Flexible ultrasonic patch for continuously monitoring multi-dimensional physiological parameters of blood vessel

Abstract

The invention discloses a flexible ultrasonic patch for continuous monitoring of multi-dimensional physiological parameters of blood vessels, relates to the technical field of medical instruments, and utilizes a designed flexible ultrasonic array to extract high-resolution carotid artery wall velocity motion waveforms by parallelly collecting multi-channel data, so that the blood vessel multi-dimensional physiological parameters can be continuously monitored, and the blood vessel multi-dimensional physiological parameters can be continuously monitored. And signal delays among different array elements can be calculated to obtain local pulse wave velocities (PWV). Moreover, the speed waveform, the heart rate, the amplitude, the calculated PWV and other key physical parameters are jointly input into a PINN (Physical Information Neural Network) framework, and the cuff blood pressure is used as a gold standard for training. According to the method integrating physical information (PWV) and deep learning, the defects that a traditional physical model is too simplified and a pure black box model lacks generalization ability are overcome, and the blood pressure prediction result is more accurate and robust and has a physical basis.

Description

Technical Field

[0001] This invention relates to the field of medical device technology, and in particular to a flexible ultrasound patch for continuous monitoring of multidimensional physiological parameters of blood vessels. Background Technology

[0002] Blood pressure monitoring is crucial in the healthcare field. Traditional cuff blood pressure monitors are widely used, but their measurements are intermittent and require the subject to remain still, failing to meet the need for dynamic, continuous monitoring.

[0003] Existing non-invasive methods, such as photoplethysmography (PPG) and arterial tonometry, also have limitations. PPG has insufficient penetration depth, making it difficult to measure deep central vascular vessels. Arterial tonometry relies on strain sensors to detect vascular pulsation and is generally only applicable to peripheral vessels supported by skeletal structures; its accuracy is also greatly affected by operator technique and patch placement.

[0004] While existing ultrasound technology allows for non-invasive observation of deep tissues, traditional ultrasound monitoring devices (such as B-mode ultrasound probes) are typically bulky and made of rigid materials. This results in unstable acoustic coupling with the tissue surface, preventing a tight and stable conformal fit with human skin and limiting their application in long-term, everyday monitoring scenarios. More importantly, existing blood pressure conversion algorithms based on flexible ultrasound have fundamental flaws.

[0005] Traditional physical modeling methods: Many methods rely on physical models (such as the Moens-Korteweg equation or its simplified form) to convert measurable changes in blood vessel diameter into blood pressure values. However, the physiological characteristics of human blood vessels are extremely complex, and these simple physical models cannot fully encompass all nonlinear influencing factors, resulting in insufficient measurement accuracy and large errors.

[0006] Traditional deep learning methods: Other approaches attempt to use deep learning models to directly map extracted motion waveforms (such as displacement or velocity waveforms) to blood pressure values. While this method avoids complex physical modeling, it lacks physical constraints, and the model may learn spurious correlations, resulting in poor generalization ability and insufficient robustness under different physiological states.

[0007] Existing methods for measuring pulse wave velocity (PWV) based on flexible ultrasound are complex to operate and can only reflect the average pulse wave velocity between two blood vessels, but cannot reflect the local vascular elasticity.

[0008] In addition, blood vessels can be tortuous, and the distance between two blood vessels measured by traditional methods is a straight-line distance on the body surface, which differs from the actual distance of the blood vessels, thus affecting the accuracy of PWV measurement.

[0009] Therefore, how to provide a non-invasive continuous blood pressure monitoring technology that can not only achieve conformal fitting and continuous monitoring in hardware, but also combine physical basis (such as arterial stiffness) with the powerful fitting ability of deep learning in algorithm, so as to achieve more accurate and robust results, is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0010] In view of the above problems, the present invention provides a flexible ultrasound patch for continuous monitoring of multidimensional physiological parameters of blood vessels to overcome or at least partially solve the above problems. It overcomes the shortcomings of traditional cuff-type blood pressure monitors, which cannot provide continuous monitoring, and the insufficient accuracy of traditional ultrasound monitoring methods due to their reliance on physical models.

[0011] This invention provides the following solution: A flexible ultrasound patch for continuous monitoring of multidimensional physiological parameters of blood vessels, comprising: Two encapsulation layers are respectively disposed on the top and bottom of the patch, and the material of the encapsulation layers includes a flexible elastomer material; A transducer array, comprising a piezoelectric unit, an upper flexible electrode, a lower flexible electrode, and a gold finger; the piezoelectric unit comprises a plurality of piezoelectric composite materials; the upper flexible electrode comprises a plurality of upper island elements; and the lower flexible electrode comprises a plurality of lower island elements. A plurality of piezoelectric composite material bodies are respectively disposed one-to-one between a plurality of upper island elements and a plurality of lower island elements, so that the transducer array forms an island-bridge structure; each upper island element is connected to the gold finger by a stretchable serpentine interconnecting wire, so that each piezoelectric composite material body is independently addressable; two adjacent lower island elements are connected by serpentine interconnecting wires to form a common ground electrode for all piezoelectric composite material bodies; the gold finger is used to connect the ultrasonic patch to an external data acquisition system; A vertical interconnect structure is used to wire a plurality of the lower flexible electrodes to the top plane, where they are aggregated together with the gold fingers; The method for monitoring blood pressure and pulse wave using the flexible ultrasound patch includes: The flexible ultrasound patch was fixed to the skin surface of the test subject's carotid artery; An AC voltage is applied to the flexible ultrasonic patch by a data acquisition system, causing the flexible ultrasonic patch to emit ultrasonic waves and receive echo data to obtain the original ultrasonic signal. The original ultrasound signal is processed by signal processing and waveform extraction to obtain the carotid artery position velocity motion waveform; Based on the carotid artery position velocity motion waveform, heart rate, local pulse wave conduction velocity, and amplitude are extracted in parallel. The carotid artery position velocity motion waveform, the heart rate, the local pulse wave conduction velocity, and the amplitude are input into a deep learning network based on the PINN framework, so that the deep learning network outputs a predicted blood pressure value.

[0012] Preferably, all of the piezoelectric composite material bodies are made of type 1-3 piezoelectric composite material and are arranged in a 1×9 matrix array.

[0013] Preferably, both the upper island and the lower island are made of polyimide and copper double-layer stacking.

[0014] Preferably, the flexible elastomer material is Ecoflex00-30 silicone rubber.

[0015] Preferably, the piezoelectric composite material body is connected to the upper island and the lower island by soldering.

[0016] Preferably, the soldering process includes a soldering iron tip temperature of 180–220°C; The soldering iron is pressed vertically from above onto the pads of the upper flexible printed circuit board, providing both pressure and heat. Under the combined effect of pressure and heat, the solder melts at the contact points, fixing the layers together.

[0017] Preferably, the signal processing includes denoising the original ultrasound signal; extracting the waveforms of the anterior and posterior walls of the carotid artery after denoising, and selecting the waveform of the side wall with the most prominent signal.

[0018] Preferably, the waveform extraction includes using a cross-correlation algorithm to estimate the motion velocity of the selected sidewall waveform to obtain a high-resolution carotid artery position velocity motion waveform.

[0019] Preferably: the heart rate is calculated by segmenting the carotid artery position velocity motion waveform into a heartbeat cycle; Cross-correlation delay calculation is performed on the heartbeat cycle waveforms collected by different array elements in the same time period to obtain the number of delayed frames; the local pulse wave conduction velocity is calculated by linear fitting based on the known physical distance between array elements. Calculate the amplitude of the carotid artery position velocity motion waveform.

[0020] Preferably, the collected cuff blood pressure values ​​are used as the gold standard to perform supervised training on the deep learning network based on the PINN framework.

[0021] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects: This application provides a flexible ultrasound patch for continuous monitoring of multidimensional physiological parameters of blood vessels. Utilizing a designed flexible ultrasound array, it acquires multi-channel data in parallel, enabling the extraction of high-resolution carotid wall velocity waveforms and the calculation of signal delays between different array elements to obtain local pulse wave velocity (PWV). Furthermore, by inputting key physical parameters such as velocity waveform, heart rate, amplitude, and the calculated PWV into a PINN (Physical Information Neural Network) framework, and training it using cuff blood pressure as the gold standard, this method, which integrates physical information (PWV) and deep learning, overcomes the shortcomings of overly simplified traditional physical models and the lack of generalization ability in pure black-box models, resulting in more accurate, robust, and physically grounded blood pressure predictions.

[0022] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly described below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0024] Figure 1 This is a schematic diagram of the structure of a flexible ultrasound patch for continuous monitoring of multidimensional physiological parameters of blood vessels provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the preparation method of the flexible ultrasound patch for continuous monitoring of multidimensional physiological parameters of blood vessels provided in the embodiments of the present invention; Figure 3 This is an overall flowchart of blood pressure and PWV monitoring provided in the embodiments of the present invention.

[0025] In the diagram: 1. Encapsulation layer; 2. Piezoelectric unit; 3. Upper flexible electrode; 4. Lower flexible electrode; 5. Gold finger; 6. Vertical interconnect structure. Detailed Implementation

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention are within the scope of protection of the present invention.

[0027] See Figure 1 This invention provides a flexible ultrasound patch for continuous monitoring of multidimensional physiological parameters of blood vessels, such as... Figure 1As shown, the patch may include: Two encapsulation layers 1 are respectively disposed on the top and bottom of the patch, and the material of the encapsulation layer 1 includes a flexible elastomer material; A transducer array, comprising a piezoelectric unit 2, an upper flexible electrode 3, a lower flexible electrode 4, and a gold finger 5. The piezoelectric unit 2 comprises several piezoelectric composite materials, the upper flexible electrode 3 comprises several upper island elements, and the lower flexible electrode 4 comprises several lower island elements. A plurality of piezoelectric composite material bodies are respectively disposed one-to-one between a plurality of upper island elements and a plurality of lower island elements, so that the transducer array forms an island-bridge structure; each upper island element is connected to the gold finger 5 by a stretchable serpentine interconnecting wire, so that each piezoelectric composite material body is independently addressable; two adjacent lower island elements are connected by serpentine interconnecting wires to form a common ground electrode for all piezoelectric composite material bodies; the gold finger 5 is used to connect the ultrasonic patch to an external data acquisition system; Vertical interconnection structure 6, which is used to wire a plurality of the lower flexible electrodes 4 to the top plane and to be gathered together with the gold fingers 5; The method for monitoring blood pressure and pulse wave using the flexible ultrasound patch includes: The flexible ultrasound patch was fixed to the skin surface of the test subject's carotid artery; An AC voltage is applied to the flexible ultrasonic patch by a data acquisition system, causing the flexible ultrasonic patch to emit ultrasonic waves and receive echo data to obtain the original ultrasonic signal. The original ultrasound signal undergoes signal processing and waveform extraction to obtain the carotid artery position and velocity motion waveform. The signal processing includes denoising the original ultrasound signal; extracting the denoised anterior and posterior wall waveforms of the carotid artery, and selecting the side wall waveform with the most prominent signal. The waveform extraction involves using a cross-correlation algorithm on the selected side wall waveform to estimate its motion velocity, resulting in a high-resolution carotid artery position and velocity motion waveform.

[0028] Based on the carotid artery position velocity motion waveform, the heart rate, local pulse wave conduction velocity, and amplitude are extracted in parallel. In specific implementation, the embodiments of this application can segment the heartbeat cycle from the carotid artery position velocity motion waveform and calculate the heart rate. Cross-correlation delay calculation is performed on the heartbeat cycle waveforms collected by different array elements in the same time period to obtain the number of delayed frames; the local pulse wave conduction velocity is calculated by linear fitting based on the known physical distance between array elements. Calculate the amplitude of the carotid artery position velocity motion waveform.

[0029] The carotid artery position velocity waveform, the heart rate, the local pulse wave velocity, and the amplitude are input into a deep learning network based on the PINN framework, so that the deep learning network outputs a predicted blood pressure value. In specific training, embodiments of this application can provide supervised training of the PINN-based deep learning network using cuff blood pressure values ​​as the gold standard.

[0030] In a specific implementation, the embodiments of this application may provide several piezoelectric composite material bodies, each made of type 1-3 piezoelectric composite material, arranged in a 1×9 matrix array. The number and arrangement of the piezoelectric composite material bodies can be set according to the actual application scenario. Both the upper and lower island elements are made of a double-layer stack of polyimide and copper.

[0031] The flexible elastomer material is Ecoflex00-30 silicone rubber. This layer has a modulus comparable to human skin, ensuring wearing comfort and a conformal fit to the skin, while also serving as a sound matching layer and a protective layer.

[0032] The piezoelectric composite material body is connected to the upper and lower island elements by soldering. The soldering process involves the soldering iron tip being heated to 180–220°C. The soldering iron is pressed vertically from above onto the pads of the upper flexible printed circuit board, providing both pressure and heat. Under the combined effect of pressure and heat, the solder melts at the contact points, fixing the layers together.

[0033] The flexible ultrasound patch for continuous monitoring of multidimensional physiological parameters of blood vessels provided in this application calculates the local pulse wave velocity (PWV). PWV is the gold standard for assessing arterial stiffness, and using it as a feature input provides strong physical constraints for deep learning models. This allows predictions to no longer rely solely on waveform morphology but incorporate the physical constitutive properties (stiffness) of blood vessels, resulting in stronger physical basis and higher accuracy compared to traditional physical models or purely black-box deep learning models.

[0034] By fusing waveform dynamics features (velocity waveform) and key physical quantities (PWV, heart rate, amplitude) through the PINN framework, the model overcomes the defect of traditional models that may learn spurious correlations, making the model more capable of generalizing to different physiological states (such as exercise and emotional changes).

[0035] The patch is thin and flexible, conforming to the skin to enable long-term, unobtrusive continuous data acquisition, meeting the needs of dynamic blood pressure monitoring and enabling continuous real-time monitoring.

[0036] It can not only provide blood pressure values, but also capture complete pulse wave morphology characteristics and provide local PWV values, providing richer data support for cardiovascular health assessment.

[0037] The following section provides a detailed description of the flexible ultrasound patch for continuous monitoring of multidimensional physiological parameters of blood vessels, using the piezoelectric units 2 arranged in a 1×9 matrix array as an example.

[0038] This application provides a flexible ultrasound patch for continuous monitoring of multidimensional physiological parameters of blood vessels. The flexible ultrasound transducer includes an elastomer encapsulation layer 1, an upper flexible electrode 3, a vertical interconnect via (VIA), a piezoelectric unit 2, and a lower flexible electrode 4.

[0039] Encapsulation layer 1: The top and bottom of the patch are encapsulated by a layer of flexible elastomer material (such as Ecoflex 00-30 silicone rubber). This layer has a modulus comparable to human skin, ensuring wearing comfort and conformal fit to the skin, while also serving as an acoustic matching layer and a protective layer.

[0040] Piezoelectric unit 2: At the core of the patch, there are 9 piezoelectric composite material bodies arranged in a 1×9 matrix. The piezoelectric composite material bodies are preferably type 1-3 piezoelectric composite materials. These piezoelectric composite material bodies constitute the "island" structure of the transducer array.

[0041] Upper flexible electrode 3: Located above the piezoelectric unit 2, this is the upper flexible electrode 3. This electrode employs a stretchable serpentine interconnect wire design to independently address each piezoelectric unit 2.

[0042] Lower flexible electrode 4: Located below the piezoelectric unit 2, this is the "bridge" structure, i.e., the lower flexible electrode 4. This lower flexible electrode 4 also adopts a serpentine interconnect design, serving as the common ground electrode for all piezoelectric units 2.

[0043] Gold finger 5: This is the external connection interface of the patch, i.e., gold finger 5. It connects to the upper flexible electrode 3 and is used to connect the ultrasonic patch to an external ultrasonic system or integrated circuit (IC).

[0044] Vertical Interconnect Structure 6 (VIA): It is used to route the common electrode of the bottom layer to the top plane, which is then connected with the gold fingers 5 to achieve a coplanar electrical connection, simplifying the packaging and connection process.

[0045] Both the upper flexible electrode 3 and the lower flexible electrode 4 are islands. Bridge structure, island The islands in the bridge structure are connected by serpentine wires; the upper islands in the upper flexible electrode 3 correspond one-to-one with each piezoelectric composite material body in the piezoelectric unit 2 and are connected by solder; the lower islands in the lower flexible electrode 4 correspond one-to-one with each piezoelectric composite material body in the piezoelectric unit 2 and are connected by solder; the piezoelectric unit 2 is located between the corresponding islands in the upper flexible electrode 3 and the corresponding islands in the lower flexible electrode 4.

[0046] The upper flexible electrode 3 and the lower flexible electrode 4 are stretchable serpentine interconnect electrodes, made of a double-layer stack of polyimide (PI) and copper (Cu), ensuring that the patch maintains circuit integrity when bent or stretched. The upper electrode contains individually addressable stimulation electrodes, each corresponding to a cell in the piezoelectric unit 2 array; the lower layer is a common ground electrode. Vertical interconnect vias (VIA) guide the top stimulation electrode and the bottom ground layer to the same plane, optimizing mechanical robustness and simplifying electrical connections.

[0047] The elastomeric encapsulation layer 1 is made of Ecoflex 00-30 silicone rubber, with a modulus comparable to human skin. As an acoustic matching layer, it matches the acoustic impedance between the flexible ultrasonic transducer and human tissue, reducing the reflectivity of ultrasound waves at the interface between the flexible ultrasonic transducer and human tissue, enhancing the ultrasound transmittance of the flexible ultrasonic transducer, reducing energy attenuation during transmission, and improving the transmission efficiency of ultrasound waves. Simultaneously, it eliminates the need for traditional liquid coupling agents, achieving a balance between mechanical robustness and acoustic emission performance.

[0048] As attached Figure 2 As shown, a key welding step in the fabrication process of an embodiment of the present invention is illustrated, which aims to solve the problem of incomplete welding that may occur in the welding process of flexible devices.

[0049] The process uses a soldering iron with the tip temperature set to approximately 200°C.

[0050] The soldering iron is pressed vertically from above onto the pads of the upper flexible printed circuit board (FPCB), providing both pressure and heat.

[0051] Under the combined action of pressure and heat, the solder melts at the contact point, firmly fixing the layers (such as FPCB pads and gold fingers 5(2) or VIA(6)) together.

[0052] The enlarged image on the right shows in detail the soldering process of the soldering iron tip, pads, and solder in the contact area.

[0053] Throughout the process, a fixing clamp is provided at the bottom to support the entire flexible ultrasonic patch assembly, ensuring the precision and stability of the welding.

[0054] The core of this invention lies in providing a method for monitoring blood pressure and pulse wave based on the above-mentioned patch array, comprising the following steps: Step 1: Place the flexible ultrasound patch on the skin surface of the carotid artery.

[0055] Step 2: Apply AC voltage to the patch through the data acquisition system to make it emit ultrasonic waves and receive echo data.

[0056] Step 3: Signal Processing and Waveform Extraction. The acquired data is denoised; the denoised waveforms of the anterior and posterior walls of the carotid artery are extracted, and the waveform of the side wall with the more prominent signal (such as the anterior wall) is selected.

[0057] Step 4: Feature extraction.

[0058] a. Velocity waveform: The cross-correlation algorithm is used to estimate the motion velocity of the selected wall waveform, and the velocity motion waveform at that position is obtained.

[0059] b. Physical parameter extraction: u1 (heart rate): Automatically segments the heartbeat cycle from the velocity motion waveform and calculates the heart rate.

[0060] u2 (Local PWV): Cross-correlation delay calculation is performed on the heart rate cycle waveforms acquired by different array elements (e.g., array element 1 and array element 9) in the same time period to obtain the number of delayed frames (time delay Δt). Based on the known physical distance Δd between array elements, the local pulse wave velocity (PWV) is calculated through linear fitting. This value directly reflects the local arterial stiffness.

[0061] u3 (amplitude): Calculates the amplitude (e.g., peak-to-peak value) of the velocity motion waveform.

[0062] Step 5: Blood Pressure Prediction.

[0063] The velocity-motion waveform obtained in step four, along with the three calculated physical quantities—u1 (heart rate), u2 (local PWV), and u3 (amplitude)—are used as feature inputs.

[0064] The combined features described above are input into a deep learning network within the PINN (Physics-Informed Neural Network) framework.

[0065] The training process uses synchronously acquired cuff blood pressure values ​​as the gold standard to ultimately output predicted continuous blood pressure values ​​and pulse wave morphology.

[0066] In practical applications, deep learning networks need to be trained, such as... Figure 3 As shown, the training process may include: Step 1: Experimental preparation and data collection.

[0067] Experimental preparation: Test subjects wore cuff-type blood pressure monitors as the gold standard reference.

[0068] Patch fixation: The flexible ultrasonic patch described in this invention is fixed to the skin surface of the carotid artery of the test subject.

[0069] Synchronous Start-up: Synchronously start the data acquisition system of the flexible ultrasonic patch and the cuff blood pressure monitor.

[0070] Simultaneous data acquisition: Data is acquired simultaneously by the patch and the cuff blood pressure monitor. The patch acquires raw ultrasound echo data, while the cuff blood pressure monitor provides the gold standard cuff blood pressure data for subsequent model training.

[0071] Step 2: Signal processing and waveform extraction.

[0072] Signal processing flow: The acquired raw ultrasound signal enters the signal processing flow.

[0073] Noise reduction: First, the signal is denoised, and then the motion waveforms of the anterior and posterior walls of the carotid artery are extracted.

[0074] Waveform filtering: Filter out arterial wall waveforms with higher signal quality and clearer appearance.

[0075] Velocity estimation: For the selected waveforms, the cross-correlation algorithm is used to estimate the motion velocity.

[0076] Waveform output: The final result is a high-resolution carotid artery position and velocity motion waveform.

[0077] Step 3: Feature Parameter Extraction. Based on the velocity-motion waveform obtained in Step 2, three key physical feature parameters are extracted in parallel: u1 (heart rate): Calculated by automatically segmenting the heartbeat cycle from the velocity motion waveform.

[0078] u2 (local PWV): This is a key step in the present invention.

[0079] u3 (amplitude): is obtained by calculating the peak-to-peak value and other indicators of the velocity motion waveform.

[0080] First, the velocity waveform is automatically segmented into heartbeat cycles.

[0081] By utilizing the 1x9 array characteristics of the patch, the heartbeat cycle waveforms of different array elements within the same time period are selected.

[0082] Cross-correlation delay calculations are performed on these waveforms to determine the number of delayed frames (i.e., time delay Δt) for the pulse wave to propagate between these array elements.

[0083] Based on the known actual physical distance between array elements (Δd) and the calculated number of delayed frames (Δt), the local PWV (u2) of the carotid artery segment is calculated through linear fitting.

[0084] Step 4: PINN model training and blood pressure prediction.

[0085] Model input: The three physical feature parameters extracted in step three (u1: heart rate, u2: PWV, u3: amplitude) and the velocity motion waveform obtained in step two are used as input features and fed into a deep learning network with a PINN framework.

[0086] Model training: Using the cuff blood pressure values ​​collected in step one as the gold standard, the PINN network was trained in a supervised manner.

[0087] Final output: The trained model can output a predicted blood pressure based on the features input in real time.

[0088] In summary, the flexible ultrasound patch for continuous monitoring of multidimensional physiological parameters of blood vessels provided in this application utilizes the designed flexible ultrasound array to acquire multi-channel data in parallel. It can not only extract high-resolution carotid wall velocity motion waveforms, but also calculate the signal delay between different array elements to obtain local pulse wave velocity (PWV).

[0089] Furthermore, by inputting key physical parameters such as velocity waveform, heart rate, amplitude, and calculated PWV into a PINN (Physical Information Neural Network) framework, and training with cuff blood pressure as the gold standard.

[0090] This method, which combines Physical Information Value (PWV) with deep learning, overcomes the shortcomings of overly simplified traditional physical models and the lack of generalization ability of pure black-box models, making blood pressure prediction results more accurate, robust, and physically based.

[0091] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0092] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of this application.

[0093] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for system or system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and relevant parts can be referred to the descriptions in the method embodiments. The systems and system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0094] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of protection of the present invention.

Claims

1. A flexible ultrasound patch for continuous monitoring of multi-dimensional physiological parameters of blood vessels, characterized in that, include: Two encapsulation layers are respectively disposed on the top and bottom of the patch, and the material of the encapsulation layers includes a flexible elastomer material; A transducer array, comprising a piezoelectric unit, an upper flexible electrode, a lower flexible electrode, and a gold finger; the piezoelectric unit comprises a plurality of piezoelectric composite materials; the upper flexible electrode comprises a plurality of upper island elements; and the lower flexible electrode comprises a plurality of lower island elements. A plurality of piezoelectric composite material bodies are respectively disposed one-to-one between a plurality of upper island elements and a plurality of lower island elements, so that the transducer array forms an island-bridge structure; each upper island element is connected to the gold finger by a stretchable serpentine interconnecting wire, so that each piezoelectric composite material body is independently addressable; two adjacent lower island elements are connected by serpentine interconnecting wires to form a common ground electrode for all piezoelectric composite material bodies; the gold finger is used to connect the ultrasonic patch to an external data acquisition system; A vertical interconnect structure is used to wire a plurality of the lower flexible electrodes to the top plane, where they are aggregated together with the gold fingers; The method for monitoring blood pressure and pulse wave using the flexible ultrasound patch includes: The flexible ultrasound patch was fixed to the skin surface of the test subject's carotid artery; An AC voltage is applied to the flexible ultrasonic patch by a data acquisition system, causing the flexible ultrasonic patch to emit ultrasonic waves and receive echo data to obtain the original ultrasonic signal. The original ultrasound signal is processed by signal processing and waveform extraction to obtain the carotid artery position velocity motion waveform; Based on the carotid artery position velocity motion waveform, heart rate, local pulse wave conduction velocity, and amplitude are extracted in parallel. The carotid artery position velocity motion waveform, the heart rate, the local pulse wave conduction velocity, and the amplitude are input into a deep learning network based on the PINN framework, so that the deep learning network outputs a predicted blood pressure value.

2. The flexible ultrasound patch for continuous monitoring of multi-dimensional physiological parameters of blood vessels of claim 1, wherein, Several of the piezoelectric composite materials are made of type 1-3 piezoelectric composite materials and are arranged in a 1×9 matrix array.

3. The flexible ultrasound patch for continuous monitoring of multi-dimensional physiological parameters of blood vessels of claim 1, wherein, Both the upper and lower islands are made of polyimide and copper double-layer stacking.

4. The flexible ultrasound patch for continuous monitoring of multi-dimensional physiological parameters of blood vessels of claim 1, wherein, The flexible elastomer material is Ecoflex00-30 silicone rubber.

5. The flexible ultrasound patch for continuous monitoring of multi-dimensional physiological parameters of blood vessels of claim 1, wherein, The piezoelectric composite material body is connected to the upper island and the lower island by soldering.

6. The flexible ultrasound patch for continuous monitoring of multi-dimensional physiological parameters of blood vessels of claim 5, wherein, The soldering process includes a soldering iron tip temperature of 180–220°C; The soldering iron is pressed vertically from above onto the pads of the upper flexible printed circuit board, providing both pressure and heat. Under the combined effect of pressure and heat, the solder melts at the contact points, fixing the layers together.

7. The flexible ultrasound patch for continuous monitoring of multi-dimensional physiological parameters of blood vessels of claim 1, wherein, The signal processing includes denoising the original ultrasound signal; extracting the waveforms of the anterior and posterior walls of the carotid artery after denoising, and selecting the waveform of the side wall with the most prominent signal.

8. The flexible ultrasound patch for continuous monitoring of multi-dimensional physiological parameters of blood vessels of claim 7, wherein, The waveform extraction includes using a cross-correlation algorithm to estimate the motion velocity of the selected sidewall waveform, thereby obtaining a high-resolution carotid artery position velocity motion waveform.

9. The flexible ultrasound patch for continuous monitoring of multi-dimensional physiological parameters of blood vessels of claim 1, wherein, The heart rate is calculated by segmenting the heartbeat cycle from the carotid artery position velocity motion waveform. The heartbeat period waveforms collected by different array elements in the same time period are subjected to cross-correlation delay calculation to obtain delay frame numbers; and the local pulse wave conduction velocity is calculated through linear fitting according to the known physical distance between the array elements; The amplitude of the velocity motion waveform of the carotid artery position is calculated.

10. The flexible ultrasound patch for continuous monitoring of multi-dimensional physiological parameters of blood vessels of claim 1, wherein, The collected cuff blood pressure value is used as a gold standard to supervise the training of the deep learning network based on the PINN framework.

Images