Dual-channel closed-loop emotion regulation method, device and equipment based on state awareness and timing pairing

Abstract

The application relates to a double-channel closed-loop emotion regulation method, device and equipment based on state sensing and timing pairing. The method comprises the following steps: performing wake-up state discrimination according to self-evaluation data of an emotional visual analog scale, heart rate variability data and task interaction response data, generating a real-time wake-up state discrimination result, performing timing pairing of physiological regulation and cognitive intervention according to the real-time wake-up state discrimination result, generating an electric stimulation mode switching instruction and a training task calling instruction, performing successful event identification according to execution completion data of the training task calling instruction and real-time changes of the heart rate variability data, and generating a reward electric stimulation pulse emission instruction. The application drives physiological-cognitive timing pairing based on a wake-up state, realizes closed-loop self-adaption, identifies and emits differentiated reward stimulation through a behavior and physiological double channel, establishes conditioned reflex reinforcement, anchors haptics through respiratory coupling modulation stimulation, and promotes skill generalization and migration.

Description

Technical Field

[0001] This application relates to the field of medical information processing technology, and in particular to a dual-pathway closed-loop emotion regulation method, device and equipment based on state perception and temporal pairing. Background Technology

[0002] In recent years, the field of mood disorder intervention technology has shown a dual development trend of physiological regulation and psychological training. Peripheral neuromodulation technology, through transcutaneous electrical stimulation of the median nerve or vagus nerve, projects to the amygdala and prefrontal cortex via the ascending reticular activating system, effectively influencing heart rate variability and inducing relaxation responses. Digital psychological intervention technologies, represented by cognitive behavioral therapy and mindfulness training, have been widely popularized through mobile applications, providing structured psychological training content and enabling 24 / 7 access. However, existing technologies face a common bottleneck when integrating these two types of interventions: the system cannot dynamically determine the execution sequence of physiological regulation and cognitive intervention based on the user's real-time arousal state. Current wearable electrical stimulation devices generally adopt an open-loop operating mode, with stimulation parameters fixed at the factory or manually switched by the user from a limited set of preset modes. The system lacks the ability to sense changes in the state of the intervention subject, let alone adjust the intervention strategy based on state feedback. Existing digital therapy apps mainly output content in a one-way manner, lacking closed-loop perception of the user's real-time physiological state. When a user is in a state of high arousal anxiety, the system fails to recognize this state and implement physiological downgrading interventions, resulting in cognitive training being difficult to engage in effectively due to the inhibition of prefrontal cortex function. Similarly, when a user is in a state of low arousal lag, the system fails to recognize this state and implement physiological activation interventions, resulting in behavioral tasks being unable to be initiated due to psychomotor retardation. This open-loop architecture of "lack of state perception - inappropriate intervention timing" keeps physiological regulation and psychological training in a state of disordered juxtaposition, rather than forming a closed-loop coordination of "homeostasis first, remodeling later" based on neurophysiological laws.

[0003] Therefore, how to accurately determine the real-time wake-up state of users based on multimodal data, and how to enforce the temporal pairing of physiological regulation and cognitive intervention based on the determination results, has become a technical problem that needs to be solved in this field. Summary of the Invention

[0004] Based on this, this application provides a dual-pathway closed-loop emotion regulation method based on state perception and temporal pairing, characterized by comprising: Based on the self-report data of the Emotion Visual Analogue Scale, heart rate variability data, and task interaction responsiveness data, arousal state discrimination is performed, and real-time arousal state judgment results are generated. Based on the real-time wake-up state determination result, the timing pairing of physiological regulation and cognitive intervention is performed to generate electrical stimulation mode switching instructions and training task invocation instructions; Based on the execution completion data of the training task invocation instruction and the real-time changes in the heart rate variability data, the successful execution event is identified, and a reward electrical stimulation pulse delivery instruction is generated.

[0005] Optionally, the step of performing arousal state discrimination based on the self-report data of the Emotion Visual Analogue Scale, heart rate variability data, and task interaction responsiveness data, and generating a real-time arousal state determination result, includes: Based on the sliding operation data received from the self-assessment interface of the Visual Analogue Scale (VAS), the self-assessment score is parsed to generate the anxiety score and depression score of the VAS. Based on the photoplethysmography pulse wave signal, perform time-frequency domain analysis of heart rate variability to generate the ratio of low-frequency power to high-frequency power and the total power value; Based on the anxiety score of the Visual Analogue Scale (VAS), the depression score of the VAS, the ratio of low-frequency power to high-frequency power, the total power value, the response delay time and key pressure sampling value in task interaction, the results are compared with preset high wake-up threshold conditions and preset low wake-up threshold conditions to generate a real-time wake-up status determination result.

[0006] Optionally, the step of performing a time-series pairing of physiological regulation and cognitive intervention based on the real-time arousal state determination result, and generating electrical stimulation mode switching instructions and training task invocation instructions, includes: Based on the real-time wake-up status determination result of a high-wake state, a preliminary physiological downgrade intervention is performed, generating a soothing electrical stimulation mode activation command and an abdominal breathing training interface call command, performing a physiological downgrade attainment determination, and generating a cognitive reconstruction task call command when the physiological downgrade attainment determination is passed. If the real-time wake-up status is determined to be low-wake state, a preliminary physiological activation intervention is performed, generating an activation-type electrical stimulation mode start command and an activation progress visualization interface call command. A physiological activation target assessment is performed, and a behavioral activation task call command is generated when the physiological activation target assessment is passed.

[0007] Optionally, the step of recognizing successful execution events and generating reward-based electrical stimulation pulse delivery instructions based on the execution completion data of the training task invocation instruction and the real-time changes in the heart rate variability data includes: Based on the task completion flag data generated after the training task call instruction is executed, the successful execution of the behavior task is determined, and a successful behavior recognition result is generated. Based on the root mean square variation amplitude of the difference between adjacent normal sinus beats within a continuous time window extracted from the photoplethysmography pulse wave signal and the ratio of low-frequency power to high-frequency power, physiological stability determination is performed to generate physiological stability recognition results. Based on the successful recognition result of the behavior or the physiological stability recognition result meeting the preset triggering conditions, the reward stimulus parameter configuration is executed and issued, generating the reward electrical stimulation pulse issuance instruction.

[0008] Optionally, the step of comparing the anxiety score, depression score, low-frequency power to high-frequency power ratio, total power value, response delay duration and key pressure sampling value in task interaction with preset high wake-up threshold conditions and preset low wake-up threshold conditions to generate a real-time wake-up state determination result includes: If the anxiety score of the Visual Analogue Scale is greater than or equal to the preset high arousal anxiety threshold, or the ratio of low-frequency power to high-frequency power is greater than the preset sympathetic excitation threshold and the duration exceeds the preset high arousal duration threshold, or the response delay duration is less than the preset high arousal response delay lower limit and the key pressure sampling value is greater than the preset high arousal pressure upper limit, a high arousal state is marked and a high arousal judgment result is generated. If the depression score on the Visual Analogue Scale is greater than or equal to a preset low-arousal depression threshold, or the total power value is less than a preset lower limit threshold for total autonomic power and the duration exceeds a preset low-arousal duration threshold, or the response delay duration is greater than a preset upper limit threshold for low-arousal response delay, a low-arousal state flag is executed, and a low-arousal determination result is generated.

[0009] Optionally, based on the real-time arousal state determination result being a high arousal state, the step of performing preliminary physiological downgrading intervention, generating a soothing electrical stimulation mode activation command and an abdominal breathing training interface call command, performing a physiological downgrading attainment determination, and generating a cognitive reconstruction task call command when the physiological downgrading attainment determination is passed, including: According to the soothing electrical stimulation mode activation command, the stimulation intensity is linearly increased and modulated, so that the stimulation current gradually increases from the perception threshold to the target intensity value locked by the user feedback at a preset stimulation intensity increase rate. The breathing frequency is extracted, and the electrical stimulation carrier frequency is set to a preset multiple of the real-time breathing frequency according to the extracted real-time breathing frequency. Based on the switching nodes of the inspiratory and expiratory phases presented in the abdominal breathing training interface, the stimulation intensity respiratory phase micro-modulation is performed, so that the stimulation current increases by a preset stimulation intensity modulation amplitude at the end of the inspiratory phase and returns to the target intensity value at the end of the expiratory phase.

[0010] Optionally, the step of executing the reward stimulus parameter configuration and issuing the reward electrical stimulation pulse issuance instruction based on the successful behavior recognition result or the physiological stability recognition result meeting the preset triggering condition includes: If the behavior is successfully identified and the preset triggering conditions are met, the first reward stimulus parameter configuration is executed, and a reward electrical stimulation pulse delivery instruction is generated with a frequency of the first preset reward frequency and a duration of the first preset reward duration. Based on the physiological stability recognition result, if the preset triggering condition is met, the second reward stimulation parameter configuration is executed to generate the reward electrical stimulation pulse delivery instruction with a frequency of the first preset reward frequency and a duration of the first preset reward duration; If the behavior success recognition result and the physiological stability recognition result simultaneously meet the preset triggering conditions, the third reward stimulus parameter configuration is executed to generate the reward electrical stimulation pulse delivery command with a frequency of the first preset reward frequency, a duration of the first preset reward duration, and a stimulation intensity increased to a preset enhancement ratio.

[0011] This application also provides a dual-pathway closed-loop emotion regulation device based on state perception and temporal pairing, characterized in that the device comprises: The wake-up state discrimination module is used to perform wake-up state discrimination based on the self-rating data of the Emotion Visual Analogue Scale, heart rate variability data, and task interaction responsiveness data, and generate real-time wake-up state judgment results. The timing pairing intervention module is used to perform timing pairing of physiological regulation and cognitive intervention based on the real-time arousal state determination result, and generate electrical stimulation mode switching instructions and training task call instructions; The behavior reinforcement delivery module is used to identify successful execution events and generate reward electrical stimulation pulse delivery commands based on the execution completion data of the training task call command and the real-time changes in the heart rate variability data.

[0012] Optionally, the timing pairing intervention module further includes: The wake-up state routing discrimination module is used to perform intervention path routing discrimination based on the real-time wake-up state discrimination result. When the discrimination result is a high wake-up state, it generates a physiological downgrade prior path selection instruction, and when the discrimination result is a low wake-up state, it generates a physiological activation prior path selection instruction. The high-arousal physiological degradation prior intervention module is used to select instructions based on the physiological degradation prior path, execute the activation of the soothing electrical stimulation mode and the call to the abdominal breathing training interface, execute the physiological degradation achievement judgment, and generate a cognitive reconstruction task call instruction when the physiological degradation achievement judgment is passed. The low-arousal physiological activation prior intervention module is used to execute the activation-type electrical stimulation mode initiation and activation progress visualization interface call according to the physiological activation prior path selection instruction, execute the physiological activation target determination, and generate the behavior activation task call instruction when the physiological activation target determination is passed.

[0013] This application also provides an electronic device for implementing any of the described dual-pathway closed-loop emotion regulation methods based on state awareness and temporal pairing, including: The processor is used to execute the complete calculation and control process from collecting self-rated data from the Visual Analogue Scale (VAS) of emotions, photoplethysmography (PPS) signals, and task interaction response data; performing arousal state discrimination to generate real-time arousal state determination results; performing time-series pairing of physiological regulation and cognitive intervention based on the real-time arousal state determination results to generate electrical stimulation mode switching instructions and training task invocation instructions; performing success event recognition to generate reward electrical stimulation pulse delivery instructions based on the execution completion data of the training task invocation instructions and the real-time changes of heart rate variability data extracted from the PPS signals; until the electrical stimulation output interface delivers electrical stimulation pulses and the training task content is presented on the display interface. The memory is used to store the preset high wake-up threshold conditions, preset low wake-up threshold conditions, physiological degradation target conditions, physiological activation target conditions, preset trigger conditions, electrical stimulation mode parameters, training task content library, target intensity values ​​locked by user feedback, intermediate data during processing, and various types of instruction data generated, all necessary for implementing the complete calculation and control process.

[0014] The beneficial effects of this application are as follows: Based on arousal state perception data, physiological-cognitive temporal pairing intervention is performed, which forcibly couples the physiological regulation of electrical stimulation with the cognitive reshaping of structured psychological training in a state-dependent order, realizing a paradigm shift from open-loop stimulation to closed-loop adaptive regulation; through the dual-pathway success event recognition of behavioral task success and physiological state stability, reward-based electrical stimulation pulses with differentiated parameters are delivered within a preset time window to strengthen the operant conditioning reflex of positive cognitive behavior; and by extracting the respiratory frequency from the photoplethysmography pulse wave signal in real time and modulating the electrical stimulation carrier frequency and intensity phase accordingly, the physiological anchors required for interoceptive training are transformed into perceptible tactile signals, realizing the generalization and transfer of psychological skills from training scenarios to real-life scenarios. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the accompanying drawings required in the description of the embodiments or the prior art are briefly introduced below. Obviously, the accompanying drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0016] Figure 1 A flowchart illustrating a specific embodiment of the dual-pathway closed-loop emotion regulation method based on state perception and temporal pairing of this application is shown. Figure 2This diagram illustrates the overall architecture of a dual-pathway closed-loop emotion regulation method based on state perception and temporal pairing, according to a specific embodiment of this application. Figure 3 This paper illustrates a dual-closed-loop adaptive decision-making flowchart of a dual-pathway closed-loop emotion regulation method based on state perception and temporal pairing according to a specific embodiment of this application. Figure 4 This paper illustrates an intervention timing diagram within a high-arousal state unit of a dual-pathway closed-loop emotion regulation method based on state perception and temporal pairing, according to a specific embodiment of this application. Figure 5 The following is a time sequence diagram illustrating the conditional reflex reinforcement mechanism of a dual-pathway closed-loop emotion regulation method based on state perception and temporal pairing according to a specific embodiment of this application. Figure 6 This diagram illustrates the respiratory coupling variable frequency stimulation principle of a dual-pathway closed-loop emotion regulation method based on state perception and temporal pairing according to a specific embodiment of this application. Figure 7 This diagram illustrates a block diagram of a dual-pathway closed-loop emotion regulation device based on state perception and temporal pairing, according to a specific embodiment of this application. Detailed Implementation

[0017] Various exemplary embodiments, features, and aspects of this application will now be described in detail with reference to the accompanying drawings. The same reference numerals in the drawings denote elements that have the same or similar functions. Although various aspects of the embodiments are shown in the drawings, they are not necessarily drawn to scale unless specifically indicated otherwise.

[0018] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0019] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments.

[0020] Furthermore, to better illustrate this application, numerous specific details are provided in the following detailed embodiments. Those skilled in the art should understand that this application can be implemented without certain specific details. In some instances, methods, means, components, and circuits well-known to those skilled in the art have not been described in detail in order to highlight the main points of this application.

[0021] This application proposes a dual-pathway closed-loop regulation method based on state perception and temporal pairing to address the systemic deficiencies in existing emotional disorder intervention techniques, such as the separation of physiological regulation and psychological training, open-loop regulatory framework, and lack of behavioral reinforcement mechanisms. The method executes a complete "perception-discrimination-pairing-reinforcement" closed-loop process within a single training unit using a device: First, it collects self-assessment data from the Visual Analogue Scale (VAS), heart rate variability data, and task interaction responsiveness data. By multi-dimensionally fusing subjective self-assessment, objective physiological indicators, and behavioral interaction characteristics, it generates a real-time arousal state determination result of high or low arousal. Subsequently, based on the determination result, it enforces state-dependent temporal pairing intervention logic. When a high arousal state is determined, a soothing electrical stimulation mode is first activated and an abdominal breathing training interface is invoked. After the physiological downgrade criteria are met, a cognitive reconstruction task invocation command is generated. When a low arousal state is determined, an activation electrical stimulation mode is first activated and an activation progress visualization interface is invoked. After the physiological activation criteria are met, a behavioral activation task invocation command is generated, thereby establishing a "perception-discrimination-pairing-reinforcement" mechanism. This application employs an irreversible intervention sequence of "prioritizing homeostasis and followed by cognitive reshaping." During cognitive or behavioral tasks, it continuously monitors real-time changes in task completion markers and heart rate variability, performing behavioral task success and physiological stability assessments respectively. When a success event in either dimension is identified, a reward-based electrical stimulation pulse of a specified frequency and duration is delivered within a preset time window. When both dimensions are simultaneously met, an enhanced reward stimulus is delivered, thus establishing a conditioned association between behavioral success and physiological pleasure. Furthermore, during the physiological degradation intervention phase, the respiratory frequency is extracted in real-time from the photoplethysmography (PPG) signal. The electrical stimulation carrier frequency is set to a preset multiple of the respiratory frequency, and the stimulation intensity is micro-modulated with a respiratory phase dependence, causing the stimulation perception to exhibit tidal fluctuations synchronized with the respiratory rhythm, providing a tactile anchor point that can be actively followed for interoceptive training. This application achieves parameterized closed-loop fusion of neuromodulation technology and psychological behavioral training on the same timescale and platform, filling the technological gap in wearable emotion intervention in terms of adaptive closed-loop architecture and dual-pathway behavioral reinforcement.

[0022] Example 1 like Figure 1 The diagram shown is a flowchart of a dual-pathway closed-loop emotion regulation method based on state perception and temporal pairing according to an embodiment of this application. The method specifically includes the following: S100 performs arousal state discrimination based on self-report data from the Emotion Visual Analogue Scale, heart rate variability data, and task interaction responsiveness data, and generates real-time arousal state determination results.

[0023] Specifically, the system receives the user's swiping operation data on the Visual Analogue Scale (VAS) self-assessment interface, performs self-assessment value analysis, and generates the VAS anxiety score and VAS depression score by mapping the termination position of the swiping trajectory. Simultaneously, it collects photoplethysmography (PPG) signals from photoelectric sensors, performs time-frequency domain analysis of the PPG signals to calculate the low-frequency power to high-frequency power ratio and total power value using continuous wavelet transform or peak interval extraction algorithms, serving as a real-time quantitative representation of autonomic nervous system function. Furthermore, it continuously monitors the user's interactive behavior in the training task interface, recording response delay duration and key pressure sampling values ​​as reference indicators for arousal status in the behavioral dimension. The VAS anxiety score, VAS depression score, low-frequency power to high-frequency power ratio, total power value, response delay duration, and key pressure sampling value are compared in multiple dimensions with preset high arousal threshold conditions and preset low arousal threshold conditions. When any indicator in any dimension meets the corresponding threshold condition and the duration exceeds the preset duration, the corresponding arousal status marker is executed, generating a high arousal judgment result or a low arousal judgment result. If none of the multidimensional indicators meet any threshold condition, an unclassified wake-up state determination result is generated, and the default is to match the general intervention path.

[0024] S200, based on the real-time wake-up state determination result, performs the timing pairing of physiological regulation and cognitive intervention, and generates electrical stimulation mode switching instructions and training task invocation instructions.

[0025] Specifically, after receiving the real-time wake-up state determination result, an intervention path routing judgment is performed. When the determination result is a high-wake state, a physiological downgrade preliminary path selection instruction is generated, and a preliminary physiological downgrade intervention is performed accordingly. A soothing electrical stimulation mode activation instruction is generated, and linear incremental modulation of stimulation intensity is performed, so that the stimulation current gradually increases from the perception threshold to the target intensity value locked by the user through interface feedback at a preset stimulation intensity incremental rate. Simultaneously, an abdominal breathing training interface call instruction is generated, and a dynamic breathing light ball containing the inspiratory and expiratory phase switching nodes is displayed on the display interface. A physiological downgrade target achievement judgment is performed in real time. When the root mean square change of the difference between adjacent normal sinus heartbeats meets a preset condition or the decrease in anxiety score on the visual analog scale meets a preset condition, a target achievement signal is generated, and a cognitive reconstruction task call instruction is generated accordingly. An automatic thought recognition or cognitive bias classification task is called from the training task content library and pushed to the display interface. When the assessment result indicates a low arousal state, a physiological activation prior path selection instruction is generated, an activation-type electrical stimulation mode initiation instruction is generated, stimulation intensity is modulated in a stepwise manner, and the activation progress visualization interface is invoked simultaneously. A physiological activation threshold assessment is then performed; upon achieving the threshold, a behavioral activation task invocation instruction is generated.

[0026] S300: Based on the completion data of the training task invocation instruction and the real-time changes in the heart rate variability data, the system performs successful event recognition and generates a reward electrical stimulation pulse delivery instruction.

[0027] Specifically, during the training task execution, the execution status of the cognitive reconstruction task invocation command or the behavioral activation task invocation command is continuously monitored. When task completion flag data is detected, the behavioral task is successfully determined. The task completion flag data includes at least one of the following: confirmation signal for complete input of automatic thought text, confirmation signal for selection of cognitive bias recognition option, confirmation signal for input of alternative evidence text, or confirmation signal for clicking the practice completion button. Based on this, a behavioral success recognition result is generated. Simultaneously, the root mean square variation amplitude of the difference between adjacent normal sinus beats within a continuous time window is continuously extracted from the photoplethysmography (PPG) signal, and the real-time trend of the ratio of low-frequency power to high-frequency power is calculated to perform physiological stability determination. When the percentage increase of the root mean square variation amplitude compared to the baseline before task execution reaches a preset condition and the ratio of low-frequency power to high-frequency power is stable within a preset range, a physiological stability recognition result is generated. The system performs real-time logical judgment on the successful behavior recognition result and the physiological stability recognition result. When either recognition result meets a preset trigger condition, the system executes the first reward stimulus parameter configuration, generating a reward electrical stimulation pulse delivery command with a frequency of the first preset reward frequency and a duration of the first preset reward duration, and delivers the reward pulse through the electrical stimulation output interface. When both recognition results simultaneously meet the preset trigger condition, the system executes the third reward stimulus parameter configuration, increasing the stimulus intensity to a preset enhancement ratio based on the first preset reward frequency and the first preset reward duration before delivery. The event characteristics of each reward stimulus delivery are recorded in memory for subsequent periodic macro-closed-loop evaluation.

[0028] In summary, this application breaks away from the traditional one-dimensional model of separating physiological regulation and psychological training in emotion intervention, and constructs a dual-pathway closed-loop regulation mechanism based on state perception and temporal pairing. In the state discrimination stage, this application achieves high-precision, multi-dimensional discrimination of the user's real-time arousal state by multi-dimensionally fusing subjective emotion visual analog scale self-report data, objective heart rate variability time-frequency domain indicators, and response delay duration and key pressure sampling values ​​in task interaction. These data are then compared in parallel with preset high and low arousal threshold conditions, effectively solving the problem of insufficient reliability of single indicators in complex emotional states. In the intervention execution phase, this application enforces a state-dependent, time-paired intervention logic based on the arousal state determination result. When a high arousal state is determined, a soothing electrical stimulation mode is first activated and the diaphragmatic breathing training interface is invoked. The stimulation carrier frequency is dynamically coupled with the real-time respiratory frequency, and the stimulation intensity is micro-modulated with a respiratory phase dependence. After physiological degradation reaches the target, a cognitive reconstruction task invocation command is generated. When a low arousal state is determined, an activation electrical stimulation mode is first activated and the activation progress visualization interface is invoked. After physiological activation reaches the target, a behavioral activation task invocation command is generated. This time-forced mechanism ensures that physiological homeostasis is always a prerequisite for cognitive remodeling, solving the deficiency of diminished efficacy caused by the disordered order of physiological and psychological interventions in traditional solutions. In the behavioral reinforcement phase, this application establishes a dual-pathway success event recognition mechanism based on behavioral task success and physiological state stability. Within a preset time window, reward electrical stimulation pulses with differentiated parameters are delivered. When both dimensions reach the target simultaneously, a reward stimulus with increased intensity is delivered, embedding the establishment of operant conditioning into each training unit, filling the technical gap in existing digital therapies that lack a closed loop for behavioral reinforcement.

[0029] As an optional implementation of this application, optionally, in step S100, based on the self-report data of the Emotion Visual Analogue Scale, heart rate variability data, and task interaction responsiveness data, arousal state discrimination is performed to generate a real-time arousal state determination result, including: 110. Based on the sliding operation data received from the self-assessment interface of the Visual Analogue Scale (VAS), perform self-assessment value analysis to generate the anxiety score and depression score of the VAS.

[0030] Specifically, the Visual Analogue Scale (VAS) self-assessment interface presents a continuous sliding track with anchor marks at both ends. One end of the track is marked as "no emotional distress," and the other end as "extreme emotional distress." When the user slides along the track and releases the touch, the system reads the corresponding coordinate value from the end position of the slide, maps the coordinate value to a preset percentage range, and generates the VAS anxiety score and VAS depression score. During the mapping process, the system performs linear interpolation based on the score range corresponding to the anchor marks at both ends of the track, converting the relative offset of the end position into the corresponding self-assessment score. For example, if the user drags the anxiety dimension slider to 70% of the track's total length and releases it, the system generates a VAS anxiety score of 70.

[0031] 120. Based on the photoplethysmography pulse wave signal, perform time-frequency domain analysis of heart rate variability to generate the ratio of low-frequency power to high-frequency power and the total power value.

[0032] Specifically, the system continuously acquires photoplethysmography (PPG) signals from photoelectric sensors. These PPG signals contain waveforms representing periodic changes in blood volume caused by heartbeats. The system performs time-frequency domain analysis of heart rate variability on the PPG signals. First, a continuous heartbeat interval sequence is extracted from the PPG waveform using a peak detection algorithm. Then, a fast Fourier transform or autoregressive model spectral estimation is performed on this sequence, dividing the power spectrum into low-frequency and high-frequency bands. The power integral values ​​within the low-frequency and high-frequency bands are calculated separately, generating a low-frequency power to high-frequency power ratio. Simultaneously, the system calculates the total power integral value across the entire power spectrum, generating a total power value. The low-frequency power to high-frequency power ratio reflects the balance between the sympathetic and parasympathetic nervous systems, while the total power value reflects the overall regulatory capacity of the autonomic nervous system. This time-frequency domain analysis process is continuously executed using a sliding window with a window length set to 60 seconds, updating the analysis results every 10 seconds to ensure statistical reliability and real-time responsiveness of the data.

[0033] 130. Based on the anxiety score of the Visual Analogue Scale (VAS), the depression score of the VAS, the ratio of low-frequency power to high-frequency power, the total power value, the response delay time and key pressure sampling value in the task interaction, the results are compared with the preset high wake-up threshold conditions and the preset low wake-up threshold conditions to generate a real-time wake-up status determination result.

[0034] Specifically, step 130, based on the anxiety and depression scores of the Visual Analogue Scale (VAS) generated in step 110, the low-frequency power to high-frequency power ratio and total power value generated in step 120, and combined with behavioral indicators recorded during task interaction, performs multi-dimensional arousal state determination. Regarding task interaction behavior monitoring, the system continuously records the user's response delay duration and key pressure sampling value in the training task interface. The response delay duration refers to the time interval between the presentation of the task content and the user's first touch operation, and the key pressure sampling value refers to the pressure intensity value sensed by the touchscreen. The system compares these multi-dimensional indicators with preset threshold conditions, executing parallel logic for high arousal state determination and low arousal state determination.

[0035] In the high arousal state determination path, the system checks whether at least one of the following three triggering conditions is met: First, the anxiety score on the Visual Analogue Scale (VAS) is greater than or equal to a preset high arousal anxiety threshold. This threshold is based on clinical sample statistics and is used to distinguish the clinical significance level of subjective anxiety experience. Second, the ratio of low-frequency power to high-frequency power is greater than a preset sympathetic excitation threshold, and the duration for which this ratio exceeds the preset sympathetic excitation threshold exceeds a preset high arousal duration threshold. This condition is set based on the fact that excessive sympathetic nerve excitation usually manifests as a sustained increase in the LF / HF ratio rather than a momentary fluctuation. Third, the response delay duration is less than a preset lower limit for high arousal response delay, and the key pressure sampling value is greater than a preset upper limit for high arousal pressure. This condition captures the common impatient operation characteristics of users in a high arousal state, i.e., responding too quickly but with excessive force. When any of the triggering conditions is met, the system executes the high arousal state marking and generates a high arousal determination result.

[0036] In the low-arousal state determination path, the system checks whether at least one of the following three triggering conditions is met: First, the depression score on the Visual Analogue Scale (VAS) is greater than or equal to a preset low-arousal depression threshold; second, the total power value is less than a preset lower limit threshold for total autonomic power, and the duration of this state exceeds a preset low-arousal duration threshold, with the persistently low total power reflecting significant inhibition of the overall autonomic nervous system's regulatory function; third, the response delay duration is greater than a preset upper limit for low-arousal response delay, which captures the typical psychomotor retardation characteristics of users in the low-arousal state, i.e., a significant delay in task response. When any of the triggering conditions is met, the system executes the low-arousal state marking and generates a low-arousal determination result.

[0037] If the triggering conditions of the above two sets of judgment paths are not met, the system generates an unclassified wake-up state judgment result and defaults to matching the general intervention path.

[0038] As an optional implementation of this application, optionally, in step S200, based on the real-time wake-up state determination result, the timing pairing of physiological regulation and cognitive intervention is performed to generate an electrical stimulation mode switching instruction and a training task invocation instruction, including: 210. Based on the real-time wake-up status determination result of being in a high-wake state, perform preliminary physiological downgrade intervention, generate a soothing electrical stimulation mode activation command and an abdominal breathing training interface call command, perform physiological downgrade attainment determination, and generate a cognitive reconstruction task call command when the physiological downgrade attainment determination is passed.

[0039] Specifically, when the real-time wake-up status determination result generated in step 130 is a high wake-up state, the system executes the complete process of prior physiological downgrade intervention.

[0040] Regarding electrical stimulation output, the system first generates a soothing electrical stimulation mode activation command, setting the output parameters to a carrier frequency of 100 Hz and a pulse width of 200 microseconds. This frequency parameter is chosen based on previous neuromodulation research showing that transcutaneous electrical stimulation in the 100 Hz band can effectively activate inhibitory pathways in the ascending reticular activating system, promoting a decrease in cortical arousal. The system performs linearly increasing modulation of stimulation intensity, gradually increasing the stimulation current from the individually calibrated perception threshold at a preset rate. During the intensity increase, the system displays an intensity adjustment slider on the interface, allowing the user to provide feedback on the current perceived comfort level through sliding. Once the user confirms and locks the slider position, the system records the current stimulation current value as the target intensity value and maintains a constant output.

[0041] Regarding respiratory rate extraction and stimulation frequency coupling, the system performs real-time respiratory rate extraction from the photoplethysmography (PPG) signal. Respiratory sinus arrhythmia components are analyzed from the PPG waveform using continuous wavelet transform or peak interval extraction algorithms. These components are frequency components in heart rate variability highly coupled to the respiratory rhythm, and their instantaneous frequency can approximate the user's real-time respiratory rate. Based on the extracted real-time respiratory rate, the system sets the electrical stimulation carrier frequency to a preset multiplier value of the real-time respiratory rate. This preset multiplier value is a configurable parameter between 2 and 4 times. For example, when the user's current respiratory rate is detected to be 0.25 Hz, the system automatically adjusts the electrical stimulation carrier frequency to 0.5 Hz, 0.75 Hz, or 1.0 Hz, with the specific multiplier value pre-selected by the user according to personal preference. This modulation dynamically synchronizes the electrical stimulation delivery rhythm with the user's voluntary respiratory rhythm, providing an actively perceptible physiological anchor point for subsequent interoceptive training.

[0042] In terms of interface guidance, the system synchronously generates commands to call up the diaphragmatic breathing training interface, displaying a dynamic breathing light sphere on the screen. The dynamic breathing light sphere performs periodic expansion and contraction animations with preset inspiratory and expiratory phase durations, with the inspiratory phase set to 4 seconds and the expiratory phase set to 6 seconds, visually guiding users to establish a smooth and slow diaphragmatic breathing rhythm.

[0043] Regarding respiratory phase micro-modulation, the system performs stimulus intensity respiratory phase micro-modulation based on the switching nodes of the inspiratory and expiratory phases presented in the diaphragmatic breathing training interface. When the dynamic respiratory sphere enters the end of the inspiratory phase, the system increases the preset stimulus intensity modulation amplitude based on the target intensity value, so that the perceived stimulus intensity gradually increases with the inspiratory process; when the dynamic respiratory sphere enters the end of the expiratory phase, the system restores the stimulus intensity to the target intensity value, so that the perceived stimulus intensity gradually decreases with the expiratory process. This modulation amplitude is controlled within the range of 5% to 10% of the target intensity value, forming a tidal stimulus perception fluctuation synchronized with the respiratory rhythm, allowing the user to clearly perceive the somatic signal mapping of increased stimulus during inspiratory and decreased stimulus during expiratory.

[0044] Regarding the determination of physiological degradation achievement, the system repeatedly collects physiological indicators and performs achievement determination at preset time intervals. When the root mean square value of the difference between adjacent normal sinus beats increases by a preset percentage threshold compared to the initial baseline value of the unit, or when the anxiety score of the Visual Analogue Scale decreases by a preset score threshold compared to the initial baseline value of the unit, the system determines that the physiological degradation achievement condition is met, generates an achievement signal, and accordingly calls a cognitive reconstruction task from the training task content library and pushes it to the display interface. The cognitive reconstruction task includes automatic thought recognition and recording, evidence verification and alternative thought generation, or cognitive bias classification training.

[0045] 220. Based on the real-time wake-up status determination result of low wake-up state, perform preliminary physiological activation intervention, generate activation-type electrical stimulation mode start command and activation progress visualization interface call command, perform physiological activation standard determination, and generate behavior activation task call command when the physiological activation standard determination is passed.

[0046] Specifically, when the real-time wake-up state determination result generated in step 130 is a low wake-up state, the system executes the complete process of prior physiological activation intervention.

[0047] The system first generates an activation command for the electrical stimulation mode, setting the output parameters to a carrier frequency of 20 Hz and a pulse width of 200 microseconds. This frequency setting is based on the fact that electrical stimulation in the 20-30 Hz band has a clear excitatory effect on the ascending reticular activating system, effectively enhancing cortical arousal. The system then performs stepwise modulation of the stimulation intensity, starting from an individually calibrated perception threshold and gradually increasing the stimulation current with preset step increments and durations. Each step is maintained for a fixed duration to observe user tolerance, until the preset target activation intensity range is reached.

[0048] In terms of interface guidance, the system generates a command to call the activation progress visualization interface. The activation progress is presented on the display interface in the form of an energy bar filling animation. The filling ratio of the energy bar increases synchronously with the increase of the stimulation intensity, providing users with real-time visual feedback on the activation process.

[0049] Regarding the determination of physiological activation achievement, the system receives user feedback signals or changes in objective indicators in real time. When a user actively clicks the confirmation button on the interface to indicate that they feel energetic, or when the energy score on the Visual Analogue Scale (VAS) rises above the initial baseline value of the unit to a preset threshold, the system determines that the physiological activation achievement condition is met, generates an achievement signal, and accordingly retrieves behavioral activation tasks from the training task content library and pushes them to the display interface. These behavioral activation tasks include arranging a predictable pleasurable activity, completing a postponed simple task, or initiating a brief contact with a friend or family member. Each task is designed to be achievable within a preset time limit and has a clear completion indicator.

[0050] As an optional implementation of this application, optionally, in step S300, based on the completion data of the training task invocation instruction and the real-time changes in the heart rate variability data, a success event recognition is performed, and a reward electrical stimulation pulse delivery instruction is generated, including: 310. Based on the task completion flag data generated after the training task call instruction is executed, the successful execution of the behavior task is determined, and a successful behavior recognition result is generated.

[0051] Specifically, after the cognitive reconstruction task invocation instruction generated in step 210 or the behavior activation task invocation instruction generated in step 220 is executed, the system continuously monitors the interaction data during task execution and extracts task completion marker data from it. The task completion marker data has different judgment forms depending on the task type. For automatic thought recognition and recording tasks, the system monitors whether the user has completely entered a description of an automatic thought in the text input area and clicked the submit confirmation button; for cognitive bias classification training tasks, the system monitors whether the user has completed the selection operation from the preset cognitive bias type option list and clicked the confirmation button; for evidence verification and alternative thought generation tasks, the system monitors whether the user has generated at least one piece of text content in the alternative evidence input area and clicked the complete button; for behavior activation tasks, the system monitors whether the user has clicked the practice completion confirmation button. When the system detects any of the above task completion marker data, the behavior task is successfully judged, and a successful behavior recognition result is generated. This judgment mechanism transforms abstract cognitive skill growth into automatically identifiable discrete behavioral events, providing a clear trigger basis for subsequent reward incentive distribution.

[0052] 320. Based on the root mean square variation amplitude of the difference between adjacent normal sinus beats within a continuous time window extracted from the photoplethysmography signal and the ratio of low-frequency power to high-frequency power, a physiological stability determination is performed, and a physiological stability identification result is generated.

[0053] Specifically, during the execution of the training task invocation instruction generated in step 210 or step 220, the system continuously extracts physiological indicators from the photoplethysmography (PPG) signal and performs physiological stability determination. The system first extracts the root mean square (RMS) variation of the difference between adjacent normal sinus beats within a continuous time window from the PPG signal using a sliding window method. This RMS variation is a core indicator for heart rate variability time-domain analysis, reflecting the real-time level of parasympathetic activity. The system compares the RMS variation of the difference between adjacent normal sinus beats during task execution with the baseline value recorded before task execution and calculates the improvement ratio. Simultaneously, the system continuously calculates the real-time trend of the low-frequency power to high-frequency power ratio, monitoring whether this ratio is stable within a preset autonomic nervous system balance range. The upper and lower limits of the preset autonomic nervous system balance range are set based on the LF / HF ratio distribution in healthy individuals at rest, used to characterize a relatively balanced state between the sympathetic and parasympathetic nervous systems. When the system detects that the root mean square change in the difference between adjacent normal sinus beats has increased by a preset percentage threshold compared to the baseline before the task, and the ratio of low-frequency power to high-frequency power remains stable within a preset autonomic nervous system balance range exceeding a preset stability duration threshold, a physiological stability state is determined to have been achieved. Furthermore, when the anxiety score or depression score on the Visual Analogue Scale (VAS) decreases by a preset percentage threshold compared to before the task ends, or when the user completes full diaphragmatic breathing training within a preset duration and the amplitude of respiratory sinus arrhythmia increases by a preset percentage threshold compared to before training, the system also generates a physiological stability recognition result. This dual-dimensional physiological stability determination mechanism comprehensively assesses the user's physiological homeostasis level during the cognitive task from two perspectives: the degree of parasympathetic activation and the autonomic nervous system balance state, providing a second triggering channel for reward stimuli independent of the behavioral task.

[0054] 330. Based on the successful recognition result of the behavior or the physiological stability recognition result meeting the preset triggering conditions, execute the reward stimulus parameter configuration and release, and generate the reward electrical stimulation pulse release instruction.

[0055] Specifically, step 330, based on the successful behavior recognition results generated in step 310 and the physiological stability recognition results generated in step 320, performs tiered reward stimulus parameter configuration and distribution. The system performs real-time logical judgment on the successful behavior recognition results and the physiological stability recognition results, and executes differentiated reward strategies according to the number of dimensions that meet the conditions.

[0056] In the single-dimensional triggering path, when only the behavior's successful recognition result meets the preset triggering conditions, the system executes the first reward stimulus parameter configuration, generating a reward electrical stimulation pulse delivery command with a frequency of the first preset reward frequency and a duration of the first preset reward duration. The first preset reward frequency is set to 80 Hz, a frequency that has been proven in basic neuroscience research to effectively activate the mesolimbic dopamine reward pathway, and its neural coding characteristics are similar to the firing patterns of dopamine neurons induced by natural reward stimuli. The first preset reward duration is set to 1.0 second, a duration chosen to ensure that the stimulus pulse is clearly perceptible without inducing neural adaptation. When only the physiological stability recognition result meets the preset triggering conditions, the system executes the second reward stimulus parameter configuration, similarly generating a reward electrical stimulation pulse delivery command with a frequency of the first preset reward frequency and a duration of the first preset reward duration. The basic reward parameters for both are the same, but the system adds different source markers during delivery for subsequent cumulative effect tracking.

[0057] In the dual-dimensional triggering path, when both the successful behavioral recognition result and the physiological stability recognition result simultaneously meet the preset triggering conditions, the system executes the third reward stimulus parameter configuration. While maintaining the frequency and duration of the first preset reward frequency and duration, the stimulus intensity is increased to a preset enhancement ratio before being released. Simultaneously, enhanced animations or auditory cues in the visual feedback interface can be optionally triggered. This dual-dimensional simultaneous triggering mechanism aims to strengthen the positive experience of successful physiological and cognitive synergy, enabling users to intrinsically link the completion of cognitive tasks with a sense of bodily stability.

[0058] After the reward stimulus pulse is issued, the system records the timestamp, triggering pathway source, and issuance parameter configuration to the memory for subsequent periodic macro-closed-loop evaluation of behavioral activation level statistics and reinforcement density analysis. The time lock interval between successful event recognition and reward stimulus pulse issuance is controlled within 0.5 seconds, which meets the optimal time schedule conditions for establishing the association between behavior and reward in operant conditioning.

[0059] Example 2 As an application example of this application, the specific application is as follows: I. Basic Contents of the Implementation Example This embodiment consists of a closed-loop workflow comprised of the following four core modules: electrical stimulation provides a calm psychophysiological basis for skill learning, while skill training consolidates the positive effects of electrical stimulation. (1) Multimodal sensing module: Integrated into wristband devices and mobile devices, it collects self-assessment data from the Visual Analogue Scale (VAS), heart rate variability (HRV), and task interaction responsiveness for real-time assessment of patient status.

[0060] (2) Dual closed-loop decision engine: Micro-loop (within the unit): Based on real-time VAS / HRV, the stimulation frequency / intensity is dynamically adjusted and matched to the difficulty of the training task.

[0061] Macro-loop (periodic): Based on the reduction rate of the Hamilton Depression Rating Scale (HAMD) / Hamilton Anxiety Rating Scale (HAMA) and the frequency of skill use, the intervention plan is reset weekly.

[0062] (3) Median nerve electrical stimulation output module: It offers two stimulation modes: soothing (100Hz, 200μs, low-frequency mode) and activating (20Hz, 200μs, high-frequency mode) to suit different arousal needs.

[0063] (4) Structured Emotion Training Content Library: It includes four modules: relaxation training, cognitive restructuring, mindfulness practice, and behavioral activation, and outputs are paired with electrical stimulation timing according to state-dependent logic.

[0064] like Figure 2 As shown, the system consists of five layers: a perception layer, a micro-loop, a macro-loop, an execution layer, and a reinforcement layer. 1. Perception Layer (Blue): Integrates multimodal inputs, including subjective emotion VAS self-rating, objective HRV monitoring, and task interaction features (such as response delay and completion rate), which are then fused and input into the decision layer.

[0065] 2. Micro-loop (intra-unit adaptive) (orange): Runs within a single training unit (seconds), dynamically modulating stimulus parameters and matching task difficulty based on real-time state discrimination (high arousal / low arousal) until the exit condition is met. This loop solves the problem of "current state mismatch".

[0066] 3. Macro-loop (Periodic Reset) (Dark Orange): Runs weekly, using scale reduction rates, skill usage frequency, and high-frequency data accumulated from micro-loops to determine symptom evolution (residual anxiety, emerging anxiety, insufficient compliance), and updates the baseline for the next cycle. This loop addresses the issue of "symptom evolution during treatment."

[0067] 4. Nested Relationship: The cyclical baseline generated by the macro-loop constrains the decision boundaries of the micro-loop (such as the upper limit of stimulus intensity and the range of task difficulty) through dashed arrows; the unit termination event of the micro-loop triggers the cyclical evaluation of the macro-loop. The two form a nested relationship where "the micro-loop obeys the macro-loop, and the macro-loop depends on the micro-loop".

[0068] 5. Execution layer (green): includes peripheral nerve electrical stimulation output (with replaceable target points) and a structured training content library, and receives control instructions from the decision layer.

[0069] 6. Reinforcement layer (purple): It contains two reinforcement mechanisms—success event-driven reward stimuli (80Hz / 1s / suprathreshold intensity) and respiratory coupling stimulation modulation (dynamically adjusting the stimulation frequency and intensity based on real-time respiratory rate). Its output is fed back to the executive layer to form a behavior shaping loop.

[0070] Solid arrows represent data / control flow, while dashed arrows represent constraints.

[0071] II. Detailed Description of the Implementation Examples 1. System Initialization: Setting the Baseline of the Solution Based on Multidimensional Evaluation Upon first use or at the start of each treatment cycle, the system in this embodiment first performs a baseline assessment process to establish a personalized initial intervention plan. This process integrates clinical scales and objective physiological indicators to form a multi-dimensional state profile.

[0072] 1.1 The evaluation index layer adopts a dual-track parallel architecture: In the subjective scale dimension, the system collects the total score and factor scores of the Hamilton Depression Rating Scale (HAMD), Hamilton Anxiety Rating Scale (HAMA) or Patient Health Questionnaire-9 (PHQ-9), and Generalized Anxiety Disorder Scale (GAD-7), focusing on analyzing clinical factors such as "psychological anxiety", "somatic anxiety", "depressive core mood" and "retardation". In terms of objective physiological dimensions, the system collects heart rate variability (HRV) under resting conditions and extracts time-frequency domain indicators such as low-frequency power (LF), high-frequency power (HF), LF / HF ratio, total power (TP), and RMSSD as quantitative representations of autonomic nervous function.

[0073] 1.2 The decision logic layer performs symptom subtype discrimination based on the above multidimensional features, distinguishing between initial and subsequent stages: (1) For the first time, discriminant logic was used (without historical scale data) to determine symptom subtype tendency based on absolute thresholds: If the LF / HF ratio is >2.0 and the resting-state HRV total power is normal (>500ms²), it is judged as "high arousal anxiety tendency"; If the total HRV power is less than 500 ms² and the LF / HF ratio is less than 1.0, it is judged as "low-arousal depression with retardation tendency"; If neither is significant (e.g., LF / HF is between 1.0 and 2.0 and the total power is normal), it is judged as "mixed / undifferentiated" and a general solution (soothing stimulation + mindfulness training) is pushed.

[0074] (2) Subsequent cycle discrimination logic (after accumulating ≥2 weeks of data), activate the trend discrimination, combining the changes in scale factor scores and physiological indicator trends: Calculate the difference between the current cycle factor score and the baseline (first time or the mean of the first 4 weeks). If the increase exceeds the preset threshold (e.g., ≥30%) and the physiological indicators match (e.g., LF / HF increase simultaneously), it is determined to be symptom activation. Based on slope analysis (e.g., if HAMD decreases by <10% for two consecutive weeks and HRV shows no improvement, it is determined to be a plateau in efficacy), the treatment plan is adjusted and the strategy is reset.

[0075] Based on the discrimination results, the system automatically generates an initial baseline scheme and configures it differently according to the population stratification.

[0076] This baseline setting is not fixed, but rather is subject to real-time correction of the micro-loop within the unit and scheme reset of the periodic macro-loop during subsequent interventions.

[0077] 1.3 Population Stratification and Differentiated Intervention: Based on the multidimensional data from the initial assessment, the system categorizes users into four levels and configures differentiated intervention goals and parameters: Table 1 Evaluation Data Table The initial baseline protocols for users at different levels show gradient differences in stimulus intensity, training difficulty, and task frequency, but the core closed-loop logic (double closed loop, state dependence, and behavioral reinforcement) remains consistent. User levels are dynamically adjusted as the intervention progresses (e.g., mild to moderate patients can be downgraded to a sub-healthy mode after improvement).

[0078] 2. Real-time adaptive control within the unit: state-dependent, second-level closed-loop intervention The temporal nesting relationship between the micro-loop within the unit and the periodic macro-loop is as follows: Figure 3 As shown, a single training unit (typically 15-25 minutes in duration) constitutes the minimum intervention closed loop in this embodiment. During unit operation, the system acquires the user's state feedback signals with high temporal resolution and dynamically switches the combination mode and execution order of stimulation parameters and training tasks based on preset state discrimination rules. This is one of the essential features that distinguishes this embodiment from existing open-loop stimulation devices.

[0079] 2.1 Physiological Degradation Priority Strategy in High Arousal State Taking the high-arousal state as an example, the complete second-level closed-loop intervention process within a single training unit is as follows: Figure 4 As shown, the system automatically determines that the user is in a high-wake state when it detects any of the following trigger conditions: Anxiety score on the Visual Analogue Scale (VAS) with an instantaneous input value ≥ 70 points (out of 100), or Real-time HRV analysis shows an LF / HF ratio >2.0 for more than 30 seconds, or The rate of rise of electrical skin activity (EDA) is >0.5 μS / s (alternative), or Abnormal task interaction characteristics (such as excessive key pressure, excessively fast response, and high error rate). The intervention sequence strictly follows the temporal principle of "physiological downgrade first, followed by cognitive reconstruction".

[0080] Phase 1: Physiological downgrade intervention (preset duration 3-5 minutes, can be exited early if the condition is satisfactory) The system immediately activates the soothing electrical stimulation mode: carrier frequency 100Hz, pulse width 200μs, stimulation intensity starts from the perception threshold and increases linearly at a rate of 0.1mA / second until the user provides feedback on the "best comfort" position via the interface slider, and then maintains that intensity.

[0081] Synchronous activation of guided abdominal breathing training: The interface displays a dynamic breathing light sphere, with an inspiratory phase of 4 seconds and an exhalation phase of 6 seconds. The breathing cycle is linked to the micro-modulation of the stimulation intensity—the stimulation intensity increases slightly by 5% at the end of inspiration and returns to the baseline at the end of exhalation.

[0082] The system reassesses the status indicators every 30 seconds. The exit condition is that any of the following is met: ① The RMSSD of the HRV time domain index increases by ≥15% compared to the initial value of this unit (the system uses the sliding window method to calculate the HRV index in real time, with a window length of 60 seconds and an update every 10 seconds to ensure the statistical reliability of the index); ② The VAS anxiety score decreases by ≥20 points compared to the initial value of this unit; ③ The physiological degradation phase has reached the preset upper limit time (5 minutes).

[0083] Phase Two: Cognitive Restructuring Intervention After the user successfully exits the physiological degradation phase, the system automatically switches to the cognitive challenge task module. This module is built on the core technologies of cognitive behavioral therapy and includes, but is not limited to, structured tasks such as "automatic thought recognition and recording," "evidence verification and alternative thought generation," "cognitive bias classification training," and "catastrophic imagination degradation exercises."

[0084] During this stage, the electrical stimulation output switches to maintenance mode: the frequency is reduced to 50Hz and the intensity is reduced to a sub-sensory level (sub-sensory intensity that is imperceptible to the user but has a neuromodulation effect). Its functional positioning changes from "active degradation" to "state maintenance" to prevent the user from experiencing a second rebound in arousal level during cognitive processing.

[0085] 2.2 Physiological activation-first strategy in low-arousal state The system automatically determines that the user is in a low wake-up / hysteresis state when it detects any of the following trigger conditions: A VAS depression score of ≥ 70 points at the moment of input, or HRV total power <500ms² for more than 60 seconds, or The task response latency is ≥1.5 times longer than the user's historical average, or User interface interaction characteristics (such as swiping pauses and input interruption frequency) indicate insufficient energy. The intervention sequence strictly follows the temporal principle of "physiological activation first, followed by behavioral tasks".

[0086] Phase 1: Physiological activation intervention (preset duration 3-4 minutes) The system initiates an activating electrical stimulation mode: frequency 20Hz, pulse width 200μs, and intensity gradually increasing from the sensory threshold to the upper limit of the sensory threshold using a stepwise strategy. The selection of this frequency parameter (20Hz) is based on the fact that previous neuromodulation studies have shown that low-frequency stimulation (below 20Hz) has an inhibitory effect on cortical excitability, while the 20-30Hz frequency band has a clear excitatory effect on the ascending reticular activating system. A synchronized visualization of the activation progress is displayed, providing immediate feedback in the form of an "energy bar."

[0087] Exit conditions: ① VAS vitality / energy score increases by ≥15 points from the initial value of this unit; ② The user actively clicks the "Feeling ready" button; ③ The activation phase runs for the maximum duration (4 minutes).

[0088] Phase Two: Behavioral Activation Intervention Upon successful activation, the system will push a behavioral activation task. Behavioral activation is one of the core techniques in evidence-based psychotherapy for depression. Its theoretical assumption is that by increasing opportunities for positive behavior to encounter positive reinforcement, the vicious cycle of "low motivation → avoidance behavior → negative emotions" can be broken. The behavioral activation task library built into this embodiment includes structured exercises such as "scheduling a predictable and enjoyable activity," "completing a simple, postponed task," "initiating a brief contact with a friend or family member," and "recording a small task completed today." Each task is designed to be completed within 5 minutes, with clear completion indicators and quantifiable feedback.

[0089] 3. Conditioned reflex establishment mechanism: behavioral reinforcement paradigm based on successful events This embodiment introduces the principle of operant conditioning into a fusion system of wearable neuromodulation and digital therapy, constructing an automated, parameter-specific, time-locked behavioral reinforcement stimulation paradigm.

[0090] 3.1 Success Event Recognition Layer (Dual Path): The system monitors the user's "success status" in real time from the following two dimensions, and a reward incentive is triggered when either dimension is met: (1) Dimension 1: Success of behavioral task Enter the complete text into the automatic thought process; Identify at least one cognitive bias; Generate at least one alternative piece of evidence; Click the "Complete this exercise" button.

[0091] (2) Dimension Two: Stable Physiological State During task execution, the RMSSD of HRV improved by ≥15% compared to the baseline for 3 consecutive minutes, and the LF / HF ratio remained stable in the range of 0.8-1.5. Or, the VAS anxiety / depression score decreased by ≥30% by the end of the task compared to before the task; Or complete a full diaphragmatic breathing exercise (6 breaths / minute, lasting 5 minutes) and the amplitude of respiratory sinus arrhythmia (RSA) increases by ≥20%.

[0092] (3) Dual-path recognition logic: Achieving the target in any dimension → triggers the standard reward stimulus; Achieving the goals in both dimensions simultaneously triggers an enhanced reward stimulus (such as increasing the stimulus intensity to 2.0 times the sensory threshold, or adding visual / auditory Easter egg effects), strengthening the positive experience of "physiological-cognitive synergistic success".

[0093] The system performs real-time logical judgment on the above events, with a recognition latency of less than 200ms.

[0094] 3.2 Reward-Incentive Execution Layer (Dual-Channel Reinforcement): (1) Channel 1: Physiological reinforcement (electrical stimulation pulse) Stimulation frequency: 80Hz (This frequency band has been shown in basic research to activate the mesolimbic dopamine reward pathway, which is similar to the neural coding characteristics of natural reward). Pulse width: 200μs Stimulation duration: 1.0 second (short, clearly perceptible, does not induce adaptation) Stimulation intensity: 1.5 times the sensory threshold (clearly higher than the perception threshold, but strictly controlled below the pain threshold; subjectively described by users as "a slight tingling sensation"). Function: Activates the mesolimbic dopamine reward pathway, providing a physiologically pleasurable experience.

[0095] (2) Channel Two: Cognitive Reinforcement (Text / Voice Feedback) While the electrical stimulation pulses are being delivered, the screen displays personalized positive feedback, such as: "Great! You've just successfully identified an automatic thought. This is a crucial step in emotion regulation!" "Your training today was 20% more effective than last time; your stability and focus have both improved!" "Physiological indicators show your body is learning to relax—this is the beginning of long-term change!" If both dimensions are met: "Double success in both physiological and cognitive aspects! Your progress is beyond imagination!" Optional voice prompts (which users can turn off) enhance emotional connection.

[0096] (3) Reinforcement mechanism principle: Physiological reinforcement provides "instant pleasure," while cognitive reinforcement provides "a sense of meaning and progress." The combination of the two forms a complete reward experience. According to self-determination theory, this design simultaneously satisfies the user's competence needs (I did it) and relational needs (the system recognizes me), thus sustainably enhancing intrinsic motivation.

[0097] 3.3 Strengthening temporal relationships: The time-locking relationship between successful event recognition and reward incentive distribution is as follows: Figure 5 As shown, the time lock interval between successful event recognition and reward stimulus distribution is less than 0.5 seconds, which meets the optimal time schedule conditions for establishing the "behavior-reward association" in operant conditioning. The system repeats this pairing relationship of "successful coping → physiological pleasure" after each successful task, aiming to increase the spontaneous frequency of positive cognitive behavior and enhance users' positive value assignment of their own coping abilities.

[0098] 3.4 Cumulative Effect Tracking: The system backend records the number of reward stimuli each user receives daily, the corresponding task types, and the characteristics of successful events, and generates a "positive response frequency curve" on a weekly basis. This data is used for the periodic macro-closed-loop evaluation described in Part (V), and can also be used to generate a user-visible "growth visualization" interface to further strengthen the positive cycle.

[0099] 4. Skill generalization-assisted anchoring: variable frequency stimulation and interoceptive training with respiratory coupling Patients with mood disorders commonly have a deficiency in interoceptive perception—that is, a decreased or distorted ability to perceive their own bodily signals (heartbeat, breathing, muscle tension).

[0100] This deficiency makes it difficult for patients to learn, master, and transfer psychological skills such as mindfulness training based on bodily awareness, progressive relaxation, and body scanning.

[0101] This embodiment provides patients with a physiological anchor point that can be actively perceived and controlled by dynamically coupling the electrical stimulation frequency with the real-time respiratory rhythm, significantly accelerating the efficiency of interoceptive training.

[0102] 4.1 Real-time extraction of respiratory rate: The system continuously acquires photoplethysmography (PPG) signals while the user wears the wristband device. Through continuous wavelet transform or peak interval extraction algorithms, the respiratory sinus arrhythmia (RSA) component is extracted. RSA is a frequency component of HRV that is highly coupled with respiratory rhythm, and its instantaneous frequency can be approximated as the user's real-time respiratory rate. The advantage of this method is that it eliminates the need for an additional breathing band sensor, achieving respiratory monitoring entirely based on existing PPG hardware.

[0103] 4.2 Stimulus Frequency Modulation Algorithm: The complete technical path of respiratory-coupled variable frequency stimulation is as follows: Figure 6 As shown, it is divided into four layers: signal processing layer, stimulus modulation layer, user perception layer, and training application layer. 1. Signal Processing Layer (Blue): Extracts respiratory sinus arrhythmia (RSA) from the PPG signal to obtain real-time respiratory rate f (e.g., 0.25 Hz) and respiratory phase (inspiratory / expiratory phase). This process does not require wearing an additional respiratory sensor.

[0104] 2. Stimulation Modulation Layer (Orange): Based on the respiratory rate f, the electrical stimulation carrier frequency is dynamically modulated to 2-4 times f (e.g., 0.5Hz-1.0Hz); simultaneously, based on the respiratory phase, the stimulation intensity is fine-tuned—increasing by 5%-10% at the end of the inspiratory phase and returning to baseline at the end of the expiratory phase. Together, these two elements generate a variable-frequency stimulation waveform that fluctuates synchronously with the respiratory rhythm.

[0105] 3. User perception layer (green): The stimulation waveform output by the wristband electrodes is perceived by the user as a tactile fluctuation synchronized with breathing, forming a subjective experience of "stimulus changing with the strength of breathing", thereby establishing an internal sensory anchor point.

[0106] 4. Training Application Layer (Purple): This anchor point is applied to multiple interoceptive training tasks, including mindfulness body scan, progressive muscle relaxation, and mindfulness breathing, to help users concretize abstract mental skills into actionable bodily sensations.

[0107] The system sets the carrier frequency of the electrical stimulation to 2 to 4 times the real-time respiratory rate (the user can select the multiplier according to their preference). For example, when the detected respiratory rate of the user is 0.25 Hz (15 breaths / minute), the stimulation frequency is automatically set to 0.5 Hz, 0.75 Hz, or 1.0 Hz. This modulation is not a fixed frequency division ratio, but a continuous dynamic response—the stimulation frequency increases synchronously when breathing speeds up, and decreases synchronously when breathing slows down.

[0108] 4.3 Stimulus Intensity and Respiratory Phase Modulation: Based on the aforementioned carrier frequency modulation, the system further applies respiratory phase-dependent micro-modulation to the stimulus intensity: at the end of the inspiratory phase, the stimulus intensity increases linearly by 5% to 10% from the baseline; at the end of the expiratory phase, the baseline intensity is restored. This modulation causes the perceived intensity of the stimulus to exhibit a tidal fluctuation synchronized with the respiratory rhythm, allowing the user to clearly perceive the body signal mapping of "stimulus enhancement during inhalation and stimulus reduction during exhalation."

[0109] 4.4 Training Scenario Application: This respiratory coupling stimulation pattern was activated during the following interoceptive training tasks: Mindfulness-based body scan: guides users to shift their attention to different parts of the body in sequence, and the synchronous breathing coupling stimulation provides multimodal perceptual cues (tactile + respiratory proprioception) for the "attention anchor point", significantly reducing mind wandering.

[0110] Progressive muscle relaxation: During the inspiratory phase, the user is guided to "feel the preparatory tension of the muscles" during the inspiratory phase and to "release the tension and feel the relaxation and sinking" during the expiratory phase, thus making the abstract "relaxation" instruction concrete into an operable physiological perception switch.

[0111] Mindful breathing: An external pacemaker that uses stimulation frequency as the respiratory rhythm to help patients with respiratory rhythm disorders (hyperventilation, tendency to sleep apnea) establish a stable and smooth breathing pattern.

[0112] 5. Periodic-level treatment plan reset: a macro-adaptive closed loop based on efficacy feedback Intra-unit adaptive regulation addresses the immediate state matching problem of a single intervention, while the treatment of mood disorders is essentially a dynamic process lasting from weeks to months. As treatment progresses, the patient's core symptoms may change—a patient initially predominantly anxious may develop a core depressive mood as anxiety subsides, and a patient initially predominantly lethargic may reveal underlying social anxiety as energy levels improve. This necessitates that the intervention program possess adaptive resetting capabilities on a macro-timescale.

[0113] This embodiment constructs a periodic macro-closed-loop control mechanism, whose operating cycle is usually set to 7 days, which is compatible with the frequency of clinical treatment follow-up.

[0114] 5.1 Evaluation Indicator Layer (Hybrid Evaluation Strategy): Weekly Simplified Scales: The system automatically pushes simplified scales (such as HAMD-7 and GAD-7, each with 7 questions) every weekend to monitor symptom fluctuations at high frequency and reduce the burden on patients.

[0115] Monthly complete scale: The system pushes out the complete scale (HAMD-17, HAMA-14) every 4 weeks to calibrate the assessment bias of the abbreviated scale and avoid the practice effect.

[0116] Objective physiological indicators: The system continuously records resting-state HRV (LF / HF, total power), sleep parameters (sleep latency, sleep efficiency) and daytime activity levels as evidence of efficacy independent of subjective scales.

[0117] Behavioral data: Statistics on the frequency of skill usage, task completion rate, and number of reward stimuli this week serve as objective proxy indicators of behavioral activation levels.

[0118] Intra-unit data summary: Summarize the average VAS improvement, HRV improvement rate, and stimulus parameter usage distribution of each training unit to form a quantitative profile of user compliance and responsiveness.

[0119] 5.2 Resetting Decision Logic (Multi-dimensional Integration): The system comprehensively judges the evolution of symptoms based on the above multi-source data. A typical decision path example is as follows: Pathway A (Residual Symptom Orientation): If the HAMD-7 reduction rate is ≥50% but the somatic anxiety factor score is still higher than the baseline 50% threshold, and the HRV LF / HF ratio remains persistently high (>2.0), it is judged as "improvement in core depressive mood, but residual somatic anxiety". The system increases the proportion of mindfulness-based body scanning and interoceptive training for the next cycle, and fixes the electrical stimulation mode of the corresponding training unit to a respiratory-coupled variable frequency mode.

[0120] Pathway B (Emerging Symptom-Oriented): If a patient initially presenting primarily with low arousal experiences an increase of ≥30% in their GAD-7 score compared to the previous cycle after improvement in energy levels, but their total HRV power remains low (<500ms²), they are classified as having "emerging anxiety symptoms after lag relief." The system resets the initial baseline protocol for the next cycle, switching from an "activating stimulus + behavioral activation" combination to a "soothing stimulus + relaxation training + cognitive restructuring" combination.

[0121] Path C (Compliance Intervention): If the frequency of skill use this week is lower than the preset threshold (e.g., ≤3 times / week) and there is no significant improvement in physiological indicators for two consecutive weeks, the system will determine it as "insufficient compliance". In the next cycle, the task difficulty level will be proactively reduced, while the frequency threshold of reward stimulus distribution will be increased to compensate for task participation through enhanced density.

[0122] Path D (Level Shift): If the symptom score is below the current level threshold for 4 consecutive weeks and the physiological indicators return to normal, the system will automatically downgrade the user to a lower intervention level (e.g., mild to moderate → subclinical) and adjust the stimulation intensity and training difficulty accordingly.

[0123] 5.3 The nesting relationship between macro-loops and micro-loops: This is the core technical feature of the dual closed-loop adaptive architecture in this embodiment, and it is also the essential difference from existing technologies that only provide real-time feedback or only provide periodic follow-up.

[0124] The micro-loop follows the macro-loop: the decision boundaries for real-time adjustment within a unit (such as the upper limit threshold of stimulus intensity and the selectable range of task difficulty) are constrained by the "cycle baseline" set by the macro-loop. For example, if the macro-loop determines that "high arousal is dominant", the upper limit of the micro-loop intensity in all units in the following week will be limited to the median value of the comfort threshold to prevent users from choosing excessive stimulation.

[0125] Macro-loop relies on micro-loop: The reset decision of the macro-loop does not rely solely on the scale input at periodic intervals, but deeply integrates the high-frequency state data accumulated in the micro-loop. For example, when determining "residual somatic anxiety," the macro-loop not only refers to the scale factor scores, but also retrieves micro-behavioral indicators such as the frequency of users actively extending the duration of the physiological degradation phase and the slope of HRV recovery in all training units this week, achieving a two-dimensional fusion decision based on clinical subjective scales and objective physiological behavioral trajectories.

[0126] Example 3 Based on the same principles as the aforementioned methods, a dual-pathway closed-loop emotion regulation device based on state perception and temporal pairing is also proposed, see [link to relevant documentation]. Figure 7 The dual-pathway closed-loop emotion regulation device 100 based on state awareness and temporal pairing in this embodiment includes: The wake-up state discrimination module 110 is used to perform wake-up state discrimination based on the self-rated data of the emotional visual analog scale, heart rate variability data and task interaction responsiveness data, and generate real-time wake-up state judgment results. The timing pairing intervention module 120 is used to perform timing pairing of physiological regulation and cognitive intervention based on the real-time arousal state determination result, and generate electrical stimulation mode switching instructions and training task call instructions. The behavior reinforcement delivery module 130 is used to identify successful events and generate reward electrical stimulation pulse delivery instructions based on the execution completion data of the training task call instruction and the real-time changes in the heart rate variability data.

[0127] As an optional implementation of this application, the timing pairing intervention module 120 may further include: The wake-up state routing discrimination module 121 is used to perform intervention path routing discrimination based on the real-time wake-up state discrimination result. When the discrimination result is a high wake-up state, it generates a physiological downgrade prior path selection instruction, and when the discrimination result is a low wake-up state, it generates a physiological activation prior path selection instruction. The high-arousal physiological degradation prior intervention module 122 is used to execute the start of the soothing electrical stimulation mode and the call of the abdominal breathing training interface according to the physiological degradation prior path selection instruction, execute the physiological degradation achievement judgment, and generate a cognitive reconstruction task call instruction when the physiological degradation achievement judgment is passed. The low-arousal physiological activation prior intervention module 123 is used to execute the activation-type electrical stimulation mode initiation and activation progress visualization interface call according to the physiological activation prior path selection instruction, execute the physiological activation target determination, and generate a behavior activation task call instruction when the physiological activation target determination is passed.

[0128] Obviously, those skilled in the art should understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the control methods described above. The modules or steps of this application described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, and thus can be stored in a storage device for execution by a computing device, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Therefore, this application is not limited to any specific hardware and software combination.

[0129] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the control methods described above. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk drive (HDD), or solid-state drive (SSD), etc.; the storage medium can also include combinations of the above types of memory.

[0130] Example 4 Furthermore, an electronic device for implementing any of the described dual-pathway closed-loop emotion regulation methods based on state awareness and temporal pairing includes: The processor is used to execute the complete calculation and control process from collecting self-rated data from the Visual Analogue Scale (VAS) of emotions, photoplethysmography (PPS) signals, and task interaction response data; performing arousal state discrimination to generate real-time arousal state determination results; performing time-series pairing of physiological regulation and cognitive intervention based on the real-time arousal state determination results to generate electrical stimulation mode switching instructions and training task invocation instructions; performing success event recognition to generate reward electrical stimulation pulse delivery instructions based on the execution completion data of the training task invocation instructions and the real-time changes of heart rate variability data extracted from the PPS signals; until the electrical stimulation output interface delivers electrical stimulation pulses and the training task content is presented on the display interface. The memory is used to store the preset high wake-up threshold conditions, preset low wake-up threshold conditions, physiological degradation target conditions, physiological activation target conditions, preset trigger conditions, electrical stimulation mode parameters, training task content library, target intensity values ​​locked by user feedback, intermediate data during processing, and various types of instruction data generated, all necessary for implementing the complete calculation and control process.

[0131] The electronic device of this disclosure includes a processor and a memory for storing processor-executable instructions. The processor is configured to implement, when executing the executable instructions, any of the aforementioned dual-pathway closed-loop emotion modulation methods based on state awareness and timing pairing.

[0132] It should be noted that the number of processors can be one or more. Furthermore, the electronic device in this embodiment may also include input devices and output devices. The processor, memory, input devices, and output devices can be connected via a bus or other means, without specific limitations herein.

[0133] The memory, serving as a computer-readable storage medium for automated fault handling and self-learning methods in modules, can be used to store software programs, computer-executable programs, and various modules, such as the program or module corresponding to the dual-pathway closed-loop emotion regulation method based on state awareness and timing pairing in this disclosure. The processor executes various functional applications and data processing of the electronic device by running the software programs or modules stored in the memory.

[0134] The various embodiments of this application have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A dual-pathway closed-loop emotion regulation method based on state perception and temporal pairing, characterized in that, include: Based on the self-report data of the Emotion Visual Analogue Scale, heart rate variability data, and task interaction responsiveness data, arousal state discrimination is performed, and real-time arousal state judgment results are generated. Based on the real-time wake-up state determination result, the timing pairing of physiological regulation and cognitive intervention is performed to generate electrical stimulation mode switching instructions and training task invocation instructions; Based on the completion data of the training task invocation instruction and the real-time changes in the heart rate variability data, the successful execution event is identified, and a reward electrical stimulation pulse delivery instruction is generated.

2. The dual-pathway closed-loop emotion regulation method based on state perception and temporal pairing as described in claim 1, characterized in that, The process involves using self-reported emotional visual analog scale data, heart rate variability data, and task interaction responsiveness data to perform arousal state determination and generate real-time arousal state determination results, including: Based on the sliding operation data received from the self-assessment interface of the Visual Analogue Scale (VAS), the self-assessment score is parsed to generate the anxiety score and depression score of the VAS. Based on the photoplethysmography pulse wave signal, perform time-frequency domain analysis of heart rate variability to generate the ratio of low-frequency power to high-frequency power and the total power value; Based on the anxiety score of the Visual Analogue Scale (VAS), the depression score of the VAS, the ratio of low-frequency power to high-frequency power, the total power value, the response delay time and key pressure sampling value in task interaction, the results are compared with preset high wake-up threshold conditions and preset low wake-up threshold conditions to generate a real-time wake-up status determination result.

3. The dual-pathway closed-loop emotion regulation method based on state perception and temporal pairing as described in claim 1, characterized in that, The step of performing a time-series pairing of physiological regulation and cognitive intervention based on the real-time arousal state determination result, generating electrical stimulation mode switching instructions and training task invocation instructions, includes: Based on the real-time wake-up status determination result of a high-wake state, a preliminary physiological downgrade intervention is performed, generating a soothing electrical stimulation mode activation command and an abdominal breathing training interface call command, performing a physiological downgrade attainment determination, and generating a cognitive reconstruction task call command when the physiological downgrade attainment determination is passed. If the real-time wake-up status is determined to be low-wake state, a preliminary physiological activation intervention is performed, generating an activation-type electrical stimulation mode start command and an activation progress visualization interface call command. A physiological activation target assessment is performed, and a behavioral activation task call command is generated when the physiological activation target assessment is passed.

4. The dual-pathway closed-loop emotion regulation method based on state perception and temporal pairing as described in claim 1, characterized in that, The step of recognizing successful execution events and generating reward electrical stimulation pulse delivery instructions based on the execution completion data of the training task invocation instruction and the real-time changes in the heart rate variability data includes: Based on the task completion flag data generated after the training task call instruction is executed, the successful execution of the behavior task is determined, and a successful behavior recognition result is generated. Based on the root mean square variation amplitude of the difference between adjacent normal sinus beats within a continuous time window extracted from the photoplethysmography pulse wave signal and the ratio of low-frequency power to high-frequency power, physiological stability determination is performed to generate physiological stability recognition results. Based on the successful recognition result of the behavior or the physiological stability recognition result meeting the preset triggering conditions, the reward stimulus parameter configuration is executed and issued, generating the reward electrical stimulation pulse issuance instruction.

5. The dual-pathway closed-loop emotion regulation method based on state perception and temporal pairing as described in claim 2, characterized in that, The method involves comparing the anxiety score of the Visual Analogue Scale (VAS), the depression score of the VAS, the ratio of low-frequency power to high-frequency power, the total power value, and the response delay and key pressure sampling value during task interaction with preset high and low wake-up threshold conditions to generate a real-time wake-up status determination result, including: If the anxiety score of the Visual Analogue Scale is greater than or equal to the preset high arousal anxiety threshold, or the ratio of low-frequency power to high-frequency power is greater than the preset sympathetic excitation threshold and the duration exceeds the preset high arousal duration threshold, or the response delay duration is less than the preset high arousal response delay lower limit and the key pressure sampling value is greater than the preset high arousal pressure upper limit, a high arousal state is marked and a high arousal judgment result is generated. If the depression score on the Visual Analogue Scale is greater than or equal to a preset low-arousal depression threshold, or the total power value is less than a preset lower limit threshold for total autonomic power and the duration exceeds a preset low-arousal duration threshold, or the response delay duration is greater than a preset upper limit threshold for low-arousal response delay, a low-arousal state flag is executed, and a low-arousal determination result is generated.

6. The dual-pathway closed-loop emotion regulation method based on state perception and temporal pairing as described in claim 3, characterized in that, The process involves determining a high arousal state based on real-time wakefulness status, performing preliminary physiological downgrading intervention, generating a soothing electrical stimulation mode activation command and an abdominal breathing training interface invocation command, performing a physiological downgrading attainment assessment, and generating a cognitive restructuring task invocation command when the physiological downgrading attainment assessment is passed, including: According to the soothing electrical stimulation mode activation command, the stimulation intensity is linearly increased and modulated, so that the stimulation current gradually increases from the perception threshold to the target intensity value locked by the user feedback at a preset stimulation intensity increase rate. The breathing frequency is extracted, and the electrical stimulation carrier frequency is set to a preset multiple of the real-time breathing frequency according to the extracted real-time breathing frequency. Based on the switching nodes of the inspiratory and expiratory phases presented in the abdominal breathing training interface, the stimulation intensity respiratory phase micro-modulation is performed, so that the stimulation current increases by a preset stimulation intensity modulation amplitude at the end of the inspiratory phase and returns to the target intensity value at the end of the expiratory phase.

7. The dual-pathway closed-loop emotion regulation method based on state perception and temporal pairing as described in claim 4, characterized in that, The step of executing the reward stimulus parameter configuration and issuing the reward electrical stimulation pulse issuance instruction based on the successful recognition result of behavior or the physiological stability recognition result meeting the preset triggering conditions includes: If the behavior is successfully identified and the preset triggering conditions are met, the first reward stimulus parameter configuration is executed, and a reward electrical stimulation pulse delivery instruction is generated with a frequency of the first preset reward frequency and a duration of the first preset reward duration. Based on the physiological stability recognition result, if the preset triggering condition is met, the second reward stimulation parameter configuration is executed to generate the reward electrical stimulation pulse delivery instruction with a frequency of the first preset reward frequency and a duration of the first preset reward duration; If the behavior success recognition result and the physiological stability recognition result simultaneously meet the preset triggering conditions, the third reward stimulus parameter configuration is executed to generate the reward electrical stimulation pulse delivery command with a frequency of the first preset reward frequency, a duration of the first preset reward duration, and a stimulation intensity increased to a preset enhancement ratio.

8. A dual-pathway closed-loop emotion regulation device based on state perception and temporal pairing, characterized in that, The device includes: The wake-up state discrimination module is used to perform wake-up state discrimination based on the self-rating data of the Emotion Visual Analogue Scale, heart rate variability data, and task interaction responsiveness data, and generate real-time wake-up state judgment results. The timing pairing intervention module is used to perform timing pairing of physiological regulation and cognitive intervention based on the real-time arousal state determination result, and generate electrical stimulation mode switching instructions and training task call instructions; The behavior reinforcement delivery module is used to identify successful execution events and generate reward electrical stimulation pulse delivery commands based on the execution completion data of the training task call command and the real-time changes in the heart rate variability data.

9. The dual-pathway closed-loop emotion regulation device based on state perception and temporal pairing according to claim 8, wherein the temporal pairing intervention module further comprises: The wake-up state routing discrimination module is used to perform intervention path routing discrimination based on the real-time wake-up state discrimination result. When the discrimination result is a high wake-up state, it generates a physiological downgrade prior path selection instruction, and when the discrimination result is a low wake-up state, it generates a physiological activation prior path selection instruction. The high-arousal physiological degradation prior intervention module is used to select instructions based on the physiological degradation prior path, execute the activation of the soothing electrical stimulation mode and the call to the abdominal breathing training interface, execute the physiological degradation achievement judgment, and generate a cognitive reconstruction task call instruction when the physiological degradation achievement judgment is passed. The low-arousal physiological activation prior intervention module is used to execute the activation-type electrical stimulation mode initiation and activation progress visualization interface call according to the physiological activation prior path selection instruction, execute the physiological activation target determination, and generate the behavior activation task call instruction when the physiological activation target determination is passed.

10. An electronic device for implementing the dual-pathway closed-loop emotion regulation method based on state awareness and temporal pairing as described in any one of claims 1 to 7, comprising: The processor is used to execute the complete calculation and control process from collecting self-rated data from the Visual Analogue Scale (VAS) of emotions, photoplethysmography (PPS) signals, and task interaction response data; performing arousal state discrimination to generate real-time arousal state determination results; performing time-series pairing of physiological regulation and cognitive intervention based on the real-time arousal state determination results to generate electrical stimulation mode switching instructions and training task invocation instructions; performing success event recognition to generate reward electrical stimulation pulse delivery instructions based on the execution completion data of the training task invocation instructions and the real-time changes of heart rate variability data extracted from the PPS signals; until the electrical stimulation output interface delivers electrical stimulation pulses and the training task content is presented on the display interface. The memory is used to store the preset high wake-up threshold conditions, preset low wake-up threshold conditions, physiological degradation target conditions, physiological activation target conditions, preset trigger conditions, electrical stimulation mode parameters, training task content library, target intensity values ​​locked by user feedback, intermediate data during processing, and various types of instruction data generated, all necessary for implementing the complete calculation and control process.

Images