Intelligent cardio-pulmonary resuscitation equipment based on physical sign information fusion and analysis system thereof

Abstract

The invention relates to the technical field of medical first-aid equipment, and provides intelligent cardio-pulmonary resuscitation equipment based on physical sign information fusion, which comprises a physical sign monitoring module which is integrated with an electrocardiogram monitoring unit, an end-tidal carbon dioxide (ETCO2) monitoring unit and a carotid artery blood flow ultrasonic sensor and is used for acquiring electrocardiosignals, ETCO2 concentration and carotid artery pulse data of a patient in real time; the circulation optimization module comprises an in-vitro double-lower-limb pressurization device, promotes lower limb vein blood backflow through periodic air pressure pressurization, and is provided with a pressure sensor for dynamically adjusting the pressurization intensity. Through the intelligent pressing mechanical arm, the pressing track and the springback strength can be dynamically adjusted according to the pressing effective coefficient (CCF), personal errors caused by fatigue of an operator and individual differences are avoided, high standardization and stability of pressing operation are achieved, cardio-pulmonary resuscitation effectiveness is remarkably improved, lower limb vein blood backflow is promoted through periodic air pressure pressurization, and the cardio-pulmonary resuscitation effect is improved. A pressure sensor is arranged to dynamically adjust the pressurization intensity, and the success rate of cardio-pulmonary resuscitation is increased.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical first-aid equipment, in particular to an intelligent cardiopulmonary resuscitation device based on sign information fusion and an analysis system thereof. BACKGROUND

[0002] As a key means for rescuing patients with cardiac arrest, the operation quality of cardiopulmonary resuscitation (CPR) directly affects the survival probability and prognosis of patients. At present, the traditional CPR device mainly relies on manual operation, which has significant defects in actual application. On the one hand, it is difficult for manual compression to ensure the stability and standardization of compression depth and frequency, and human errors caused by operator fatigue and individual differences can easily reduce the effectiveness of cardiopulmonary resuscitation; on the other hand, the injection of emergency drugs such as adrenaline also relies on manual operation, which not only easily interrupts the compression process, but also makes it difficult to ensure the precise 3-5 minute drug interval, delaying the best treatment opportunity.

[0003] In terms of monitoring technology, the existing devices mostly use single monitoring method, which can only monitor single indicators such as electrocardiogram or chest pressure, and lack comprehensive monitoring and feedback of key hemodynamic parameters such as carotid perfusion and venous return. This makes it difficult for medical staff to fully understand the physiological state of patients during resuscitation and to adjust the CPR strategy in a timely manner according to the actual situation. At the same time, after the heart stops, 15%-20% of blood is retained in the lower extremity veins, and the traditional CPR technology cannot actively improve the venous return condition, further affecting the success rate of cardiopulmonary resuscitation.

[0004] In summary, the existing cardiopulmonary resuscitation devices and technologies have deficiencies in operation standardization, comprehensive monitoring and timeliness of treatment, and there is an urgent need for a cardiopulmonary resuscitation device and analysis system that can integrate multiple sign information and achieve intelligent monitoring and precise treatment to improve the quality and efficiency of cardiopulmonary resuscitation and improve patient prognosis. SUMMARY

[0005] In view of the deficiencies of the prior art, the present application provides an intelligent cardiopulmonary resuscitation device based on sign information fusion and an analysis system thereof, which solves the problem of deficiencies in operation standardization, comprehensive monitoring and timeliness of treatment of existing cardiopulmonary resuscitation devices and technologies.

[0006] To achieve the above purpose, the present application realizes the following technical scheme: an intelligent cardiopulmonary resuscitation device based on sign information fusion, comprising:

[0007] a sign monitoring module: integrating an electrocardiogram monitoring unit, an end-tidal carbon dioxide (ETCO2) monitoring unit and a carotid blood flow ultrasound sensor, for real-time acquisition of patient electrocardiogram signals, ETCO2 concentration and carotid pulse data;

[0008] Circulatory optimization module: contains an in-vitro double lower limb compression device, promotes the return of venous blood in the lower limbs through periodic air pressure compression, and is configured with a pressure sensor to dynamically adjust the compression strength;

[0009] Intelligent drug delivery module: automatic injector with pre-filled adrenaline cartridge, supports preset time interval (3-5 minutes) or dynamic adjustment of drug delivery timing based on ETCO2 concentration;

[0010] Central control module: receives data from the vital sign monitoring module, generates control instructions through a fusion algorithm, synchronously coordinates the compression rhythm of the circulatory optimization module and the injection action of the intelligent drug delivery module, and ensures that the chest compression interruption time is ≤10 seconds.

[0011] Preferably, the carotid artery blood flow ultrasonic sensor is coupled with an infrared spectrum analyzer to calculate the cerebral perfusion index (CPI) in real time; the ECG monitoring unit has an AI classifier built-in to automatically identify asystole, pacing signals, and shockable rhythms, and trigger alarms; the data sampling frequency is ≥100Hz, and is transmitted to the central control module through Bluetooth 5.0.

[0012] Preferably, the in-vitro double lower limb compression device uses a pneumatic array air bag, the compression frequency (0.5-1Hz) is synchronized with the diastolic period of the chest compression machine (LUCAS), and is configured with a pressure feedback closed loop to automatically increase the compression strength when the carotid artery blood flow rate is below a threshold value, supports linkage with an impedance threshold device (ITD) to enhance the chest cavity negative pressure during the compression rebound period.

[0013] Preferably, the automatic injector integrates an accelerometer to predict the compression pause period through the motion trajectory of the chest compression machine, complete the injection within a ≤5 second window, and has a dose self-adaptive algorithm: automatically calculates the child adrenaline dose (0.01mg / kg) after inputting the body weight, with an error rate of ≤±2%, wherein the cartridge is equipped with an RFID tag to record the drug delivery time, dose and upload to the medical database.

[0014] Preferably, the central control module uses a reinforcement learning algorithm, takes the ETCO2 rising slope and carotid artery pulse amplitude as the reward function, and dynamically optimizes the compression and drug delivery strategy; the output end is connected to the ECMO device, which automatically starts the venous-arterial shunt mode when ROSC fails; and the built-in emergency protocol prioritizes triggering the defibrillator discharge when ventricular fibrillation is detected, and delays the compression action for 3 seconds after defibrillation.

[0015] Preferably, the intelligent cardiopulmonary resuscitation equipment analysis system based on vital sign information fusion includes: a multi-source vital sign acquisition module, a dynamic decision analysis module, and a closed-loop execution control module;

[0016] The multi-source vital sign acquisition module is used to acquire vital sign data in real time, wherein the vital sign data includes:

[0017] Compression sign: Chest compression depth, frequency and recoil status are obtained by acceleration sensor;

[0018] Circulation sign: Pulse strength and blood flow velocity are detected by carotid artery blood flow monitoring unit (ultrasound / infrared sensor);

[0019] Metabolic sign: ETCO2 value and waveform trend are monitored by end-tidal carbon dioxide (ETCO2) sensor;

[0020] Drug response sign: Heart rhythm changes and pulseless electrical activity signals are captured by electrocardiogram monitor;

[0021] Among them, the pressure sensor is integrated into the double lower limb compression device, which can real-time feedback venous return resistance and compression efficiency;

[0022] The dynamic decision analysis module adopts a dynamic decision analysis engine to construct a multi-dimensional fusion algorithm, dynamically correlating the following parameters:

[0023] Compression quality index: Based on compression depth, frequency and ETCO2 rising slope, compression effective coefficient (CCF) is calculated;

[0024] Circulation recovery prediction model: Combining carotid artery blood flow acceleration and ETCO2 peak change, the probability of spontaneous circulation recovery (ROSC) is predicted;

[0025] Drug precise control strategy: According to ETCO2 trend and heart rhythm type, the interval (3-5min dynamic window) and dose (±0.1mg floating threshold) of adrenaline administration are adjusted adaptively;

[0026] Generate lower limb compression instructions: According to venous return resistance data, control the inflation pressure (40-80mmHg), timing (synchronized with chest compression diastolic period) and frequency (0.5-1Hz) of double lower limb compression device;

[0027] Among them, the dynamic decision analysis engine integrates a transfer learning model, which optimizes prediction parameters through historical rescue data, compensates for compression depth based on different body types (BMI>30), and corrects the adrenaline dose formula for patients with acidosis (basic dose x ETCO2 / 15);

[0028] The closed-loop control module is used for linkage intelligent pressing mechanical arm, full-automatic drug injection system and extracorporeal circulation auxiliary interface; the intelligent pressing mechanical arm adjusts pressing track and rebounding force dynamically according to CCF index, the full-automatic drug injection system triggers pre-filled adrenaline injector based on decision instruction, and ensures that drug administration and pressing intermittent period are strictly aligned (error <2s), and the extracorporeal circulation auxiliary interface activates ECMO equipment access protocol when ETCO2 is continuously <10mmHg and carotid blood flow velocity is <5cm / s.

[0029] Preferably, the carotid blood flow monitoring unit adopts a double-probe redundant design, combines with an AI noise reduction algorithm to eliminate chest compression motion artifacts, and outputs a cerebral perfusion index (CPI) in real time, and the CPI formula is:

[0030]

[0031] Wherein K1, K2 are dynamic weight coefficients, and a pressing device strength improvement instruction is triggered when CPI <0.7.

[0032] Preferably, the drug precision control strategy includes an anti-interference mechanism: when a pacing electrical signal or pulseless ventricular tachycardia is detected, drug administration is suspended and electrocardiogram pattern analysis is started to exclude false ROSC signals.

[0033] Preferably, the double lower limb pressing device adopts a zoned pressure control: high pressure pulse (70-80mmHg) is applied to the thigh, and low pressure steady state (40-50mmHg) is maintained in the lower leg, forming a ladder type venous blood driving wave.

[0034] Preferably, an intelligent cardiopulmonary resuscitation method based on sign information fusion includes the following steps:

[0035] S1, synchronously collecting chest compression mechanics data, ETCO2 metabolic waveform, carotid blood flow velocity and lower limb venous pressure;

[0036] S2, inputting multiple source signs into a fusion analysis model to calculate in real time:

[0037] S3, circulation efficiency index (CEI): product of ETCO2 increment and carotid blood flow acceleration;

[0038] S4, drug response urgency: when CEI continuously decreases by >15% for 2 minutes, adrenaline administration interval is shortened to 3 minutes;

[0039] S5, dynamically controlling the double lower limb pressing device: pulse pressure is applied during chest compression rebounding period to drive venous blood backflow;

[0040] S6, if CEI < threshold value and there is pulseless electrical activity, automatically switching to ACD-CPR mode and activating ECMO preparation state.

[0041] The present application provides an intelligent cardiopulmonary resuscitation device based on sign information fusion and an analysis system thereof. The present application has the following beneficial effects:

[0042] 1. The present application receives sign monitoring module data through a central control module, generates control instructions through a fusion algorithm, synchronously coordinates the actions of each module, ensures that the chest compression interruption time is ≤10 seconds, and simultaneously, the intelligent compression mechanical arm can dynamically adjust the compression trajectory and rebound force according to the compression effective coefficient (CCF), avoids human errors caused by operator fatigue and individual differences, realizes the high standardization and stability of compression operation, and significantly improves the effectiveness of cardiopulmonary resuscitation.

[0043] 2. The present application integrates multiple sign monitoring modules such as an electrocardiogram monitoring unit, an end-tidal carbon dioxide (ETCO2) monitoring unit, and a carotid artery blood flow ultrasonic sensor, and collects multi-dimensional sign information such as patient electrocardiogram signals, ETCO2 concentration, and carotid artery pulsation data in real time. The carotid artery blood flow ultrasonic sensor is coupled with an infrared spectrum analyzer to calculate the cerebral perfusion index (CPI), and the AI classifier built-in the electrocardiogram monitoring unit can automatically identify multiple heart rhythms and trigger alarms, comprehensively grasps the physiological state of the patient during resuscitation, and provides sufficient basis for medical personnel to adjust the CPR strategy.

[0044] 3. The present application uses an automatic injector with a pre-filled epinephrine cartridge built-in the intelligent drug delivery module, supports preset time intervals (3-5 minutes) or dynamic adjustment of drug delivery timing based on ETCO2 concentration, integrates an accelerometer in the automatic injector, completes injection within a ≤5 second window by predicting the compression pause period, automatically calculates the epinephrine dose for children according to body weight through the built-in dose self-adaptive algorithm, with an error rate of ≤±2%, and records the drug delivery information for uploading to a medical database, which ensures the timeliness of drug delivery, guarantees the accuracy of the dose, and does not affect the compression process.

[0045] 4. The present application contains an extracorporeal double lower limb pressurizing device through a cycle optimization module, promotes the return of venous blood in the lower limbs through periodic air pressure pressurization, and dynamically adjusts the pressurization intensity through a pressure sensor. The pressurization frequency is synchronized with the diastolic period of the external chest compression machine, and can also be automatically adjusted according to the carotid artery blood flow rate, supports linkage with impedance threshold devices, enhances the chest cavity negative pressure during the compression rebound period, effectively improves the venous return condition, and improves the success rate of cardiopulmonary resuscitation. BRIEF DESCRIPTION OF DRAWINGS

[0046] Figure 1 The present application is a resuscitation device block diagram;

[0047] Figure 2 The present application is a system block diagram. DETAILED DESCRIPTION

[0048] With reference to the accompanying drawings, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0049] Embodiments:

[0050] Please refer to the accompanying Figure 1 and the accompanying Figure 2 The embodiments of the present application provide an intelligent cardiopulmonary resuscitation device based on fusion of vital sign information, which comprises a vital sign monitoring module: an integrated electrocardiogram monitoring unit, an end-tidal carbon dioxide (ETCO2) monitoring unit, and a carotid artery blood flow ultrasound sensor, which are used to collect electrocardiogram signals, ETCO2 concentration, and carotid artery pulse data of a patient in real time, wherein the carotid artery blood flow ultrasound sensor is coupled with an infrared spectrum analyzer to calculate a cerebral perfusion index (CPI) in real time; an AI classifier is built in the electrocardiogram monitoring unit to automatically identify asystole, pacing signals, and shockable rhythms and trigger an alarm; the data sampling frequency is greater than or equal to 100 Hz, and the data is transmitted to a central control module through Bluetooth 5.0; when in use, the patient is placed on a hard plane, a mechanical compression machine is started, and then the vital sign monitoring module is deployed, wherein the carotid artery ultrasound probe is fixed at the bilateral carotid arteries, the carotid artery ultrasound neckband is buckled, and is automatically calibrated, the ETCO2 sensor is connected to the tracheal cannula interface of the breathing circuit, the electrocardiogram electrode patch is attached to the chest of the patient, and the lead is automatically guided.

[0051] The extracorporeal double lower limb compression device in the circulation optimization module is bound with the thigh and the calf through a double lower limb compression belt, is connected to the host air pump port through a pneumatic pipeline, promotes the return of venous blood in the lower limbs through periodic air pressure compression, and is configured with a pressure sensor to dynamically adjust the compression strength, wherein the extracorporeal double lower limb compression device adopts a pneumatic array type air bag, the compression frequency (0.5-1 Hz) is synchronized with the diastolic period of a chest compression machine (LUCAS), a pressure feedback closed loop is configured, the compression strength is automatically increased when the carotid artery blood flow rate is lower than a threshold value, the device is supported to be linked with an impedance threshold device (ITD), and the thoracic cavity negative pressure is enhanced during the compression rebound period.

[0052] The intelligent drug delivery module supports preset time intervals (3-5 minutes) or dynamic adjustment of drug delivery timing based on ETCO2 concentration through an automatic injector with a built-in pre-filled epinephrine cartridge. The automatic injector integrates an accelerometer to predict the compression pause period through the motion trajectory of the chest compression machine, complete the injection within a ≤5 second window, and has a built-in dose self-adaptive algorithm: automatically calculate the child epinephrine dose (0.01 mg / kg) after inputting the body weight, with an error rate ≤±2%. The cartridge is equipped with an RFID tag to record the drug delivery time, dose, and upload to the medical database. During use, the pre-filled cartridge is inserted into the injector card slot, the infusion line is connected to the patient's venous access through the IO needle, and the accelerometer is adsorbed to the surface of the mechanical compression machine (LUCAS).

[0053] The central control module receives data from the vital sign monitoring module, generates control instructions through a fusion algorithm, synchronously coordinates the compression rhythm of the cyclic optimization module and the injection action of the intelligent drug delivery module, and ensures that the chest compression interruption time is ≤10 seconds. The central control module uses a reinforcement learning algorithm, takes the ETCO2 rising slope and carotid artery pulse amplitude as the reward function, and dynamically optimizes the compression and drug delivery strategy. The output end is connected to the ECMO device, which automatically starts the venous-arterial shunt mode when ROSC fails. The built-in emergency protocol: when ventricular fibrillation is detected, the defibrillator discharge is triggered first, and the compression action is delayed for 3 seconds after defibrillation.

[0054] The use process of the device includes:

[0055] Step 1: Place the patient on a hard plane, start the mechanical compression machine (LUCAS), fix the carotid ultrasound probe at the bilateral carotid arteries, buckle the carotid ultrasound neck strap, and automatically calibrate. Connect the ETCO2 sensor to the endotracheal tube interface of the breathing circuit, paste the electrocardiogram electrode on the patient's chest, and automatically lead the lead.

[0056] Step two: bind through double lower limb compression belts and large thighs, connect the host air pump port through the pneumatic pipeline, promote lower limb venous blood return through periodic air pressure compression, and configure a pressure sensor to dynamically adjust the compression intensity. The lower limb compression device frequency is 0.5-1Hz, synchronized with the diastolic period of the compression machine, and the pressure gradient is 120mmHg at the distal end (ankle) → 80mmHg at the proximal end (thigh), simulating the principle of "muscle pump". At the same time, the central control screen displays the carotid blood flow rate, compression intensity, and venous return efficiency in real time.

[0057] Step three, no interruption collaborative drug delivery, automatic injector for adults, 1mg epinephrine every 3-5 minutes, immediately when ETCO2 concentration <10mmHg and slope decreases. The central module issues instructions during the compression rebound period, the injector injects the drug through the pre-filled cartridge, with an error ≤±2%, and the drug delivery is recorded synchronously.

[0058] Step four, emergency treatment, such as ventricular fibrillation: ECG AI alarm (ventricular fibrillation! Suggest defibrillation), central module immediately suspends compression / drug, triggers defibrillator charging, automatically resumes circulation optimization 3 seconds after defibrillation, if carotid artery pulse + ETCO2 rapid rise > 40mmHg for 10 seconds in a row→ device enters perfusion maintenance mode.

[0059] The intelligent cardiopulmonary resuscitation device analysis system based on sign information fusion includes:

[0060] The multi-source sign acquisition module is used for real-time acquisition of sign data, wherein the sign data includes:

[0061] Compression signs: chest compression depth, frequency and rebound state are obtained through an acceleration sensor;

[0062] Circulatory signs: pulsation intensity and blood flow velocity are detected through a carotid artery blood flow monitoring unit (ultrasound / infrared sensor);

[0063] Metabolic signs: ETCO2 value and waveform trend are monitored through an end-tidal carbon dioxide (ETCO2) sensor;

[0064] Drug response signs: heart rhythm changes and pulseless electrical activity signals are captured through an electrocardiograph;

[0065] Among them, the pressure sensor is integrated into the double lower limb compression device, which can real-time feedback venous return resistance and compression efficiency, the carotid artery blood flow monitoring unit adopts a double-probe redundancy design, combines with an AI noise reduction algorithm to eliminate chest compression motion artifacts, and real-time outputs a cerebral perfusion index (CPI), and the CPI formula is:

[0066]

[0067] Wherein K1, K2 are dynamic weight coefficients, and the CPI <0.7 triggers a compression device strength improvement instruction.

[0068] The dynamic decision analysis module adopts a dynamic decision analysis engine to construct a multi-dimensional fusion algorithm, and dynamically correlates the following parameters:

[0069] Compression quality index: based on compression depth, frequency and ETCO2 rise slope, a compression effective coefficient (CCF) is calculated;

[0070] Circulatory recovery prediction model: combining carotid artery blood flow acceleration and ETCO2 peak value change, a spontaneous circulation recovery (ROSC) probability is predicted;

[0071] Drug precision control strategy: According to the ETCO2 trend and the type of cardiac rhythm, the interval (3-5 min dynamic window) and dose (±0.1 mg floating threshold) of adrenaline administration are adaptively adjusted. The drug precision control strategy includes an anti-interference mechanism: when a pacing electrical signal or pulseless ventricular tachycardia is detected, drug administration is suspended and electrocardiographic pattern analysis is initiated to exclude false ROSC signals;

[0072] Lower extremity compression instruction generation: According to the venous return resistance data, the inflation pressure (40-80 mmHg), timing (synchronized with the diastolic period of chest compression) and frequency (0.5-1 Hz) of the double lower extremity compression device are controlled. The double lower extremity compression device uses zoned pressure control: high pressure pulses (70-80 mmHg) are applied to the thigh, and low pressure steady state (40-50 mmHg) is maintained in the lower leg, forming a stepped venous blood drive wave;

[0073] The dynamic decision analysis engine integrates a transfer learning model to optimize prediction parameters based on historical rescue data, and a compression depth compensation coefficient for different body types (BMI>30), and an adrenaline dose correction formula for acidosis patients (basic dose x ETCO2 / 15);

[0074] The closed-loop execution control module is used to link the intelligent compression mechanical arm, the fully automatic drug injection system and the extracorporeal circulation auxiliary interface. The intelligent compression mechanical arm dynamically adjusts the compression trajectory and rebound force according to the CCF index. The fully automatic drug injection system triggers the pre-filled adrenaline injector based on the decision instruction to ensure that the drug administration is strictly aligned with the compression interval (error <2s). When ETCO2 is continuously <10 mmHg and carotid blood flow velocity is <5 cm / s, the extracorporeal circulation auxiliary interface activates the ECMO device access protocol.

[0075] An intelligent cardiopulmonary resuscitation method based on the fusion of physical information, comprising the following steps:

[0076] S1, synchronously collecting chest compression mechanics data, ETCO2 metabolic waveform, carotid blood flow velocity and lower extremity venous pressure;

[0077] S2, inputting multiple source signs into a fusion analysis model to calculate in real time:

[0078] S3, cycle efficiency index (CEI): the product of ETCO2 increment and carotid blood flow acceleration;

[0079] S4, drug response urgency: when CEI decreases by >15% for 2 consecutive minutes, shorten the adrenaline administration interval to 3 minutes;

[0080] S5, dynamically control the double lower extremity compression device: apply pulse pressure during the rebound period of chest compression to drive venous blood return;

[0081] S6. If CEI < threshold and there is no pulsatile electrical activity, automatically switch to ACD-CPR mode and activate ECMO readiness state.

[0082] While embodiments of the application have been shown and described, it is to be understood that the embodiments described are merely divergences of the principles and spirit of the application and that numerous modifications, changes, substitutions, and alterations can be made thereto without departing from the spirit and scope of the application as defined by the appended claims and their equivalents.

Claims

1. An intelligent cardiopulmonary resuscitation device based on vital sign information fusion, characterized in that, include: Vital signs monitoring module: integrates an electrocardiogram monitoring unit, an end-tidal carbon dioxide (ETCO2) monitoring unit, and a carotid artery blood flow ultrasound sensor to collect patients' electrocardiogram signals, ETCO2 concentration, and carotid artery pulsation data in real time; Circulation optimization module: Includes an external dual lower limb compression device, which promotes venous blood return in the lower limbs through periodic pneumatic compression, and is equipped with a pressure sensor to dynamically adjust the compression intensity; Intelligent drug delivery module: An autoinjector with a built-in pre-filled epinephrine cartridge, which supports dynamic adjustment of drug delivery timing based on preset time intervals (3-5 minutes) or ETCO2 concentration; Central control module: Receives data from the vital signs monitoring module, generates control commands through a fusion algorithm, and synchronously coordinates the pressurization rhythm of the cycle optimization module and the injection action of the intelligent drug delivery module to ensure that the chest compression interruption time is ≤10 seconds.

2. The intelligent cardiopulmonary resuscitation device based on vital sign information fusion according to claim 1, characterized in that, The carotid artery blood flow ultrasound sensor is coupled with an infrared spectral analyzer to calculate the cerebral perfusion index (CPI) in real time; the electrocardiogram monitoring unit has a built-in AI classifier to automatically identify pulseless ventricular tachycardia, pacing signals and shockable rhythms, and trigger an alarm; the data sampling frequency is ≥100Hz and is transmitted to the central control module via Bluetooth 5.

0.

3. The intelligent cardiopulmonary resuscitation device based on vital sign information fusion according to claim 1, characterized in that, The external dual lower limb compression device uses a pneumatic array airbag, and the compression frequency (0.5-1Hz) is synchronized with the diastolic phase of the chest compression machine (LUCAS). It is equipped with a pressure feedback closed loop, which automatically increases the compression intensity when the carotid artery blood flow velocity is lower than the threshold. It supports linkage with the impedance threshold device (ITD) to enhance the negative pressure in the thoracic cavity during the compression rebound phase.

4. The intelligent cardiopulmonary resuscitation device based on vital sign information fusion according to claim 1, characterized in that, The automatic injector integrates an accelerometer to predict the compression interval based on the movement trajectory of the chest compression machine, and completes the injection within a ≤5-second window. It has a built-in dose adaptive algorithm: after inputting the weight, it automatically calculates the child's adrenaline dose (0.01mg / kg) with an error rate of ≤±2%. The cartridge is equipped with an RFID tag to record the administration time and dosage and upload it to the medical database.

5. The intelligent cardiopulmonary resuscitation device based on vital sign information fusion according to claim 1, characterized in that, The central control module employs a reinforcement learning algorithm, using the ETCO2 rise slope and carotid artery pulsation amplitude as reward functions to dynamically optimize pressurization and drug delivery strategies. The output is connected to the ECMO device, which automatically initiates the venous-arterial shunt mode when ROSC fails. It has a built-in emergency protocol: when ventricular fibrillation is detected, the defibrillator is triggered first, and the pressurization action is delayed until 3 seconds after defibrillation.

6. An intelligent cardiopulmonary resuscitation (CPR) device analysis system based on vital sign information fusion, using the intelligent CPR device based on vital sign information fusion as described in any one of claims 1-5, characterized in that, include: Multi-source vital sign acquisition module, dynamic decision analysis module, and closed-loop execution control module; The multi-source vital sign acquisition module is used to acquire vital sign data in real time, wherein the vital sign data includes: Chest compression signs: Chest compression depth, frequency, and rebound status are obtained using an accelerometer; Circulatory signs: Pulsation intensity and blood flow velocity were detected using a carotid artery blood flow monitoring unit (ultrasound / infrared sensor); Metabolic signs: ETCO2 values ​​and waveform trends were monitored using an end-tidal carbon dioxide (ETCO2) sensor; Drug response signs: changes in heart rhythm morphology and pulseless electrical activity signals are captured by electrocardiogram monitoring; Among them, a pressure sensor is integrated into the dual lower limb compression device to provide real-time feedback on venous return resistance and compression efficiency; The dynamic decision analysis module uses a dynamic decision analysis engine to construct a multi-dimensional fusion algorithm, dynamically associating the following parameters: Compression quality indicators: The compression effectiveness factor (CCF) is calculated based on compression depth, frequency, and ETCO2 rise slope. Circulation recovery prediction model: Integrating changes in carotid artery blood flow acceleration and peak ETCO2, predicting the probability of spontaneous circulation recovery (ROSC); Precise drug regulation strategy: Adaptively adjust the adrenaline dosing interval (3-5 min dynamic window) and dose (±0.1 mg floating threshold) based on ETCO2 trends and heart rhythm type; Generate lower limb compression commands: Based on venous return resistance data, control the inflation pressure (40-80 mmHg), timing (synchronized with the diastolic phase of chest compressions), and frequency (0.5-1 Hz) of the dual lower limb compression devices; Among them, the dynamic decision analysis engine integrates a transfer learning model, optimizes prediction parameters through historical rescue data, and uses the compression depth compensation coefficient for different body types (BMI>30) to adjust the adrenaline dose for patients with acidosis (basal dose × ETCO2 / 15). The closed-loop execution control module is used to link the intelligent compression robotic arm, the fully automated drug injection system, and the extracorporeal circulation auxiliary interface. The intelligent compression robotic arm dynamically adjusts the compression trajectory and rebound force according to the CCF index. The fully automated drug injection system triggers the pre-filled adrenaline injector based on decision commands to ensure strict alignment between drug administration and compression intervals (error <2s). The extracorporeal circulation auxiliary interface activates the ECMO device access protocol when ETCO2 is consistently <10mmHg and carotid artery blood flow velocity is <5cm / s.

7. The intelligent cardiopulmonary resuscitation equipment analysis system based on vital sign information fusion according to claim 1, characterized in that, The carotid artery blood flow monitoring unit adopts a dual-probe redundancy design, combined with an AI noise reduction algorithm to eliminate chest compression motion artifacts, and outputs the cerebral perfusion index (CPI) in real time. The CPI formula is: K1 and K2 are dynamic weighting coefficients. When CPI < 0.7, the pressure boosting device intensity increase command is triggered.

8. The intelligent cardiopulmonary resuscitation equipment analysis system based on vital sign information fusion according to claim 1, characterized in that, The drug precision control strategy includes an anti-interference mechanism: when a pacing signal or pulseless ventricular tachycardia is detected, drug administration is paused and electrocardiogram morphology analysis is initiated to exclude spurious ROSC signals.

9. The intelligent cardiopulmonary resuscitation equipment analysis system based on vital sign information fusion according to claim 1, characterized in that, The dual lower limb compression device adopts zoned pressure control: high-pressure pulses (70-80 mmHg) are applied to the thighs, while low-pressure steady state (40-50 mmHg) is maintained in the calves, forming a step-like venous blood flow wave.

10. An intelligent cardiopulmonary resuscitation method based on vital sign information fusion, using the intelligent cardiopulmonary resuscitation device analysis system based on vital sign information fusion as described in any one of claims 6-9, characterized in that, Includes the following steps: S1. Simultaneously collect chest compression data, ETCO2 metabolic waveform, carotid artery blood flow velocity, and lower extremity venous pressure. S2. Input multi-source vital signs into the fusion analysis model for real-time calculation: S3, Circulatory Efficiency Index (CEI): The product of ETCO2 increment and carotid artery blood flow acceleration; S4. Drug response urgency: When CEI decreases by more than 15% for 2 consecutive minutes, shorten the epinephrine dosing interval to 3 minutes; S5. Dynamic control of bilateral lower limb compression device: Apply pulse pressure during the rebound phase of external chest compressions to drive venous blood return; S6. If CEI < threshold and pulseless electrical activity is present, automatically switch to ACD-CPR mode and activate ECMO preparation state.

Images