A user mind intervention intelligent watch and system based on double heart medicine

Abstract

The present application relates to the field of health informatics, and discloses a user mental intervention intelligent watch and system based on double-heart medicine, which comprises a user concentration hierarchical evaluation module, a variable ratio reinforcement strategy library, a user input interface, an intervention logic and scheduling engine, a strategy migration engine and an acupoint electric stimulation control interface, the engine detects a flow state signal at a non-deterministic random trigger time, if the signal exists, the engine stops and generates a new random trigger time, the strategy migration engine monitors the number of occurrences of the stop, and automatically updates the evaluation state data in the evaluation module, the present application constructs a closed loop of information processing, avoids habituation of intervention, protects the user's concentration flow, and realizes long-period dynamic self-adaptation of intervention strategy.

Description

Technical Field

[0001] This invention relates to a smartwatch and system for user mental intervention based on dual-heart medicine, belonging to the field of healthcare informatics technology. Background Technology

[0002] Currently, dual-heart medicine is a discipline system proposed based on interdisciplinary research in cardiology and psychology. This discipline focuses on the interaction between cardiovascular disease and mental and psychological state, improving patients' physical heart disease while also emphasizing the intervention of psychological problems to enhance treatment effectiveness and patient prognosis, and achieve comprehensive physical and mental rehabilitation. With the introduction of the dual-heart medical concept, cognitive health training has gradually shifted from single brain function regulation to synergistic intervention of the brain and heart. Dual-heart emphasizes the bidirectional regulatory relationship between brain activity and physiological signals such as heart rate variability and skin conductance response. By comprehensively considering psychological behavior and physiological feedback, dual-heart medicine achieves synchronous intervention of cognitive and physiological states during training processes such as attention maintenance.

[0003] In information processing to assist users in cognitive health training or improve concentration, a common technical approach is to provide time block management logic based on deterministic timing. For example, the system executes a fixed-duration focus countdown and sends a rest instruction to the user when the countdown ends. This information processing method is widely used because of its simple logic and ease of implementation on portable devices. However, when this information processing method is applied to cognitive behavioral training tasks, the determinism of the information scheduling logic leads to a diminishing of the intervention effect in principle. A fixed-cycle, completely predictable intervention signal will quickly become habitual for the user, leading to a decrease in the intervention signal and a gradual weakening of the system's information intervention function.

[0004] To avoid the aforementioned limitations, real-time analysis of a user's heart rate variability or skin conductance data can be used to determine when their concentration declines and trigger intervention at that moment. For example, Chinese utility model patent CN204813868U discloses a smartwatch that integrates ECG, blood oxygen, and blood pressure, which integrates multiple sensors such as ECG and blood oxygen into the watch casing to achieve composite collection of physiological parameters. However, this type of technology is a passive collection of physiological data and does not involve how to utilize this data or build effective active intervention information processing logic that can avoid the aforementioned habituation problem without relying on this complex data.

[0005] Therefore, the technical problem to be solved by this invention is how to design an information processing system that can technically correlate a user's personalized assessment status with a non-deterministic intervention scheduling sequence without relying on complex real-time physiological monitoring. Summary of the Invention

[0006] This invention provides a smartwatch and system for user mental intervention based on dual-heart medicine. Its main purpose is to solve the problem of how to technically associate the user's personalized assessment status with a non-deterministic intervention scheduling sequence without relying on complex real-time physiological monitoring, so as to avoid the problem of weakened intervention effect due to the predictability of intervention logic.

[0007] To achieve the above objectives, this invention provides a user mental intervention system based on dual-heart medicine. The system includes a user attention level assessment module, a variable ratio reinforcement strategy library, a user input interface, an intervention logic and scheduling engine, a strategy transfer engine, and an acupoint electrical stimulation control interface.

[0008] The user attention level assessment module is used to store assessment status data that characterizes the degree of poor user attention.

[0009] A variable ratio reinforcement strategy library is used to store the set of intervention parameters corresponding to the evaluation state data. The set of intervention parameters at least defines the random intervention time interval.

[0010] The user input interface is used to receive user-initiated flow state signals that represent the current state of focused flow.

[0011] The intervention logic and scheduling engine connects the user attention level assessment module, the variable ratio reinforcement strategy library, and the user input interface. It is used to retrieve and load the corresponding intervention parameter set from the variable ratio reinforcement strategy library based on the assessment status data; generate non-deterministic random trigger times within the random intervention time interval; detect the existence of a flow state signal when the random trigger time arrives; generate an intervention trigger command if no flow state signal exists; and stop generating the current intervention trigger command and immediately return to regenerate a new random trigger time different from the random trigger time within the random intervention time interval if a flow state signal exists.

[0012] The strategy migration engine connects the intervention logic and scheduling engine with the user attention level assessment module. It is used to monitor and count the number of times the intervention logic and scheduling engine stops generating intervention trigger commands when a random trigger time arrives and a flow state signal exists. When the number of occurrences exceeds the preset migration threshold, the assessment status data in the user attention level assessment module is automatically updated, and the data is migrated to the assessment status corresponding to a higher attention level.

[0013] The acupoint electrical stimulation control interface connects the intervention logic and the scheduling engine, and is used to receive intervention trigger commands and output control signals.

[0014] Preferably, the system further includes: a physiological baseline data interface for acquiring the user's real-time physiological baseline parameters; the intervention logic and scheduling engine is also used to: compare the real-time physiological baseline parameters with the stored nominal baseline parameters before retrieving and loading the corresponding intervention parameter set from the variable ratio reinforcement strategy library based on the assessment state data; if the real-time physiological baseline parameters do not exceed a preset stress threshold, then continue to execute the logic of retrieving and loading the corresponding intervention parameter set and the subsequent generation of non-deterministic random trigger times and processing flow state signals; if the real-time physiological baseline parameters exceed the preset stress threshold, then stop executing the logic of retrieving and loading the corresponding intervention parameter set and the subsequent generation of non-deterministic random trigger times and processing flow state signals, and reset to generate an instruction for initiating a preset relaxation intervention.

[0015] Preferably, the intervention logic and scheduling engine are also used to: automatically return to regenerate a non-deterministic random trigger time within the random intervention time interval after generating the intervention trigger command.

[0016] Preferably, the variable ratio reinforcement strategy library stores at least two sets of intervention parameters corresponding to different assessment state data; the random intervention time interval defined by the first set of parameters is different from the range of the random intervention time interval defined by the second set of parameters.

[0017] Preferably, the intervention parameter set also defines the total focus cycle duration; the intervention logic and scheduling engine are also used to: after the total focus cycle duration ends, pause the execution of the logic loop of generating indeterminate random trigger times, detecting flow state signals, and generating or terminating intervention trigger commands based on the detection results at random intervention time intervals; generate rest intervention commands and start preset rest cycle electrical stimulation through the acupoint electrical stimulation control interface.

[0018] Preferably, the intervention logic and scheduling engine is used to invoke the pseudo-random number generator when a non-deterministic random trigger moment occurs within the random intervention time interval; the strategy transfer engine is used to make judgments based on heuristic rules when automatically updating the evaluation status data in the user attention level assessment module, wherein when the number of occurrences... With the preset migration threshold satisfy At that time, automatic updates are performed, where The cumulative number of times the intervention logic and scheduling engine stop generating intervention trigger commands when a random trigger time arrives and a flow state signal exists is determined. This is a preset integer threshold determined based on the evaluation status data.

[0019] Preferably, the assessment status data stored in the user attention level assessment module is the result of quantifying the degree of poor user attention based on standardized assessment tools. The assessment status data is stored as one of three logical states: severe poor, moderate poor, and mild poor.

[0020] Preferably, the system further includes: a strategy cruise engine, which is connected to the user attention level assessment module and is configured with an interactive interface; the strategy cruise engine is used to: actively obtain the user's confirmation response regarding fatigue status through the interactive interface before the end of the rest cycle electrical stimulation and the start of the next total attention cycle; statistically analyze the response preference within the preset assessment cycle; and automatically update the assessment status data in the user attention level assessment module when the response preference meets the preset migration conditions.

[0021] Preferably, the system further includes: an entropy pool collector for extracting timing jitter or sensor data noise bits from the accompanying information stream of at least one hardware component of the system to generate a high entropy value; the intervention logic and scheduling engine is also used to: acquire the high entropy value before the non-deterministic random triggering moment within the random intervention time interval, and inject it as a seed into the pseudo-random number generator; the acupoint electrostimulation control interface is a hardware abstraction layer, which is used to: receive abstract logic instructions from the intervention logic and scheduling engine; and translate the abstract logic instructions into specific electrical signals or bus commands required by the hardware module.

[0022] A smartwatch for user mental intervention based on dual-heart medicine, comprising a processor, memory, acupoint electrical stimulation hardware module, and user input interface:

[0023] The processor, memory, acupoint electrical stimulation hardware module, and user input interface are all integrated into the smartwatch casing.

[0024] Memory, used to store a user mental intervention system based on dual-heart medicine;

[0025] The processor connects to the memory, the acupoint electrical stimulation hardware module, and the user input interface. The processor is used to execute a user mental intervention system based on dual-heart medicine.

[0026] Compared with the prior art, the beneficial effects of the present invention are:

[0027] 1. Through the collaborative operation of the user attention level assessment module and the variable ratio reinforcement strategy library, the assessment status data representing the degree of poor user attention is technically linked to the random intervention time interval in the intervention parameter set. Based on this link, the intervention logic and scheduling engine generate non-deterministic trigger times within the specific time interval corresponding to the user's status. This information processing method utilizes the non-determinism of the trigger times to repeatedly stimulate neural feedback, promote the replay of brain neural signals, and avoid user habituation and behavior decline caused by traditional fixed-cycle interventions. By directly constraining the random intervals with the assessment status, the timing logic of intervention scheduling automatically adapts to the user's personalized health status, solving the technical limitations of a one-size-fits-all approach to intervention strategies and achieving personalized adaptation at the information scheduling logic level.

[0028] 2. By coordinating the user input interface with the intervention logic and scheduling engine, an information processing path is provided that balances the intervention principle with the user's immediate experience. When the system receives a signal indicating that the user is in a state of focused flow, the scheduling engine does not simply skip or pause the intervention. Instead, it suspends the currently generated random trigger moment and returns to regenerate a completely new, non-deterministic trigger moment that is different from the previous one. Logically, this achieves a dual effect: it immediately suspends intervention actions that may disrupt the user's current state of focus, and by immediately resetting the random number, it maintains the non-deterministic nature of the intervention scheduling, ensuring the continuation of the core training principle of variable ratio reinforcement, and resolving the conflict between random training and focus protection.

[0029] 3. By utilizing the information generated by the focus flow protection mechanism, a long-term adaptive information closed loop is constructed. The strategy migration engine monitors and counts the frequency of user-initiated input flow status signals, using interactive behavior as implicit data representing the improvement of the user's focus level. When the number of counts meets the preset migration conditions, the system automatically updates the evaluation status data in the user evaluation module. This design utilizes the byproduct information of users' daily interactions, enabling the system to dynamically adjust the set of basic intervention parameters called in the strategy library without the user's awareness. This upgrades the information flow from static personalization to dynamic adaptation, avoiding the problem of the original intervention strategy becoming rigid and ineffective due to the improvement of the user's status. Attached Figure Description

[0030] Figure 1 This is a diagram of the closed-loop information processing architecture of the mental intervention system of the present invention;

[0031] Figure 2 This is a diagram showing the correspondence between focus levels and intervention interval parameters in this invention;

[0032] Figure 3 This is a timing diagram of the multi-entity physiological data acquisition interaction of the system of the present invention. Detailed Implementation

[0033] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0034] This invention provides a smartwatch and system for user mental intervention based on dual-heart medicine. As an information processing system deployed in portable devices such as smartwatches, the system includes the following components at the information processing level: a user attention level assessment module for storing personalized assessment states; a variable ratio reinforcement strategy library for storing intervention timing parameters corresponding to the states; an intervention logic and scheduling engine for loading parameters from the strategy library based on the assessment states to generate non-deterministic intervention timing sequences; a user input interface for receiving real-time flow state signals from the user for arbitration intervention; a strategy migration engine for automatically updating the assessment states based on the user's interaction history; and an acupoint electrical stimulation control interface for converting the engine's logical instructions into hardware control signals. The user attention level assessment module receives and stores assessment state data characterizing the degree of poor user attention during system initialization or the user's first use. This assessment state data is based on standardized assessment tools, such as the Attention Network Test (ANT) or Consistency Performance Test (CPT), to quantify the degree of poor user attention. The system stores the evaluation result as a logical state in local memory. Specifically, it stores it as one of three logical states: severely poor, moderately poor, or mildly poor. This state data serves as the initial index for all subsequent information scheduling logic. The variable ratio reinforcement strategy library is a structured parameter table or database that stores intervention parameter sets corresponding to different evaluation state data in the user's attention stratification evaluation module. For example, the first parameter set corresponding to the severely poor state and the second parameter set corresponding to the moderately poor state have different defined key parameters. To support the implementation, the intervention parameter set at least defines a random intervention time interval. The random intervention time interval defined by the first parameter set, for example [4min, 6min], is different from the random intervention time interval defined by the second parameter set, for example [10min, 15min]. In other implementations, the intervention parameter set can further define the total attention cycle duration and rest cycle electrical stimulation parameters, etc.

[0035] Existing intervention logic based on deterministic timing is prone to user habituation, leading to diminishing intervention effectiveness. To address this issue, this invention employs a non-deterministic scheduling timing combined with user rejection in its intervention logic and scheduling engine information processing path. The intervention logic and scheduling engine connect the user attention level assessment module, the variable ratio reinforcement strategy library, and the user input interface. Based on assessment status data obtained from the assessment module (e.g., moderately poor), the engine retrieves and loads the corresponding intervention parameter set from the strategy library, such as a random intervention time interval of [10 min, 15 min]. This ensures that the subsequently generated random times have high quality. To mitigate unpredictability and prevent the pseudo-random sequence from being predicted, the system may also include an entropy pool collector. This collector extracts timing jitter or noise data from the least significant noise bits of data streams from system hardware components, such as microsecond timestamps from touchscreen interactions or sensor data (e.g., PPG photoelectric sensors), to generate a high entropy value. Then, before executing the next step, the intervention logic and scheduling engine acquires this high entropy value and injects it as a seed into the pseudo-random number generator. The engine calls this seeded pseudo-random number generator to generate a non-deterministic random trigger time within the loaded random intervention time interval [10 min, 15 min]. Taking 12 minutes and 43 seconds as an example; at the random trigger time... When the time reaches 12 minutes and 43 seconds, the engine immediately checks whether a flow state signal received from the user input interface exists. If no flow state signal exists, the engine generates an intervention trigger command and sends the command to the acupoint electrical stimulation control interface. If a flow state signal is detected, it indicates that the user is currently in deep focus and does not want to be interrupted, so the engine stops generating the current intervention trigger command. This stoppage is not a simple skip, but an immediate return to the step of obtaining a new entropy seed. Within the random intervention time interval [10 minutes, 15 minutes], a new, non-deterministic random trigger time different from the previous one (12 minutes and 43 seconds) is generated. This mechanism protects the user's focus flow while maintaining the non-deterministic nature of intervention scheduling by immediately resetting the random number. Correspondingly, after the absence of a flow state signal and the generation of an intervention trigger command, the intervention logic and the scheduling engine automatically return to regenerate the next non-deterministic random trigger time.

[0036] To address the issue of existing intervention strategies becoming ineffective after user focus has improved, this invention utilizes information generated by a focus flow protection mechanism to construct a long-term adaptive information closed loop. A strategy migration engine connects the intervention logic and scheduling engine with the user focus stratification assessment module. This strategy migration engine monitors and counts the cumulative number of times the intervention logic and scheduling engine stops generating intervention trigger commands when a random trigger time arrives and a flow state signal exists. The engine obtains a preset migration threshold corresponding to the current evaluation state data from the evaluation module. ,Should It is a preset integer determined based on the current state, such as It can be set to 15 times; the strategy migration engine determines the cumulative number of occurrences. When the preset migration conditions are met, the specific heuristic rules are as follows: At this point, the engine automatically updates the assessment status data in the user's attention level assessment module to migrate to an assessment status corresponding to a higher attention level, such as from moderately poor to slightly poor. Utilizing byproduct information from the user's daily interactions, namely rejection behavior, as implicit data representing the improvement in the user's attention level, the system can dynamically adjust the set of basic intervention parameters subsequently called from the strategy library without the user's awareness. Upon the next startup, it will automatically load the sparser [20min, 30min] interval corresponding to slightly poor, achieving an information flow upgrade from static personalization to dynamic adaptation. In a preferred embodiment, to establish an arbitration gating mechanism for information processing to ensure that intervention measures match the user's immediate physiological context, the system may further include: physiological baseline data reception... The system is used to acquire the user's real-time physiological baseline parameters, such as the real-time resting heart rate obtained through a PPG photoelectric sensor. Before retrieving and loading the corresponding intervention parameter set based on the assessment status data, the intervention logic and scheduling engine execute a pre-arbitration logic: comparing the real-time physiological baseline parameters with the stored nominal baseline parameters, such as the user's historical average resting heart rate; if the real-time physiological baseline parameters do not exceed the preset stress threshold, such as nominal baseline + 2 standard deviations, the state is deemed suitable, and the subsequent variable ratio reinforcement scheduling logic continues; if the real-time physiological baseline parameters exceed the preset stress threshold, it indicates that the user's current physiological state is not suitable for focused training, and the engine will terminate the original variable ratio reinforcement scheduling steps and reset the information flow to generate instructions for initiating preset relaxation interventions, such as guided breathing.

[0037] In other implementations, to construct a structured training and rest process, the set of intervention parameters in the variable ratio reinforcement strategy library can also define the total focus cycle duration, to... =85 minutes as an example; the intervention logic and scheduling engine are also used to: after the total focus cycle duration (85 minutes) ends, pause the execution of the logic loop of generating random trigger times, detecting flow state signals, and generating or terminating intervention trigger commands within the random intervention time interval; and generate rest intervention commands, and start the preset rest cycle electrical stimulation through the acupoint electrical stimulation control interface; as an alternative or supplementary adaptive mechanism to the strategy migration engine, the system may also include: a strategy cruise engine, connected to the user's focus level assessment module and configured with an interactive interface; the strategy cruise engine is used to: before the end of the rest cycle electrical stimulation and the start of the next total focus cycle duration, actively obtain the user's confirmation response regarding fatigue status through the interactive interface, such as whether they feel fatigued (yes / no); the engine then statistically analyzes the response preference in the preset assessment period, such as 7 days; when the response preference, such as the proportion of no response, meets the preset migration conditions, such as no > 80%, the strategy cruise engine automatically updates the assessment status data in the user's focus level assessment module to achieve another form of long-term dynamic adaptation; the core protected by this invention lies in the information processing architecture and scheduling logic, acupoint electrical stimulation The control interface functions as a Hardware Abstraction Layer (HAL) in the system architecture; it receives abstract logic instructions from the intervention logic and scheduling engine, such as initiating short-term intervention or initiating rest intervention, and translates these abstract logic instructions into specific electrical signals or bus commands required by downstream hardware modules, decoupling the information processing logic of healthcare informatics from the hardware implementation of A61N (physical stimulation). This invention also provides a smartwatch for user mental intervention based on dual-heart medicine, used to implement any of the above-described embodiments of the system. The smartwatch's hardware structure includes: a processor, a memory, an acupoint electrical stimulation hardware module, and a user input interface integrated into the smartwatch casing. The memory stores all logic modules and data of a user mental intervention system based on dual-heart medicine, such as an evaluation module, a strategy library, a scheduling engine, and a migration engine. The processor connects to the memory, the acupoint electrical stimulation hardware module, and the user input interface. The processor executes instructions from the memory to implement all information processing functions of the system, including loading parameters, generating random times, detecting flow state signals, counting rejections, updating evaluation states, and driving the hardware module to output control signals through the control interface.

[0038] Example 1: In a specific application scenario, when a user whose initial assessment status is moderately poor starts the system of this invention, the intervention logic and scheduling engine, during initialization, obtains the user's real-time resting heart rate through the physiological baseline data interface and compares it with the stored nominal baseline parameters. If the real-time physiological baseline parameters do not exceed the preset stress threshold, the arbitration logic allows the process to proceed. The engine then continues to retrieve and load the corresponding intervention parameter set from the variable ratio reinforcement strategy library based on the moderately poor state. This parameter set defines a random intervention time interval of [10 min, 15 min]. During system operation, the intervention logic and scheduling engine randomly... Within the intervention time interval, a pseudo-random number generator, whose seed has been injected with a high entropy value through the entropy pool collector, is invoked to generate a non-deterministic random trigger time, such as 12 minutes and 43 seconds. When this time arrives, the user is in a highly focused state and immediately inputs a flow state signal through the user input interface. The scheduling engine detects this flow state signal, therefore suspends the generation of this intervention trigger command and immediately returns to regenerate a non-deterministic random trigger time different from 12 minutes and 43 seconds within the same interval. This information processing path, by suspending and immediately resetting the random number, protects the user's immediate focused experience without destroying the temporal non-determinism required for the variable ratio reinforcement.

[0039] During this process, the strategy migration engine operates synchronously, monitoring and counting the number of times intervention trigger commands are generated during this interruption in the background. After setting a preset evaluation period of 7 days, the policy migration engine determines the cumulative number of occurrences for the user. The preset migration threshold of 15 times for binding with the moderately poor state has been exceeded. This judgment indicates that the frequency with which users can maintain deep focus is increasing; the strategy transfer engine automatically updates the assessment status data in the user's focus stratification assessment module, migrating from moderately poor to slightly poor; when the user restarts the system, the intervention logic and scheduling engine, after arbitration through physiological baseline, automatically retrieves and loads the intervention parameter set corresponding to slightly poor, and the random intervention time interval is adjusted accordingly to a range of [20min, 30min], becoming more sparse; the system utilizes the byproduct information generated by the user's focus flow during short-term interactions. This enables a closed-loop information processing mechanism that automatically updates the status over long periods, allowing the timing logic of intervention scheduling to automatically and dynamically match the user's improved focus ability, thus solving the problem of user churn caused by rigid strategies in information intervention systems.

[0040] Example 2: This example constructs a computation-based simulation test environment to compare the state changes of the information processing system of this invention, an existing fixed-cycle intervention system, and a single variable-ratio intervention system during long-term operation. The test environment is constructed as a software platform simulating user attention behavior responses. The platform incorporates a simulated user state model, which includes two core parameters: a habituation index and a focus level. The habituation index quantifies the degree of decrease in the simulated user's sensitivity to predictable intervention signals; its value increases with the increase in predictable signals. The focus level characterizes the simulated user's concentration ability; its value slowly increases with effective, non-deterministic intervention. The experiment sets up three test groups, running in the simulation environment for 30 days: a control group... A is used to simulate existing technologies, employing a fixed-cycle intervention logic with 25 minutes of focus followed by 5 minutes of rest, ensuring the intervention signal is fully predictable. Control group B, serving as a partially missing control group, utilizes the variable-ratio reinforcement strategy library and intervention logic and scheduling engine of this invention, without enabling the strategy transfer engine. The system loads a random intervention time interval of [10 min, 15 min] based on the simulated user's initial moderately poor state, maintaining this interval parameter unchanged for 30 days. The sample group of this invention uses the complete system of this invention, including a user focus level assessment module, a variable-ratio reinforcement strategy library, a user input interface, intervention logic and scheduling engine, and a strategy transfer engine. The simulated user's initial state is also set to moderately poor, with a preset transfer threshold. The simulation is set to trigger a flow state signal 15 times. When the simulated user's focus level increases, the signal will be automatically triggered to simulate the behavior of real users protecting deep focus. During the simulation, the habituation index of each group of simulated users is recorded. For the sample group of this invention, the cumulative number of occurrences of the flow state signal is recorded. The system policy status after the policy migration engine is executed is shown in Table 1.

[0041] Table 1: Comparison of Simulated User States under Different Information Scheduling Strategies

[0042]

[0043] As shown in Table 1, the habituation index of control group A (fixed period) increased to 78.6% on day 30, indicating that its intervention signal had essentially failed. The habituation indices of both control group B (fixed variable ratio) and the present invention's sample group remained low on day 5, at 3.1% and 3.3% respectively, indicating that the non-deterministic timing of variable ratio reinforcement addressed the habituation problem caused by predictability. However, over a long period, control group B triggered simulated flow state signals 21 times on day 30, indicating that its fixed [10-15 min] interval was too frequent for the improved simulated users, resulting in policy mismatch. In contrast, the collaborative mechanism of the data display system in the present invention's sample group detected the cumulative number of flow state signals by the policy migration engine on day 18. Reaching 16 times, exceeding the preset migration threshold of 15 times. The system automatically updates the evaluation status data and shifts the operating strategy from the moderate to the mild [20-30 min] interval. Finally, on the 30th day, the habituation index of the sample group of this invention remained at a low level of 4.2%, and the operating strategy was matched with the mild state of the simulated user.

[0044] Example 3: This example combines Figures 1 to 3 A description of a smartwatch and system for user mental intervention based on dual-heart medicine, such as... Figure 1 As shown, the intervention logic and scheduling engine, as the core processing unit, loads assessment status data from the user's attention level assessment module, retrieves the intervention parameter set from the variable ratio reinforcement strategy library, receives real-time physiological baseline parameters from the physiological baseline data interface to perform stress threshold checks, and obtains high-entropy values ​​from the entropy pool collector as seeds for the pseudo-random number generator. After performing nondeterministic scheduling and arbitration, the engine stops if it detects a flow state signal input from the user input interface; otherwise, it generates an intervention trigger command to the acupoint electrical stimulation control interface or resets to initiate a preset relaxation intervention when the real-time physiological baseline parameters exceed the stress threshold. The strategy transfer engine monitors the number of times the flow state of the intervention logic and scheduling engine is stopped. Under the condition that > The evaluation status in the user's attention level assessment module is automatically updated at that time.

[0045] like Figure 2As shown in the figure, this bar chart displays the relationship between the intervention time interval (min) as the vertical axis and the attention level as the horizontal axis. The chart shows three different attention levels: severely poor, moderately poor, and mildly poor, each corresponding to a different lower and upper limit (min) of the intervention time interval. As the attention level progresses from severely poor to mildly poor, both the lower and upper limits of the corresponding random intervention time interval increase. The interval range correspondingly shifts from shorter intervals (e.g., 4 min, 6 min) to moderate intervals (e.g., 10 min, 15 min), and finally to longer intervals (e.g., 20 min, 30 min). Figure 3 As shown, the process begins with the user wearing the smartwatch and starting monitoring. The smartwatch then activates its sensor module to collect data from multiple sensors. During continuous monitoring, the sensor module collects raw physiological data, including heart rate, blood pressure, blood oxygen saturation, skin conductance, and body temperature. The data is then transmitted to the data processing unit. The data processing unit performs data filtering and preprocessing, feature extraction, and outlier detection. After that, the processed data is returned to the smartwatch to display the real-time monitoring results to the user, and the data packets are uploaded to the cloud server for data storage and backup. The smartwatch also receives confirmation signals from the cloud server.

[0046] Example 4: This example provides a standardized engineering calibration procedure for determining the specific values ​​of each intervention parameter set in the variable ratio reinforcement strategy library and the preset migration threshold in the strategy migration engine. This procedure is executed offline before system deployment to ensure that the initial parameters of the information processing logic have an objective basis. It addresses two technical problems: first, determining the correlation between the three assessment state data (severely poor, moderately poor, and mildly poor) and statistically valid random intervention time intervals; second, determining the preset migration threshold used to trigger automatic updates of the assessment state. The specific values; the first stage of the calibration process is used to determine the random intervention time interval; based on the standardized assessment tool used by the user attention stratification assessment module, in this embodiment, continuous performance testing (CPT) is used to recruit and screen representative user queues corresponding to the three states of severe poor, moderate poor, and mild poor, respectively; taking the moderate poor queue as an example, standard attention tasks are performed in a controlled environment, and the system runs an intervention logic version with adjustable parameters, allowing the back-end operator to dynamically adjust the upper and lower limits of the random intervention time interval. In the initial setting of a relatively wide interval, for example, [5min, 25min], through multiple test cycles, combined with the subjective feedback data collected after the cycle regarding the timing of intervention being too frequent or too infrequent, this interval is iteratively converged; when this interval is converged to a range that is statistically reported as unexpected and helpful by most moderate poor queue users, i.e., [10min, 15min], the interval [10min, 15min] is determined as the intervention parameter associated with the moderate poor state to be stored in the variable ratio reinforcement strategy library.

[0047] The second stage of the calibration process is used to determine the preset migration threshold. Taking the aforementioned moderately poor cohort as an example, the long-term tracking test, set at 14 days, continued, running according to the random intervention time intervals [10min, 15min] corresponding to the moderately poor state determined in the first phase. Users were instructed to actively input a flow status signal through the user input interface when they perceived an intervention trigger but considered it unnecessary or an interruption during the focus process. The system then executed the logic of terminating and regenerating a random time interval, with the background strategy migration engine accumulating the number of such terminations. At the end of the 14-day period, the CPT test was performed on the queue of users to identify a subset of users whose assessment status data had objectively improved to slightly poor; the data of this user subset over the 14 days were retrieved. The statistical value, calculated as the mean or median, is determined 15 times. These 15 values ​​are used to determine the preset migration threshold corresponding to the migration from moderately poor to slightly poor, which is stored in the policy migration engine. .

[0048] Example 5: Upon initial system activation, the intervention logic and scheduling engine automatically trigger a standardized pre-calibration procedure to establish the nominal baseline parameters in the physiological baseline data interface. The engine guides the user into a resting state through the system's interactive interface, maintaining this state for a preset calibration duration (300 seconds in this example). Within this calibration duration, the physiological baseline data interface uses a photoplethysmography sensor to continuously collect the user's photoelectric signals at a sampling frequency of 100Hz, calculating instantaneous resting heart rate data points. After the calibration duration ends, the engine processes the data point sequence, removing outliers exceeding a preset standard deviation, and calculates the average value of the effective sequence (72 times / minute in this example). This average value is stored by the user's attention stratification assessment module and used as the nominal baseline parameter for subsequent arbitration decisions by the user. Information exchange between the intervention logic and scheduling engine and the acupoint electrical stimulation control interface is achieved through structured data packets. Intervention trigger commands, rest intervention commands, or relaxation intervention commands generated by the scheduling engine are all manifested in the data packet containing at least an instruction identifier to distinguish the command type. 0x01 represents variable ratio triggering, 0x02 represents rest intervention, and 0x03 represents relaxation intervention; and a duration parameter for defining the intervention duration, in milliseconds; the data packet also contains a logic level parameter for mapping intensity, which can be one of three levels: 1, 2, or 3; after receiving this structured data packet, the acupoint electrical stimulation control interface translates the abstract parameters in the data packet, taking an instruction with a duration of 10000ms and a logic level of 3 as an example, into bus commands for driving the specific electrical pulse width, frequency, or amplitude required by the downstream hardware module.

[0049] Example 6: This example discloses the specific implementation procedure of the entropy pool collector at the information processing level. It extracts and mixes noise data from multiple system-accompanying information streams to generate a high entropy value. The entropy pool collector maintains an internal fixed-width entropy pool register variable in the system background, which is set with an initial value during system initialization. When the system captures an event from a preset noise source, such as the least significant bit of a timestamp of a touchscreen interaction or the lowest two noise bits of a PPG sensor raw data reading, the entropy pool collector extracts the noise data of the event, performs a cyclic shift bit operation on the current entropy pool register variable and the noise data, and writes the operation result back to the entropy pool register variable. By continuously and asynchronously folding and mixing timing jitter or noise data from different hardware components into the entropy pool register variable in this way, the state undergoes unpredictable changes. When the intervention logic and scheduling engine need to generate random trigger moments and call the pseudo-random number generator, the current instantaneous value of the entropy pool register is extracted from the entropy pool collector and injected as a seed into the pseudo-random number generator.

[0050] This embodiment also provides a long-cycle adaptive recalibration procedure for nominal baseline parameters to address slow changes in the user's physiological state. The intervention logic and scheduling engine automatically perform baseline data sampling once every 24 hours in a preset background maintenance cycle. The sampling constraint is triggered only when the user is in a non-training state and the physiological baseline data interface detects that the user's physiological parameters, such as heart rate and activity level, are below the resting threshold for a continuous period of time. The specific triggering condition can be that this state is detected during the user's sleep at night. The engine collects the real-time physiological baseline parameters during this period and stores them as new data points in a historical baseline database with a sliding time window, such as 7 days. The engine recalculates the average value of all valid data points in the database and uses this new calculation to obtain an average value that reflects the user's recent physiological average value, automatically updating the nominal baseline parameter stored in the user's attention stratification assessment module.

[0051] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0052] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A user mental health intervention system based on dual-heart medicine, characterized in that, The system includes a user attention level assessment module, a variable ratio enhancement strategy library, a user input interface, an intervention logic and scheduling engine, a strategy transfer engine, and an acupoint electrical stimulation control interface. The user attention level assessment module is used to store assessment status data that characterizes the degree of poor user attention. The variable ratio reinforcement strategy library is used to store the set of intervention parameters corresponding to the evaluation state data. The set of intervention parameters at least defines the random intervention time interval. The user input interface is used to receive user-initiated flow state signals that indicate the current state of focused flow; The intervention logic and scheduling engine connect the user attention level assessment module, the variable ratio reinforcement strategy library, and the user input interface. It is used to retrieve and load the corresponding intervention parameter set from the variable ratio reinforcement strategy library based on the assessment status data; generate an indeterminate random trigger time within the random intervention time interval; detect the existence of a flow state signal when the random trigger time arrives; generate an intervention trigger command if no flow state signal exists; and stop generating the current intervention trigger command and immediately return to regenerate a new random trigger time different from the random trigger time within the random intervention time interval if a flow state signal exists. The strategy migration engine connects the intervention logic and scheduling engine with the user attention level assessment module. It is used to monitor and count the number of times the intervention logic and scheduling engine stops generating intervention trigger commands when a random trigger time arrives and a flow state signal exists. When the number of occurrences exceeds the preset migration threshold, the evaluation status data in the user's attention level assessment module is automatically updated, and the data is migrated to the evaluation status corresponding to a higher attention level. The acupoint electrical stimulation control interface connects the intervention logic and the scheduling engine, and is used to receive intervention trigger commands and output control signals. Specifically, the intervention logic and scheduling engine is used to invoke the pseudo-random number generator when a non-deterministic random trigger moment occurs within a random intervention time interval; the strategy transfer engine is used to make judgments based on heuristic rules when automatically updating the evaluation status data in the user attention level assessment module, where the number of occurrences... With the preset migration threshold satisfy At that time, automatic updates are performed, where The cumulative number of times the intervention logic and scheduling engine stop generating intervention trigger commands when a random trigger time arrives and a flow state signal exists is determined. This is a preset integer threshold determined based on the evaluation status data; The system includes: an entropy pool collector for extracting timing jitter or sensor data noise bits from the accompanying information stream of at least one hardware component of the system to generate a high entropy value; an intervention logic and scheduling engine for: acquiring the high entropy value before the non-deterministic random triggering moment within the random intervention time interval, and injecting it as a seed into a pseudo-random number generator; and an acupoint electrostimulation control interface, which is a hardware abstraction layer for: receiving abstract logic instructions from the intervention logic and scheduling engine; and translating the abstract logic instructions into specific electrical signals or bus commands required by the hardware modules. The system also includes: a physiological baseline data interface for acquiring the user's real-time physiological baseline parameters; and an intervention logic and scheduling engine for: comparing the real-time physiological baseline parameters with the stored nominal baseline parameters before retrieving and loading the corresponding intervention parameter set from the variable ratio reinforcement strategy library based on the assessment status data; if the real-time physiological baseline parameters do not exceed a preset stress threshold, then the logic for retrieving and loading the corresponding intervention parameter set and subsequently generating non-deterministic random trigger times and processing flow state signals continues; if the real-time physiological baseline parameters exceed the preset stress threshold, then the logic for retrieving and loading the corresponding intervention parameter set and subsequently generating non-deterministic random trigger times and processing flow state signals is stopped, and the system is reset to generate an instruction to initiate a preset relaxation intervention.

2. The user mental health intervention system based on dual-heart medicine according to claim 1, characterized in that, The intervention logic and scheduling engine are also used to: automatically return to regenerate a non-deterministic random trigger time within the random intervention time interval after generating the intervention trigger command.

3. The user mental health intervention system based on dual-heart medicine according to claim 1, characterized in that, The variable ratio reinforcement strategy library stores at least two sets of intervention parameters corresponding to different assessment state data; the random intervention time interval defined by the first set of parameters is different from the range of the random intervention time interval defined by the second set of parameters.

4. A user mental health intervention system based on dual-heart medicine according to claim 1, characterized in that, The intervention parameter set also defines the total focus cycle duration; the intervention logic and scheduling engine are also used to: pause the execution of the logic loop of generating indeterminate random trigger times, detecting flow state signals, and generating or terminating intervention trigger commands based on the detection results after the total focus cycle duration ends; generate rest intervention commands and start preset rest cycle electrical stimulation through the acupoint electrical stimulation control interface.

5. A user mental health intervention system based on dual-heart medicine according to claim 1, characterized in that, The assessment status data stored in the user attention level assessment module is the result of quantifying the degree of poor user attention based on standardized assessment tools. The assessment status data is stored as one of three logical states: severe poor, moderate poor, and mild poor.

6. A user mental health intervention system based on dual-heart medicine according to claim 4, characterized in that, The system also includes: a strategy cruise engine, which connects to the user's attention level assessment module and is configured with an interactive interface; the strategy cruise engine is used to: actively obtain the user's confirmation response regarding fatigue status through the interactive interface before the end of the rest cycle electrical stimulation and the start of the next total attention cycle; statistically analyze the response preference within the preset assessment cycle; and automatically update the assessment status data in the user's attention level assessment module when the response preference meets the preset migration conditions.

7. A smartwatch for user mental intervention based on dual-psychology medicine, used to implement the user mental intervention system based on dual-psychology medicine as described in claim 1, characterized in that, A smartwatch includes a processor, memory, an acupoint electrical stimulation hardware module, and a user input interface. The processor, memory, acupoint electrical stimulation hardware module, and user input interface are all integrated into the smartwatch casing. Memory, used to store a user mental intervention system based on dual-heart medicine; The processor connects to the memory, the acupoint electrical stimulation hardware module, and the user input interface. The processor is used to execute a user mental intervention system based on dual-heart medicine.

Images