Self-adaptive control system, method and device based on ST segment correction

Abstract

The invention relates to a self-adaptive control system, method and device based on ST segment correction, and belongs to the technical field of medical instruments. The system comprises a signal processing module, a trigger control module, an exception adaptation module and an execution module, the process of the method realizes a complete closed loop of signal acquisition, feature extraction, decision output, exception handling and execution feedback, and the device comprises a device body provided with the self-adaptive control system based on ST segment correction. The problems that a traditional method is low in synchronization precision, insufficient in pressure maintaining time and the like are solved by establishing a reference time sequence framework through R waves, correcting inflation points in real time through ST sections and optimizing the pressure maintaining stage, the blood flow pressurization efficiency is remarkably improved, and the targeted treatment effect on coronary heart disease and myocardial ischemia patients is remarkably improved.

Description

Technical Field

[0001] The present invention relates to an adaptive control system, method and device based on ST segment correction, and belongs to the technical field of medical equipment. Background Art

[0002] Traditional external counterpulsation devices generally rely on the ECG R wave to trigger the inflation and deflation of the airbag. The principle is to determine the end point of the heart's systole by detecting the peak of the R wave, and then drive the airbag to increase pressurization during the diastole to enhance coronary blood flow. For example, Chinese patent publication number CN118059383A discloses a core algorithm for synchronizing the pressure application of an external counterpulsation device by combining ECG and pulse information, including: S1: collecting the ECG signal and pulse signal of the user, synchronously processing the ECG signal and pulse signal, and determining the systole and diastole of the heart; S2: determining the actual start time of the external counterpulsation device to apply pressure based on the synchronous processing results of the ECG signal and pulse signal, obtaining the reaction time difference between the systole and diastole of the heart and the pressure sensor, and resetting and correcting the actual start time based on the reaction time difference threshold; setting a pressure sensor at the counterpulsation action surface, the pressure sensor is connected to the controller, obtaining the reaction time difference between the systole and diastole of the heart and the pressure sensor, and resetting and correcting the actual start time based on the reaction time difference threshold, restarting, returning the cycle error to 0, and re-circulating the counterpulsation cycle to improve the execution accuracy of the counterpulsation device.

[0003] However, such methods have significant limitations in clinical application. R-wave triggering is susceptible to arrhythmias (such as atrial fibrillation and premature ventricular contractions) and electromagnetic interference in the patient's room, resulting in asynchrony between balloon movement and diastole, reducing blood flow pressurization efficiency. Furthermore, R-wave triggering cannot dynamically adapt to the patient's real-time ST segment excursion (a key electrophysiological marker of myocardial ischemia), leading to a delayed response to acute myocardial ischemia and hindering personalized targeted treatment for patients with coronary artery disease. Furthermore, conventional devices lack intelligent anti-interference mechanisms. When patients experience complex rhythms (such as atrial fibrillation with dynamic ST segment changes), system stability deteriorates significantly, leading to large synchronization errors and severely impacting treatment efficacy and safety. Furthermore, existing technologies lack explicit pressure-holding control, resulting in insufficient blood flow pressurization time. To address these issues, a dual-mode control approach based on ECG R-wave cycle prediction and dynamic ST-segment correction is urgently needed. By establishing a baseline timing framework and dynamically correcting key trigger points, this approach overcomes the technical bottlenecks of conventional R-wave triggering and achieves precise control of the three phases of inflation, pressure maintenance, and deflation, thereby improving treatment accuracy, stability, and clinical adaptability. Summary of the Invention

[0004] The purpose of the present invention is to provide an adaptive control system, method and device based on ST segment correction. By establishing a reference timing framework and dynamically correcting key trigger points, it breaks through the technical bottleneck of traditional R-wave triggering and realizes precise control of the three stages of inflation / pressure maintenance / exhaust, so as to improve the accuracy, stability and clinical adaptability of treatment.

[0005] The adaptive control system based on ST segment correction described in the present invention includes a signal processing module, a trigger control module, an abnormal adaptation module and an execution module.

[0006] The signal processing module is used to pre-process the electrical signal and extract the ST segment features, determine the starting position, offset and morphological characteristics of the ST segment, and perform confidence assessment based on the stability of the ST segment slope;

[0007] The trigger control module dynamically calculates the balloon inflation delay time based on the ST segment morphological characteristics, sets an adjustable time window after the ST segment onset, and implements a hybrid trigger mode;

[0008] The abnormality adaptation module is used to immediately switch to the sliding window averaging algorithm to eliminate transient noise interference when a persistent ST segment abnormality is detected. If ST segment recognition fails, it will prioritize switching to the cycle prediction mode. Only when the cycle prediction fails will the R wave trigger mode be activated as a backup logic and an alarm will be issued at the same time.

[0009] The execution module is used to receive the inflation instruction from the trigger control module in real time and control the inflation and deflation of the airbag.

[0010] Through a closed-loop architecture design encompassing signal processing, trigger control, anomaly adaptation, and execution modules, the system achieves real-time linkage between electrical signal processing and mechanical intervention, significantly improving the accuracy and response speed of ST segment correction compared to traditional single-parameter control systems. The multi-module collaborative mechanism provides the system with autonomous fault tolerance, automatically switching to backup logic when ST segment recognition fails, and minimizing the probability of application interruption.

[0011] Preferably, the signal processing module includes an adaptive filtering unit, a cycle prediction unit and an ST segment feature extraction unit.

[0012] Adaptive filtering unit: eliminates baseline drift and high-frequency noise by dynamically adjusting filter parameters;

[0013] Cycle prediction unit: calculates the average cardiac cycle T through the R wave interval of the previous 3-5 cardiac cycles and generates benchmark timing parameters;

[0014] ST segment feature extraction unit: uses baseline correction technology to align isoelectric lines, and combines the slope threshold method to accurately determine the ST segment starting position, offset and morphological characteristics, while performing confidence assessment based on the ST segment slope stability.

[0015] The adaptive filtering unit effectively suppresses myoelectric interference and baseline drift through dynamic parameter adjustment (cutoff frequency 0.5-40Hz, adjustable in real time). The cycle prediction unit uses a historical cycle weighting algorithm (with decreasing weights for the first 3-5 cycles) to reduce the error in generating the reference timing parameters, providing a reliable timing reference for airbag control. The ST segment feature extraction unit's slope threshold method, combined with a confidence assessment mechanism, achieves an ST segment onset determination accuracy of 97.6% and an offset measurement error of <0.1mV.

[0016] Preferably, the trigger control module includes a delay calculation unit, a timing optimization unit and a mode switching unit.

[0017] Delay calculation unit: dynamically calculates the airbag inflation delay time based on ST segment morphological characteristics;

[0018] Timing optimization unit: Sets an adjustable time window after the ST segment starting point in combination with the hemodynamic model;

[0019] Mode switching unit: executes mixed trigger mode, when ST confidence ≥ 80: inflation point = ST starting point + 5ms; when 40≤ST confidence < 80: wait for ST signal within 10ms before the preset inflation starting point, and switch to periodic prediction mode if timeout; when ST confidence < 40: force the use of preset inflation starting point.

[0020] The delay calculation unit integrates ST morphological characteristics with a hemodynamic model to minimize inflation delay calculation errors. The timing optimization unit's adjustable time window, combined with the hemodynamic model, ensures a high degree of consistency between the balloon's action phase and cardiac diastole. The mode switching unit's three-level confidence decision mechanism (80 / 40 dual-threshold design) enables smooth transitions between treatment strategies and reduces the incidence of overinflation compared to fixed-threshold systems.

[0021] Preferably, the abnormal adaptation module includes an anti-interference unit and a dynamic fault tolerance unit.

[0022] Anti-interference unit: When a persistent ST segment abnormality is detected, it immediately switches to the sliding window averaging algorithm to perform weighted smoothing on the ST segment data of the last three cycles to eliminate transient noise interference;

[0023] Dynamic fault-tolerant unit: When ST segment recognition fails within three consecutive cardiac cycles, it will switch to cycle prediction mode first. Only when the cycle prediction mode fails, the R wave trigger mode will be enabled as a backup logic and the alarm mode will be issued at the same time.

[0024] The anti-interference unit's sliding window averaging algorithm achieves an 85% transient noise suppression rate, improving the stability of ST segment morphology recognition. The dynamic fault-tolerant unit's dual-mode switching strategy (cycle prediction priority + R-wave backup) enhances adaptability and high system availability. The alarm trigger mechanism (activated within three cycles of ST segment recognition failure) provides sufficient warning time for intervention.

[0025] Preferably, under the R-wave trigger mode backup logic, the dynamic fault-tolerant unit automatically switches to the R-wave trigger mode after the cycle prediction mode fails; the system continuously monitors the ST segment status and re-enables the ST segment correction mode when the ST confidence level is ≥60 for 5 consecutive cycles.

[0026] The automatic switching design of the R-wave triggered mode ensures that it can continue to work under extremely abnormal heart rhythms; the ST segment confidence recovery mechanism (automatic switching after 5 consecutive cycles ≥ 60 minutes) realizes intelligent fallback of the treatment strategy and avoids unnecessary mode oscillation.

[0027] Preferably, in the hybrid trigger mode, a sliding window averaging algorithm is used to smooth the ST segment data of three consecutive cardiac cycles; if a valid ST signal is detected within 10 ms before the preset inflation start point, the cycle prediction timing is covered.

[0028] The fusion design of the sliding window averaging algorithm and cycle prediction improves anti-interference capabilities while ensuring timing accuracy; the signal priority waiting mechanism (10ms timeout protection) improves ST segment utilization and reduces invalid cycle predictions.

[0029] Preferably, the reference timing parameters are: preset inflation starting point = T×0.25, pressure holding end point = T×0.8, exhaust trigger point = T×0.9.

[0030] The golden ratio design of the inflation starting point (T×0.25), the pressure-maintaining end point (T×0.8) and the exhaust trigger point (T×0.9) is designed according to the hemodynamic characteristics of the heart's diastole to increase the working pressure (i.e., increase the coronary artery perfusion pressure).

[0031] The adaptive control method based on ST segment correction described in the present invention is applied to the above-mentioned adaptive control system based on ST segment correction, and includes the following steps:

[0032] S1, real-time acquisition and filtering of electrical signals, calculation of average period T by R wave detection, and generation of reference timing parameters;

[0033] S2. Extract the ST segment waveform through baseline correction and determine the starting point and offset according to the slope threshold method;

[0034] S3. Calculate the ST confidence score. When the ST confidence score is ≥80, the inflation point is equal to the ST starting point + 5ms. When the ST confidence score is 40≤ and the ST confidence score is <80, the ST signal is prioritized to be waited for within 10ms before the preset inflation starting point. If the signal times out, the cycle prediction is switched. When the ST confidence score is <40, the preset inflation starting point is used forcibly.

[0035] S4. If persistent ST segment abnormalities are detected, a sliding window averaging algorithm is used to optimize the data.

[0036] S5: When ST recognition fails for three consecutive cardiac cycles, the system switches to the cycle prediction mode. Only when the cycle prediction mode fails, the R wave trigger mode is enabled and an alarm is issued.

[0037] S6. Drive the multi-channel airbag to perform inflation and deflation actions according to the trigger signal.

[0038] The method achieves a complete closed-loop process: signal acquisition (S1) → feature extraction (S2) → decision output (S3) → exception handling (S4-S5) → execution feedback (S6), shortening the operating cycle. Seamless switching between cycle prediction mode and R-wave triggering mode enables the system to continuously operate under complex electrical signals (maintaining therapeutic effectiveness in complex rhythms such as atrial fibrillation).

[0039] Preferably, the cycle prediction mode calculates the average cardiac cycle T through the R wave intervals of the previous 3-5 cardiac cycles and generates the reference timing parameters.

[0040] Moving average filtering (window length 3-5 cycles) combined with weighted coefficient optimization is used to reduce the cycle prediction error and provide a high-precision timing benchmark for airbag control.

[0041] The adaptive control device based on ST segment correction described in the present invention includes a device body, in which the above-mentioned adaptive control system based on ST segment correction is arranged.

[0042] The modular system packaging design (all modules are integrated into the device body) enables plug-and-play, and the multi-channel airbag drive interface supports personalized configuration to meet different needs.

[0043] Compared with the existing technology, the adaptive control system, method and device based on ST segment correction of the present invention have the following beneficial effects:

[0044] By establishing a baseline timing framework through R-wave prediction, and optimizing the ST-segment inflation point and pressure-holding phase in real time, this approach addresses the problems of low synchronization accuracy and insufficient pressure-holding time in traditional methods, significantly improving blood flow pressurization efficiency and achieving targeted treatment effects for patients with coronary heart disease and myocardial ischemia. This invention achieves the functions of establishing a baseline timing framework through R-wave prediction cycles, using the ST-segment as a dynamic correction signal, optimizing airbag triggering timing, adaptive fault tolerance for abnormal heart rhythms, and precise multi-channel control. It offers significant advantages such as high therapeutic synchronization, strong anti-interference capabilities, wide clinical adaptability, and excellent safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is a flow chart of an adaptive control method based on ST segment correction according to the present invention;

[0046] Figure 2 Schematic diagram of electrical signals. DETAILED DESCRIPTION

[0047] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0048] Example 1

[0049] This embodiment discloses an adaptive control system based on ST segment correction, which includes a signal processing module, a trigger control module, an abnormal adaptation module and an execution module.

[0050] The signal processing module, including an adaptive filtering unit, an ST-segment feature extraction unit, and a cycle prediction unit, uses an ECG sensor to collect the patient's ECG signal in real time as the input electrical signal for the signal processing module. The electrical signal is first input to the adaptive filtering unit, which dynamically adjusts the filter parameters to eliminate baseline drift and high-frequency noise. The cycle prediction unit calculates the average cardiac cycle T using the R-wave interval of the previous 3-5 cardiac cycles and generates benchmark timing parameters: preset inflation start point = T × 0.25, pressure hold end point = T × 0.8, and deflation trigger point = T × 0.9. The preprocessed electrical signal is transmitted to the ST-segment feature extraction unit, which uses baseline correction technology to align the isoelectric lines and combines the slope threshold method to accurately determine the ST-segment starting position, offset, and morphological characteristics. Confidence assessment is also performed based on the stability of the ST-segment slope.

[0051] The trigger control module includes a delay calculation unit, a timing optimization unit and a mode switching unit. The delay calculation unit dynamically calculates the airbag inflation delay time based on the ST segment morphological characteristics (including offset amplitude, duration and waveform slope); the generated delay parameters are input into the timing optimization unit, and an adjustable time window is set after the ST segment starting point in combination with the hemodynamic model; the mode switching unit executes a hybrid trigger logic. When the ST confidence level is ≥80, the inflation point = ST starting point + 5ms; when the ST confidence level is 40≤<80, the ST signal is preferentially waited for within 10ms before the preset inflation starting point. If the timeout exceeds the limit, the cycle prediction mode is switched; when the ST confidence level is <40, the preset inflation starting point is forcibly adopted.

[0052] The abnormal adaptation module includes an anti-interference unit and a dynamic fault-tolerant unit. When the anti-interference unit detects a persistent ST segment abnormality, it immediately switches to the sliding window averaging algorithm and performs weighted smoothing on the ST segment data of the last three cycles to eliminate transient noise interference. If the dynamic fault-tolerant unit detects that ST segment recognition fails within three consecutive cardiac cycles, it will prioritize switching to the cycle prediction mode. Only when the cycle prediction mode fails, the R-wave trigger mode is enabled as a backup logic. At the same time, an early warning signal is sent to medical staff through the sound and light alarm module to ensure treatment continuity and safety.

[0053] The execution module includes a multi-channel airbag and an embedded processor. The embedded processor receives inflation commands from the trigger control module in real time and then, through a multi-channel driver circuit, time-sharingly controls the inflation and deflation of the airbags in the calf and thigh regions. During early diastole, airbags are inflated synchronously according to a preset pressure gradient. At late diastole, airbags are rapidly deflated via high-speed solenoid valves to maximize blood flow boost efficiency. The processor also integrates a pressure feedback unit, which uses a piezoelectric sensor to monitor airbag pressure fluctuations in real time and dynamically adjust inflation duration to mitigate the risk of ischemia.

[0054] The signal processing module, trigger control module, abnormal adaptation module and execution module together constitute a real-time control system, which supports ST segment correction mode, cycle prediction mode, mixed mode and R wave trigger mode.

[0055] This technical solution has the following characteristics:

[0056] Double breakthrough in treatment synchronization and accuracy: through the R-wave cycle prediction to establish a benchmark timing framework (preset inflation starting point = T×0.25, pressure holding end point = T×0.8, exhaust trigger point = T×0.9) and the dual-mode collaborative mechanism of real-time fine-tuning of ST segment dynamic correction, the timing error bottleneck of traditional single R-wave triggering is broken through.

[0057] Robustness under complex heart rhythms is significantly enhanced: the innovative fusion of sliding window averaging algorithm (3-cycle weighted smoothing) and dynamic threshold expansion technology effectively suppresses noise interference such as atrial fibrillation and premature beats. When the ST segment is transiently abnormal, trigger continuity is maintained through noise smoothing. The dynamic ST confidence assessment mechanism significantly reduces the false trigger rate, and the treatment interruption rate under complex heart rhythms is reduced.

[0058] Wide clinical adaptability: It supports dynamic adaptation of multiple ST segment forms, such as horizontal and downward-sloping ST segments, and has a wide adjustable range of inflation delay time, meeting the individualized treatment needs of different scenarios such as coronary heart disease, myocardial ischemia, and postoperative rehabilitation.

[0059] It has a dynamic fault-tolerant mechanism: if ST segment recognition fails within three consecutive cardiac cycles, it will prioritize switching to cycle prediction mode. Only when cycle prediction fails will R wave triggering be enabled, and a two-level audio-visual warning (flashing yellow light + buzzer alarm) will be triggered simultaneously to ensure treatment continuity and greatly reduce clinical interruption rate.

[0060] Fast-response exhaust design: High-speed solenoid valve is used to achieve synchronous exhaust at the end of diastole, avoiding blood flow backflow and maximizing boosting efficiency.

[0061] Example 2

[0062] like Figure 1 and Figure 2 As shown, this embodiment discloses an adaptive control method based on ST segment correction, which is applied to the adaptive control system based on ST segment correction described in Example 1, and includes the following steps:

[0063] S1, real-time acquisition and filtering of electrical signals, calculation of average period T by R wave detection, and generation of reference timing parameters;

[0064] S2. Extract the ST segment waveform through baseline correction and determine the starting point and offset according to the slope threshold method;

[0065] S3. Calculate the ST confidence score. When the ST confidence score is ≥80, the inflation point is equal to the ST starting point + 5ms. When the ST confidence score is 40≤ and the ST confidence score is <80, the ST signal is preferentially waited for within 10ms before the preset inflation starting point. If the ST signal times out, the cycle prediction mode is switched. When the ST confidence score is <40, the preset inflation starting point is used forcibly.

[0066] S4. If persistent ST segment abnormalities are detected, a sliding window averaging algorithm is used to optimize the data.

[0067] S5: When ST recognition fails for three consecutive cardiac cycles, the system switches to the cycle prediction mode. Only when the cycle prediction mode fails, the R wave trigger mode is enabled and an alarm is issued.

[0068] S6. Drive the multi-channel airbag to perform inflation and deflation according to the trigger signal, synchronizing the cardiac diastole.

[0069] This technical solution focuses on real-time assessment and personalized control of external counterpulsation to improve lower limb blood perfusion based on multi-physiological signal analysis. The signal processing module provides key input signals for the hybrid triggering mechanism through dynamic filtering and feature analysis. This means that after the embedded processor acquires the electrical signals in real time, it first uses an adaptive filtering algorithm to eliminate baseline drift and high-frequency noise. Subsequently, the cycle prediction unit calculates the average cycle T based on the R-wave intervals of the previous 3-5 cardiac cycles to generate baseline timing parameters. The preprocessed signals are simultaneously transmitted to the ST segment feature extraction unit, which uses baseline correction technology to align isoelectric lines and combines the slope threshold method to accurately determine the ST segment origin, offset, and morphological characteristics. The ST confidence score is simultaneously calculated, providing dynamic parameters and reliability assessment for trigger control. The trigger control module implements intelligent hybrid triggering logic based on the ST confidence score and feature parameters. This means the system dynamically decides based on the ST confidence value: when the confidence level is ≥80, the precise "ST start point + 5ms" is directly used as the inflation point; when the confidence level is between 40 and 80, the system prioritizes waiting for a valid ST signal within 10ms before the preset inflation start point, seamlessly switching to the baseline timing generated by cycle prediction if the timeout exceeds; when the confidence level is <40, the preset inflation start point is forcibly adopted. The delay calculation unit further dynamically fine-tunes the inflation delay time based on ST segment morphological characteristics (such as offset amplitude and slope), and the timing optimization unit, combined with the hemodynamic model, sets an adjustable time window after the ST start point to ensure that the airbag action is strictly synchronized with the diastolic blood flow boost demand. When the abnormality adaptation module detects persistent ST segment abnormalities or recognition failures, it activates a multi-level fault tolerance mechanism. This means that when the anti-interference unit detects ST signal instability, it immediately switches to a sliding window averaging algorithm, performing weighted smoothing on the ST data of the last three cycles to suppress transient noise. If the dynamic fault-tolerant unit detects ST recognition failure for three consecutive cardiac cycles, it prioritizes switching to cycle prediction mode to maintain treatment. Only if cycle prediction also fails does the final R-wave trigger mode backup logic activate, simultaneously triggering a two-stage audible and visual warning (flashing yellow light + beeping alarm). This ensures that in extreme noise or signal loss scenarios, the system can maintain treatment continuity through a cascade of backup strategies and prompt medical staff intervention through an early warning mechanism. After receiving the final trigger command, the execution module uses a multi-channel driver circuit to time-share the movement of the calf and thigh airbags. In early diastole, the airbags are synchronously inflated according to a preset pressure gradient. Pressure maintenance ends at the pressure hold end point (T×0.8) generated by cycle prediction or the equivalent time determined by ST dynamic assessment. Rapid deflation occurs via a high-speed solenoid valve at the deflation trigger point (T×0.9) or at the end of diastole. A pressure feedback unit integrated into the embedded processor monitors airbag pressure fluctuations in real time and dynamically compensates for inflation duration if local pressure exceeds the limit. This means that the system not only achieves precise closed-loop control of the three stages of inflation / pressure maintenance / exhaust, maximizing blood flow pressurization efficiency, but also effectively avoids the risk of local tissue ischemia damage.

[0070] The key program pseudo code is as follows:

[0071]

[0072]

[0073]

[0074]

[0075]

[0076] This technical solution is a control method for an adaptive external counterpulsation device based on ECG ST segment correction, involving components such as a signal processing module, a trigger control module, an abnormal adaptation module, and an execution module. It realizes functions such as R-wave period reference framework construction, dynamic ST segment correction, adaptive fault tolerance for abnormal heart rhythms, and multi-channel precise control. It has significant advantages such as high therapeutic synchronization, strong anti-interference ability, wide clinical adaptability, and excellent safety.

[0077] Example 3

[0078] The present embodiment discloses an adaptive control device based on ST segment correction, including a device body, in which the adaptive control system based on ST segment correction described in Example 1 is arranged.

[0079] The modular system packaging design (all modules are integrated into the device body) enables plug-and-play, and the multi-channel airbag drive interface supports personalized configuration to meet different needs.

[0080] The above is only a preferred specific implementation method of this embodiment, but the protection scope of this embodiment is not limited to this. Any technician familiar with this technical field can make equivalent replacements or changes based on the technical solution and inventive concept of this embodiment within the technical scope disclosed in this embodiment, and they should be covered by the protection scope of this embodiment.

Claims

1. An adaptive control system based on ST segment correction, characterized in that: Including signal processing module, trigger control module, abnormal adaptation module and execution module, The signal processing module is used to pre-process the electrical signal and extract the ST segment features, determine the starting position, offset and morphological characteristics of the ST segment, and perform confidence assessment based on the stability of the ST segment slope; The trigger control module dynamically calculates the balloon inflation delay time based on the ST segment morphological characteristics, sets an adjustable time window after the ST segment onset, and implements a hybrid trigger mode; The abnormality adaptation module is used to immediately switch to the sliding window averaging algorithm to eliminate transient noise interference when a persistent ST segment abnormality is detected. If ST segment recognition fails, it will prioritize switching to the cycle prediction mode. Only when the cycle prediction fails will the R wave trigger mode be activated as a backup logic and an alarm will be issued at the same time. The execution module is used to receive the inflation instruction from the trigger control module in real time and control the inflation and deflation of the airbag.

2. The adaptive control system based on ST segment correction according to claim 1, characterized in that: The signal processing module includes an adaptive filtering unit, a cycle prediction unit and an ST segment feature extraction unit. Adaptive filtering unit: eliminates baseline drift and high-frequency noise by dynamically adjusting filter parameters; Cycle prediction unit: calculates the average cardiac cycle T through the R wave interval of the previous 3-5 cardiac cycles and generates benchmark timing parameters; ST segment feature extraction unit: uses baseline correction technology to align isoelectric lines, and combines the slope threshold method to accurately determine the ST segment starting position, offset and morphological characteristics, while performing confidence assessment based on the ST segment slope stability.

3. The adaptive control system based on ST segment correction according to claim 2, characterized in that: The trigger control module includes a delay calculation unit, a timing optimization unit and a mode switching unit. Delay calculation unit: dynamically calculates the airbag inflation delay time based on ST segment morphological characteristics; Timing optimization unit: Sets an adjustable time window after the ST segment starting point in combination with the hemodynamic model; Mode switching unit: executes mixed trigger mode, when ST confidence ≥ 80: inflation point = ST starting point + 5ms; when 40≤ST confidence < 80: wait for ST signal within 10ms before the preset inflation starting point, and switch to periodic prediction mode if timeout; when ST confidence < 40: force the use of preset inflation starting point.

4. The adaptive control system based on ST segment correction according to claim 3, characterized in that: The abnormal adaptation module includes an anti-interference unit and a dynamic fault-tolerant unit. Anti-interference unit: When a persistent ST segment abnormality is detected, it immediately switches to the sliding window averaging algorithm to perform weighted smoothing on the ST segment data of the last three cycles to eliminate transient noise interference; Dynamic fault-tolerant unit: When ST segment recognition fails within three consecutive cardiac cycles, the system will switch to the cycle prediction mode first. Only when the cycle prediction mode fails will the R wave trigger mode be enabled as a backup logic and an alarm will be issued at the same time.

5. The adaptive control system based on ST segment correction according to claim 4, characterized in that: Under the R-wave trigger mode backup logic, the dynamic fault-tolerant unit automatically switches to the R-wave trigger mode after the cycle prediction mode fails; the system continuously monitors the ST segment status and re-enables the ST segment correction mode when the ST confidence level is ≥60 for 5 consecutive cycles.

6. The adaptive control system based on ST segment correction according to claim 4, characterized in that: In the hybrid trigger mode, the sliding window averaging algorithm smoothes the ST segment data of three consecutive cardiac cycles; if a valid ST signal is detected within 10 ms before the preset inflation start point, the cycle prediction timing is covered.

7. The adaptive control system based on ST segment correction according to claim 4, characterized in that: The reference timing parameters are: preset inflation starting point = T×0.25, pressure holding end point = T×0.8, exhaust trigger point = T×0.

9.

8. An adaptive control method based on ST segment correction, applied to the adaptive control system based on ST segment correction according to any one of claims 1 to 7, characterized in that: The steps include: S1, real-time acquisition and filtering of electrical signals, calculation of average period T by R wave detection, and generation of reference timing parameters; S2. Extract the ST segment waveform through baseline correction and determine the starting point and offset according to the slope threshold method; S3. Calculate the ST confidence score. When the ST confidence score is ≥80, the inflation point is equal to the ST starting point + 5ms. When the ST confidence score is 40≤ and the ST confidence score is <80, the ST signal is preferentially waited for within 10ms before the preset inflation starting point. If the ST signal times out, the cycle prediction mode is switched. When the ST confidence score is <40, the preset inflation starting point is used forcibly. S4. If persistent ST segment abnormalities are detected, a sliding window averaging algorithm is used to optimize the data. S5: When ST recognition fails for three consecutive cardiac cycles, the system switches to the cycle prediction mode. Only when the cycle prediction mode fails, the R wave trigger mode is enabled and an alarm is issued. S6. Drive the multi-channel airbag to perform inflation and deflation actions according to the trigger signal.

9. The adaptive control method based on ST segment correction according to claim 8, characterized in that: The cycle prediction mode calculates the average cardiac cycle T through the R wave intervals of the previous 3-5 cardiac cycles and generates reference timing parameters.

10. An adaptive control device based on ST segment correction, characterized in that: The device comprises a device body, wherein the adaptive control system based on ST segment correction according to any one of claims 1 to 7 is arranged in the device body.

Images