Multi-modal intelligent skin patch for multi-scene cardiopulmonary function monitoring and use method thereof

Abstract

The invention discloses a multi-modal intelligent skin patch for multi-scene cardiopulmonary function monitoring and a use method thereof. The multi-modal intelligent skin patch comprises a layered structure body, a flexible circuit board, a multi-modal sensing assembly, a power management assembly and a wireless communication module. The layered structure body comprises an upper Ecoflex material layer, a lower Ecoflex material layer, a lithium battery and a wireless charging coil, wherein the lithium battery and the wireless charging coil are packaged between the two Ecoflex material layers. A layered structure in which the upper Ecoflex material layer and the lower Ecoflex material layer package the lithium battery and the wireless charging coil is adopted, and a polyimide flexible circuit board and a double-bridge snake-shaped structure sensing wire are combined, so that the equipment is endowed with excellent tensile tolerance and skin fitting property; in cooperation with the sweat-proof adhesiveness of the 3M2477P medical silica gel double-faced adhesive tape and the high conductivity of the high-adhesiveness conductive hydrogel layer, the problems of electrode falling and signal distortion caused by sweating and skin stretching in violent activities such as high-intensity intermittent movement are fundamentally solved, and stable acquisition of electrocardio, PPG and respiratory wave signals is ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cardiopulmonary function monitoring, more particularly to a multi-modal intelligent skin patch for multi-scene cardiopulmonary function monitoring and a use method thereof. BACKGROUND

[0002] Traditional cardiopulmonary function detection methods have certain limitations: conventional electrocardiogram, lung function instrument, etc. are mostly static detection, which can only reflect the function of a single organ (heart or lung) in a quiet state, cannot simulate the synergistic effect of heart-lung-skeletal muscle during daily activities or exercise, and is difficult to explain symptoms such as unexplained dyspnea and decreased exercise tolerance. Although cardiopulmonary exercise test is the gold standard for evaluating overall cardiopulmonary function, it has the following limitations: (1) Limited to people with acute high-risk conditions: exercise load may induce disease deterioration or even endanger life; (2) The accuracy of the results depends on the subject's subjective cooperation: the subject needs to reach a true state of exhaustion; (3) The index lacks absolute specificity and needs to be combined with clinical judgment: CPET indicators are mostly functional indicators, not etiologically specific indicators; (4) High operating conditions and poor comfort: not only do you need to have high-precision gas metabolism analyzers, adjustable load treadmills / power cars, real-time electrocardiogram monitors, and other high-cost equipment, but also medical personnel need to monitor the subject's vital signs in real time, handle any discomfort during exercise, and require professional interpretation of complex indicators. Ordinary medical personnel cannot complete this independently, making it difficult to popularize in primary medical institutions; (5) Single detection, limited scene: unable to fully reflect the dynamic changes in the cardiopulmonary function of the subject under different daily activities, various exercises, or sleep scenarios. SUMMARY

[0003] To solve the above problems, the present application provides a multi-modal intelligent skin patch for multi-scene cardiopulmonary function monitoring and a use method thereof.

[0004] The present application provides a multi-modal intelligent skin patch for multi-scene cardiopulmonary function monitoring, which comprises a layered structure main body, a flexible circuit board, a multi-modal sensing assembly, a power management assembly, and a wireless communication module. The layered structure main body comprises an upper layer of Ecoflex material, a lower layer of Ecoflex material, and a lithium battery and a wireless charging coil encapsulated between the two layers of Ecoflex material. The flexible circuit board substrate is polyimide, and the circuit board is provided with a double-bridge serpentine structure sensing lead. The multi-modal sensor assembly comprises a respiration sensing electrode, two electrocardio sensing electrodes, a PPG sensor, an IMU module, and a respiration wave detection module. The respiration sensing electrode is surface-adhered to a PDMS material layer, and the two electrocardio sensing electrodes are surface-adhered to a high-adhesion conductive hydrogel layer; the lower Ecoflex material layer is cut according to the shape of each electrode and sensor position, so that each electrode and sensor is exposed to directly contact the human skin; the surface of the lower Ecoflex material layer is adhered to a 3M 2477P medical silicone double-sided adhesive tape in a non-sensing area. The power management assembly comprises an NMOS switch, a power path control chip, a low-dropout linear regulator, a low-dropout linear regulator, and a low-dropout linear regulator; the NMOS switch is connected to the power management module through the power path control chip to control the power paths of each module; the wireless communication module comprises a Bluetooth antenna and an NFC coil.

[0005] Preferably, the top layer circuit integration area of the flexible circuit board is integrated with a wireless charging and power supply module, a power management module, an automatic power control module, an NFC dynamic tag module, a power detection module, an electrocardiogram front-end analog module, and a low-power Bluetooth integrated master control module. The electrocardiogram front-end analog module communicates with the MCU through an SPI interface, and is used to collect electrocardio signals and transmit them to the MCU; the IMU module and the respiration wave detection module communicate with the MCU through an I2C interface, and are respectively used to collect electrocardiogram signals and respiration wave signals and transmit them to the MCU; the PPG sensor signal acquisition unit is electrically connected to the electrocardiogram front-end analog module.

[0006] Preferably, in the power management assembly, the output voltage of the low-dropout linear regulator satisfies The low-dropout linear regulator with an output voltage satisfying powers the IMU module and the respiration wave detection module; and the low-dropout linear regulator with an output voltage satisfying powers the PPG sensor light source driving unit. The MCU controls the power switch of the power management assembly through an I2C interface; the power detection module is electrically connected to the MCU, and the MCU collects the remaining capacity Q of the lithium battery in real time and transmits it to the MCU, and the MCU adjusts the power supply strategy of each module based on Q.

[0007] Preferably, the wireless communication module supports BLE mode and NFC mode independent operation; in the BLE mode, the MCU transmits the heart rate HR, the respiration rate RR, and the blood oxygen saturation parameter processed by a conventional filtering algorithm in real time through the Bluetooth antenna; In the NFC mode, the MCU transmits the heart rate HR, the respiration rate RR, and the blood oxygen saturation parameter through the NFC coil at an interval of Update heart rate HR, respiratory rate RR, blood oxygen saturation parameters, single maximum support duration data reading; the MCU controls the skin patch to enter a deep sleep state when detecting no Bluetooth connection within the duration The deep sleep state is only the NFC module and the MCU that remain working. The MCU is also configured with a synchronous data acquisition firmware to trigger the multi-modal sensing assembly to synchronously acquire physiological signals.

[0008] Preferably, a use method of a multi-modal intelligent skin patch for multi-scene cardiopulmonary function monitoring, characterized in that the use method comprises the following steps: S1, device attachment and position calibration: attach the Ecoflex material layer at the lower layer of the skin patch to the skin of the user in the monitoring area in front of the chest close to the heart and lungs through 3M2477P medical silicone double-sided adhesive; after attachment, the MCU acquires initial electrocardiogram signals ECG0 through the electrocardiogram front-end analog module, acquires initial respiratory wave signals RESp0 through the respiratory wave detection module, calculates the signal-to-noise ratio SNRECG of ECG0 and the amplitude ARESP of RESp0; if SNRECG < SNR0 or ARESP < A0, the MCU pushes a position adjustment prompt to the mobile phone through Bluetooth until SNRECG ≥ SNR0 or ARESP ≥ A0. S2, if the skin patch is in a deep sleep state, place the mobile phone close to the NFC coil of the skin patch to capture electromagnetic induction energy to wake up the MCU; the MCU acquires initial motion data ACC0 through the IMU module, and determines the scene in combination with a preset scene threshold: when ACC0 < Ath1, it is determined to be a night sleep apnea monitoring scene; when ACC0 ≤ Ath2, it is determined to be a home rest rehabilitation monitoring scene; when ACC0 ≥ Ath2, it is determined to be a high-intensity intermittent exercise scene; after scene recognition, the MCU configures a power management assembly voltage output strategy based on the scene; S3, the MCU initializes the electrocardiogram front-end analog module through the SPI interface, configures the sampling rate fECG according to the scene and establishes a data transmission link; initializes the IMU module through the I2C interface, configures the sampling frequency fACG according to the scene; initializes the respiratory wave detection module, sets the capacitive coupling detection sensitivity level according to the scene; initializes the PPG sensor, adjusts the output precision of the output voltage stabilizer and the driving current Idrv of the output voltage stabilizer; at the same time, initializes the communication parameters of the Bluetooth antenna and the NFC coil; S4, multi-modal signal acquisition and effectiveness verification: the MCU calls the synchronous data acquisition firmware to trigger each sensing module to synchronously acquire signals according to the initialization parameters; during the acquisition process, the MCU determines the signal effectiveness through multi-modal coordination logic: S5, after the multi-modal signal acquisition and effectiveness verification, selecting the BLE mode or the NFC mode for data transmission, specifically: BLE mode: the MCU adopts a conventional filtering algorithm to process the effective signal, and then calculates the heart rate (HR), respiratory rate (RR), and oxygen saturation (SpO2); NFC mode: at this time, the Bluetooth communication is closed, the current time is synchronized to the NFC dynamic tag module through the NFC function of the mobile phone, and the MCU adopts a parameter calculation algorithm to process the effective signal; S6, after S5, data storage and low-power management are performed: Data storage: in the BLE mode, the raw signal data is stored in the local of the mobile phone APP, and the processed heart rate (HR), respiratory rate (RR), and oxygen saturation (SpO2) are stored in the cloud server; in the NFC mode, the data is first stored in the storage unit of the NFC dynamic tag module, and when the mobile phone reads the data, the relevant parameters are stored in the local, and the raw signal snapshot is uploaded to the cloud, and the cloud data is associated through the unique number of the device.

[0009] Low-power management: in the BLE mode, if no Bluetooth connection is detected for a time T3, and the IMU module determines that the current is a non-motion scene, the MCU will control the NMOS switch to disconnect the power supply of each sensor module, only the NFC module and itself enter deep sleep state; if the IMU module determines that it is a motion scene, it will still be in deep sleep state after T7 time without connection. In the NFC mode, if no data reading interaction is detected for a time T3, the MCU will control to disconnect the power supply of each sensor module, and enter deep sleep state.

[0010] During the deep sleep process, the IMU module will wake up once every T8 time, and continuously collect motion data for T9 time, if the motion data ACCwake is greater than or equal to Ath1, the MCU will wake up each sensor module to restore the monitoring function.

[0011] Preferably, the position calibration of step S1 further includes skin state adaptation: The MCU collects the skin contact impedance through the electrocardio sensing electrode, if the collected skin contact impedance is greater than a preset impedance threshold Zth, the mobile phone will prompt to apply medical conductive gel in the area where the sensing electrode is attached until the skin contact impedance is less than or equal to the impedance threshold.

[0012] The calculation method of the skin contact impedance is: the electrocardiogram front-end analog module outputs a test voltage, and simultaneously collects a test current, and the skin contact impedance is equal to the test voltage divided by the test current.

[0013] Preferably, the scene recognition of step S2 further comprises an artificial correction mechanism: If the scene recognized by the MCU does not match the actual scene of the user, the user can manually select the scene through the mobile phone APP, and the MCU reconfigures the voltage output strategy of the power management component and the parameters of each module after receiving the manual instruction.

[0014] Preferably, the multi-modal signal validity verification of step S4 further comprises the content of abnormal signal supplement: If the single ECG is determined to be invalid, the current R-wave time is estimated by combining the pulse wave peak time of PPG with the historical average, specifically, the current estimated R-wave time is equal to the pulse wave peak time of PPG minus the historical average, thereby supplementing the electrocardiogram signal data.

[0015] If the single RESP is determined to be invalid, the current respiratory rate is calculated by the ACC of the IMU module.

[0016] Preferably, the BLE mode of step S5 further sets a data supplement transmission mechanism: when the Bluetooth connection is interrupted, the skin patch temporarily stores the valid signal data during the interruption in the local cache; If the interruption duration is less than or equal to , after the connection is restored, the MCU will supplement the temporarily stored data to the mobile phone in chronological order, and these supplemented data will be marked with a supplement identifier; if the interruption duration is greater than , the MCU will trigger the encrypted storage of the local cache data, and the next time the Bluetooth is connected, the encrypted storage data will be preferentially supplemented, and after the supplement is completed, the local encrypted cache data will be deleted.

[0017] Preferably, the low-power consumption management of step S6 further comprises the content of adaptive adjustment of the power: the MCU collects the remaining power of the lithium battery in real time through the power detection module.

[0018] When the remaining power is less than the preset power threshold , the MCU will automatically reduce the sampling frequency of each module; when the remaining power is less than the preset power threshold , the MCU only retains the collection functions of the electrocardiogram front-end analog module and the PPG sensor, as well as the NFC communication function, while turning off the IMU module and the respiratory wave detection module; when the remaining power is greater than or equal to the preset power threshold , the MCU will restore the normal sampling frequency and power supply of each module. The size relationship of the three power thresholds is < < .

[0019] Beneficial effects: It adopts the layered structure of the upper Ecoflex material layer, the lower Ecoflex material layer, encapsulates the lithium battery and the wireless charging coil, combines the polyimide flexible circuit board and the double-bridge serpentine structure sensing wire, gives the device excellent tensile resistance and skin adhesion, cooperates with the sweat-proof adhesion of the 3M 2477P medical silicone double-sided adhesive and the high adhesion conductive hydrogel layer, fundamentally solves the electrode falling and signal distortion problems caused by sweating and skin stretching in high-intensity intermittent exercise and other intense activities, and ensures stable acquisition of electrocardio, PPG and respiratory wave signals. Secondly, by integrating multi-modal sensing components and configuring synchronous data acquisition firmware, combining the high-speed transmission of the SPI interface and the low-power consumption characteristics of the I2C interface, synchronous capture and correlation analysis of multi-dimensional signals are realized, the diagnostic accuracy of sleep apnea is significantly improved, and the risk of missed diagnosis of traditional single-mode monitoring is effectively avoided. The power management strategy and the dual-mode communication mechanism enable the device to dynamically adjust the power supply strategy of each module according to the remaining power, and realize data backtracking through offline storage and non-contact reading in the NFC mode in a network-free environment, and the single endurance is prolonged. The intelligent scene adaptive mechanism and multi-modal signal cooperative verification logic ensure the optimization of parameter configuration and signal processing of the device in different application scenarios, and combined with data retransmission and encrypted storage functions, a full-link reliable monitoring system from signal acquisition, processing, transmission to storage is built, providing a precise, continuous and convenient integrated solution for multi-scene cardiopulmonary function monitoring. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 is a schematic diagram of the patch of the present application; Figure 2 is a flowchart of the use method of the present application; DETAILED DESCRIPTION

[0021] As Figure 1 shown: the present application provides a multi-modal intelligent skin patch for multi-scene cardiopulmonary function monitoring, comprising a layered structure main body, a flexible circuit board, multi-modal sensing components, a power management component and a wireless communication module. The layered structure main body comprises an upper Ecoflex material layer, a lower Ecoflex material layer and a lithium battery and a wireless charging coil encapsulated between the two Ecoflex material layers. The lithium battery capacity satisfies C=300mAh, and the voltage satisfies U1=3.7V; the wireless charging coil inductance value satisfies L1=15.8μH, and is adapted to Qi protocol to realize wireless charging of the lithium battery. The flexible circuit board substrate is polyimide, and the double-bridge serpentine structure sensing wire is provided on the circuit board. The double-bridge serpentine structure length satisfies L3=15.91 mm, the width satisfies W1=7.41 mm, the pitch satisfies D1=2.02 mm, is arranged in an upper and lower double layer and is composed of a 270° semicircle, the semicircle ring width satisfies W2=7.62 mm; the circuit board is divided into a top layer circuit integration area and a bottom layer electrode / sensor arrangement area, and the top layer circuit integration area area satisfies S1=51 mm×31 mm; The multi-modal sensing assembly includes a respiration sensing electrode, two electrocardio sensing electrodes, a PPG sensor, an IMU module, and a respiration wave detection module. The respiration sensing electrode and the electrocardio sensing electrode both satisfy a semicircle with a diameter of R=28.9 mm; The respiration sensing electrode is attached with a PDMS material layer, and the two electrocardio sensing electrodes are attached with a high-adhesion conductive hydrogel layer; the lower Ecoflex material layer is cut according to the shapes of the electrodes and the sensors at the positions of the electrodes and the sensors, so that the electrodes and the sensors are exposed to directly contact the human skin; and the surface of the lower Ecoflex material layer is attached with 3M 2477P medical silicone double-sided adhesive tape in a non-sensing area. The power management assembly includes an NMOS switch, a power path control chip, a low-dropout linear regulator, a low-dropout linear regulator, and a low-dropout linear regulator; the NMOS switch is connected with the power management module through the power path control chip to control the power paths of the modules; and the wireless communication module includes a Bluetooth antenna and an NFC coil.

[0022] In the embodiment, the above scheme can realize special effects in a high-intensity intermittent exercise scene: the sweat-resistant adhesion of 3M 2477P medical silicone double-sided adhesive tape and the high-conductivity of the hydrogel are combined, and the stretch resistance of the polyimide double-bridge serpentine structure ensures that the electrodes do not separate from the skin during exercise, the multi-modal sensing assembly can stably collect electrocardio, PPG, and respiration wave signals, and solves the problem of signal drop in the traditional device in the exercise scene.

[0023] It also needs to be explained that the IMU module model is LIS2DH12, the respiratory wave detection module model is FDC2214RGHR, the power path control chip model is ADP196ACBZ-R7, the output voltage meets U2=1.8V of the low dropout linear regulator model is TPS7A0218PDBVR, the output voltage meets U3=3.3V of the low dropout linear regulator model is TLV75733PDBVR, the output voltage meets U4=5V of the low dropout linear regulator model is PAM2401YPADJ, and the NFC module model is ST25DV64K-JFR6D3; the IMU module collects the seismocardiogram signal, the respiratory wave detection module collects the respiratory wave signal, the electrocardio sensing electrode cooperates with the electrocardiogram front-end analog module to collect the electrocardio signal, the PPG sensor collects the chest photoelectric volume signal, and the NFC module transmits the heart rate, RR interval, respiratory rate and blood oxygen parameters.

[0024] As an optional embodiment: the top layer circuit integration area of the flexible circuit board integrates a wireless charging and power supply module, a power management module, an automatic power control module, an NFC dynamic tag module, an electric quantity detection module, an electrocardiogram front-end analog module and a low-power Bluetooth integrated master control module. The electrocardiogram front-end analog module communicates with the MCU through an SPI interface, is used for collecting the electrocardio signal and transmitting to the MCU; the IMU module and the respiratory wave detection module communicate with the MCU through an I2C interface, are respectively used for collecting the electrocardiogram signal and the respiratory wave signal and transmitting to the MCU; and the PPG sensor signal acquisition unit is electrically connected with the electrocardiogram front-end analog module.

[0025] The light source driving unit is electrically connected with the voltage stabilizer meeting ; In the embodiment, the above scheme can realize special effects in the night sleep apnea monitoring scene: the high-speed transmission characteristics of the SPI interface ensure that the electrocardio signal is uploaded in real time, the low-power characteristics of the I2C interface adapt to the low-power demand in the sleep scene, and the two cooperate to make the MCU can synchronously capture the heart rate-RR interval extension+respiratory wave amplitude drop associated signal during sleep apnea, provide multi-dimensional data support for sleep apnea diagnosis, and solve the missed diagnosis problem of traditional single-mode monitoring.

[0026] It should be noted that the electrocardiogram front-end analog module model is MAX30003CTI+, the MCU model is ESP32-PICO-D4, the IMU module model is LIS2DH12, and the respiratory wave detection module model is FDC2214RGHR; the electrocardiogram front-end analog module transmits the electrocardio signal to the MCU through the SPI interface, the IMU module transmits the electrocardiogram signal to the MCU through the I2C interface, the respiratory wave detection module transmits the respiratory wave signal to the MCU through the I2C interface, and the PPG sensor signal acquisition unit receives the thoracic photoelectric volume signal and transmits it to the electrocardiogram front-end analog module. The light source driving unit provides a 5V driving voltage for the PPG sensor to ensure stable output of the light signal.

[0027] As an optional embodiment: in the power management component, the output voltage satisfies The voltage regulator with an output voltage satisfying The voltage regulator with an output voltage satisfying The voltage regulator with an output voltage satisfying The MCU controls the power switch of the power management component through the I2C interface, the power detection module is electrically connected with the MCU, the remaining power Q of the lithium battery is collected in real time and transmitted to the MCU, and the MCU adjusts the power supply strategy of each module based on Q.

[0028] It should be noted that specifically: When , the duty cycle of the IMU module and the respiratory wave detection module is reduced; when , only the electrocardiogram front-end analog module, the PPG sensor, and the NFC module are powered, wherein ; The parameters satisfy , ), the duty cycle of the IMU module and the respiratory wave detection module is reduced from 100% to 50%.

[0029] The above scheme can achieve special effects in the home resting rehabilitation monitoring scene: when the remaining power of the lithium battery (Q<50mAh), only the core monitoring module is powered, and the non-contact data reading of the NFC mode is used, so that the single endurance of the device is extended to 72 hours, meeting the 7-day rehabilitation monitoring needs of discharged patients at home, and solving the problem of short endurance and frequent charging of traditional devices in the home scene.

[0030] The voltage regulator with an output voltage satisfying The voltage regulator with an output voltage satisfying The voltage regulator with an output voltage satisfying The voltage regulator model is PAM2401YPADJ; When powered by a voltage regulator, the ECG front-end analog module stably acquires ECG signals, and the PPG sensor signal acquisition unit stably acquires chest photoplethysmography signals. When the voltage regulator is in operation, the IMU module collects the electrocardiogram signal and the respiratory wave detection module collects the respiratory wave signal; the power detection module transmits the remaining power signal of the lithium battery to the MCU in real time to ensure dynamic adjustment of the power supply strategy.

[0031] Initial ECG signal This is the first 10-second continuous ECG signal collected after the device was attached. This is the first 10-second continuous respiratory wave signal collected after the device was fitted. for Signal-to-noise ratio, preset signal-to-noise ratio threshold It is 5. for amplitude, To preset the respiratory wave amplitude threshold, and The acceleration thresholds are 0.2 and 0.5, respectively. As an optional embodiment: the wireless communication module supports independent operation in BLE mode and NFC mode; in BLE mode, the MCU transmits heart rate (HR), respiratory rate (RR), and blood oxygen saturation parameters processed by conventional filtering algorithms in real time through the Bluetooth antenna; In the NFC mode, the MCU uses the NFC coil at intervals. Update heart rate (HR), respiratory rate (RR), and blood oxygen saturation parameters; maximum duration of single test. Data reading; the MCU detects the duration When there is no Bluetooth connection, the control skin patch enters a deep sleep state, in which only the NFC module and MCU remain working. MCU deep sleep current meets ; The MCU is also configured with synchronous data acquisition firmware to trigger the multimodal sensing components to synchronously acquire physiological signals.

[0032] The parameters satisfy ( =10s), =7h), =30s), the frequency of the physiological signal synchronously acquired by the multimodal sensing components meets (f=250Hz), and the conventional filtering algorithm is 50Hz notch filter + 0.5-100Hz bandpass filter (that is, the filtering function satisfies (H(f)=1), (0.5Hz≤f≤100Hz) and (f≠50Hz); (H(f)=0), other f); The above scheme can achieve special effects in outdoor emergency monitoring scenarios: real-time transmission of BLE mode (delay ≤100 ms) can be used for instant health warning during outdoor sports, and offline storage + contactless reading of NFC mode (without mobile phone Bluetooth connection) can read the historical data within 7 hours by approaching the patch to the mobile phone in a network-free environment, providing key cardiopulmonary function information for outdoor first aid and solving the limitations of traditional devices relying on network / Bluetooth connection.

[0033] The NFC module is ST25DV64K-JFR6D3, and the MCU is ESP32-PICO-D4; after processing the electrocardiogram signal, the respiratory wave signal, and the chest photoplethysmogram signal by the conventional filtering algorithm, the heart rate HR, the respiratory rate RR, and the oxygen saturation SpO2 are calculated; when the synchronous acquisition frequency f=250 Hz, the multi-modal sensing component synchronously acquires the electrocardiogram signal, the seismocardiogram signal, the respiratory wave signal, and the chest photoplethysmogram signal; the parameters updated at intervals T1=10 s in the NFC mode include the RR interval RR{int} derived from the electrocardiogram signal, the oxygen saturation SpO2 derived from the chest photoplethysmogram signal, and the respiratory rate RR derived from the respiratory wave signal.

[0034] As an optional embodiment, the application further includes a use method of the multi-modal intelligent skin patch for cardiopulmonary function monitoring in multiple scenarios; The use method includes the following steps: S1, device attachment and position calibration: the lower Ecoflex material layer of the skin patch is attached to the skin of the user in the monitoring area in front of the chest near the heart and lungs by 3M2477P medical silicone double-sided adhesive; after attachment, the MCU acquires the initial electrocardiogram signal ECG0 by the electrocardiogram front-end analog module and the initial respiratory wave signal RESp0 by the respiratory wave detection module, calculates the signal-to-noise ratio SNRECG of ECG0 and the amplitude ARESP of RESp0; if SNRECG<SNR0 or ARESP<A0, the MCU pushes a position adjustment prompt to the mobile phone through Bluetooth until SNRECG≥SNR0 or ARESP≥A0; Ensure that the respiratory sensing electrode, the electrocardio sensing electrode, and the PPG sensor are in close contact with the skin without air bubbles; S2, if the skin patch is in a deep sleep state, the mobile phone is placed close to the NFC coil of the skin patch to wake up the MCU through electromagnetic induction energy capture; the MCU acquires the initial motion data ACC0 by the IMU module, and determines the scene in combination with the preset scene threshold: when ACC0<Ath1, it is determined as a night sleep apnea monitoring scene; when ACC0≤Ath2, it is determined as a home rest rehabilitation monitoring scene; when ACC0≥Ath2, it is determined as a high-intensity intermittent exercise scene; after scene recognition, the MCU configures the voltage output strategy of the power management component based on the scene. S3, the MCU initializes the electrocardiogram front-end analog module through the SPI interface, configures the sampling rate fECG according to the scene and establishes a data transmission link; initializes the IMU module through the I2C interface, configures the sampling frequency fACG according to the scene; initializes the respiratory wave detection module, sets the capacitive coupling detection sensitivity level according to the scene; initializes the PPG sensor, adjusts the output precision of the output voltage stabilizer and the driving current Idrv of the output voltage stabilizer; and initializes the communication parameters of the Bluetooth antenna and the NFC coil at the same time; It should be noted that in the sleep scene, the electrocardiogram front-end analog module sampling rate is 125Hz, the IMU module sampling frequency is 50Hz, and the respiratory wave detection module capacitive coupling detection sensitivity level is 3; in the resting scene, the electrocardiogram front-end analog module sampling rate is 200Hz, the IMU module sampling frequency is 80Hz, and the respiratory wave detection module capacitive coupling detection sensitivity level is 4; in the HIIT scene, the electrocardiogram front-end analog module sampling rate is 250Hz, the IMU module sampling frequency is 100Hz, and the respiratory wave detection module capacitive coupling detection sensitivity level is 5; the Bluetooth antenna communication baud rate is 9600bps, and the NFC coil communication distance is ≤5cm; S4, multi-modal signal acquisition and effectiveness verification: the MCU calls the synchronous data acquisition firmware to trigger each sensing module to synchronously collect signals according to the initialization parameters; during the collection process, the MCU determines the signal effectiveness through multi-modal coordination logic: The specific calculation steps are as follows: first, determine the effectiveness of ECG and PPG signals. First find the R-wave peak occurrence time of the ECG signal and the pulse wave peak occurrence time of the PPG signal, then calculate the time difference between the two time points, which is equal to the PPG pulse wave peak time minus the ECG R-wave peak time. If the calculated time difference is within the interval of the pre-set minimum allowed time difference to the maximum allowed time difference, it is determined that both the ECG signal and the PPG signal are valid.

[0035] Then determine the effectiveness of RESP and ACC signals. The periodical trend of the RESP signal and the chest fluctuation trend reflected by the ACC signal are analyzed respectively, and then the consistency of the two trends is calculated. If the calculated trend consistency reaches or exceeds the pre-set threshold P0, it is determined that both the RESP signal and the ACC signal are valid.

[0036] Finally, the invalid signal processing mechanism. In the above determination process, as long as the number of any one signal (one of ECG, PPG, RESP, and ACC) being continuously determined as invalid reaches the pre-set N1 times, the corresponding module will be reinitialized; S5, after the multi-modal signal acquisition and effectiveness verification, the BLE mode or the NFC mode is selected for data transmission, and specifically: BLE mode: the MCU adopts a conventional filtering algorithm to process the effective signal, and then calculates the heart rate (HR), respiratory rate (RR), and blood oxygen saturation (SpO2); Among them, the calculation method of heart rate is 60 divided by the average value of 5 consecutive RR intervals; the calculation method of respiratory rate is 60 divided by the average value of 3 consecutive respiratory periods; the blood oxygen saturation is calculated based on the amplitude ratio of the red light and infrared light of PPG, and the processed data is transmitted to the mobile phone through the Bluetooth antenna, and the transmission format is to send a packet of data every interval T4 time, and each packet of data contains a time stamp, a scene identifier and a signal effectiveness mark; NFC mode: at this time, the Bluetooth communication is closed, the current time is synchronized to the NFC dynamic tag module through the NFC function of the mobile phone, and the MCU adopts a parameter calculation algorithm to process the effective signal; The heart rate, RR interval, respiratory rate, and blood oxygen saturation are updated every T1 time, and the heart rate scatter plot and the RR scatter plot are generated, and the data is stored in the storage unit of the NFC dynamic tag module as a segment every interval T5 time; S6, after S5, data storage and low-power management are performed: Data storage: in the BLE mode, the original signal data is stored in the local of the mobile phone APP, and the processed heart rate (HR), respiratory rate (RR), and blood oxygen saturation (SpO2) are stored in the cloud server; in the NFC mode, the data is first stored in the storage unit of the NFC dynamic tag module, and when the mobile phone reads the data, the relevant parameters are stored in the local, and the original signal snapshot is uploaded to the cloud, and the cloud data is associated through the unique number of the device.

[0037] Low-power management: in the BLE mode, if no Bluetooth connection is detected for a time of T3, and the IMU module determines that the current is a non-motion scene, the MCU will control the NMOS switch to disconnect the power supply of each sensor module, and only the NFC module and itself will enter a deep sleep state; if the IMU module determines that it is a motion scene, it will still be in a deep sleep state after a time of T7 without connection. In the NFC mode, if no data reading interaction is detected for a time of T3, the MCU will control to disconnect the power supply of each sensor module, and enter a deep sleep state.

[0038] During the deep sleep process, the IMU module will wake up every T8 time, and continuously collect motion data for a time of T9, if the motion data ACCwake is greater than or equal to Ath1, the MCU will wake up each sensor module, and restore the monitoring function.

[0039] The relevant parameters are as follows: a signal-to-noise ratio SNR0 is 5 decibels, an A0 is 0.1 volts, an Ath1 is 0.2g (g is the acceleration of gravity), an Ath2 is 0.5g, a voltage deviation ΔU is plus or minus 1%, a driving current Idrv is 20 milliamps, a Δtmin is 0.1 seconds, a Δtmax is 0.5 seconds, a P0 is 0.8, an N1 is 3, a T1 is 10 seconds, a T4 is 1 second, a T5 is 10 seconds, a T3 is 30 seconds, a T7 is 60 seconds, a T8 is 5 minutes, and a T9 is 10 seconds; It also needs to be explained that the unique effect is produced in three special scenarios: in the HIIT scenario, the flexible circuit board of the device + the high sampling rate initialization of the method, which realizes the suppression of motion artifact (the signal-to-noise ratio is improved to more than 8dB); In the sleep scenario, the low-power power management of the device + the intermittent power supply of the method, which realizes 7 hours of continuous monitoring and power consumption ≤80mAh; In the resting rehabilitation scenario, the dual-mode communication of the device + the cloud storage of the method, which realizes real-time viewing of the patient's heart and lung function trend by the doctor remotely, and solves the problems of poor adaptation and data isolation in the traditional monitoring scenario. The model of the electrocardiogram front-end analog module is MAX30003CTI+, the model of the IMU module is LIS2DH12, the model of the respiratory wave detection module is FDC2214RGHR, the model of the NFC module is ST25DV64K-JFR6D3, and the model of the MCU is ESP32-PICO-D4; In the multi-modal collaborative logic, the trend consistency is calculated by the Pearson correlation coefficient, the PPG red light wavelength is 660nm, the infrared light wavelength is 940nm, the storage unit capacity of the NFC dynamic tag module is 64KB, and the device unique number is the built-in UUID of the MCU; The initial electrocardiogram signal (ECG0) and the initial respiratory wave signal (RESP0) collected in step S1 are used to calibrate the fitting position, the signals collected synchronously in step S4 include the electrocardiogram signal ECG, the acceleration signal ACC, the respiratory wave signal RESP, and the PPG signal, and the parameter calculation algorithm in step S5 is consistent with the conventional filtering algorithm logic (both are 50Hz notch filtering + 0.5-100Hz band-pass filtering to calculate physiological parameters).

[0040] As an optional embodiment, the position calibration of step S1 also includes skin state adaptation: The MCU collects the skin contact impedance through the electrocardiogram sensing electrode. If the collected skin contact impedance is greater than a preset impedance threshold Zth, the mobile phone will prompt to apply medical conductive gel in the area where the sensing electrode is fitted until the skin contact impedance is less than or equal to the impedance threshold.

[0041] The skin contact impedance is calculated in the following manner: the electrocardiogram front-end analog module outputs a test voltage, and the test current is collected at the same time, and the skin contact impedance is equal to the test voltage divided by the test current.

[0042] The relevant parameters are as follows: the impedance threshold Zth is 5000 ohms, and the test voltage is 0.5 volts.

[0043] It should be noted that the skin contact impedance is reduced to below 5kΩ by the conductive gel, the electrocardio signal amplitude is increased to above 0.2V, the problem of weak signal caused by dry skin is solved, and the device can still be stably monitored when the skin condition is not good. The electrocardiogram front-end analog module is MAX30003CTI+, and the MCU is ESP32-PICO-D4. When the electrocardio sensing electrode collects the skin contact impedance, the test current is collected based on the test voltage output by the electrocardiogram front-end analog module, the impedance value is accurately calculated, and the accuracy of the skin condition adaptation is ensured; after applying the medical conductive gel, the conductive contact between the electrocardio sensing electrode and the skin is enhanced, and the signal-to-noise ratio of the electrocardio signal collection is improved by ≥3dB.

[0044] As an optional embodiment, the scene recognition of step S2 further includes an artificial correction mechanism: If the scene recognized by the MCU does not match the actual scene of the user, the user can manually select the scene through the mobile phone APP, and the MCU reconfigures the voltage output strategy of the power management component and the parameters of each module after receiving the manual instruction.

[0045] It should be noted that the artificial correction avoids the scene misjudgment of the device caused by the fluctuation of the motion data, quickly adapts the module parameters to the actual scene, ensures the continuity and accuracy of the monitoring data in the transition scene, and solves the hysteresis problem of traditional automatic scene recognition. The MCU is ESP32-PICO-D4, the wireless communication module includes a Bluetooth antenna and an NFC coil (the NFC module is ST25DV64K-JFR6D3); the artificial correction instruction is transmitted to the MCU through the Bluetooth antenna through the mobile phone APP, the voltage output strategy of the power management component reconfigured by the MCU after receiving the instruction is consistent with the power supply logic of the automatically recognized scene, such as the intermittent power supply of (U4) voltage stabilizer corresponding to the resting scene, the continuous power supply of (U4) voltage stabilizer corresponding to the motion scene), and the reconfigured parameters of each module are consistent with the scene parameters of step S3.

[0046] As an optional embodiment, the multi-modal signal effectiveness verification of step S4 further includes the content of abnormal signal supplement: If the ECG is determined to be invalid, the current R wave time is estimated by combining the pulse wave peak time of the PPG with the historical average, specifically, the current estimated R wave time is equal to the pulse wave peak time of the PPG minus the historical The mean value of the adjacent peak interval of the filtered ACC signal is extracted, and the estimated respiratory rate is obtained by dividing 60 by the mean value.

[0047] If the RESP is determined to be invalid once, the current respiratory rate is calculated by the ACC of the IMU module.

[0048] The specific process is: low-pass filtering the ACC to obtain a filtered ACC signal, extracting the mean value of the adjacent peak interval of the filtered signal, and then dividing 60 by the mean value to obtain the estimated respiratory rate.

[0049] The relevant parameters are explained: the average value of the historical Δt is the average value of the previous 5 valid Δt.

[0050] It should be noted that through the cross-complement of multi-modal signals, the accuracy of filling invalid signals is improved, the monitoring interruption caused by the loss of single signal is avoided, and the problem of poor anti-interference ability of traditional single-mode monitoring is solved.

[0051] The model of the IMU module is LIS2DH12, the model of the ECG front-end analog module is MAX30003CTI+, and the model of the MCU is ESP32-PICO-D4; when the ECG is invalid, the pulse wave peak time (tP) of the PPG is obtained from the chest photoelectric volume signal collected by the PPG sensor.

[0052] As an optional embodiment: the BLE mode of step S5 also sets a data retransmission mechanism: when the Bluetooth connection is interrupted, the skin patch will temporarily store the valid signal data during the interruption in the local cache; If the interruption duration is less than or equal to After the connection is restored, the MCU will retransmit the temporarily stored data to the phone in time sequence, and these retransmitted data will be marked with a retransmission identifier; if the interruption duration is greater than The MCU will trigger the encrypted storage of the local cache data, and the next time the Bluetooth connection is established, the encrypted storage data will be preferentially retransmitted, and after the retransmission is completed, the local encrypted cache data will be deleted.

[0053] The relevant parameters are: 5 minutes.

[0054] It should be noted that the model of the MCU is ESP32-PICO-D4, the wireless communication module includes a Bluetooth antenna; the local cache capacity is 16KB, which can store valid signal data (including ECG signal, PPG signal, and respiratory wave signal derived physiological parameters) during the interruption; the encrypted storage adopts the AES-128 algorithm to ensure that the local cache data cannot be illegally read; after the retransmission data is marked with a retransmission identifier, the phone APP can distinguish real-time data from retransmission data, and integrate to form a complete monitoring record according to the timestamp.

[0055] As an optional embodiment: the low-power consumption management of the step S6 further comprises the content of adaptive adjustment of the power: the MCU collects the residual power of the lithium battery in real time through the power detection module.

[0056] When the residual power is less than a preset power threshold , the MCU automatically reduces the sampling frequency of each module; when the residual power is less than a preset power threshold , the MCU only retains the collection functions of the electrocardiogram front-end analog module, the PPG sensor and the NFC communication function, and simultaneously closes the IMU module and the respiratory wave detection module; when the residual power is greater than or equal to a preset power threshold , the MCU restores the normal sampling frequency and power supply of each module. The size relationship of the three power thresholds is < < .

[0057] The related parameters satisfy: the power threshold is 100 mA, is 50 mA, is 200 mA.

[0058] It should be further pointed out that the scene threshold experimental data supplement test object is 100 people, of which 60 are healthy people aged 20 to 40; 40 are patients with heart and lung diseases, aged 50 to 80, including patients with hypertension and coronary heart disease. The test method is 24-hour dynamic collection, covering 8 hours of sleep, 8 hours of home rest, 1 hour of high-intensity exercise, and 7 hours of other activities, recording the IMU acceleration average value every 10 seconds. The threshold is determined by K-means clustering analysis. The acceleration average value of the sleep scene is concentrated in 0 to 0.2g, and the threshold Ath1 is taken as 0.2g; the home rest is concentrated in 0.2 to 0.5g, and the threshold Ath2 is taken as 0.5g; the high-intensity exercise is concentrated in 0.5 to 2g, and Ath2 is taken as 0.5g as the boundary. In terms of population adaptation, children (6 to 12 years old, 20 people tested) have a smaller exercise amplitude, and the threshold is lowered by 10%; the elderly (65 years old and above, 20 people tested) have weak acceleration signals due to muscle relaxation, and the threshold is raised by 10%.

[0059] The above is only a preferred embodiment of the present application, and the protection scope of the present application is not limited to the above-mentioned embodiments. Any technical solution falling within the scope of the present application is within the protection scope of the present application. It should be noted that, for ordinary technical personnel in the technical field, some improvements and refinements without departing from the principles of the present application are also considered to be within the protection scope of the present application.

Claims

1. A multimodal smart skin patch for multi-scenario cardiopulmonary function monitoring, characterized in that, It includes a layered main body, a flexible circuit board, multimodal sensing components, power management components, and a wireless communication module; The layered structure includes an upper Ecoflex material layer, a lower Ecoflex material layer, and a lithium battery and a wireless charging coil encapsulated between the two Ecoflex material layers. The flexible circuit board substrate is polyimide, and the circuit board is provided with a double-bridge serpentine structure sensing wire. The multimodal sensing component includes a respiratory sensing electrode, two electrocardiogram sensing electrodes, a PPG sensor, an IMU module, and a respiratory wave detection module. The surface of the respiratory sensing electrode is bonded with a PDMS material layer, and the surface of the two electrocardiogram sensing electrodes is bonded with a highly adhesive conductive hydrogel layer. The lower Ecoflex material layer is cut according to the shape of each electrode and sensor position, so that each electrode and sensor is exposed to directly contact human skin. The non-sensing areas on the surface of the lower Ecoflex material layer are bonded with 3M2477P medical silicone double-sided tape. The power management component includes an NMOS switch, a power path control chip, a low dropout linear regulator, and a low dropout linear regulator. The NMOS switch is connected to the power management module through the power path control chip to control the power path of each module. The wireless communication module includes a Bluetooth antenna and an NFC coil.

2. The multimodal smart skin patch for multi-scenario cardiopulmonary function monitoring according to claim 1, characterized in that, The top circuit integration area of ​​the flexible circuit board integrates a wireless charging and power supply module, a power management module, an automatic power control module, an NFC dynamic tag module, a power detection module, an electrocardiogram front-end simulation module, and a low-power Bluetooth integrated main control module. The ECG front-end simulation module communicates with the MCU via an SPI interface to acquire ECG signals and transmit them to the MCU; the IMU module and the respiratory wave detection module communicate with the MCU via an I2C interface to acquire acceleration signals and respiratory wave signals, respectively, and transmit them to the MCU; the PPG sensor signal acquisition unit is electrically connected to the ECG front-end simulation module.

3. A multimodal smart skin patch for multi-scenario cardiopulmonary function monitoring according to claim 2, characterized in that, In the power management component, the output voltage meets the following requirements. The voltage regulator powers the ECG front-end analog module and the PPG sensor signal acquisition unit; the output voltage meets the requirements. The voltage regulator powers the IMU module and the respiratory wave detection module; the output voltage meets the requirements. The voltage regulator powers the PPG sensor light source drive unit; The MCU controls the power switch of the power management component through the I2C interface. The power detection module is electrically connected to the MCU, collects the remaining power Q of the lithium battery in real time and transmits it to the MCU. The MCU adjusts the power supply strategy of each module based on Q.

4. A multimodal smart skin patch for multi-scenario cardiopulmonary function monitoring according to claim 3, characterized in that, The wireless communication module supports independent operation in BLE mode and NFC mode; in BLE mode, the MCU transmits heart rate (HR), respiratory rate (RR), and blood oxygen saturation parameters processed by conventional filtering algorithms in real time through the Bluetooth antenna. In the NFC mode, the MCU uses the NFC coil at intervals. Update heart rate (HR), respiratory rate (RR), and blood oxygen saturation parameters; maximum duration of single test. Data reading; the MCU detects the duration When there is no Bluetooth connection, the control skin patch enters a deep sleep state. In the deep sleep state, the NFC module, MCU and IMU module continue to work intermittently. The MCU is also configured with synchronous data acquisition firmware to trigger the multimodal sensing components to synchronously acquire physiological signals.

5. A method for using a multimodal smart skin patch for multi-scenario cardiopulmonary function monitoring, characterized in that, The method of using the multimodal smart skin patch for multi-scenario cardiopulmonary function monitoring, as described in any one of claims 1 to 4, includes the following steps: S1. Device Fitting and Position Calibration: The lower layer of Ecoflex material on the skin patch is attached to the user's chest monitoring area near the heart and lungs using 3M 2477P medical silicone double-sided adhesive. After attachment, the MCU acquires the initial ECG signal through the ECG front-end analog module. The respiratory wave detection module acquires the initial respiratory wave signal. ,calculate signal-to-noise ratio and amplitude ;like or The MCU will then push a location adjustment prompt to the mobile phone via Bluetooth until... or ; If the calculated skin contact impedance Z is greater than the preset impedance threshold Zth, the phone will prompt you to apply medical conductive gel to the contact area of ​​the sensing electrode until the skin contact impedance Z is less than or equal to the preset threshold Zth. Once the impedance detection is complete, the MCU will send a 0x05 command to turn off the test voltage, and then the module will switch to ECG acquisition mode. S2. If the skin patch is in deep sleep mode, bring the phone close to the NFC coil of the skin patch to wake up the MCU through electromagnetic induction energy capture; the MCU collects initial motion data through the IMU module. Determine the scenario by combining preset scenario thresholds: when When, it is determined to be a nocturnal sleep apnea monitoring scenario; when At that time, it was determined to be a home-based rest and rehabilitation monitoring scenario; when When the scene is identified as a high-intensity intermittent exercise scenario, the MCU configures the power management component's voltage output strategy based on the scenario. S3, the MCU initializes the ECG front-end simulation module via the SPI interface and configures the sampling rate according to the scenario. A data transmission link is established; the IMU module is initialized via the I2C interface, and the sampling frequency is configured according to the scenario. Initialize the respiratory wave detection module and set the capacitive coupling detection sensitivity level according to the scenario; initialize the PPG sensor and adjust the output accuracy and drive current of the output voltage regulator. Simultaneously initialize the communication parameters of the Bluetooth antenna and NFC coil; S4. Perform multimodal signal acquisition and validity verification: The MCU calls the synchronous data acquisition firmware to trigger each sensor module to synchronously acquire signals according to the initialization parameters; during the acquisition process, the MCU determines the validity of the signal through multimodal collaborative logic: S5. After multimodal signal acquisition and validity verification, select BLE mode or NFC mode for data transmission. Specifically: BLE mode: The MCU uses a conventional filtering algorithm to process the effective signal, and then calculates the heart rate, respiratory rate and blood oxygen saturation. NFC mode: Bluetooth communication is turned off in this mode. The current time is synchronized to the NFC dynamic tag module through the NFC function of the mobile phone. The MCU uses parameter calculation algorithm to process the valid signal. S6. Data storage and low-power management are performed after S5: Data storage: In BLE mode, the raw signal data is stored locally on the mobile app, and the processed heart rate (HR), respiratory rate (RR), and blood oxygen saturation (SpO2) are simultaneously stored on the cloud server; in NFC mode, the data is first stored in the storage unit of the NFC dynamic tag module. When the mobile phone reads the data, it will store the relevant parameters locally and upload a snapshot of the raw signal to the cloud. The cloud data is associated with the device's unique number. Low power management: In BLE mode, if a duration of [duration missing] is detected... If there is no Bluetooth connection within a certain time and the IMU module determines that the current situation is not a motion scenario, the MCU will control the NMOS switch to disconnect the power to all sensor modules, leaving only the NFC module and itself in a deep sleep state; if the IMU module determines that the situation is a motion scenario, the time will be extended to a certain duration. If no connection is established after a certain period, it will enter deep sleep mode; in NFC mode, if a connection is detected for a period of time... If there is no data reading or interaction within a certain period of time, the MCU will control the power supply of each sensor module to be disconnected and enter a deep sleep state. During deep sleep, the IMU module will wake up every T8 time interval and continuously collect data for a duration of [duration missing]. The motion data, if the motion data Greater than or equal to This will trigger the MCU to wake up each sensor module and restore the monitoring function.

6. The method of using a multimodal smart skin patch for multi-scenario cardiopulmonary function monitoring according to claim 5, characterized in that, The position calibration in step S1 also includes skin condition adaptation: The MCU collects skin contact impedance through ECG sensing electrodes. If the collected skin contact impedance is greater than the preset impedance threshold Zth, the phone will prompt you to apply medical conductive gel to the area where the sensing electrodes are attached until the skin contact impedance is less than or equal to the impedance threshold. The skin contact impedance is calculated as follows: the ECG front-end simulation module outputs a test voltage and simultaneously collects a test current. The skin contact impedance is equal to the test voltage divided by the test current.

7. The method of using a multimodal smart skin patch for multi-scenario cardiopulmonary function monitoring according to claim 5, characterized in that, The scene recognition in step S2 also includes a manual correction mechanism: If the scenario automatically identified by the MCU does not match the user's actual scenario, the user can manually select the scenario through a mobile APP. After receiving the manual command, the MCU will reconfigure the voltage output strategy of the power management component and the parameters of each module.

8. The method of using a multimodal smart skin patch for multi-scenario cardiopulmonary function monitoring according to claim 7, characterized in that, The multimodal signal validity verification in step S4 also includes supplementing abnormal signals: If a single ECG is deemed invalid, then the peak pulse wave time of the PPG, combined with historical data, will be used. The current R-wave time is estimated using the mean of the PPG pulse wave peak time. Specifically, the current estimated R-wave time is equal to the PPG pulse wave peak time minus the historical values. The mean value is used to supplement the electrocardiogram signal data; If a single RESP is deemed invalid, the current respiratory rate is calculated using the ACC function of the IMU module.

9. The method of using a multimodal smart skin patch for multi-scenario cardiopulmonary function monitoring according to claim 8, characterized in that, Step S5's BLE mode also includes a data retransmission mechanism: when the Bluetooth connection is interrupted, the skin patch will temporarily store the valid signal data during the interruption in the local cache; If the interruption duration is less than or equal to After the connection is restored, the MCU will retransmit the temporarily stored data to the phone in chronological order, and this retransmitted data will be marked with a retransmission identifier; if the interruption duration is longer than... The MCU will trigger the encrypted storage of local cache data. When the Bluetooth connection is connected again, it will prioritize the retransmission of this encrypted data. After the retransmission is completed, the local encrypted cache data will be deleted.

10. The method of using a multimodal smart skin patch for multi-scenario cardiopulmonary function monitoring according to claim 9, characterized in that, The low-power management in step S6 also includes adaptive power adjustment: the MCU collects the remaining power of the lithium battery in real time through the power detection module. When the remaining battery power is less than the preset battery power threshold When the remaining battery power is less than the preset battery power threshold, the MCU will automatically reduce the sampling frequency of each module; At this time, the MCU retains only the ECG front-end analog module, PPG sensor acquisition function, and NFC communication function, while shutting down the IMU module and respiratory wave detection module; when the remaining battery power is greater than or equal to the preset battery power threshold... When this happens, the MCU will restore the normal sampling frequency and power supply of each module.

Images