Implantable nanogenerator-driven smart breathing training and monitoring patch

Abstract

This invention discloses an implantable nanogenerator-driven intelligent respiratory training and monitoring patch, belonging to the field of implantable medical devices and intelligent respiratory rehabilitation. It employs a stretchable, layered structure, including an energy harvesting layer, a monitoring and sensing layer, a control and intervention layer, and a biocompatible layer. The energy harvesting layer includes a triboelectric nanogenerator unit, an airflow guiding channel, and an energy storage module. The monitoring and sensing layer includes a laser-engraved graphene strain sensor array, a signal conditioning module, and a data preprocessing module. The control and intervention layer includes a BLE communication module, a data analysis module, a neuromuscular electrical stimulation unit, and a trigger control module. The stretchable, layered structure also includes a first stretchable adjustment layer and a second stretchable adjustment layer. This invention innovatively proposes a stretchable, layered structure, enabling reliable and stable self-powered operation and achieving an integrated function of "stable self-powered operation - precise dynamic monitoring - safe and personalized intervention."

Description

Technical Field

[0001] This invention relates to the field of implantable medical devices and intelligent respiratory rehabilitation, specifically to a nanogenerator-driven intelligent respiratory training and monitoring patch that integrates adaptive airflow acquisition, respiratory phase perception weighting, multi-physiological linkage intervention, and cross-layer energy management. It is suitable for long-term respiratory function monitoring and personalized training intervention for patients with severe respiratory dysfunction (such as after mechanical ventilation weaning, high paraplegia with diaphragmatic paralysis). Background Technology

[0002] The rehabilitation process of critically ill patients with respiratory dysfunction (such as those weaned from mechanical ventilation or those with high-level paraplegia and diaphragmatic paralysis) relies heavily on long-term, accurate monitoring and real-time, safe intervention of diaphragmatic movement. External respiratory monitoring devices (such as chest strap strain gauges and nasal / oral airflow sensors) are easily affected by body movement, clothing obstruction, and environmental airflow interference, often resulting in data error rates exceeding 20%, and cannot meet the continuous monitoring needs of patients at night or during activity. Implantable devices, due to their direct proximity to the diaphragm projection area and strong anti-interference capabilities, have become the preferred clinical option for these patients. However, existing technologies still have the following core limitations:

[0003] Poor energy harvesting stability: Chinese invention patent 202011470564.3 discloses a wireless lung function detection system and method, which discloses a self-driven respiratory monitoring system based on triboelectric nanogenerator (TENG). However, it uses a rigid plastic airflow channel. After implantation, the periodic movement of the chest wall tissue can easily cause the channel to deform and become blocked, resulting in a fluctuation of more than 35% in the utilization rate of respiratory airflow. This causes the TENG output voltage to fluctuate between 2.1-6.8V, which cannot provide a continuous and stable power supply for the low-power sensing module and requires external battery to supplement power.

[0004] Insufficient accuracy in diaphragm monitoring: Chinese invention patent 202310233882.5, "A Respiratory Sensor Based on Graphene Composite Material and Its Preparation Method," proposes a flexible graphene sensing structure for respiratory monitoring. However, it uses a globally unified signal fusion algorithm that does not consider the variability of diaphragm strain during inhalation (peak strain in the central region reaches 1.2%) and exhalation (peak strain in the peripheral region reaches 0.9%). Furthermore, it does not address the interference of respiratory humidity on the sensing signal, resulting in displacement calculation errors often exceeding 0.9 mm. This makes it impossible to distinguish between physiological fluctuations and pathological abnormalities of the diaphragm.

[0005] Low safety profile of the intervention: Chinese invention patent 202411629424.4 discloses a respiratory neuromuscular stimulation control method and system, which discloses an electrical stimulation modulation scheme based on muscle contraction signals. However, it only triggers neuromuscular electrical stimulation (NMES) based on a single muscle signal and does not consider the patient's circulatory and oxygenation status. Clinical data show that when the patient's blood oxygen saturation is <93%, blind stimulation may cause a sudden increase in respiratory muscle oxygen consumption, leading to a worsening of hypoxemia, with a risk rate of up to 27%.

[0006] Lack of cross-layer collaboration: Existing implantable respiratory devices, such as the external phrenic nerve electrical stimulator for respiratory synchronization described in Chinese Invention Patent 202510754097.3, have independent closed-loop designs for energy harvesting, monitoring, and intervention modules. The energy management module is only responsible for storing electrical energy and does not establish linkage with the functional modules. When the TENG power generation is insufficient (supercapacitor voltage <2.7V), the intervention module still maintains full power operation, resulting in a power outage on average every 48 hours, interrupting the monitoring data link. Summary of the Invention

[0007] In view of this, the present invention provides an implantable nanogenerator-driven intelligent respiratory training and monitoring patch. Through four core innovations, namely adaptive airflow acquisition, respiratory phase weighted monitoring, multi-physiological linkage intervention, and cross-layer energy management, it solves the problems of unstable energy supply, low monitoring data accuracy, high intervention operation risk, and discontinuous system operation, and realizes the integrated function of "stable self-powered power supply - accurate dynamic monitoring - safe and personalized intervention".

[0008] To achieve the above objectives, the present invention provides the following technical solution:

[0009] An implantable nanogenerator-driven smart respiratory training and monitoring patch adopts a stretchable layered structure, which includes an energy harvesting layer, a monitoring and sensing layer, a control and intervention layer and a biocompatible layer from top to bottom. Each layer is integrated through a flexible connection structure.

[0010] The energy harvesting layer includes a triboelectric nanogenerator unit, an airflow guiding channel, and an energy storage module. The triboelectric nanogenerator unit adopts a layered structure composed of a flexible polydimethylsiloxane film and graphene electrodes, with 3-5 groups arranged symmetrically at the edge of the patch. The airflow guiding channel is an arc-shaped microchannel structure, with one end connected to the implanted area next to the trachea and the other end open in the chest wall tissue space, used to guide respiratory airflow through the triboelectric nanogenerator unit. The energy storage module is a miniature supercapacitor, connected to the triboelectric nanogenerator unit through a flexible wire.

[0011] The monitoring sensing layer includes a laser-engraved graphene strain sensing array, a signal conditioning module, and a data preprocessing module. The graphene strain sensing array is a sensing unit array, and the coverage area of ​​the sensing unit array matches the projection area of ​​the diaphragm on the chest wall. The signal conditioning module integrates an operational amplifier and a filter circuit. The data preprocessing module is a micro MCU that converts the strain signal into diaphragm contraction displacement data through a bilinear interpolation algorithm and generates a diaphragm contraction vector map by combining a region-weighted fusion algorithm.

[0012] The control intervention layer includes a Bluetooth Low Energy communication module, a data analysis module, a neuromuscular electrical stimulation unit, and a trigger control module. The data analysis module pre-stores diaphragmatic movement characteristic parameters of normal breathing patterns and evaluates respiratory efficiency through a respiratory efficiency index calculation algorithm, the formula of which is:

[0013]

[0014] in, The maximum diaphragmatic contraction displacement within 3-5 respiratory cycles, f is the respiratory rate, σ is the standard deviation of the displacement data during the same period, and k is the correction factor. Synchronicity is determined using the formula for synergistic chest and abdominal movements.

[0015] ,

[0016] in, For chest displacement data, For abdominal displacement data, Phase difference, normal phase difference The threshold is set to 30-60°.

[0017] The neuromuscular electrical stimulation unit includes 3-5 sets of microelectrodes and a stimulation signal generator. The trigger control module can trigger the neuromuscular electrical stimulation unit when an abnormal breathing pattern is detected.

[0018] The biocompatible layer uses a medical-grade polyetheretherketone film as the outer layer and a polylactic acid-glycolic acid copolymer coating as the inner layer, with both layers having a thickness of 0.08-0.12 mm.

[0019] The stretchable stacked structure further includes a first stretchable adjustment layer disposed between the energy harvesting layer and the monitoring and sensing layer, and a second stretchable adjustment layer disposed between the monitoring and sensing layer and the control intervention layer. The first stretchable adjustment layer is a super-elastic SMA fiber woven mesh and a porous PLGA composite buffer layer, and the second stretchable adjustment layer includes a flexible hinge array and multiple low-friction PTFE positioning grooves.

[0020] Furthermore, in the energy harvesting layer, the area of ​​each triboelectric nanogenerator unit is 12-18mm × 8-12mm, the diameter of the airflow guiding channel is 1.0-1.4mm, and the capacitance of the micro supercapacitor is 400-600μF; a single triboelectric nanogenerator unit outputs a voltage of 3.2-5.3V under a breathing airflow velocity of 0.4-2.2m / s, and the output voltage satisfies:

[0021]

[0022] in, For airflow velocity, The coefficients are experimental fitting values, and the total power consumption of the device is ≤120μW.

[0023] Furthermore, in the monitoring sensing layer, each sensing unit of the graphene strain sensing array has an area of ​​1.8-2.2 mm × 1.8-2.2 mm, a sensing unit spacing of 1.2-1.8 mm, an initial resistance value R of 4-6 kΩ, and the relationship between strain and resistance change satisfies:

[0024]

[0025] in, This is the change in resistance. The initial resistance is K = 120-150, the gauge coefficient is K = 120-150, and the strain value is ε.

[0026] The operational amplifier in the signal conditioning module has a gain of 800-1200 times, and the cutoff frequency of the filter circuit is 8-12Hz. The data preprocessing module uses an STM32L476 or an equivalent MCU. In the bilinear interpolation algorithm, the fitting coefficients are solved by solving a system of equations based on the coordinates and displacement data of four adjacent sensing units. After regional weighted fusion, the vector vector displacement accuracy is 0.3-0.6mm.

[0027] Furthermore, in the control intervention layer, the Bluetooth Low Energy communication module uses the nRF52832 or a similarly performing chip with a transmission rate of 0.8-1.2 Mbps and a transmission distance ≤12 m; when the data analysis module calculates REI, Take the maximum displacement over 3-5 respiratory cycles. Take the standard deviation of displacement data from the same period:

[0028] ,

[0029] This represents the displacement value for a single cycle. The average displacement; in the determination of synchronicity of chest and abdominal movements, the normal phase difference is... The threshold is set to 30-60°.

[0030] Furthermore, in the control intervention layer, the microelectrode diameter of the neuromuscular electrical stimulation unit is 0.6-1.0 mm, and the exposed end of the electrode protrudes 0.15-0.25 mm from the biocompatible layer; the pulse signal parameters generated by the stimulation signal generator are: frequency 15-55 Hz, current intensity 0.08-0.55 mA, and pulse width 180-220 μs; the stimulation intensity is correlated with the diaphragm functional state and meets the following requirements.

[0031] ,

[0032] in Only when Enabled at any time.

[0033] Furthermore, the fabrication process of the energy harvesting layer includes: fabricating a PDMS flexible film with a thickness of 45-55 μm using a molding method; depositing a graphene electrode with a thickness of 8-12 nm on the film surface by magnetron sputtering; assembling the electrode into a triboelectric nanogenerator unit; fabricating an arc-shaped airflow guiding channel made of PLGA material using 3D printing; bonding the channel to the triboelectric nanogenerator unit with medical adhesive; and welding a miniature supercapacitor with dimensions of 4-6 mm × 2.5-3.5 mm × 0.8-1.2 mm to the triboelectric nanogenerator unit with gold wire and connecting it to a rectifier circuit. The rectified output voltage meets the requirements. .

[0034] Furthermore, the fabrication process of the monitoring sensing layer includes: fabricating a graphene strain sensing array on a polyimide substrate with a thickness of 20-30 μm using laser engraving technology; electrically connecting the signal conditioning module and the data preprocessing module to the sensing array via a flexible PCB board; the data preprocessing module has built-in program code for bilinear interpolation algorithm and region weighted fusion algorithm, and is entirely encapsulated in a PI film, with the response time of the sensing unit ≤15ms after encapsulation.

[0035] Furthermore, the biocompatible layer is applied to the surface of the control intervention layer using hot-pressing technology and sealed via laser welding; the weld strength meets the requirements of tensile testing. The waterproof and tissue fluid intrusion resistance meets the leakage rate requirements in the immersion test. .

[0036] The beneficial effects of this invention are as follows:

[0037] 1. This invention adopts a composite structure that matches a multi-layer functional structure with a multi-layer telescopic structure to achieve a telescopic stacked structure, so that the thickness and shape change significantly with human breathing movements, making it easier to accept the energy in this change and making self-generated power possible.

[0038] 2. This invention significantly reduces the fluctuation amplitude of the input airflow of the triboelectric nanogenerator unit by using a composite airflow channel of "nickel-titanium shape memory alloy spring + medical silicone rubber tube" and a pressure regulating valve. Combined with the TPS73633 voltage regulator circuit, the output voltage is stabilized at 3.3V±0.05V, meeting the continuous power supply requirements of the entire module.

[0039] 3. This invention uses a respiratory phase identification unit and an adaptive weighting algorithm to focus on collecting high strain signals at the center of the diaphragm during inhalation and focus on edge signals during exhalation. Combined with a bilinear interpolation algorithm, the displacement monitoring accuracy is improved to 0.1-0.3 mm, which is far below the clinical error threshold of 0.5 mm.

[0040] Meanwhile, the sensor array is encapsulated within a PI film, effectively isolating respiratory moisture and reducing resistance drift, ensuring long-term reliability of monitoring data. Based on precise diaphragm contraction and displacement data, doctors can clearly distinguish between "physiological shallow breathing caused by transient fatigue" and "pathological abnormalities due to decreased diaphragm strength after weaning from mechanical ventilation," avoiding misadjustments to rehabilitation plans and providing data support for personalized treatment.

[0041] 4. This invention constructs a three-dimensional assessment matrix of "diaphragm displacement-heart rate-blood oxygenation". Intervention is triggered only when multiple indicators are abnormal at the same time (such as displacement <5mm + heart rate exceeding resting value by 15%). The trigger threshold is dynamically adjusted with blood oxygenation to achieve precise triggering with "safety first".

[0042] The graded stimulation strategy and real-time feedback regulation mechanism (reducing flow by 20% when displacement fluctuation is >30%) further avoid overstimulation: low-intensity stimulation of 0.08-0.2mA is used for mild paradoxes, and the intensity is increased to 0.2-0.55mA only for severe paradoxes, which significantly reduces the risk of respiratory stress under hypoxic conditions. It is particularly suitable for the physiological vulnerability of critically ill patients (such as those in the recovery period of COPD and ARDS) and significantly improves the safety of intervention.

[0043] 5. The present invention adds a cross-layer energy linkage module, which dynamically adjusts the function mode according to the supercapacitor voltage: when the voltage is 2.5-3.0V, the NMES redundancy function is turned off (power consumption is reduced by 40%), and when the voltage is <2.5V, intervention is paused but monitoring and communication are retained (power consumption ≤50μW), while a low power warning is sent.

[0044] 6. The biocompatible layer of this invention adopts a composite structure of "PEEK film + PLGA coating + nano silver antibacterial coating", which significantly improves the antibacterial rate and significantly reduces the leakage rate after laser welding and sealing, avoiding the risk of tissue fluid intrusion and infection; each layer is connected by flexible polyimide hinges, which can withstand ±20° deformation, perfectly adapting to the periodic movement of the chest wall and reducing the foreign body sensation after implantation.

[0045] In summary, this invention innovatively proposes a scalable, stacked structure that enables reliable and stable self-powered operation, significantly improving user comfort and compliance. It is applicable to four core innovations: post-mechanical ventilation weaning, high-level cutoff with adaptive airflow acquisition, respiratory phase weighted monitoring, multi-physiological linkage intervention, and cross-layer energy management. It solves the problems of unstable energy supply, low monitoring data accuracy, high intervention operation risk, and discontinuous system operation, achieving an integrated function of "stable self-powered operation - accurate dynamic monitoring - safe and personalized intervention". Attached Figure Description

[0046] To make the objectives, technical solutions, and advantages of the present invention clearer, the preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, wherein:

[0047] Figure 1 This is a cross-sectional view of the present invention;

[0048] Figure 2 This is a cross-sectional view of the energy harvesting layer;

[0049] Figure 3 A cross-sectional view of the monitoring sensing layer;

[0050] Figure 4 A cross-sectional view of the control intervention layer;

[0051] Figure 5 This is a cross-sectional view of the biocompatible layer;

[0052] Figure 6 This is a cross-sectional view of the first telescopic adjustment layer;

[0053] Figure 7 This is a cross-sectional view of the second telescopic adjustment layer.

[0054] Explanation of reference numerals in the attached figures:

[0055] 1-Energy harvesting layer; 2-Monitoring and sensing layer; 3-Control and intervention layer; 4-Biocompatible layer; 5-First stretching adjustment layer; 6-Second stretching adjustment layer; 7-Triboelectric nanogenerator unit; 8-Airflow guiding channel; 9-Energy storage module; 10-Laser-engraved graphene strain sensing array; 11-Signal conditioning module; 12-Data preprocessing module; 13-BLE communication module; 14-Data analysis module; 15-Neuromuscular electrical stimulation unit; 16-Trigger control module; 17-PEEK film; 18-PLGA coating; 19-Superelastic SMA fiber braided mesh; 20-Porous PLGA buffer layer; 21-Flexible hinge array; 22-Low-friction PTFE positioning groove. Detailed Implementation

[0056] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0057] The accompanying drawings are for illustrative purposes only and are schematic diagrams, not actual pictures. They should not be construed as limiting the invention. To better illustrate the embodiments of the invention, some parts in the drawings may be omitted, enlarged, or reduced, and do not represent the actual product dimensions. It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0058] like Figure 1-7 As shown, this embodiment provides an implantable nanogenerator-driven intelligent respiratory training and monitoring patch, which adopts a stretchable layered structure. From top to bottom, it includes an energy harvesting layer 1, a monitoring and sensing layer 2, a control and intervention layer 3, and a biocompatible layer 4. Each layer is integrated through a flexible connection structure, and the overall thickness is ≤2.8mm. The specific structure and parameters are as follows:

[0059] like Figure 2 As shown, the energy harvesting layer 1 in this embodiment is used to convert the mechanical energy of the breathing airflow into electrical energy to power the entire device. It includes a triboelectric nanogenerator unit 7, an airflow guiding channel 8, and an energy storage module 9. The triboelectric nanogenerator unit 7 is a layered structure composed of a flexible polydimethylsiloxane (PDMS) film and graphene electrodes, with 3-5 groups symmetrically distributed along the edge of the patch. Each triboelectric nanogenerator unit 7 has an area of ​​12-18 mm × 8-12 mm and can output a voltage of 3.2-5.3 V under a breathing airflow velocity of 0.4-2.2 m / s. The output voltage and the airflow velocity μ satisfy a fitting relationship (fitting coefficients a and b are obtained through experimental calibration). (Fixed); Airflow guiding channel 8: an arc-shaped microchannel structure made of polylactic acid-glycolic acid copolymer (PLGA), with a diameter of 1.0-1.4mm; one end is connected to the implanted area next to the trachea, and the other end is open in the interstitial space of the chest wall tissue, used to guide respiratory airflow through the triboelectric nanogenerator unit 7; Energy storage module 9: a miniature supercapacitor with a capacity of 400-600μF and a size of 4-6mm×2.5-3.5mm×0.8-1.2mm; connected to the triboelectric nanogenerator unit 7 through a flexible gold wire and connected in series with a rectifier circuit, the output voltage after rectification is stabilized at 3.0-3.3V, which powers the subsequent monitoring sensing layer 2 and control intervention layer 3.

[0060] like Figure 3 As shown, the monitoring sensing layer 2 in this embodiment is used to collect strain signals of diaphragmatic movement and convert them into quantified contraction displacement data. It includes a laser-engraved graphene strain sensing array 10, a signal conditioning module 11, and a data preprocessing module 12. The laser-engraved graphene strain sensing array 10 is fabricated based on a 20-30 μm thick polyimide (PI) substrate, with a sensor unit array of 5×5-7×7 dimensions. The coverage area perfectly matches the projection area of ​​the diaphragm on the chest wall. Each sensor unit has an area of ​​1.8-2.2 mm × 1.8-2.2 mm, a spacing of 1.2-1.8 mm, and an initial resistance of 4-6 kΩ. The strain and resistance changes satisfy the following relationship: (Where K=120-150 is the gauge coefficient, and ε is the strain value,) (where ΔR is the initial resistance and ΔR is the change in resistance); Signal conditioning module 11: integrates operational amplifier and filter circuit; the operational amplifier gain is 800-1200 times, used to amplify weak strain signals; the filter circuit cutoff frequency is 8-12Hz, used to filter out high-frequency noise other than respiratory signals; Data preprocessing module 12: adopts STM32L476 or equivalent micro MCU (microcontroller unit), with built-in bilinear interpolation algorithm and region weighted fusion algorithm; the bilinear interpolation algorithm solves the fitting coefficient by solving a system of equations based on the coordinates and displacement data of four adjacent sensing units, converting the strain signal into diaphragmatic contraction displacement; the weight coefficient of the region weighted fusion algorithm is negatively correlated with the distance d from the sensing unit to the diaphragm center, finally generating a diaphragmatic contraction vector diagram with a displacement accuracy of 0.3-0.6mm; the module is encapsulated in a PI film, with a response time ≤15ms.

[0061] like Figure 4 As shown, the control intervention layer 3 in this embodiment is used to evaluate respiratory efficiency, determine the respiratory pattern, and trigger intervention when abnormal breathing is detected. It includes a Bluetooth Low Energy communication module (BLE communication module 13), a data analysis module 14, a neuromuscular electrical stimulation unit 15, and a trigger control module 16. The BLE communication module 13 uses an nRF52832 or similar chip, with a transmission rate of 0.8-1.2 Mbps and a transmission distance ≤12 m; it is used to feed back diaphragmatic displacement data, respiratory efficiency parameters, and intervention results to the nurse's terminal in real time. The data analysis module 14 pre-stores diaphragmatic movement characteristic parameters of the normal respiratory pattern and performs evaluation using two algorithms: Respiratory Efficiency Index (REI) calculation. (in The maximum contractile displacement of the diaphragm within 3-5 respiratory cycles, where f is the respiratory rate, σ is the standard deviation of the displacement data during the same period, Di is the displacement value per cycle, is the average displacement, and k is the correction factor); the neuromuscular electrical stimulation unit 15 includes 3-5 sets of microelectrodes and a stimulation signal generator; the electrodes are made of platinum-iridium alloy or medical-grade pure silver, with a diameter of 0.6-1.0 mm, and the exposed end protrudes a biocompatible layer of 40.15-0.25 mm, with the position corresponding one-to-one with the nerve innervation areas related to diaphragmatic movement; the stimulation signal generator generates pulse signals with parameters of frequency 15-55 Hz, current intensity 0.08-0.55 mA, and pulse width 180-220 μs; the stimulation intensity is correlated with REI and meets the following requirements. (in (Only enabled when REI < 40); Trigger control module 16: Abnormal breathing mode determination condition is "phase difference" "Continued for 2-4 seconds" or "Other abnormal features lasting 4-6 seconds"; when the trigger condition is met (paradoxical breathing lasting ≥4-6 seconds), the neuromuscular electrical stimulation unit 15 is activated, and the parameters are adjusted through a graded stimulation algorithm: current intensity (Only activated if the anomalous pattern is not corrected within 5-8 seconds after stimulation at time t-1, and) Stimulation duration T = 10 + 2 × (t - 1) (t is the number of stimulations, t ≥ 1); Simultaneously, diaphragmatic displacement is collected in real time using a feedback adjustment algorithm; if the displacement fluctuation amplitude... ( Displacement upon stimulation To stimulate forward displacement, the current intensity is automatically reduced by ΔI = 0.03-0.08 mA, and the adjusted current... .

[0062] like Figure 5 As shown, the biocompatible layer 4 in this embodiment is used to protect the internal structure and ensure the biocompatibility of long-term implantation. It adopts a double-layer structure: outer layer: medical-grade polyetheretherketone film (PEEK film 17), thickness 0.08-0.12mm; inner layer: coated with polylactic acid-glycolic acid copolymer coating (PLGA coating 18), thickness 0.08-0.12mm; it is covered on the surface of the control intervention layer 3 by thermoforming technology and sealed by laser welding; the welding strength meets the requirement of F>8N in the tensile test, and the waterproof and tissue fluid intrusion prevention performance meets the leakage rate requirement in the immersion test. .

[0063] In particular, such as Figure 6 , 7As shown, in this embodiment, to better receive fluctuations caused by respiration and achieve self-powered operation, the stretchable stacked structure further includes a first stretchable adjustment layer 5 and a second stretchable adjustment layer 6. The first stretchable adjustment layer 5 is disposed between the energy harvesting layer 1 and the monitoring and sensing layer 2, and the second stretchable adjustment layer 6 is disposed between the monitoring and sensing layer 2 and the control and intervention layer 3. Specifically, the first stretchable adjustment layer 5 is a super-elastic SMA fiber woven mesh 19 + a porous PLGA buffer layer 20. This structure has strong elastic deformation and will change shape when receiving motion impact and respiratory impact, thereby generating deformation energy. The second stretchable adjustment layer 6 includes a flexible hinge array 21 (medical... Using high-grade TPU material and low-friction PTFE positioning grooves 22, the flexible hinge array 21 is laid flat across the entire surface, forming a plane that can change at a certain angle. This changes with the shape of the upper energy harvesting layer 1 and the monitoring and sensing layer 2, and fully releases this energy, thereby ensuring that the energy harvesting layer 1 receives more impact energy. The low-friction PTFE positioning grooves 22 are also provided in this layer, and positioning protrusions that match the low-friction PTFE positioning grooves 22 are provided on the surface of the control intervention layer 3, thereby forming a fixed structure with a certain constraint force, which constrains or limits the movement or impact from the upper surface.

[0064] Meanwhile, the functional and structural layers are integrated with medical-grade flexible adhesives, resulting in extremely low overall power consumption. It can achieve continuous self-powering through the energy harvesting layer 1, eliminating the need for external power supply or battery replacement, making it simple and stable to use.

[0065] In particular, this embodiment also proposes a preparation method for the above structure:

[0066] I. Preparation of Energy Harvesting Layer 1 (Core Function: Mechanical Energy to Electrical Energy Conversion)

[0067] 1. Fabrication of triboelectric nanogenerator unit 7

[0068] Preparation of flexible PDMS film: The PDMS prepolymer and curing agent (mass ratio 10:1) are mixed evenly by molding method and poured into a customized mold (corresponding to the size of a single set of triboelectric nanogenerator unit 7, 12-18mm×8-12mm). After vacuum degassing, the film is cured in an oven at 80℃ for 2h to form a PDMS film with a thickness of 45-55μm. The film is then demolded and ready for use.

[0069] Graphene electrode deposition: Graphene electrodes with a thickness of 8-12 nm are deposited on one side of a PDMS thin film using magnetron sputtering technology. The sputtering power is controlled at 150-200 W, and the vacuum level is [not specified]. This ensures that the electrodes are evenly covered and firmly bonded to PDMS.

[0070] Triboelectric nanogenerator unit 7 cutting: Cut the PDMS film with deposited electrodes into 12-18mm×8-12mm sizes, and prepare 3-5 sets for later use.

[0071] 2. Preparation of airflow guiding channel 8

[0072] Using 3D printing technology (photopolymerization), PLGA was used as the printing material. The model was designed with an arc-shaped microchannel structure with a diameter of 1.0-1.4 mm. The printing speed was 5-8 mm / s and the layer thickness was 50 μm. After printing, the model was post-treated in a vacuum drying oven at 60℃ for 4 hours to remove residual support material.

[0073] 3. Assembly of Energy Harvesting Layer 1

[0074] Three to five sets of triboelectric nanogenerator units 7 were symmetrically attached to the edge of a flexible PI substrate (20 μm thick), with a spacing of 5 to 8 mm between adjacent units, and fixed with medical-grade epoxy adhesive (elastic modulus ≤ 1 MPa after curing).

[0075] Align one end of the PLGA airflow guiding channel 8 with the gap of the triboelectric nanogenerator unit 7, and the other end with the preset "chest wall tissue gap opening end", and fix it with medical glue to ensure that the channel is in contact with the surface of the triboelectric nanogenerator unit 7 (so that the airflow can pass through smoothly).

[0076] A miniature supercapacitor with dimensions of 4-6mm×2.5-3.5mm×0.8-1.2mm and a capacitance of 400-600μF is used. It is then welded to the electrode of the triboelectric nanogenerator unit 7 via a flexible gold wire with a diameter of 20-30μm. A miniature rectifier circuit (output voltage 3.0-3.3V) is connected in series to complete the assembly of the energy harvesting layer 1.

[0077] II. Fabrication of the second stretching adjustment layer 6 (core function: interlayer stretching adaptation + positioning constraint)

[0078] 1. Fabrication of flexible hinge array 21

[0079] A medical-grade TPU film with a thickness of 50-80μm is taken and cut into a "cross-shaped" flexible hinge array 21 structure (the width of a single hinge is 0.5-0.8mm and the spacing is 2-3mm) using laser engraving technology to form an array covering the energy harvesting layer 1.

[0080] 2. Preparation of low-friction PTFE positioning groove 22

[0081] Using PTFE as raw material, low-friction PTFE positioning grooves 22 with a diameter of 1.5-2.0 mm and a depth of 0.3-0.5 mm are prepared by molding. They are then fixed to the surface of TPU flexible hinge array 21 with medical adhesive at a spacing of 5-6 mm to "match the positioning protrusions of the subsequent monitoring and sensing layer 2".

[0082] 3. Interlayer bonding

[0083] The prepared second stretchable adjustment layer 6 is adhered to the surface of the PI substrate of the energy harvesting layer 1, ensuring that the low-friction PTFE positioning groove 22 is accurately positioned. After adhesion, a pressure of 0.1 MPa is applied and cured at room temperature for 1 hour.

[0084] III. Preparation of Monitoring Sensing Layer 2 (Core Function: Acquisition and Conversion of Diaphragmatic Strain Signals)

[0085] 1. Fabrication of a laser-engraved graphene strain sensor array 10

[0086] A PI film with a thickness of 20-30μm is used as a substrate, and a laser-engraved graphene strain sensor array 10 with a size of 5×5-7×7 is etched on its surface using laser engraving technology. The size of a single unit is 1.8-2.2mm×1.8-2.2mm, and the unit spacing is 1.2-1.8mm.

[0087] Graphene with a thickness of 5-10 nm was grown on the etched PI substrate using chemical vapor deposition (CVD). Electrode leads were then defined using photolithography to form a laser-engraved graphene strain sensing array 10 with an initial resistance of 4-6 kΩ.

[0088] 2. The signal conditioning module 11 and the data preprocessing module 12 are integrated.

[0089] A micro operational amplifier (gain 800-1200 times) and a filter circuit (cutoff frequency 8-12Hz) are soldered onto a flexible PCB board to form a signal conditioning module 11; an STM32L476 MCU (with built-in bilinear interpolation and region weighted fusion algorithm program) is soldered onto a reserved position on the same flexible PCB board to form a data preprocessing module 12.

[0090] The output of the signal conditioning module 11 is connected to the electrode leads of the laser-engraved graphene strain sensing array 10 via ultra-fine copper wires with a diameter of 50μm. The data preprocessing module 12 is electrically connected to the signal conditioning module 11 to complete the module integration.

[0091] 3. Monitoring and Sensing Layer 2 Encapsulation and Positioning Protrusion Fabrication

[0092] The integrated module and the laser-engraved graphene strain sensor array 10 are packaged using a PI film with a thickness of 10-15μm. During the packaging process, it is ensured that the leads are not pulled, and the edges are sealed by laser welding. The response time after packaging is tested to be ≤15ms.

[0093] On the lower surface of the encapsulated monitoring and sensing layer 2 (corresponding to the low-friction PTFE positioning groove 22 of the second telescopic adjustment layer 6), a PTFE positioning protrusion with a diameter of 1.4-1.9 mm and a height of 0.3-0.5 mm is prepared by 3D printing to ensure that the protrusion is precisely matched with the low-friction PTFE positioning groove 22, thus completing the preparation of the monitoring and sensing layer 2.

[0094] IV. Preparation of the first stretching adjustment layer 5 (core function: impact amplification + energy absorption)

[0095] 1. Preparation of Super-Elastic SMA Fiber Braided Mesh 19

[0096] Superelastic SMA fibers (nickel-titanium alloy) with a diameter of 50-80μm are woven into superelastic SMA fiber woven mesh 19 (mesh size 1-2mm×1-2mm) using a warp and weft weaving method. After weaving, the mesh is annealed at 500℃ for 30min to enhance elastic recovery performance.

[0097] 2. Preparation of porous PLGA buffer layer 20

[0098] PLGA powder and porogen (sodium chloride particles, particle size 100-200μm) are mixed at a mass ratio of 3:1, pressed into a film with a thickness of 100-150μm, and soaked in deionized water for 24h to remove the porogen, forming a porous PLGA buffer layer 20 (porosity 40%-50%).

[0099] 3. Composite layer preparation

[0100] The super-elastic SMA fiber woven mesh 19 and the porous PLGA buffer layer 20 are composited with medical polyurethane adhesive at a composite pressure of 0.05 MPa and cured at 60°C for 1.5 h to ensure that the two do not peel off, thus completing the preparation of the first stretch adjustment layer 5; then it is pasted onto the upper surface of the monitoring and sensing layer 2 and fixed with the edges aligned.

[0101] V. Preparation of Control Intervention Layer 3 (Core Functions: Respiratory Assessment + Abnormal Intervention)

[0102] 1. Preparation of neuromuscular electrical stimulation unit 15

[0103] Take a platinum-iridium alloy wire (0.6-1.0 mm in diameter), cut it into electrodes with a length of 2-3 mm, grind one end into a hemispherical shape (exposed end), and weld a flexible wire with a diameter of 50 μm to the other end; prepare a total of 3-5 sets for later use.

[0104] Take a miniature stimulation signal generator (which can output signals with a frequency of 15-55Hz and a pulse width of 180-220μs), solder it onto a flexible PCB board, connect the electrode wires of the neuromuscular electrical stimulation unit 15 to the output end of the generator, and ensure that the exposed end of the electrode faces the preset "protruding biocompatible layer 4" direction.

[0105] 2. Integration of BLE communication module 13 and data analysis module 14

[0106] The nRF52832BLE communication module 13 (transmission rate 0.8-1.2Mbps) and the data analysis module 14 (compatible with STM32L476MCU) are soldered onto the same flexible PCB board and connected to the flexible PCB board of the neuromuscular electrical stimulation unit 15 via ribbon cables to form the core circuit of the control intervention layer 3.

[0107] Test circuit functions: ensure that the transmission distance of BLE communication module 13 is ≤12m, data analysis module 14 can normally calculate REI and determine the phase difference, and neuromuscular electrical stimulation unit 15 can output 0.08-0.55mA current. After meeting the standards, it can be used as a backup.

[0108] 3. Control intervention layer 3 fixed

[0109] The prepared control intervention layer 3 is attached to the upper surface of the first stretching adjustment layer 5 to ensure that the electrode position of the neuromuscular electrical stimulation unit 15 corresponds to the "diaphragmatic nerve innervation area" (refer to the preoperative CT positioning size). The edges are fixed with medical adhesive to prevent displacement.

[0110] VI. Preparation and overall integration of biocompatible layer 4 (core functions: biosafety + structural protection)

[0111] 1. Preparation of biocompatible layer 4

[0112] Preparation of inner PLGA coating 18: PLGA is dissolved in dichloromethane (concentration 10%) and uniformly coated onto the surface of control intervention layer 3 by spraying technology. The coating thickness is 0.08-0.12 mm. It is then dried in a fume hood (room temperature 25℃, time 2h) to form the inner layer.

[0113] Preparation of outer PEEK film 17: Using hot pressing technology, medical-grade PEEK particles are heated to 340℃ to melt and pressed into a PEEK film 17 with a thickness of 0.08-0.12mm, which is then cut into a shape that matches the overall size of the patch.

[0114] 2. Overall sealing and integration

[0115] A PEEK film 17 was applied to the surface of the PLGA coating 18, and the edges were sealed using laser welding technology (laser power 5-8W, scanning speed 10mm / s) to ensure that the welding strength met the tensile test requirement of F>8N. After welding, an immersion test (simulating a body fluid environment) was performed to verify the leakage rate. .

[0116] The gaps between each layer are filled with medical-grade flexible silicone (thickness ≤ 0.1 mm), cured at room temperature for 1 hour to ensure an overall thickness ≤ 2.5 mm; finally, aseptic treatment is performed (ethylene oxide sterilization, sterilization dose 25 kGy) to complete the entire patch preparation.

[0117] VII. Finished Product Performance Testing

[0118] Performance test of energy harvesting layer 1: With an airflow of 0.4-2.2m / s, the output voltage of the triboelectric nanogenerator unit 7 was tested to be 3.2-5.3V, and the power supply stability of the energy storage module 9 was tested (continuous output of 3.0-3.3V).

[0119] Accuracy test of monitoring sensor layer 2: Simulating diaphragm contraction, the displacement accuracy of laser-engraved graphene strain sensor array 10 was tested to be 0.3-0.6mm, and the response time of data preprocessing module 12 was ≤15ms.

[0120] Intervention layer 3 intervention function test: Simulates abnormal breathing pattern (REI < 25-35 or phase difference > 160-200°), tests the trigger accuracy of neuromuscular electrical stimulation unit 15, and the current regulation conforms to... .

[0121] Strength and elasticity test: Applying ±30% tensile deformation, the test showed no delamination between the first stretch adjustment layer 5, the second stretch adjustment layer 6 and other layers, and all functions were normal.

[0122] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. An implantable nanogenerator-driven smart respiratory training monitoring patch, characterized in that, Adopting a scalable laminated structure, from top to bottom includes energy harvesting layer, monitoring sensing layer, control intervention layer and biocompatible layer, each layer is integrated through flexible connection structure; The energy harvesting layer includes triboelectric nanogenerator unit, airflow guide channel and energy storage module, the triboelectric nanogenerator unit adopts layered structure composed of polydimethylsiloxane flexible film and graphene electrode, 3-5 groups are arranged and symmetrically distributed at the edge of the patch, the airflow guide channel is arc-shaped microchannel structure, one end is communicated with trachea implantation area, the other end is open in the interstitial space of chest wall tissue, for guiding respiratory airflow to pass through the triboelectric nanogenerator unit, the energy storage module is micro super capacitor, connected with the triboelectric nanogenerator unit through flexible wire; The monitoring sensing layer includes laser engraved graphene strain sensing array, signal conditioning module and data preprocessing module, the graphene strain sensing array is sensing unit array, the sensing unit array covers the projection area of diaphragm on the chest wall, the signal conditioning module integrates operational amplifier and filter circuit, the data preprocessing module is micro MCU, strain signal is converted into diaphragm contraction displacement data through bilinear interpolation algorithm, and diaphragm contraction vector diagram is generated by combining regional weighted fusion algorithm; The control intervention layer comprises a Bluetooth low-power communication module, a data analysis module, a neuromuscular electrical stimulation unit and a trigger control module, the data analysis module pre-stores diaphragm movement characteristic parameters of normal breathing mode, and evaluates breathing efficiency through a breathing efficiency index calculation algorithm, and the formula is: , wherein, is the maximum diaphragm displacement in 3-5 respiratory cycles, f is the respiratory frequency, σ is the standard deviation of the displacement data in the same period, k is the correction coefficient, and the synchronization is determined by the chest and abdominal movement synchronization formula: , wherein, is the chest displacement data, is the abdominal displacement data, is the phase difference, normal phase difference the threshold is set to 30-60°, The neuromuscular electrical stimulation unit includes 3-5 groups of micro electrodes and stimulation signal generator, the trigger control module can trigger the neuromuscular electrical stimulation unit when detecting abnormal breathing mode; The biocompatible layer adopts medical grade polyether ether ketone film as outer layer, and polylactic acid-glycolic acid copolymer coating is coated on the inner layer, the thickness of the two layers is 0.08-0.12mm, The scalable laminated structure further includes first stretch adjustment layer arranged between the energy harvesting layer and the monitoring sensing layer and second stretch adjustment layer arranged between the monitoring sensing layer and the control intervention layer, the first stretch adjustment layer is super-elastic SMA fiber woven mesh and porous PLGA composite buffer layer, the second stretch adjustment layer includes flexible hinge array and multiple low-friction PTFE positioning grooves.

2. The implantable nanogenerator-driven smart respiratory training monitoring patch of claim 1, wherein, In the energy collection layer, the area of each group of triboelectric nanogenerator units is 12-18 mm x 8-12 mm, the diameter of the airflow guide channel is 1.0-1.4 mm, and the capacity of the micro supercapacitor is 400-600 μF; a single group of triboelectric nanogenerator units outputs a voltage of 3.2-5.3 V under a respiratory airflow with a flow rate of 0.4-2.2 m / s, and the output voltage satisfies: , wherein, is the gas flow rate, is the experimental fitting coefficient, the total power consumption of the device is ≤ 120 μW.

3. The implantable nanogenerator-driven smart respiratory training monitoring patch of claim 1, wherein, In the monitoring sensing layer, each sensing unit of the graphene strain sensing array has an area of 1.8-2.2mm*1.8-2.2mm, a spacing of 1.2-1.8mm, an initial resistance value R of 4-6kΩ, and a strain-resistance change relationship satisfying: , wherein, is the resistance change amount, is the initial resistance, K = 120-150 is the gauge coefficient, and ε is the strain value; The gain of operational amplifier in the signal conditioning module is 800-1200 times, and the cut-off frequency of filter circuit is 8-12Hz; the data preprocessing module adopts STM32L476 or equivalent performance model MCU, the fitting coefficient in bilinear interpolation algorithm is solved by equation group of adjacent 4 sensing unit coordinates and displacement data, and the displacement accuracy of vector diagram after regional weighted fusion is 0.3-0.6mm.

4. The implantable nanogenerator-driven smart respiratory training monitoring patch of claim 1, wherein, In the control intervention layer, the Bluetooth low-power communication module adopts nRF52832 or similar performance model chip, the transmission rate is 0.8-1.2Mbps, and the transmission distance is ≤12m; when calculating REI, Take the maximum value of displacement in 3-5 breath cycles, Take the standard deviation of the displacement data at the same time: , for a single cycle displacement value, for average displacement; normal phase difference in thoraco-abdominal motion synchrony determination Threshold set at 30-60°.

5. The implantable nanogenerator-driven smart respiratory training monitoring patch of claim 1, wherein, In the control intervention layer, the microelectrode diameter of the neuromuscular electrical stimulation unit is 0.6-1.0 mm, and the electrode bare end protrudes from the biocompatible layer by 0.15-0.25 mm; the pulse signal parameters generated by the stimulation signal generator are: frequency 15-55 Hz, current intensity 0.08-0.55 mA, pulse width 180-220 μs; the stimulation intensity is associated with the diaphragm function state, satisfying , wherein , only when is enabled.

6. The implantable nanogenerator-driven smart respiratory training monitoring patch of claim 1, wherein, The preparation process of the energy collection layer comprises: using a molding method to prepare a PDMS flexible film with a thickness of 45-55 μm, depositing a graphene electrode with a thickness of 8-12 nm on the surface of the film by magnetron sputtering, and assembling a triboelectric nanogenerator unit; using 3D printing to prepare an arc-shaped airflow guide channel made of PLGA, and bonding the triboelectric nanogenerator unit through medical glue; welding a micro-supercapacitor with a size of 4-6 mm×2.5-3.5 mm×0.8-1.2 mm to the triboelectric nanogenerator unit through gold wire, and connecting a rectifier circuit, and the output voltage after rectification satisfies .

7. The implantable nanogenerator-driven smart respiratory training monitoring patch of claim 1, wherein, The preparation process of the monitoring sensing layer includes: preparing graphene strain sensing array on polyimide substrate with thickness of 20-30μm through laser engraving technology; the signal conditioning module and the data preprocessing module are electrically connected with the sensing array through flexible PCB board, the data preprocessing module is embedded with program code of bilinear interpolation algorithm and regional weighted fusion algorithm, and the whole is packaged in PI film, the response time of sensing unit after packaging is ≤15ms.

8. The implantable nanogenerator-driven smart respiratory training monitoring patch of claim 1, wherein, The biocompatible layer is covered on the surface of the control intervention layer by hot press forming technology, and is sealed by laser welding. The welding strength meets the tensile test , and the waterproof and tissue fluid invasion resistance meets the leakage rate in the immersion test.

Images