control scheme in the template control scheme sequence.

[0032] In this embodiment, the server has a control configuration file, which contains the index relationship between various template integration status data sequences and corresponding template control scheme sequences.

[0033] After acquiring the integrated status data sequence, the server begins searching the control configuration file for the reference control scheme corresponding to each integrated status data. For example, for a certain integrated status data, the server finds the matching template integrated status data in the file, retrieves the corresponding template control scheme based on the index, and uses it as the reference control scheme for that integrated status data.

[0034] Assume the integrated state data sequence contains a comprehensive picture of a patient's neural activity and electrical stimulation response at different times. For example, state data 1 indicates relatively active neural activity but a weak electrical stimulation response, while state data 2 indicates relatively stable neural activity and a good electrical stimulation response. The server locates template integrated state data that matches these integrated state data in the control configuration file and obtains the corresponding reference control scheme. For example, the reference control scheme for state data 1 might be to increase the intensity of electrical stimulation, while the reference control scheme for state data 2 might be to keep the current electrical stimulation parameters unchanged. The server repeats this process to obtain the corresponding reference control scheme for each integrated state data in the integrated state data sequence, ultimately generating a reference control scheme sequence.

[0035] Step S140: Based on the regulation selection strategy, determine the target regulation scheme under the target implanted neural monitoring trigger node in the reference regulation scheme sequence.

[0036] In this embodiment, the server determines the target modulation scheme under the target implanted neural monitoring trigger node from the reference modulation scheme sequence based on the modulation selection strategy.

[0037] Assume the modulation selection strategy is based on a scheme selection strategy. The server first obtains the prior modulation scheme of the forward-looking implanted neural monitoring trigger node. Assume the prior modulation scheme used in the previous round of monitoring trigger nodes was to moderately increase the frequency of electrical stimulation.

[0038] The server outputs redundant control schemes that are identical to the prior control scheme in the reference control scheme sequence and removes them from the reference control scheme sequence, generating a clean control scheme sequence. Then, the server obtains the scheme knowledge depth for each reference control scheme in the clean control scheme sequence. Scheme knowledge depth can be determined by evaluating factors such as scheme complexity, the degree of impact on the nervous system, and long-term effects.

[0039] For example, a reference modulation scheme might involve fine-tuning electrical stimulation parameters and the synergistic effects of multiple neurotransmitters, exhibiting a high level of scheme knowledge depth; while another scheme might simply increase the intensity of electrical stimulation, with a relatively low level of scheme knowledge depth. The server retrieves the scheme with the highest scheme knowledge depth from the scheme knowledge depth sequence and outputs its corresponding reference modulation scheme as the target modulation scheme under the target implanted neural monitoring trigger node.

[0040] Alternatively, we can assume that the modulation selection strategy is based on the selection strategy of integrated state data. The server first obtains the neuromodulation effect characteristics of the target electronic medical device at the target implantable neural monitoring trigger node based on the modulation selection strategy. For example, the neuromodulation effect characteristics may include the degree of improvement in patient symptoms (such as the magnitude of tremor reduction, the degree of improvement in motor flexibility), and the stability of the nervous system (such as the regularity and stability of neural activity).

[0041] The server loads neuromodulation effect features and integrated state data sequences into a deep learning network. This deep learning network is trained using a large number of example neuromodulation effect features and example integrated state data sequences. The deep learning network can generate a neuromodulation effect index corresponding to each integrated state data point based on the input data.

[0042] Assuming the integrated state data sequence contains multiple distinct state data points, the server uses a deep learning network to calculate the neural modulation effect index corresponding to each state data point. Then, the server identifies the maximum neural modulation effect index from the sequence and outputs its corresponding integrated state data as the target integrated state data related to the neural modulation effect characteristics. The reference modulation scheme corresponding to this target integrated state data is then output as the target modulation scheme.

[0043] Through the above steps, under the target implantable neural monitoring trigger node, the most suitable target regulation plan for the current patient state can be determined by comprehensively considering physiological signal data, electrical stimulation response data, regulation configuration files, and regulation selection strategies, so as to achieve effective monitoring and regulation of the patient's nervous system and improve the patient's symptoms and quality of life.

[0044] Let's refine this scenario further. In step S110, the patient is performing a daily activity, such as walking. The targeted implanted neural monitoring trigger node is activated, and the device monitors bioelectrical signals in areas of the patient's brain related to motor control. These signals show abnormal fluctuations in the frequency of neural activity, while electrical stimulation response data indicates insufficient coordination of muscle contractions.

[0045] In step S120, when the server integrates these data, it discovers that abnormal bioelectrical signals and uncoordinated electrical stimulation responses occur simultaneously at a specific moment, forming a series of complex integrated state data.

[0046] In step S130, the template integrated state data sequence in the modulation configuration file covers a variety of possible combinations of neural activity and electrical stimulation responses. The server accurately finds a corresponding reference modulation scheme for each integrated state data, forming a rich sequence of reference modulation schemes.

[0047] In step S140, if the selection strategy based on the scheme is adopted, the server finds that the previous round of prior regulation scheme is not effective in this monitoring. After analyzing the purification regulation scheme sequence, a regulation scheme based on deeper neuroscience principles is selected.

[0048] Alternatively, if a selection strategy based on integrated state data is adopted, the deep learning network can accurately calculate the neuromodulation effect index for each integrated state data point based on the input neuromodulation effect characteristics and the integrated state data sequence. The final determined target modulation scheme can more accurately adapt to the patient's current neurological state, and is expected to further optimize the treatment effect.

[0049] Furthermore, assuming that the patient's condition improves after a period of treatment, under the new targeted implantable neural monitoring trigger node, the bioelectrical signals and electrical stimulation response data acquired in step S110 show a relatively stable trend. During the integration process in step S120, the server detects relatively small changes in the integrated state data. In step S130, the reference control scheme in the control configuration file is also adjusted accordingly, focusing more on maintaining the stability of the existing treatment effect. In step S140, the target control scheme determined according to the control selection strategy may involve fine-tuning the electrical stimulation parameters to maintain the good state of the patient's nervous system.

[0050] Alternatively, an unexpected situation may occur during treatment, such as a patient experiencing drastic fluctuations in neural activity due to external stimulation. In this case, the bioelectrical signals and electrical stimulation response data acquired in step S110 will show significant anomalies. The integration in step S120 will highlight these abnormal states. In step S130, the reference control plan in the control configuration file may include emergency response measures, such as temporarily stopping electrical stimulation or significantly adjusting stimulation parameters. In step S140, the server will quickly determine the most suitable target control plan to deal with the current emergency based on the control selection strategy, in order to ensure patient safety and treatment effectiveness.

[0051] It can also be assumed that, as treatment progresses, the server accumulates a large amount of patient data. In step S110, the acquired data not only comes from the current monitoring trigger node but can also be compared and analyzed with historical data to gain a more comprehensive understanding of the changing trends in the patient's neurological state. The integration in step S120 can better capture subtle changes during long-term treatment. In step S130, the control configuration profile is continuously updated and improved to adapt to more diverse patient states and treatment needs. In step S140, the control selection strategy can also be optimized based on richer data and more complex algorithms to improve the accuracy and effectiveness of the target control plan.

[0052] Consider a more complex case where the patient has other neurological disorders in addition to Parkinson's disease. In step S110, the acquired bioelectrical signals and electrical stimulation response data need to simultaneously consider the influencing factors of multiple diseases. The server needs to analyze and process this data more finely to accurately identify key information related to the target electronic medical device.

[0053] In step S120, when integrating this complex data, the server may need to use more advanced algorithms and models to fully consider the interactions and effects between different diseases and generate a more accurate and comprehensive integrated state data sequence.

[0054] In step S130, the template integration status data sequence and template regulation scheme sequence in the regulation configuration file need to cover multiple disease combinations to provide the server with a wider and more diverse reference.

[0055] In step S140, the control selection strategy needs to be more flexible and intelligent, capable of accurately determining the most suitable target control plan from numerous reference control plans based on the patient's complex condition and comprehensive treatment goals. This may require combining multiple assessment indicators and expert knowledge, as well as utilizing machine learning algorithms for more in-depth analysis and judgment.

[0056] In a medical research project, the server not only needs to determine the target modulation plan for a single patient, but also needs to aggregate and analyze data from multiple patients. In step S110, the server simultaneously receives data from multiple patients at the target implanted neural monitoring trigger node.

[0057] In step S120, when integrating this large amount of data, the server needs to consider the individual differences and commonalities among patients in order to extract valuable information.

[0058] In step S130, the control configuration profile can be optimized and updated based on population data to provide a more targeted reference control plan for each patient.

[0059] In step S140, the regulation selection strategy can draw on the experience and trends of group treatment while taking into account the unique circumstances of each patient to determine the most suitable individual target regulation plan. This helps to improve the overall effectiveness of treatment and the scientific rigor of research.

[0060] Assuming that in the future, implantable neural monitoring and modulation technology achieves a major breakthrough, enabling devices to monitor more types of neural signals and physiological indicators, the bioelectrical signals and electrical stimulation response data acquired by the server in step S110 will be richer and more multidimensional.

[0061] In step S120, integrating this complex and multi-dimensional data will require more powerful computing capabilities and more advanced algorithms.

[0062] In step S130, the control configuration profile will need to be continuously expanded and updated to adapt to new monitoring indicators and treatment needs.

[0063] In step S140, the control selection strategy will also become more complex and precise, making full use of this new data and technology to provide patients with more optimized treatment options.

[0064] In a telemedicine scenario, a patient is located in a remote area and transmits data from an implantable neural monitoring device to a server via remote communication. In step S110, the server needs to ensure the stability and accuracy of the data transmission and handle potential delays and data loss.

[0065] In step S120, even if the data is somewhat incomplete, the server still needs to make every effort to effectively integrate it in order to provide valuable information.

[0066] In step S130, the control configuration file can take into account the characteristics of telemedicine and provide some reference control schemes suitable for limited data conditions.

[0067] In step S140, the control selection strategy needs to be more robust, capable of making reasonable decisions even with imperfect data, and able to be dynamically adjusted and optimized based on subsequent supplementary data.

[0068] Suppose that during a clinical trial, the server needs to evaluate and optimize a newly developed implantable neural monitoring and modulation system. In step S110, the server collects a large amount of data from patients at different stages of treatment, including bioelectrical signals and electrical stimulation responses at the beginning, middle, and end of treatment.

[0069] In step S120, the data from these different stages are integrated and compared to evaluate the performance of the system and the changing trends of the treatment effect.

[0070] In step S130, the configuration file can be adjusted and improved based on the results of clinical trials, providing a more reliable reference for subsequent clinical applications.

[0071] In step S140, the regulatory selection strategy can be optimized based on data feedback from clinical trials to improve the system's adaptability and effectiveness in practical applications. This helps to accelerate the maturation and widespread application of new medical technologies.

[0072] In an era of personalized medicine, each patient's treatment plan would need to be customized based on their unique genes, lifestyle, and environmental factors. In step S110, in addition to acquiring data from the implantable neural monitoring device, the server would also collect multi-source information such as the patient's genetic testing results and daily activity records.

[0073] In step S120, these multi-source information are deeply integrated with bioelectrical signals and electrical stimulation response data to construct a more comprehensive patient state model.

[0074] In step S130, the regulation configuration file will fully consider personalized factors and provide a reference regulation plan that is highly matched with the individual characteristics of the patient.

[0075] In step S140, the regulation selection strategy will accurately select the most suitable target regulation plan for the patient from a large number of reference regulation plans based on the patient's individual characteristics and overall condition, so as to achieve truly personalized medical services and improve treatment effects and patient satisfaction.

[0076] In a region with limited medical resources, the server needs to optimize resource allocation and utilization efficiency while ensuring treatment effectiveness. In step S110, the server will focus more on monitoring key data indicators and anomalies.

[0077] In step S120, valuable information is quickly extracted through an efficient integration algorithm, reducing unnecessary consumption of computing and storage resources.

[0078] In step S130, the control configuration file will prioritize control schemes that are cost-effective and easy to implement.

[0079] In step S140, the regulation selection strategy will comprehensively consider factors such as treatment effect, resource demand and operability to determine the target regulation scheme that can both meet the patient's treatment needs and save medical resources to the greatest extent, so as to achieve the rational allocation and optimized utilization of medical resources.

[0080] In an emergency rescue scenario, such as a patient suddenly experiencing a neurological crisis requiring immediate intervention, in step S110, the server acquires real-time bioelectrical signals and electrical stimulation response data as quickly as possible, capturing even the slightest changes.

[0081] In step S120, a real-time, fast integration algorithm is used to generate an integrated state data sequence in a very short time.

[0082] In step S130, the control configuration file will pre-set a series of reference control schemes for emergency situations to deal with various possible crisis situations.

[0083] In step S140, the regulation selection strategy follows the principle of "saving lives first, then treating diseases," quickly determining the target regulation plan that can stabilize the patient's vital signs and alleviate the crisis, thus buying time and conditions for further treatment.

[0084] Assuming that during a long-term rehabilitation process, the patient needs to gradually restore the function of their nervous system. In step S110, the server will continuously monitor the patient's bioelectrical signals and electrical stimulation response data at different stages of rehabilitation, and observe their changing trends.

[0085] In step S120, these long-term data are integrated and analyzed in stages to assess the progress of rehabilitation.

[0086] In step S130, the regulation configuration file will provide corresponding reference regulation plans according to different rehabilitation stages, such as focusing on stimulating nerve regeneration in the early stage and focusing on strengthening nerve connections in the later stage.

[0087] In step S140, the control selection strategy will dynamically adjust the target control plan according to the patient's recovery progress and current status to adapt to the treatment needs at different stages and promote the patient's comprehensive recovery.

[0088] In a multidisciplinary medical team, the server not only provides decision support for neurologists but also shares and exchanges data with physical therapists, rehabilitation specialists, and others. In step S110, the acquired data becomes the basis for joint analysis and discussion by the multidisciplinary team.

[0089] In step S120, the integrated data will be presented to professionals from different disciplines in an intuitive and easy-to-understand manner, so that they can understand and assess the patient's condition.

[0090] In step S130, the regulatory configuration file will fully consider multidisciplinary treatment recommendations and experience to provide a comprehensive reference regulatory plan.

[0091] In step S140, the formulation of the regulation selection strategy will combine the professional knowledge and perspectives of multiple disciplines to ensure the scientific nature and comprehensiveness of the target regulation plan, promote collaboration and communication among multidisciplinary teams, and provide patients with better medical services.

[0092] In a medical big data application scenario, a server can aggregate and analyze a large amount of patient data to uncover potential patterns and trends. In step S110, the data collected by the server becomes part of the big data.

[0093] In step S120, the integrated data can be compared and correlated with data from other patients to identify commonalities and individual characteristics.

[0094] In step S130, the control configuration file can be optimized and updated based on the analysis results of big data to provide a more universal and targeted reference control scheme.

[0095] In step S140, the regulation selection strategy can leverage big data insights to develop more precise and effective target regulation plans for patients, while also providing valuable reference and guidance for medical research and clinical practice.

[0096] Based on the above steps, this embodiment of the application achieves comprehensive and dynamic monitoring of the patient's neural activity by real-time monitoring and integration of physiological signals and electrical stimulation response data. This not only improves the accuracy and timeliness of the monitoring data but also generates a detailed integrated state data sequence through intelligent integration, providing rich data support for subsequent medical decision-making. By using the indexing mechanism of the control configuration file, a reference control scheme sequence is quickly matched and generated, effectively shortening the response time of medical intervention. Finally, based on a flexible control selection strategy, the target control scheme is determined, ensuring personalized and precise medical control, significantly improving the performance of electronic medical devices, and thus providing safer and more effective neuromodulation services.

[0097] In one possible implementation, step S110 includes:

[0098] Step S111: At the target implantable neural monitoring trigger node, acquire the real-time bioelectrical signal and real-time neural stimulation feedback data of the target electronic medical device. The real-time bioelectrical signal represents the neural activity signal in the patient's body acquired by the signal acquisition module of the electronic medical device at the target implantable neural monitoring trigger node. The real-time neural stimulation feedback data represents the response data of the patient's nerves after stimulation acquired by the stimulation feedback acquisition module of the electronic medical device at the target implantable neural monitoring trigger node.

[0099] Step S112: Determine the physiological signal data sequence based on the real-time bioelectrical signal, and determine the electrical stimulation response data sequence based on the real-time neural stimulation feedback data.

[0100] In this embodiment, we continue to assume that there is a Parkinson's disease patient, Mr. Li, who has a target electronic medical device implanted in his body to monitor and regulate neural activity. At a certain moment, the target implanted neural monitoring trigger node is activated.

[0101] The server first acquires the real-time bioelectrical signals and real-time neural stimulation feedback data from the target electronic medical device at this trigger node. At this point, the signal acquisition module of the electronic medical device begins to work, collecting real-time neural activity signals from Mr. Li's body; these signals are the real-time bioelectrical signals. For example, the acquired real-time bioelectrical signals show an abnormally high frequency and amplitude of neural activity in the motor control area of ​​Mr. Li's brain.

[0102] Simultaneously, the stimulation feedback acquisition module of the electronic medical device is also operating, collecting data on Mr. Li's nerve response after stimulation; this is the real-time nerve stimulation feedback data. Assuming that in this triggering node, the device electrically stimulated Mr. Li's nerves, the stimulation feedback data shows that Mr. Li's muscle contraction response was slightly enhanced, but still not reaching the ideal state.

[0103] Next, the server determines the physiological signal data sequence based on the acquired real-time bioelectrical signals. The server first acquires the standard prior bioelectrical signal sequence under the target implanted neural monitoring trigger node.

[0104] Assuming Mr. Li has already experienced multiple prior implantable neural monitoring trigger nodes, such as node Ya, the server calculates the node difference between node Ya and the current target implantable neural monitoring trigger node. Simultaneously, it acquires the effective duration parameter corresponding to the prior bioelectrical signal Za collected at node Ya. If the node difference is greater than the effective duration parameter, the signal characteristics of Za at the current target implantable neural monitoring trigger node are determined to be abnormal; conversely, if the node difference is less than the effective duration parameter, the signal characteristics of Za are determined to be normal.

[0105] By making such a judgment on all prior bioelectric signals, the server identifies prior bioelectric signals with normal signal characteristics under the current target implanted neural monitoring trigger node and uses them as standard prior bioelectric signals.

[0106] The server then retrieves the neural activity patterns jointly characterized by these associated prior bioelectrical signals and the current real-time bioelectrical signals. For example, it was found that Mr. Li's neural activity pattern was characterized by periodic high-frequency discharges, and the discharge amplitude showed a certain pattern of variation.

[0107] Next, the server acquires standard prior physiological signal data from the target implantable neural monitoring trigger node. Assume there are multiple prior physiological signal data, such as Kb data. The server first determines the prior implantable neural monitoring trigger node corresponding to the Kb data, i.e., the data acquisition node. Then, it searches for the validity period corresponding to the Kb data in the first data validity period bitmap.

[0108] Based on the data collection node and validity period, the server determines the validity period of the KB data. For example, if the KB data was collected at a node a week ago and has a validity period of five days, then at the current triggering node, if more than five days have passed, the KB data will no longer be considered valid.

[0109] After making this judgment, the server determines the prior physiological signal data of the current target implanted neural monitoring trigger node after determining the effective period, combines it with the previously determined real-time physiological signal data, and finally outputs a physiological signal data sequence.

[0110] In determining the electrical stimulation response data sequence, the server first serializes the acquired instantaneous neural stimulation feedback data to generate an instantaneous neural stimulation data sequence.

[0111] Then, the server obtains the stimulus response pattern data template sequence and calculates the matching degree between each stimulus response pattern data template and the instantaneous neural stimulation data sequence, generating a matching degree sequence.

[0112] Assuming a threshold of 80%, if a match rate greater than 80% exists in the matching sequence, the server identifies the highest match rate among these high-match rates and uses the corresponding stimulus response pattern data template as the instantaneous electrical stimulation response data. For example, if the highest match rate is 90%, the corresponding stimulus response pattern data template shows a strong and stable muscle contraction response.

[0113] If no matching degree greater than 80% exists in the matching sequence, the server will use the stimulus response pattern data template reflecting the abnormal stimulus response pattern as the instantaneous electrical stimulation response data. For example, this abnormal pattern may indicate unstable or excessively weak muscle contraction response.

[0114] Next, the server acquires standard prior electrical stimulation response data for the target implanted neural monitoring trigger node. Assume there are multiple prior electrical stimulation response data points, such as Kc data.

[0115] The server retrieves the prior implanted neural monitoring trigger node corresponding to the Kc data, i.e., the data acquisition node. Then, it looks up the validity period corresponding to the Kc data in the second data validity period bitmap.

[0116] Based on the data acquisition node and validity period, the server determines the validity period of the Kc data.

[0117] Finally, the server combines the prior electrical stimulation response data of the current target implanted neural monitoring trigger node after the determined effective time period with the previously determined instantaneous electrical stimulation response data to output an electrical stimulation response data sequence.

[0118] Through the above series of operations, the server successfully acquired the physiological signal data sequence and electrical stimulation response data sequence corresponding to the target electronic medical device at the target implantable neural monitoring trigger node, providing basic data support for subsequent integration, analysis and determination of control schemes.

[0119] In one possible implementation, step S112 includes:

[0120] Step S1121: Obtain the standard prior bioelectrical signal sequence under the target implantable neural monitoring trigger node. Each standard prior bioelectrical signal in the standard prior bioelectrical signal sequence represents a prior bioelectrical signal with normal signal characteristics under the target implantable neural monitoring trigger node. The prior bioelectrical signal represents the bioelectrical signal corresponding to the target electronic medical device acquired by the signal acquisition module when the prior implantable neural monitoring trigger node is activated. The prior implantable neural monitoring trigger node precedes the target implantable neural monitoring trigger node.

[0121] Step S1122: Output the standard prior bioelectric signal that has signal feature correlation with the instantaneous bioelectric signal in the standard prior bioelectric signal sequence as the associated standard prior bioelectric signal.

[0122] Step S1123: Obtain the neural activity pattern jointly characterized by the associated standard prior bioelectrical signal and the real-time bioelectrical signal; output the pattern signal data reflecting the neural activity pattern as real-time physiological signal data under the target implantable neural monitoring trigger node. The neural activity pattern is used to represent the activity pattern exhibited by the target electronic medical device under the monitoring of the target electronic medical device.

[0123] Step S1124: Obtain standard prior physiological signal data at the target implantable neural monitoring trigger node. The standard prior physiological signal data represents prior physiological signal data following the effective period at the target implantable neural monitoring trigger node. The prior physiological signal data represents signal data obtained after data acquisition of the prior bioelectrical signal at the prior implantable neural monitoring trigger node.

[0124] Step S1125: Output the sequence formed by the real-time physiological signal data and the standard prior physiological signal data as the physiological signal data sequence.

[0125] In this embodiment, as in other examples, we continue to assume that there is a patient named Ms. Wang suffering from epilepsy who has had the target electronic medical device implanted in her body. During one monitoring session, the target implanted neural monitoring trigger node is activated.

[0126] The server first acquires the standard prior bioelectrical signal sequence of the target implantable neural monitoring trigger node. Prior to this, Ms. Wang had already experienced multiple prior implantable neural monitoring trigger nodes, such as X1, X2, and X3. Within these prior implantable neural monitoring trigger nodes, the corresponding prior bioelectrical signals B1, B2, and B3 were acquired using the signal acquisition module.

[0127] The server uses specific algorithms and judgment criteria to determine whether the signal characteristics of these prior bioelectrical signals are normal at the current target implanted neural monitoring trigger node. For example, by analyzing the frequency, amplitude, waveform, and other characteristics of the signals, it determines that the signal characteristics of prior bioelectrical signal B1 are normal at the current node, while the signal characteristics of B2 are abnormal. Prior bioelectrical signals with normal signal characteristics, such as B1, are identified as standard prior bioelectrical signals, forming a standard prior bioelectrical signal sequence.

[0128] The server then compares and performs correlation analysis between the standard prior bioelectrical signal sequence and the currently acquired instantaneous bioelectrical signal. If the features displayed by the instantaneous bioelectrical signal are similar to those of the standard prior bioelectrical signal B1, the server outputs B1 as the associated standard prior bioelectrical signal.

[0129] Next, the server acquires the neural activity pattern jointly characterized by the prior bioelectrical signal B1 and the real-time bioelectrical signal. It is assumed that this pattern indicates that neurons in a specific region of Ms. Wang's brain exhibit intermittent high-frequency firing at the current monitoring moment, with a certain regularity in the firing intervals. The server outputs the pattern signal data reflecting this neural activity pattern as real-time physiological signal data under the target implanted neural monitoring trigger node.

[0130] Next, the server acquires standard prior physiological signal data from the target implantable neural monitoring trigger node. Assume prior physiological signal data P1, P2, P3, etc., were previously collected and processed at different prior implantable neural monitoring trigger nodes. The server first determines the prior implantable neural monitoring trigger node corresponding to each prior physiological signal data, and then determines its validity based on its validity period. For example, prior physiological signal data P1 was collected at a node three days ago, and its validity period is two days; therefore, P1 is no longer valid at the current trigger node. Prior physiological signal data P2, however, was collected at a node yesterday, and its validity period is three days; therefore, it is still valid at the current trigger node. The server identifies prior physiological signal data like P2, whose validity period is after the current target implantable neural monitoring trigger node, as standard prior physiological signal data.

[0131] Finally, the server combines the previously determined real-time physiological signal data with the standard prior physiological signal data P2, etc., to form a sequence, which is then output as a physiological signal data sequence. This physiological signal data sequence integrates real-time information at the current monitoring moment with recently valid prior information, providing a more comprehensive and accurate data foundation for subsequent analysis and processing.

[0132] In one possible implementation, the prior implantable neural monitoring trigger nodes include X nodes, the prior bioelectric signals include X signals, the X prior implantable neural monitoring trigger nodes include a prior implantable neural monitoring trigger node Ya, and the X prior bioelectric signals include a prior bioelectric signal Za under the prior implantable neural monitoring trigger node Ya. X and a are both positive integers.

[0133] Step S1121 includes:

[0134] Step S1121-1: Obtain the node difference between the prior implanted neural monitoring trigger node Ya and the target implanted neural monitoring trigger node, and generate the node difference corresponding to the prior implanted neural monitoring trigger node Ya.

[0135] Step S1121-2: Obtain the effective duration parameter corresponding to the prior bioelectric signal Za, and compare the node difference corresponding to the prior implanted neural monitoring trigger node Ya with the effective duration parameter corresponding to the prior bioelectric signal Za.

[0136] Step S1121-3: When the node difference corresponding to the prior implanted neural monitoring trigger node Ya is greater than the effective duration parameter corresponding to the prior bioelectric signal Za, the signal characteristics of the prior bioelectric signal Za under the target implanted neural monitoring trigger node are output as abnormal signal characteristics.

[0137] Step S1121-4: When the node difference corresponding to the prior implanted neural monitoring trigger node Ya is less than the effective duration parameter corresponding to the prior bioelectric signal Za, the signal characteristics of the prior bioelectric signal Za under the target implanted neural monitoring trigger node are output as normal signal characteristics.

[0138] Step S1121-5: When determining the signal characteristics of each of the X prior bioelectric signals under the target implantable neural monitoring trigger node, the prior bioelectric signal whose signal characteristics under the target implantable neural monitoring trigger node are normal signal characteristics is output as the standard prior bioelectric signal under the target implantable neural monitoring trigger node.

[0139] In this embodiment, we continue to assume that there is a patient named Mr. Zhang suffering from Alzheimer's disease who has a target electronic medical device implanted in his body for long-term monitoring and regulation of neural activity.

[0140] In this monitoring system, there are 10 prior implanted neural monitoring trigger nodes, labeled Y1, Y2, Y3, Y4, Y5, Y6, Y7, Y8, Y9, and Y10. Correspondingly, there are also 10 prior bioelectrical signals, namely Z1, Z2, Z3, Z4, Z5, Z6, Z7, Z8, Z9, and Z10.

[0141] The goal is to obtain the standard prior bioelectrical signal sequence at the target implantable neural monitoring trigger node (let's say Y11).

[0142] The server first obtains the node difference between the prior implanted neural monitoring trigger node Y1 and the target implanted neural monitoring trigger node Y11. For example, by calculating the difference in time intervals or monitoring cycles, the node difference is determined to be 5 units.

[0143] Then, the server obtains the effective duration parameter corresponding to the prior bioelectric signal Z1, assuming that this parameter is 4 units.

[0144] Since the node difference (5 units) corresponding to the prior implantable neural monitoring trigger node Y1 is greater than the effective duration parameter (4 units) corresponding to the prior bioelectric signal Z1, the server outputs the signal characteristics of the prior bioelectric signal Z1 under the target implantable neural monitoring trigger node Y11 as abnormal signal characteristics.

[0145] Next, the server performs the same processing on the prior implanted neural monitoring trigger node Y2 and the target implanted neural monitoring trigger node Y11. Assume the resulting node difference is 3 units, while the effective duration parameter corresponding to the prior bioelectrical signal Z2 is 4 units.

[0146] Because the node difference (3 units) corresponding to the prior implantable neural monitoring trigger node Y2 is less than the effective duration parameter (4 units) corresponding to the prior bioelectric signal Z2, the server outputs the signal characteristics of the prior bioelectric signal Z2 under the target implantable neural monitoring trigger node Y11 as normal signal characteristics.

[0147] In this manner, the eight prior implantable neural monitoring trigger nodes Y3 to Y10 are processed sequentially. The node differences between them and the target implantable neural monitoring trigger node Y11 and the effective duration parameters of the corresponding prior bioelectrical signals are compared to determine the signal characteristics of each prior bioelectrical signal under the target implantable neural monitoring trigger node Y11.

[0148] Once the signal characteristics of each of the 10 prior bioelectric signals are determined under the target implantable neural monitoring trigger node Y11, the server will output the prior bioelectric signals with normal signal characteristics under the target implantable neural monitoring trigger node Y11, such as Z2, Z5, Z7, etc., as the standard prior bioelectric signals under the target implantable neural monitoring trigger node Y11.

[0149] In this way, the server completes the acquisition of the standard prior bioelectrical signal sequence under the target implanted neural monitoring trigger node Y11. With this standard prior bioelectrical signal sequence, the currently monitored bioelectrical signals can be analyzed and processed more accurately, providing more valuable evidence for subsequent diagnosis and treatment.

[0150] For example, if the current real-time bioelectrical signal is found to have similar signal characteristics to certain signals in the standard prior bioelectrical signal sequence, it may mean that the patient's condition is in a relatively stable stage; conversely, if the difference is large, it may indicate that the condition has changed or the treatment needs to be adjusted.

[0151] For example, by accumulating and analyzing standard prior bioelectrical signal sequences under different target implantable neural monitoring trigger nodes over a long period, the server can discover the development trend and pattern of a patient's condition, thus providing strong support for doctors to formulate more personalized and precise treatment plans. It may be found that certain seasons or specific environmental factors have a significant impact on a patient's neural activity, allowing doctors to take preventative measures or adjust treatment plans in advance to better control the condition and improve the patient's quality of life.

[0152] In one possible implementation, the prior physiological signal data includes Z data points. These Z prior physiological signal data points contain prior physiological signal data Kb, where Z and b are both positive integers.

[0153] Step S1124 includes:

[0154] Step S1124-1: Obtain the prior implantable neural monitoring trigger node corresponding to the prior physiological signal data Kb, and output the prior implantable neural monitoring trigger node corresponding to the prior physiological signal data Kb as the data acquisition node corresponding to the prior physiological signal data Kb.

[0155] Step S1124-2: In the first data validity period bitmap, obtain the validity period corresponding to the prior physiological signal data Kb. The first data validity period bitmap includes an index relationship between the template physiological signal data sequence and the first template validity period sequence. An index relationship exists between a template physiological signal data in the template physiological signal data sequence and a template validity period in the first template validity period sequence.

[0156] Step S1124-3: Based on the data acquisition node corresponding to the prior physiological signal data Kb and the validity period corresponding to the prior physiological signal data Kb, determine the validity period corresponding to the prior physiological signal data Kb.

[0157] Step S1124-4: When the effective time period corresponding to each of the Z prior physiological signal data is determined, the prior physiological signal data whose effective time period is after the target implantable neural monitoring trigger node is output as the standard prior physiological signal data under the target implantable neural monitoring trigger node.

[0158] In this embodiment, we assume a patient named Mr. Sun suffering from a spinal cord injury. To monitor and regulate his neurological condition, a target electronic medical device was implanted in his body. After a period of monitoring, a large amount of prior physiological signal data was accumulated. There are 10 prior physiological signal data points, labeled K1, K2, K3, K4, K5, K6, K7, K8, K9, and K10.

[0159] First, the server obtains the prior implanted neural monitoring trigger node corresponding to the prior physiological signal data K1. Assuming that a database query reveals K1 was acquired at monitoring trigger node X, the server outputs this monitoring trigger node X as the data acquisition node corresponding to K1.

[0160] Next, the server retrieves the validity period corresponding to K1 from the first data validity period bitmap. The first data validity period bitmap establishes a clear index relationship between the template physiological signal data sequence and the first template validity period sequence. Using this index, the server finds the template physiological signal data corresponding to K1 and then determines the associated first template validity period. Assume this validity period is 10 days.

[0161] Then, the server determines the validity period of K1 based on the data collection node X corresponding to K1 and its corresponding validity period of 10 days. Assuming that the time of data collection node X is 8 days ago, then the current time is still within the validity period of K1.

[0162] Following the same steps, the server obtains the prior implanted neural monitoring trigger node corresponding to the prior physiological signal data K2, assuming it is monitoring trigger node Y, and outputs it as a data acquisition node. In the first data validity period bitmap, the validity period of K2 is determined to be 15 days. If the time of data acquisition node Y is 12 days ago, then K2 is also within the validity period.

[0163] For K3, the server determines that its corresponding prior implanted neural monitoring trigger node is Z, and the validity period is 7 days. However, if the data acquisition node Z is 10 days ago, then K3 is no longer within the validity period.

[0164] Similarly, the server processes data from K4 to K10 in the same way. For example, K4 corresponds to node A, with a validity period of 12 days and a collection time of 5 days ago, so it is still valid; K5 corresponds to node B, with a validity period of 8 days and a collection time of 9 days ago, so it is no longer valid; K6 corresponds to node C, with a validity period of 18 days and a collection time of 15 days ago, so it is still valid; K7 corresponds to node D, with a validity period of 5 days and a collection time of 7 days ago, so it is no longer valid; K8 corresponds to node E, with a validity period of 10 days and a collection time of 3 days ago, so it is still valid; K9 corresponds to node F, with a validity period of 6 days and a collection time of 8 days ago, so it is no longer valid; and K10 corresponds to node G, with a validity period of 14 days and a collection time of 10 days ago, so it is still valid.

[0165] Once the server identifies the valid time period corresponding to each of the 10 prior physiological signal data points, it performs a filtering process. Assuming the current target implantable neural monitoring trigger node is M, and its time is later than data acquisition nodes X, Y, A, E, and G, then K1, K2, K4, K8, and K10 are identified as prior physiological signal data whose valid time period is later than that of the target implantable neural monitoring trigger node M.

[0166] The server outputs these data, namely K1, K2, K4, K8, and K10, as standard prior physiological signal data under the target implantable neural monitoring trigger node M.

[0167] With this standard prior physiological signal data, the server can more accurately analyze and assess the patient's current neurological condition. For example, if the standard prior physiological signal data shows that certain characteristics of neural activity have gradually stabilized over a period of time, while the currently monitored data shows significant fluctuations, this may indicate a new change in the patient's condition, requiring further examination and adjustment of the treatment plan.

[0168] For example, by comparing standard prior physiological signal data with current data, the neurological recovery of a patient after receiving a certain treatment can be observed. If the current data shows a significant improvement compared to the standard prior physiological signal data, such as increased nerve conduction velocity or enhanced signal intensity, this may indicate that the treatment is having a positive effect, and the doctor can consider maintaining or optimizing the current treatment plan.

[0169] For example, if both standard prior physiological signal data and current data show that the patient's neurological condition has not improved significantly or has even worsened, the server can assist doctors in comprehensively reassessing the treatment plan, including adjusting drug dosage, changing electrical stimulation parameters, or trying new treatment methods.

[0170] In summary, by accurately acquiring, analyzing, and filtering the validity of prior physiological signal data, the server provides valuable reference for the medical team, helping to develop more precise and personalized treatment plans for patients, thereby improving treatment outcomes and patients' quality of life.

[0171] Furthermore, during long-term monitoring, the server can continuously accumulate and update data, further optimizing the index relationships and validity period settings in the initial data validity period bitmap to better align with the actual changes in the patient's condition and treatment needs. This will make subsequent monitoring and analysis more accurate and effective, providing stronger technical support for the rehabilitation treatment of spinal cord injury patients.

[0172] Meanwhile, if a patient's living environment, dietary habits, or exercise status change, the server can also take these factors into account when analyzing prior physiological signal data and current data, providing a more comprehensive and integrated assessment report. For example, if a patient has recently increased their exercise volume but the neural signal data has not improved accordingly, this may indicate that the intensity or type of exercise needs to be adjusted; or if a patient changes their dietary structure and certain physiological signal data show positive changes, this can provide a reference for subsequent nutritional support.

[0173] In terms of interaction with other medical devices and systems, the standard prior physiological signal data output by the server can seamlessly interface with hospital electronic medical record systems, rehabilitation therapy equipment, and other systems to achieve information sharing and collaborative work. For example, rehabilitation therapy equipment can automatically adjust training parameters based on this data to better adapt to the patient's neurological recovery stage; the electronic medical record system can record this data completely, providing rich information for doctors' consultations and academic research.

[0174] Furthermore, regarding data security and privacy protection, the server strictly adheres to relevant laws, regulations, and ethical standards when processing this prior physiological signal data, employing measures such as encryption and access control to ensure that patients' personal information and medical data are not leaked or misused. Only authorized medical personnel can access and use this data to safeguard patients' rights and safety.

[0175] In other words, through careful processing and analysis of prior physiological signal data, the server plays an indispensable role in the neurorehabilitation treatment of patients, laying a solid foundation for achieving more effective medical intervention and rehabilitation results.

[0176] In one possible implementation, step S112 includes:

[0177] Step S1121: The instantaneous neural stimulation feedback data is serialized to generate an instantaneous neural stimulation data sequence corresponding to the instantaneous neural stimulation feedback data.

[0178] Step S1122: Obtain the stimulus response pattern data template sequence, determine the matching degree between each stimulus response pattern data template in the stimulus response pattern data template sequence and the instantaneous neural stimulation data sequence, and generate a matching degree sequence.

[0179] Step S1123: If there is a matching degree greater than a set threshold value in the matching degree sequence, then obtain the maximum matching degree among the matching degrees greater than the set threshold value, and output the stimulus response pattern data template corresponding to the maximum matching degree as the instantaneous electrical stimulation response data corresponding to the instantaneous neural stimulation feedback data.

[0180] Step S1124: If there is no matching degree greater than the set threshold value in the matching degree sequence, then the stimulation response pattern data template reflecting the abnormal stimulation response pattern will be output as the instantaneous electrical stimulation response data corresponding to the instantaneous neural stimulation feedback data.

[0181] Step S1125: Obtain standard prior electrical stimulation response data at the target implantable neural monitoring trigger node. The standard prior electrical stimulation response data represents the prior electrical stimulation response data following the effective period at the target implantable neural monitoring trigger node. The prior electrical stimulation response data represents the signal data obtained after acquiring prior neural stimulation feedback data at the prior implantable neural monitoring trigger node. The prior neural stimulation feedback data represents the neural stimulation feedback data corresponding to the target electronic medical device acquired using the stimulation feedback acquisition module at the prior implantable neural monitoring trigger node. The prior implantable neural monitoring trigger node follows the target implantable neural monitoring trigger node.

[0182] Step S1126: Output the sequence formed by the instantaneous electrical stimulation response data and the standard prior electrical stimulation response data as the electrical stimulation response data sequence.

[0183] In this embodiment, we continue to assume that there is a patient named Mr. Liu who suffers from the sequelae of a stroke and has a target electronic medical device implanted in his body for nerve monitoring and stimulation.

[0184] The server receives real-time neural stimulation feedback data. First, the server serializes this real-time neural stimulation feedback data to generate a real-time neural stimulation data sequence corresponding to the real-time neural stimulation feedback data. For example, the real-time neural stimulation feedback data includes detailed information such as the intensity, frequency, duration of the stimulus, and the patient's nerve and muscle responses. After serialization, a real-time neural stimulation data sequence arranged in chronological order is formed, such as: [(Intensity: 5 volts, Frequency: 100 Hz, Duration: 2 seconds, Muscle response: slight contraction), (Intensity: 6 volts, Frequency: 120 Hz, Duration: 3 seconds, Muscle response: significant contraction), (Intensity: 7 volts, Frequency: 110 Hz, Duration: 2.5 seconds, Muscle response: moderate contraction)].

[0185] Next, the server acquires a sequence of stimulus-response pattern data templates. This sequence contains various common stimulus-response pattern data templates, such as normal response patterns, overreaction patterns, and underreaction patterns. The server uses specific algorithms and metrics to determine the matching degree between each stimulus-response pattern data template and the immediate neural stimulation data sequence, generating a matching degree sequence. Assume that the set matching degree calculation metrics include the ratio of stimulus intensity to response degree, the correlation between frequency and muscle contraction frequency, etc. The calculated matching degree sequence might be: [(Normal response pattern: 80%), (Overreaction pattern: 50%), (Underreaction pattern: 30%)].

[0186] The server then checks the matching sequence. If a matching score in the sequence exceeds a set threshold (let's say 70%), the server selects the maximum matching score from those scores. In the example above, if the matching score for the normal response pattern is 80%, which is the maximum matching score, the server outputs the normal response pattern data template corresponding to the maximum matching score as the instantaneous electrical stimulation response data corresponding to the instantaneous neural stimulation feedback data.

[0187] If there is no match in the matching sequence that is greater than the set threshold, the server will output the data template reflecting the abnormal stimulus response pattern (such as the abnormal response pattern) as the real-time electrical stimulation response data corresponding to the real-time neural stimulation feedback data.

[0188] Next, the server acquires standard prior electrical stimulation response data for the target implantable neural monitoring trigger node. Prior to this, multiple prior implantable neural monitoring trigger nodes have already performed monitoring and stimulation. The server filters out prior electrical stimulation response data from those prior implantable neural monitoring trigger nodes that are later than the current target implantable neural monitoring trigger node. It is assumed that these prior electrical stimulation response data include different stimulation parameters and corresponding neural and muscle responses. Based on the data acquisition time and pre-set validity period rules, the server determines which prior electrical stimulation response data are valid standard prior electrical stimulation response data for the current target implantable neural monitoring trigger node. For example, if a prior electrical stimulation response data was acquired 5 days ago and has a validity period of 7 days, then it is considered valid standard prior electrical stimulation response data.

[0189] Finally, the server combines the immediate electrical stimulation response data with the standard prior electrical stimulation response data to form a sequence, outputting the electrical stimulation response data sequence. Assuming the immediate electrical stimulation response data represents the normal response pattern, and the standard prior electrical stimulation response data includes one slightly overreacting pattern and one normal response pattern, the output electrical stimulation response data sequence might be: [(Immediate: Normal Response Pattern), (Prior: Slightly Overreacting Pattern), (Prior: Normal Response Pattern)].

[0190] With this electrical stimulation response data sequence, the medical team can gain a more comprehensive understanding of how patients respond to neural stimulation. For example, if the sequence mostly shows normal response patterns, the current stimulation protocol may be appropriate and does not require significant adjustments; if many abnormal response patterns appear, it may be necessary to reassess the stimulation parameters, such as reducing the stimulation intensity or adjusting the stimulation frequency.

[0191] During long-term treatment, the server can discover trends in the patient's adaptive response to stimuli by continuously accumulating and analyzing such data sequences. For example, over time, the patient's response to the same stimulus parameters may gradually weaken, suggesting that the stimulus intensity needs to be appropriately increased to maintain the treatment effect.

[0192] Furthermore, if the patient's condition changes, such as an improvement in nerve recovery, the electrical stimulation response data sequence will also change accordingly. The server can provide timely alerts to the medical team based on these changes, allowing them to adjust treatment strategies, such as reducing the stimulation frequency or shortening the stimulation duration, to avoid overstimulation and nerve damage.

[0193] Furthermore, when new treatment methods or devices are introduced, the server can compare and analyze the new stimulus response data with historical data to assess the effectiveness and safety of the new method. If the electrical stimulation response data sequence corresponding to the new method shows better therapeutic effects and fewer adverse reactions, the medical team can consider wider application of this new method.

[0194] Therefore, by comprehensively processing and analyzing real-time neural stimulation feedback data and prior electrical stimulation response data, important data support is provided for the neurorehabilitation treatment of stroke sequelae, which helps to optimize treatment plans, improve treatment effects, and promote the recovery of patients' neurological function.

[0195] In one possible implementation, the prior electrical stimulation response data includes W data points, and the W prior electrical stimulation response data points contain prior electrical stimulation response data Kc, where W and c are both positive integers.

[0196] Step S1125 includes:

[0197] Step S1125-1: Obtain the a priori implantable neural monitoring trigger node corresponding to the a priori electrical stimulation response data Kc, and output the a priori implantable neural monitoring trigger node corresponding to the a priori electrical stimulation response data Kc as the data acquisition node corresponding to the a priori electrical stimulation response data Kc.

[0198] Step S1125-2: In the second data validity period bitmap, obtain the validity period corresponding to the prior electrical stimulation response data Kc. The second data validity period bitmap includes an index relationship between the stimulus response pattern data template sequence and the second template validity period sequence. An index relationship exists between a stimulus response pattern data template in the stimulus response pattern data template sequence and a template validity period in the second template validity period sequence.

[0199] Step S1125-3: Based on the data acquisition node corresponding to the prior electrical stimulation response data Kc and the validity period corresponding to the prior electrical stimulation response data Kc, determine the validity period corresponding to the prior electrical stimulation response data Kc.

[0200] Step S1125-4: When the effective time period corresponding to each of the W prior electrical stimulation response data is determined, the prior electrical stimulation response data whose effective time period is after the target implantable neural monitoring trigger node is output as the standard prior electrical stimulation response data under the target implantable neural monitoring trigger node.

[0201] In this embodiment, we continue to assume that there is a patient named Mr. Chen suffering from ALS. To help alleviate his condition, a target electronic medical device was implanted in his body for nerve monitoring and stimulation. During the long-term treatment process, a large amount of prior electrical stimulation response data was accumulated.

[0202] Let's assume there are 8 prior electrical stimulus response data, labeled K1, K2, K3, K4, K5, K6, K7, and K8.

[0203] For the prior electrical stimulation response data K3, the server first obtains its corresponding prior implantable neural monitoring trigger node. Assuming that a database query reveals that the prior implantable neural monitoring trigger node for K3 is node P, the server outputs node P as the data acquisition node corresponding to K3.

[0204] Next, the server retrieves the validity period corresponding to K3 from the second data validity period bitmap. The second data validity period bitmap establishes an index relationship between the stimulus-response pattern data template sequence and the second template validity period sequence. Using this index, the server finds the stimulus-response pattern data template corresponding to K3 and then determines the associated second template validity period. Assume this validity period is 5 days.

[0205] Then, the validity period of K3 is determined based on the data acquisition node P corresponding to K3 and its corresponding validity period of 5 days. Assuming that the time of data acquisition node P is 3 days ago, then the current time is still within the validity period of K3.

[0206] The server processes K1 in the same way. The prior implanted neural monitoring trigger node corresponding to K1 is node Q, and its validity period is determined to be 4 days in the second data validity period bitmap. Assuming the data acquisition node Q was acquired 2 days ago, K1 is also within its validity period.

[0207] For K2, its corresponding prior implantable neural monitoring trigger node is node R, which is valid for 6 days. The data acquisition node R was 7 days ago, so K2 is no longer valid.

[0208] Similarly, the server processes data from K4 to K8 in the same way. For example, K4 corresponds to node S, has a validity period of 3 days, and was collected 1 day ago, so it is still valid; K5 corresponds to node T, has a validity period of 7 days, and was collected 8 days ago, so it is no longer valid; K6 corresponds to node U, has a validity period of 5 days, and was collected 4 days ago, so it is still valid; K7 corresponds to node V, has a validity period of 4 days, and was collected 6 days ago, so it is no longer valid; K8 corresponds to node W, has a validity period of 6 days, and was collected 2 days ago, so it is still valid.

[0209] Once the server identifies the valid time period corresponding to each of the eight prior electrical stimulation response data points, it performs a filtering process. Assuming the current target implantable neural monitoring trigger node is X, and its time is later than data acquisition nodes P, Q, S, U, and W, then K1, K3, K4, K6, and K8 are identified as prior electrical stimulation response data whose valid time period is later than that of the target implantable neural monitoring trigger node X.

[0210] The server outputs these data, namely K1, K3, K4, K6, and K8, as standard prior electrical stimulation response data under the target implantable neural monitoring trigger node X.

[0211] With this standard prior electrical stimulation response data, the server can provide the medical team with more valuable references. For example, if the standard prior electrical stimulation response data shows that certain stimulation parameters have produced good therapeutic effects over a period of time, the medical team can consider using or fine-tuning these parameters in the current treatment.

[0212] For example, if standard prior electrical stimulation response data indicates that certain stimulation patterns are likely to cause discomfort or adverse reactions in patients, then these patterns can be avoided in the current treatment.

[0213] During long-term treatment, the server continuously updates and analyzes this data to detect trends in the patient's response to electrical stimulation. For example, as the condition progresses, the patient may respond differently to the same stimulation parameters, or their adaptation to certain stimulation patterns may change. The server can promptly capture these changes and provide early warnings to the medical team so that the treatment plan can be adjusted.

[0214] Furthermore, if new research findings or treatment concepts emerge, the server can compare and integrate them with existing prior electrical stimulation response data, providing medical teams with more forward-looking recommendations. For example, if new research suggests that a novel electrical stimulation modality may be beneficial for ALS patients, the server can assess the feasibility and potential risks of applying this new modality to the patient based on existing prior data.

[0215] Furthermore, the server can also compare and analyze these prior electrical stimulation response data with data from other patients. If a patient's response is found to differ significantly from that of most patients, this may indicate the need for more in-depth individual studies of that patient, or for adjustments to the treatment strategy to suit their specific circumstances.

[0216] Meanwhile, regarding data management and security, the server strictly adheres to relevant laws and ethical requirements to ensure the confidentiality and integrity of this prior electrical stimulation response data. Only authorized medical personnel can access and use this data to protect patient privacy and rights.

[0217] In other words, by accurately acquiring, analyzing, and screening the effectiveness of prior electrical stimulation response data, the server provides strong support for the treatment of ALS patients, helping to improve the accuracy and effectiveness of treatment and improve patients' quality of life.

[0218] In one possible implementation, the control selection strategy is a scheme-based selection strategy.

[0219] Step S140 includes:

[0220] Step A110: Obtain the prior modulation scheme of the forward prior implantable neural monitoring trigger node. The forward prior implantable neural monitoring trigger node is used to represent the previous round of implantable neural monitoring trigger nodes of the target implantable neural monitoring trigger node.

[0221] Step A120: Output the reference control schemes in the reference control scheme sequence that are the same as the prior control scheme as the redundant control scheme.

[0222] Step A130: Delete the redundant control schemes in the reference control scheme sequence to generate a purified control scheme sequence.

[0223] Step A140: Obtain the scheme knowledge depth corresponding to each reference control scheme in the purification control scheme sequence, and generate a scheme knowledge depth sequence.

[0224] Step A150: Obtain the maximum scheme knowledge depth in the scheme knowledge depth sequence, and output the reference control scheme corresponding to the maximum scheme knowledge depth as the target control scheme under the target implanted neural monitoring trigger node.

[0225] In this embodiment, we continue to assume that there is a patient named Ms. Zhou suffering from epilepsy who has had a target electronic medical device implanted in her body for nerve monitoring and regulation.

[0226] The selection strategy for regulation is based on the selection strategy of the plan.

[0227] First, the server obtains the prior modulation scheme of the forward-looking implanted neural monitoring trigger node. It is assumed that the target implanted neural monitoring trigger node represents the current monitoring and modulation moment, while the prior modulation scheme used in the previous round of implanted neural monitoring trigger nodes was to increase the intensity and frequency of electrical stimulation.

[0228] The server then compares the current reference control scheme sequence with this prior control scheme. Assume the reference control scheme sequence includes schemes such as "increasing electrical stimulation intensity," "adjusting the electrical stimulation waveform," and "changing the pulse width of the electrical stimulation." After comparison, it is found that the scheme "increasing electrical stimulation intensity" is the same as the prior control scheme, and the server outputs it as a redundant control scheme.

[0229] Next, the server removes the redundant control scheme "increase electrical stimulation intensity" from the reference control scheme sequence, thereby generating a purified control scheme sequence. At this point, the purified control scheme sequence contains schemes such as "adjust electrical stimulation waveform" and "change electrical stimulation pulse width".

[0230] The server then retrieves the protocol knowledge depth for each reference protocol in the purification protocol sequence. Protocol knowledge depth can be determined by assessing factors such as the complexity of the neuroscience principles involved, the potential impact on the nervous system, and the effectiveness and experience of previous clinical applications. For example, the protocol knowledge depth for "adjusting the electrical stimulation waveform" is assessed as relatively deep, while the protocol knowledge depth for "changing the pulse width of the electrical stimulation" is relatively shallow.

[0231] Finally, the server obtains the maximum scheme knowledge depth in the scheme knowledge depth sequence, which is the depth corresponding to "adjusting the electrical stimulation waveform". The server uses the "adjusting the electrical stimulation waveform" reference control scheme corresponding to this maximum scheme knowledge depth as the target control scheme under the target implanted neural monitoring trigger node.

[0232] With this targeted control plan, the medical team can adjust Ms. Zhou's nerve stimulation accordingly. For example, if the "adjusting the electrical stimulation waveform" plan is selected as the targeted control plan, the medical device may change the waveform shape of the electrical stimulation according to specific algorithms and parameter settings to more precisely target Ms. Zhou's nervous system, thereby achieving a better therapeutic effect.

[0233] During subsequent treatment, the server will continuously record and analyze the effect data of each modulation. If it is found that the "adjusting the electrical stimulation waveform" plan does not achieve the expected effect, or if the patient's condition changes, the server will re-determine a new target modulation plan according to the above steps at the next implanted neural monitoring trigger point.

[0234] At the same time, the server will also conduct comprehensive analysis of Ms. Zhou's treatment data and data from other similar patients, continuously optimizing the evaluation criteria and algorithms for the depth of knowledge of the treatment plan, so as to improve the accuracy and effectiveness of determining the target control plan.

[0235] In addition, the server will also consider the impact of individual patient differences and lifestyle habits on the treatment plan. For example, if Ms. Zhou has been under a lot of stress recently or has poor sleep quality, the server will take these factors into account when determining the target treatment plan, and may choose a treatment plan with a milder impact on the nervous system.

[0236] Furthermore, if new neuroscience research findings or clinical experience emerge, the server will promptly update its assessment system for the depth of knowledge of the proposed protocols to ensure that the identified target regulation protocols keep pace with the latest medical advancements and treatment concepts.

[0237] Therefore, we can provide Ms. Zhou with the most suitable target regulation plan for her current condition, continuously optimize the treatment effect, and improve her quality of life.

[0238] Alternatively, in one possible implementation, the control selection strategy is a selection strategy based on integrated state data.

[0239] Step S140 includes:

[0240] Step B110: Based on the aforementioned modulation selection strategy, obtain the neuromodulation effect characteristics of the target electronic medical device at the target implantable neural monitoring trigger node. These neuromodulation effect characteristics represent the state characteristics of the target electronic medical device within the patient's body.

[0241] Step B120 involves loading the neural modulation effect features and the integrated state data sequence into a deep learning network. Using the neural modulation effect features and the deep learning network, a neural modulation effect index corresponding to each integrated state data point in the integrated state data sequence is generated. The deep learning network is obtained by learning network parameters from an initial deep learning network using both the sample neural modulation effect features and the sample integrated state data sequence.

[0242] Step B130: Obtain the maximum neural modulation effect index from the neural modulation effect index sequence, output the integrated state data corresponding to the maximum neural modulation effect index as target integrated state data related to the neural modulation effect feature, and output the reference modulation scheme corresponding to the target integrated state data as the target modulation scheme.

[0243] In this embodiment, we assume that a patient named Mr. Wu, who suffers from migraines, has a target electronic medical device implanted in his body for nerve monitoring and regulation.

[0244] The control selection strategy is based on the selection strategy that integrates state data.

[0245] First, based on the modulation selection strategy, the server acquires the neuromodulation effect characteristics of the target electronic medical device at the target implanted neural monitoring trigger node. These characteristics may include the frequency of Mr. Wu's migraine attacks, the intensity of pain, and the stability of neural activity, which are used to represent the state characteristics of the target electronic medical device in Mr. Wu's body.

[0246] The server then loads these neuromodulation effect features along with the previously integrated state data sequence into the deep learning network. This deep learning network is obtained by learning the network parameters of an initial deep learning network using a large number of example neuromodulation effect features and example integrated state data sequences.

[0247] For example, suppose the integrated state data sequence contains physiological signal data (such as the frequency and amplitude of neural activity) at different times, electrical stimulation response data (such as the degree of muscle contraction and the amount of neurotransmitter release), and their integration. When the neuromodulation effect characteristics (such as three migraine attacks in the past week, each with a pain level of 7, and unstable neural activity) and this integrated state data sequence are input into a deep learning network, the deep learning network will generate a neuromodulation effect index corresponding to each integrated state data in the integrated state data sequence through the complex patterns and relationships it has learned.

[0248] Assume the generated sequence of neural modulation effect indicators is: [(Integrated state data 1: 0.8), (Integrated state data 2: 0.6), (Integrated state data 3: 0.9), (Integrated state data 4: 0.7)]

[0249] Next, the server obtains the maximum neuromodulation effect index from the neuromodulation effect index sequence. In the example above, the maximum neuromodulation effect index is 0.9, and the corresponding integrated state data is integrated state data 3.

[0250] The server outputs the integrated state data 3 corresponding to this maximum neuromodulation effect index as target integrated state data related to the neuromodulation effect characteristics. Then, it finds the reference modulation scheme corresponding to integrated state data 3, assuming it is "appropriately increasing the frequency and intensity of electrical stimulation and adjusting the duration of stimulation", and outputs this reference modulation scheme as the target modulation scheme.

[0251] With this targeted control plan, the medical team can adjust Mr. Wu's nerve stimulation accordingly. For example, the medical equipment will provide more precise electrical stimulation to Mr. Wu's nervous system based on the parameter settings in the plan.

[0252] During subsequent treatment, the server will continuously collect new data and constantly update the neuromodulation effect characteristics and integrated state data sequences. It will then analyze and optimize the data again through a deep learning network to ensure that the determined target modulation plan is always the most suitable for Mr. Wu's current condition.

[0253] At the same time, the server also considers the impact of other factors on the modulation effect. For example, if Mr. Wu's recent life stress has increased or his sleep quality has deteriorated, these factors may affect the effectiveness of neuromodulation. When acquiring the characteristics of neuromodulation effects, the server will take this additional information into account, thereby more accurately determining the target modulation plan.

[0254] Furthermore, if new research findings or clinical experience are available regarding migraine treatment, the server can incorporate this knowledge into the training of the deep learning network, continuously improving the network's performance and predictive capabilities to provide better target control solutions.

[0255] Furthermore, the server will seamlessly interface with other medical devices and systems to enable information sharing and collaborative work. For example, it can transmit the determined target control plan to medical devices in real time to ensure the timeliness and accuracy of treatment.

[0256] Therefore, we can provide Mr. Wu, a migraine patient, with a personalized, precise and effective target regulation plan, continuously improve the treatment effect and improve his quality of life.

[0257] In one possible implementation, the method further includes:

[0258] Step C110: Obtain the prior modulation scheme of the forward prior implanted neural monitoring trigger node. The forward prior implanted neural monitoring trigger node is used to represent the previous round of implanted neural monitoring trigger nodes of the target implanted neural monitoring trigger node.

[0259] Step C120: Obtain the implantable neural monitoring termination node corresponding to the prior modulation scheme.

[0260] Step C130: If the implantable neural monitoring termination node corresponding to the prior control scheme is later than the target implantable neural monitoring trigger node, then determine the execution interval parameter between the target implantable neural monitoring trigger node and the implantable neural monitoring termination node corresponding to the prior control scheme, obtain the scheme timeout parameter corresponding to the target control scheme, and when the execution interval parameter is less than the scheme timeout parameter, output the implantable neural monitoring termination node corresponding to the prior control scheme as the implantable neural monitoring trigger node of the target control scheme.

[0261] Step C140: If the implantable neural monitoring termination node corresponding to the prior modulation scheme is earlier than the target implantable neural monitoring trigger node, then the target implantable neural monitoring trigger node is output as the implantable neural monitoring trigger node of the target modulation scheme.

[0262] In this embodiment, we continue to assume that there is a patient named Ms. Lin suffering from depression who has a target electronic medical device implanted in her body for neural monitoring and regulation.

[0263] The server first obtains the prior modulation scheme of the implanted neural monitoring trigger node. It is assumed that the prior modulation scheme used by the previous implanted neural monitoring trigger node was to adjust the frequency and intensity of electrical stimulation.

[0264] Next, the server obtains the implantable neural monitoring termination node corresponding to this prior modulation scheme. For example, the implantable neural monitoring termination node corresponding to this prior modulation scheme is at a specific time point T1.

[0265] If the implanted neural monitoring termination node T1 corresponding to this prior control scheme is after the current target implanted neural monitoring trigger node T2, the server determines the execution interval parameter between T2 and T1. Simultaneously, it obtains the scheme timeout parameter corresponding to the current target control scheme.

[0266] Assuming the execution interval parameter is 3 hours and the scheme timeout parameter is 5 hours, since the execution interval parameter is less than the scheme timeout parameter, the server will output the implantable neural monitoring termination node T1 corresponding to the prior control scheme as the implantable neural monitoring trigger node of the target control scheme.

[0267] If the termination node of the implantable neural monitoring corresponding to the prior control scheme, such as time point T3, is earlier than the current target implantable neural monitoring trigger node T4, the server will directly output the current target implantable neural monitoring trigger node T4 as the implantable neural monitoring trigger node of the target control scheme.

[0268] This process allows for the accurate determination of the implantable neural monitoring trigger point of the target regulation plan based on the execution status and relevant parameters of the prior regulation plan, thereby providing more precise and effective neural regulation services for Ms. Lin's treatment of depression.

[0269] Figure 2 The following is a schematic diagram of the hardware structure of a deep learning system 100 for implementing the above-described method for monitoring state data of electronic medical devices, as provided in an embodiment of this application. Figure 2 As shown, the deep learning system 100 may include a processor 110, a machine-readable storage medium 120, a bus 130, and a communication unit 140.

[0270] In an alternative embodiment, the deep learning system 100 may be a single server or a group of servers. The server group may be centralized or distributed (e.g., the deep learning system 100 may be a distributed system). In an alternative embodiment, the deep learning system 100 may be local or remote. For example, the deep learning system 100 may access information and / or data stored in machine-readable storage medium 120 via a network. As another example, the deep learning system 100 may directly connect to machine-readable storage medium 120 to access stored information and / or data. In an alternative embodiment, the deep learning system 100 may be implemented on a cloud platform. By way of example only, the cloud platform may include private clouds, public clouds, aggregated clouds, community clouds, distributed clouds, internal clouds, multi-tiered clouds, etc., or any integration thereof.

[0271] Machine-readable storage medium 120 can store data and / or instructions. In an alternative embodiment, machine-readable storage medium 120 can store data acquired from an external terminal. In an alternative embodiment, machine-readable storage medium 120 can store data and / or instructions used by the deep learning system 100 to perform or complete the exemplary methods described in this application. In an alternative embodiment, machine-readable storage medium 120 may include mass storage, removable storage, volatile read-write storage, read-only storage, etc., or any combination thereof. Exemplary mass storage may include disks, optical disks, solid-state drives, etc. Exemplary removable storage may include flash drives, floppy disks, optical disks, memory cards, compact disks, magnetic tapes, etc.

[0272] In the specific implementation process, multiple processors 110 execute computer-executable instructions stored in the machine-readable storage medium 120, so that the processors 110 can execute the status data monitoring method for electronic medical devices as described in the above method embodiment. The processors 110, the machine-readable storage medium 120, and the communication unit 140 are connected through the bus 130. The processors 110 can be used to describe the sending and receiving actions of the communication unit 140.

[0273] The specific implementation process of processor 110 can be found in the various method embodiments executed by the deep learning system 100 described above. The implementation principle and technical effect are similar, and will not be repeated here.

[0274] Furthermore, this application embodiment also provides a readable storage medium, wherein computer-executable instructions are pre-installed in the readable storage medium, and when the processor executes the computer-executable instructions, the above-described method for monitoring status data of electronic medical devices is implemented.

[0275] Similarly, it should be noted that, in order to simplify the description disclosed in this application and thus aid in the understanding of one or more embodiments of the invention, multiple features may sometimes be combined into one embodiment, drawing, or description thereof in the foregoing description of the embodiments of this application.

Claims

1. A method for monitoring status data of electronic medical devices, characterized in that, The method includes: At the target implantable neural monitoring trigger node, physiological signal data sequences and electrical stimulation response data sequences corresponding to the target electronic medical device are acquired. The physiological signal data sequences are obtained by collecting bioelectrical signals at the target implantable neural monitoring trigger node. The electrical stimulation response data sequences are obtained by collecting neural stimulation feedback data at the target implantable neural monitoring trigger node. The bioelectrical signals represent neural activity signals monitored by the implanted device in the patient's body. The neural stimulation feedback represents the response data generated after the implanted device stimulates the patient's nerves. The physiological signal data sequence is integrated with the electrical stimulation response data sequence to generate an integrated state data sequence under the target implantable neural monitoring trigger node; In the control configuration file, the reference control scheme corresponding to each integrated state data in the integrated state data sequence is obtained, and a reference control scheme sequence is generated; the control configuration file includes the index relationship between the template integrated state data sequence and the template control scheme sequence; there is an index relationship between a template integrated state data in the template integrated state data sequence and a template control scheme in the template control scheme sequence. Based on the regulation selection strategy, the target regulation scheme under the target implanted neural monitoring trigger node is determined in the reference regulation scheme sequence.

2. The method for monitoring status data of electronic medical devices according to claim 1, characterized in that, The acquisition of physiological signal data sequences and electrical stimulation response data sequences corresponding to the target electronic medical device at the target implanted neural monitoring trigger node includes: At the target implantable neural monitoring trigger node, real-time bioelectrical signals and real-time neural stimulation feedback data of the target electronic medical device are acquired; the real-time bioelectrical signals are used to represent the neural activity signals in the patient's body collected by the signal acquisition module of the electronic medical device at the target implantable neural monitoring trigger node; the real-time neural stimulation feedback data are used to represent the patient's nerve response data after stimulation collected by the stimulation feedback acquisition module of the electronic medical device at the target implantable neural monitoring trigger node. The physiological signal data sequence is determined based on the real-time bioelectrical signal, and the electrical stimulation response data sequence is determined based on the real-time neural stimulation feedback data.

3. The method for monitoring status data of electronic medical devices according to claim 2, characterized in that, The determination of the physiological signal data sequence based on the real-time bioelectrical signal includes: A standard prior bioelectrical signal sequence is acquired under the target implantable neural monitoring trigger node; each standard prior bioelectrical signal in the standard prior bioelectrical signal sequence is used to represent a prior bioelectrical signal with normal signal characteristics under the target implantable neural monitoring trigger node; the prior bioelectrical signal is used to represent the bioelectrical signal corresponding to the target electronic medical device acquired by the signal acquisition module at the prior implantable neural monitoring trigger node; the prior implantable neural monitoring trigger node precedes the target implantable neural monitoring trigger node; The standard prior bioelectric signal that has a signal feature correlation with the instantaneous bioelectric signal in the standard prior bioelectric signal sequence is output as the associated standard prior bioelectric signal. The neural activity pattern jointly characterized by the prior bioelectrical signal of the associated standard and the real-time bioelectrical signal is obtained, and the pattern signal data reflecting the neural activity pattern is output as real-time physiological signal data under the target implantable neural monitoring trigger node; the neural activity pattern is used to represent the activity pattern exhibited by the target electronic medical device under the monitoring of the target electronic medical device. Acquire standard prior physiological signal data under the target implantable neural monitoring trigger node; the standard prior physiological signal data is used to represent prior physiological signal data after the effective period at the target implantable neural monitoring trigger node; the prior physiological signal data is used to represent signal data obtained after data acquisition of the prior bioelectrical signal under the prior implantable neural monitoring trigger node; The sequence formed by the real-time physiological signal data and the standard prior physiological signal data is output as the physiological signal data sequence.

4. The method for monitoring status data of electronic medical devices according to claim 3, characterized in that, The prior implantable neural monitoring trigger nodes include X nodes, and the prior bioelectric signals include X signals. The X prior implantable neural monitoring trigger nodes include a prior implantable neural monitoring trigger node Ya, and the X prior bioelectric signals include a prior bioelectric signal Za under the prior implantable neural monitoring trigger node Ya; X and a are both positive integers. The acquisition of the standard prior bioelectrical signal sequence under the target implantable neural monitoring trigger node includes: Obtain the node difference between the prior implanted neural monitoring trigger node Ya and the target implanted neural monitoring trigger node, and generate the node difference corresponding to the prior implanted neural monitoring trigger node Ya; Obtain the effective duration parameter corresponding to the prior bioelectric signal Za, and compare the node difference corresponding to the prior implanted neural monitoring trigger node Ya with the effective duration parameter corresponding to the prior bioelectric signal Za. When the node difference corresponding to the prior implanted neural monitoring trigger node Ya is greater than the effective duration parameter corresponding to the prior bioelectric signal Za, the signal characteristics of the prior bioelectric signal Za under the target implanted neural monitoring trigger node are output as abnormal signal characteristics. When the node difference corresponding to the prior implanted neural monitoring trigger node Ya is less than the effective duration parameter corresponding to the prior bioelectric signal Za, the signal characteristics of the prior bioelectric signal Za under the target implanted neural monitoring trigger node are output as normal signal characteristics. When the signal characteristics of each of the X prior bioelectric signals under the target implantable neural monitoring trigger node are determined, the prior bioelectric signal whose signal characteristics under the target implantable neural monitoring trigger node are normal signal characteristics is output as the standard prior bioelectric signal under the target implantable neural monitoring trigger node.

5. The method for monitoring status data of electronic medical devices according to claim 3, characterized in that, The prior physiological signal data includes Z items; the Z prior physiological signal data includes prior physiological signal data Kb, where Z and b are both positive integers; The acquisition of standard prior physiological signal data under the target implanted neural monitoring trigger node includes: Obtain the prior implantable neural monitoring trigger node corresponding to the prior physiological signal data Kb, and output the prior implantable neural monitoring trigger node corresponding to the prior physiological signal data Kb as the data acquisition node corresponding to the prior physiological signal data Kb. In the first data validity period bitmap, the validity period corresponding to the prior physiological signal data Kb is obtained; the first data validity period bitmap includes an index relationship between the template physiological signal data sequence and the first template validity period sequence; there is an index relationship between a template physiological signal data in the template physiological signal data sequence and a template validity period in the first template validity period sequence. Based on the data acquisition node corresponding to the prior physiological signal data Kb and the validity period corresponding to the prior physiological signal data Kb, the validity period corresponding to the prior physiological signal data Kb is determined. When the effective time period corresponding to each of the Z prior physiological signal data is determined, the prior physiological signal data whose effective time period is after the target implantable neural monitoring trigger node is output as the standard prior physiological signal data under the target implantable neural monitoring trigger node.

6. The method for monitoring status data of electronic medical devices according to claim 2, characterized in that, The step of determining the electrical stimulation response data sequence based on the instantaneous neural stimulation feedback data includes: The real-time neural stimulation feedback data is serialized to generate a real-time neural stimulation data sequence corresponding to the real-time neural stimulation feedback data. Obtain a stimulus response pattern data template sequence, determine the matching degree between each stimulus response pattern data template in the stimulus response pattern data template sequence and the instantaneous neural stimulation data sequence, and generate a matching degree sequence; If there is a matching degree greater than a set threshold in the matching degree sequence, then the maximum matching degree is obtained from the matching degree greater than the set threshold, and the stimulus response pattern data template corresponding to the maximum matching degree is output as the instantaneous electrical stimulation response data corresponding to the instantaneous neural stimulation feedback data. If there is no matching degree greater than the set threshold value in the matching degree sequence, the stimulus response pattern data template reflecting the abnormal stimulus response pattern will be output as the instantaneous electrical stimulation response data corresponding to the instantaneous neural stimulation feedback data. Acquire standard prior electrical stimulation response data at the target implantable neural monitoring trigger node; the standard prior electrical stimulation response data is used to represent prior electrical stimulation response data after the effective period at the target implantable neural monitoring trigger node; the prior electrical stimulation response data is used to represent signal data obtained after data acquisition of prior neural stimulation feedback data at the prior implantable neural monitoring trigger node; the prior neural stimulation feedback data is used to represent neural stimulation feedback data corresponding to the target electronic medical device acquired by the stimulation feedback acquisition module at the prior implantable neural monitoring trigger node; the prior implantable neural monitoring trigger node is after the target implantable neural monitoring trigger node. The sequence formed by the instantaneous electrical stimulation response data and the standard prior electrical stimulation response data is output as the electrical stimulation response data sequence.

7. The method for monitoring status data of electronic medical devices according to claim 6, characterized in that, The prior electrical stimulation response data includes W data points, and each of the W prior electrical stimulation response data points contains prior electrical stimulation response data Kc, where W and c are both positive integers. The acquisition of standard prior electrical stimulation response data at the target implantable neural monitoring trigger node includes: Obtain the a priori implantable neural monitoring trigger node corresponding to the a priori electrical stimulation response data Kc, and output the a priori implantable neural monitoring trigger node corresponding to the a priori electrical stimulation response data Kc as the data acquisition node corresponding to the a priori electrical stimulation response data Kc. In the second data validity period bitmap, the validity period corresponding to the prior electrical stimulation response data Kc is obtained; the second data validity period bitmap includes an index relationship between the stimulus response pattern data template sequence and the second template validity period sequence; there is an index relationship between a stimulus response pattern data template in the stimulus response pattern data template sequence and a template validity period in the second template validity period sequence. Based on the data acquisition node corresponding to the prior electrical stimulation response data Kc and the validity period corresponding to the prior electrical stimulation response data Kc, the validity period corresponding to the prior electrical stimulation response data Kc is determined. When the effective time period corresponding to each of the W prior electrical stimulation response data is determined, the prior electrical stimulation response data whose effective time period is after the target implantable neural monitoring trigger node is output as the standard prior electrical stimulation response data under the target implantable neural monitoring trigger node.

8. The method for monitoring status data of electronic medical devices according to claim 1, characterized in that, The control selection strategy is a selection strategy based on the scheme; The step of determining the target modulation scheme under the target implanted neural monitoring trigger node from the reference modulation scheme sequence based on the modulation selection strategy includes: Obtain the prior control scheme of the forward prior implantable neural monitoring trigger node; the forward prior implantable neural monitoring trigger node is used to represent the previous round of implantable neural monitoring trigger node of the target implantable neural monitoring trigger node; The reference control schemes that are the same as the prior control schemes in the reference control scheme sequence are output as redundant control schemes. The redundant control schemes in the reference control scheme sequence are deleted to generate a clean control scheme sequence; Obtain the scheme knowledge depth corresponding to each reference control scheme in the purification control scheme sequence, and generate a scheme knowledge depth sequence; Obtain the maximum scheme knowledge depth in the scheme knowledge depth sequence, and output the reference control scheme corresponding to the maximum scheme knowledge depth as the target control scheme under the target implanted neural monitoring trigger node; Alternatively, the control selection strategy may be a selection strategy based on integrated state data; The step of determining the target modulation scheme under the target implanted neural monitoring trigger node from the reference modulation scheme sequence based on the modulation selection strategy includes: Based on the aforementioned regulation selection strategy, the neuromodulation effect characteristics of the target electronic medical device under the target implantable neural monitoring trigger node are obtained; the neuromodulation effect characteristics are used to represent the state characteristics of the target electronic medical device in the patient's body. The neural modulation effect features and the integrated state data sequence are loaded into a deep learning network. Through the neural modulation effect features and the deep learning network, a neural modulation effect index corresponding to each integrated state data in the integrated state data sequence is generated. The deep learning network is obtained by learning the network parameters of an initial deep learning network together with the sample neural modulation effect features and the sample integrated state data sequence. The maximum neuromodulation effect index is obtained from the neuromodulation effect index sequence. The integrated state data corresponding to the maximum neuromodulation effect index is output as target integrated state data related to the neuromodulation effect feature. The reference modulation scheme corresponding to the target integrated state data is output as the target modulation scheme.

9. The method for monitoring status data of electronic medical devices according to claim 1, characterized in that, The method further includes: Obtain the prior control scheme of the forward prior implantable neural monitoring trigger node; the forward prior implantable neural monitoring trigger node is used to represent the previous round of implantable neural monitoring trigger node of the target implantable neural monitoring trigger node; Obtain the implantable neural monitoring termination node corresponding to the prior modulation scheme. If the implantable neural monitoring termination node corresponding to the prior control scheme is later than the target implantable neural monitoring trigger node, then the execution interval parameter between the target implantable neural monitoring trigger node and the implantable neural monitoring termination node corresponding to the prior control scheme is determined, and the scheme timeout parameter corresponding to the target control scheme is obtained. When the execution interval parameter is less than the scheme timeout parameter, the implantable neural monitoring termination node corresponding to the prior control scheme is output as the implantable neural monitoring trigger node of the target control scheme. If the implantable neural monitoring termination node corresponding to the prior modulation scheme precedes the target implantable neural monitoring trigger node, then the target implantable neural monitoring trigger node will be output as the implantable neural monitoring trigger node of the target modulation scheme.

10. A deep learning system, characterized in that, The deep learning system includes a processor and a machine-readable storage medium storing machine-executable instructions, which are loaded and executed by the processor to implement the status data monitoring method for electronic medical devices as described in any one of claims 19.

Images