Multi-section-configuration wearable ultrasonic patch applied to CICU intensive care and detection alarm system

Abstract

The invention discloses a multi-section-configuration wearable ultrasonic patch applied to CICU intensive care and a detection alarm system, relates to the technical field of flexible device manufacturing, and aims to solve the problems of poor comfort, incomplete data and insufficient anti-interference capability in existing non-invasive and real-time heart monitoring. The patch comprises a flexible substrate, a multi-section linear array ultrasonic sensor, a phased array control electrode and an acoustic matching layer, and can provide non-invasive, comfortable and stable heart monitoring. The invention also provides a system realized based on the patch. The system comprises a main power supply system, an array sensor acquisition system, a front-end core control unit, a middle-end preprocessing unit and a rear-end signal processing system. The main power supply system provides reference voltage, the array sensor collects multiple paths of signals, the front-end core control unit carries out distributed collection, the middle-end preprocessing unit carries out self-adaptive noise reduction, and the rear-end signal processing system generates physiological parameters. When an abnormal signal is detected, the system outputs a reminding signal, and the system is suitable for heart function monitoring in CICU intensive care.

Description

Technical Field

[0001] The invention relates to the technical field of flexible device manufacturing, and in particular to medical monitoring. Summary of the invention

[0002] In critical care, especially for patients with cardiovascular diseases, commonly used clinical methods include invasive and non-invasive detection methods. Invasive monitoring technology represented by Swan-Ganz catheters can provide accurate hemodynamic parameters, but the risks brought by its invasiveness make this method unsuitable for long-term application. Swan-Ganz catheters require puncture of blood vessels and deep into the heart, so there is a risk of complications such as infection and thrombosis, especially when repeated catheterization is required, which affects the safety of patients.

[0003] In contrast, non-invasive monitoring technologies, such as echocardiography (heart B-ultrasound), have high safety and wide applicability in evaluating cardiac function. Echocardiography can evaluate the progression of coronary heart disease, cardiomyopathy, valvular heart disease and other diseases in real time, especially by calculating the ejection fraction to judge cardiac function. However, traditional echocardiography requires professional equipment and technicians to operate, and cannot achieve continuous monitoring, which limits its real-time application value in critical care.

[0004] In recent years, the rapid development of wearable medical devices has brought new solutions to the field of critical care. Wearable medical devices can provide continuous, non-invasive monitoring of physiological parameters, and their flexibility and comfort allow patients to wear them for a long time. In particular, wearable ultrasound devices based on flexible sensor technology are soft, comfortable, and easy to wear, showing broad application prospects in cardiac monitoring. For example, foreign research teams have tried to use flexible ultrasound sensors to monitor physiological signals of the lungs and heart, but these devices can usually only obtain ultrasound data in a single direction and cannot fully evaluate the complex geometric structure of the heart, resulting in a lack of accuracy and reliability of the data.

[0005] There are still several major problems in the application of traditional ultrasound equipment: first, ultrasound probes are generally rigid and need to be fixed to the patient's body surface with tape, which is not only inconvenient to wear but also affects the patient's comfort; second, existing wearable ultrasound devices mostly obtain heart images based on a single cross-section, which makes it difficult to accurately depict the complex structure of the heart, especially the geometric changes of the left ventricle under different states; in addition, existing equipment also has limitations in data transmission and anti-interference capabilities, which affects the detection accuracy and application stability of the equipment.

[0006] Therefore, existing technologies still face the following problems in non-invasive, real-time, and long-term monitoring of cardiac function:

[0007] 1. Although existing invasive methods (such as Swan-Ganz catheters) can provide accurate data, the risks brought by invasiveness and the limitation of long-term use have affected their widespread clinical application.

[0008] 2. Traditional non-invasive ultrasound equipment relies on professional operation and it is difficult to achieve continuous real-time monitoring.

[0009] 3. Existing wearable ultrasound devices are mostly rigid in design, which affects the wearing comfort of patients, and can only obtain ultrasound data in a single direction, and cannot comprehensively evaluate the geometric morphology and functional parameters of the heart.

[0010] 4. There is a problem of insufficient anti-interference ability during data transmission, which affects the accuracy and real-time performance of monitoring. Summary of the invention

[0012] In order to solve the technical defects existing in the prior art, that is, the prior art is not good at non-invasive, real-time, and long-term monitoring of cardiac function, but is insufficient in wearing comfort, comprehensiveness of data acquisition, and anti-interference stability, the technical solution provided by the present invention adopts a flexible design and a multi-section configuration, combined with multi-section data fusion and wireless transmission technology, to achieve real-time, non-invasive, and continuous monitoring of important organs such as the heart, specifically:

[0013] A multi-section wearable ultrasound patch for CICU intensive care, the patch comprising:

[0014] A flexible substrate for attaching to the patient's skin surface;

[0015] A multi-section linear array ultrasonic sensor for acquiring two-dimensional ultrasonic images of multiple sections;

[0016] The electrode capable of realizing the phased array control function comprises sub-array elements arranged in a double-layer array, and a bottom common-pole electrode and an upper separated collection electrode, wherein the bottom common-pole electrode is used for common grounding of the sub-array elements, and the upper separated collection electrode is used for independently receiving the signal of each sub-array element;

[0017] Acoustic matching layer, used to provide acoustic impedance gradient.

[0018] Furthermore, a preferred embodiment is provided, wherein the flexible substrate, the multi-section linear array ultrasonic sensor, the phased array control electrode and the double-layer acoustic matching layer are tightly fitted in sequence.

[0019] Furthermore, a preferred embodiment is provided, wherein the flexible substrate is made of a flexible material having adhesiveness.

[0020] Furthermore, a preferred embodiment is provided, wherein the multi-section linear array ultrasonic sensor comprises at least two groups of linear arrays in different directions.

[0021] Furthermore, a preferred embodiment is provided, wherein the acoustic matching layer is a double-layer acoustic matching layer.

[0022] Based on the same inventive concept, the present invention also provides a signal processing method for a multi-section wearable ultrasound patch applied to CICU intensive care, the method being applied to the patch, comprising:

[0023] The steps of collecting and denoising the patient's heart scan data;

[0024] A step of reconstructing three-dimensional graphics on the scan data;

[0025] A step of identifying the outline of the three-dimensional figure;

[0026] The step of analyzing the contour according to a preset analysis method.

[0027] Based on the same inventive concept, the present invention also provides a multi-section wearable ultrasound patch detection alarm system for CICU intensive care, the system is implemented based on the patch, and includes:

[0028] The main power supply system, including the AC / DC mains conversion system and the voltage stabilization system, is used as the standard reference voltage value input to the array sensor acquisition system;

[0029] The arrayed sensor acquisition system is used to collect multi-channel signals;

[0030] A front-end core control unit performs distributed acquisition of the arrayed sensor acquisition system under a unified clock;

[0031] The mid-end pre-processing unit, as an adaptive noise reduction system, pre-processes the collected signals;

[0032] A back-end signal processing system, used to process the pre-processed signals and generate information containing physiological parameters;

[0033] When the signal detected by the piezoresistive sensor in the arrayed sensor acquisition system is different from the standard value, a reminder signal is output.

[0034] Further, a preferred embodiment is provided, which also includes a wireless signal transmission module, including:

[0035] A signal acquisition module, connected to the ultrasonic patch, for collecting ultrasonic signals;

[0036] An analog-to-digital conversion unit, used to convert analog ultrasonic signals into digital signals, providing a basis for subsequent processing;

[0037] A data processing unit, used for pre-processing the collected digital signals, including compression and encoding;

[0038] A wireless communication module, used for transmitting the pre-processed digital signal to a receiving device;

[0039] Anti-interference and calibration modules reduce external interference and ensure signal stability through hardware and software optimization;

[0040] Low noise amplifier, used to amplify ultrasound signals.

[0041] Based on the same inventive concept, the present invention also provides a computer storage medium for storing a computer program. When the computer program is read by a computer, the computer executes the method described.

[0042] Based on the same inventive concept, the present invention also provides a computer, including a processor and a storage medium. When the processor reads a computer program stored in the storage medium, the computer executes the method described.

[0043] Compared with the prior art, the technical solution provided by the present invention is beneficial in that:

[0044] The flexible micro-ultrasound sensor design significantly improves the wearing comfort and adaptability to long-term monitoring. Compared with traditional rigid ultrasound probes, flexible ultrasound patches can fit tightly on the skin surface, avoiding the restrictions on patient activities caused by the rigidity of traditional devices, while reducing the discomfort caused by long-term wearing. This flexible design greatly enhances the adaptability of the device in patients' daily lives, especially for patients who need continuous monitoring in intensive care.

[0045] The design of the multi-section configuration solves the problem that a single-section ultrasound cannot accurately depict the heart's morphology by scanning the heart's geometric structure from multiple perspectives. Compared with existing wearable devices that can only obtain ultrasound images in a single direction, this solution uses orthogonal multi-section technology to obtain two-dimensional ultrasound images of the apical four chambers and two chambers, making the volume estimation of the left ventricle more accurate. This design enables the calculation of ejection fraction with high accuracy even in the distorted geometry of the left ventricle, effectively improving the reliability of monitoring data and the value of clinical diagnosis.

[0046] The multi-section data fusion processing method significantly improves the signal-to-noise ratio and the accuracy of monitoring data. By fusing ultrasound image data from different sections, this solution can compensate for the impact of noise, making the final cardiac image clearer and more accurate. Compared with the single-source processing of traditional single-section images, the multi-source fusion method of this solution effectively reduces the error caused by noise, thereby making the extracted cardiac ejection fraction and other physiological parameters more accurate, which is crucial for timely diagnosis and treatment decisions for patients in intensive care.

[0047] The realization of phased array control reduces the number of electrodes and improves the reliability of the system through the combination of the bottom common electrode and the upper separated collection electrode. Compared with other studies with complex electrode design, high manufacturing cost and poor stability, the phased array control method of this scheme can more efficiently control the signal transmission and reception of each sub-array element, ensuring the real-time performance and collection accuracy of the data. At the same time, the simplified electrode structure reduces the manufacturing cost and improves the stability of the system in long-term applications.

[0048] The introduction of wireless signal transmission modules makes data transmission more stable and efficient. Compared with the traditional equipment that needs to collect data through cable connections, wireless modules achieve efficient data transmission while ensuring low power consumption and anti-interference performance, which is particularly suitable for the high requirements for real-time data monitoring in intensive care. This improvement avoids the inconvenience of movement and infection risks caused by cable connections, allowing patients to be monitored in a relatively free state, while ensuring the reliability and real-time nature of the monitoring data.

[0049] Suitable for use in intensive care work. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 In the figure, the left side is a schematic diagram of the common ground electrode on the backing side, which introduces the parasitic capacitance crosstalk of the human body, and the right side is a schematic diagram of the test results of the experiment on the matching side;

[0051] Figure 2 Schematic diagram of the composition and attachment method of the ultrasound patch;

[0052] Figure 3 It is a schematic diagram of the signal processing flow of the ultrasound patch;

[0053] Figure 4 It is a schematic diagram of the traditional cardiac volume algorithm;

[0054] Figure 5 Schematic diagram of the three-dimensional volume algorithm formed by the Simpson method for multiple orthogonal sections;

[0055] Figure 6 It is a schematic diagram of the function and calculation process of the ultrasound patch;

[0056] Figure 7 This is a schematic diagram of the monitoring and alarm system of the ultrasonic patch; DETAILED DESCRIPTION

[0057] In order to make the advantages and benefits of the technical solution provided by the present invention more clearly reflected, the technical solution provided by the present invention is now further described in detail with reference to the accompanying drawings, specifically:

[0058] Embodiment 1: This embodiment provides a multi-section wearable ultrasound patch for CICU intensive care, the patch comprising:

[0059] A flexible substrate for attaching to the patient's skin surface;

[0060] A multi-section linear array ultrasonic sensor for acquiring two-dimensional ultrasonic images of multiple sections;

[0061] The electrode capable of realizing phased array control comprises sub-array elements arranged in a double-layer array, and a bottom common-pole electrode and an upper separated collection electrode, wherein the bottom common-pole electrode is used for common grounding of the sub-array elements, and the upper separated collection electrode is used for independently receiving the signal of each sub-array element;

[0062] Acoustic matching layer, used to provide acoustic impedance gradient.

[0063] Specifically, they include:

[0064] 1. Flexible substrate: Made of flexible material with adhesiveness, used to adhere to the patient's body surface to improve wearing comfort;

[0065] 2. Multi-section linear array ultrasonic sensor: arranged on the flexible substrate, including multiple sets of directional linear arrays, capable of simultaneously acquiring two-dimensional ultrasonic images of multiple sections;

[0066] 3. Electrodes capable of realizing phased array control: including a bottom common electrode and an upper separated collection electrode. The bottom common electrode is used for common grounding of multiple sub-array elements, and the upper separated collection electrode is used for independently receiving signals from each sub-array element to realize phased array control of multiple sub-array elements.

[0067] 4. Double-layer acoustic matching layer: It is set on the outside of the multi-section linear array ultrasonic sensor to provide the necessary acoustic impedance gradient to ensure the smooth transmission of ultrasonic waves and the effective detection of reflected waves;

[0068] 5. Signal processing module: connected to the multi-section linear array ultrasonic sensor, configured to receive the two-dimensional ultrasonic image collected by the ultrasonic sensor, and perform denoising and clarity processing on the image through adaptive wavelet transform filtering and SAR-BM3-D joint filtering;

[0069] 6. Three-dimensional reconstruction module: connected to the signal processing module, configured to perform three-dimensional reconstruction on the filtered two-dimensional cross-sectional data based on the ray casting method and the Ray Casting algorithm to generate a three-dimensional image of the heart;

[0070] 7. Calculation and analysis module: including an improved Simpson method volume calculation unit for calculating the volume of the left ventricle based on multi-orthogonal cross-sectional data, and including a contour recognition unit based on the Canny algorithm and a machine learning model for cardiac function assessment, wherein the machine learning model is Gaussian naive Bayes or decision tree;

[0071] 8. Wireless signal transmission module: connected to the calculation and analysis module, configured to transmit the analysis results to the remote receiving end in real time through Wi-Fi 6 technology, ensuring efficient and low-power data transmission;

[0072] 9. Monitoring and alarm system: integrated into the device, used to monitor the patient's physiological parameters in real time, and issue an alarm prompt when an abnormal signal is detected to remind medical staff to intervene.

[0073] Here is how to use it:

[0074] 1. Prepare a multi-section linear array ultrasound patch with a flexible substrate, and attach the ultrasound patch to the patient's body surface to collect ultrasound signals in multiple directions;

[0075] 2. Through the combination of the bottom common electrode and the upper separated collection electrode, the phased array control of multiple sub-array elements is realized to obtain two-dimensional ultrasound images of the heart in multiple cross sections;

[0076] 3. Perform adaptive wavelet transform filtering and SAR-BM3-D joint filtering on the collected ultrasonic signals to eliminate noise and improve image quality;

[0077] 4. Perform three-dimensional reconstruction of the filtered two-dimensional cross-sectional data based on the ray casting method and the Ray Casting algorithm to generate a three-dimensional image of the heart;

[0078] 5. Calculate the left ventricular volume on the reconstructed three-dimensional image using the improved Simpson method, including dividing the left ventricle into multiple cylinders using images obtained from multiple orthogonal sections and calculating the volume of each cylinder;

[0079] 6. Perform contour recognition on the reconstructed cardiac image based on the Canny algorithm, and evaluate cardiac function parameters using a machine learning model, wherein the machine learning model is Gaussian naive Bayes or a decision tree;

[0080] 7. The calculation and analysis results are transmitted to the remote receiving end through the wireless signal transmission module, and the abnormal physiological signals are alarmed through the integrated monitoring and alarm system.

[0081] The device achieves comfort for patients to wear for a long time through flexible design. Combined with multi-section ultrasound acquisition, phased array control, signal processing, three-dimensional reconstruction, wireless data transmission and monitoring and alarm functions, it provides a comprehensive and efficient cardiac function monitoring solution for intensive care, effectively solving the problems of insufficient monitoring accuracy, inconvenience in wearing and lack of real-time alarm in existing technologies.

[0082] Embodiment 2: This embodiment further limits the multi-section configuration wearable ultrasound patch for CICU intensive care provided in embodiment 1. The flexible substrate, the multi-section linear array ultrasound sensor, the electrode capable of realizing phased array control, and the double-layer acoustic matching layer are tightly fitted in sequence.

[0083] Embodiment 3: This embodiment further limits the multi-section wearable ultrasound patch for CICU intensive care provided in embodiment 1, wherein the flexible substrate is prepared by a flexible material having adhesiveness.

[0084] Embodiment 4: This embodiment further limits the multi-section wearable ultrasound patch for CICU intensive care provided in embodiment 1, wherein the multi-section linear array ultrasound sensor includes at least two groups of directional linear arrays.

[0085] Embodiment 5: This embodiment further limits the multi-section wearable ultrasound patch for CICU intensive care provided in embodiment 1, and the acoustic matching layer is a double-layer acoustic matching layer.

[0086] Embodiment 6: This embodiment provides a signal processing method for a multi-section wearable ultrasound patch for CICU intensive care. The method is applied to the patch provided in embodiment 1, including:

[0087] The steps of collecting and denoising the patient's heart scan data;

[0088] A step of reconstructing three-dimensional graphics on the scan data;

[0089] A step of identifying the outline of the three-dimensional figure;

[0090] The step of analyzing the contour according to a preset analysis method.

[0091] Embodiment 7: This embodiment provides a multi-section wearable ultrasound patch detection alarm system for CICU intensive care. The system is implemented based on the patch provided in embodiment 1, and includes:

[0092] The main power supply system, including the AC / DC mains conversion system and the voltage stabilization system, is used as the standard reference voltage value input to the array sensor acquisition system;

[0093] The arrayed sensor acquisition system is used to collect multi-channel signals;

[0094] A front-end core control unit performs distributed acquisition of the arrayed sensor acquisition system under a unified clock;

[0095] The mid-end pre-processing unit, as an adaptive noise reduction system, pre-processes the collected signals;

[0096] A back-end signal processing system, used to process the pre-processed signals and generate information containing physiological parameters;

[0097] When the signal detected by the piezoresistive sensor in the arrayed sensor acquisition system is different from the standard value, a reminder signal is output.

[0098] The above contents are modified from the Claims.

[0099] Specifically, the system includes:

[0100] 1. Signal acquisition module: connected to the multi-section linear array ultrasonic sensor of the flexible substrate, configured to collect ultrasonic signals of multiple sections;

[0101] 2. Signal preprocessing module: connected to the signal acquisition module, configured to suppress noise on the collected ultrasonic signal through adaptive wavelet transform filtering, and to clarify the image through SAR-BM3-D joint filtering to improve the quality and accuracy of the signal;

[0102] 3. A three-dimensional reconstruction module: connected to the signal preprocessing module, and configured to perform three-dimensional reconstruction on the preprocessed two-dimensional ultrasound image data based on a ray casting method and a RayCasting algorithm to generate a three-dimensional image of the heart;

[0103] 4. Contour recognition module: connected to the 3D reconstruction module, configured to use the Canny algorithm to perform cardiac cavity contour recognition on the reconstructed 3D image to extract the geometric morphology information of the heart;

[0104] 5. Volume calculation module: connected to the contour recognition module, configured to calculate the volume of the left ventricle of the heart based on the improved Simpson method, using the image data of multiple orthogonal sections to divide the left ventricle into multiple cylindrical units and calculate their volumes;

[0105] 6. Functional evaluation module: connected to the volume calculation module, configured to use Gaussian naive Bayes or decision tree model to analyze the volume calculation results to evaluate the functional parameters of the heart, including physiological indicators such as ejection fraction;

[0106] 7. Wireless signal transmission module: connected to the functional evaluation module and configured to transmit the calculation and evaluation results to a remote receiving device via Wi-Fi 6 technology to achieve real-time data sharing.

[0107] Through the integrated design of signal acquisition, preprocessing, three-dimensional reconstruction, contour recognition, volume calculation, function evaluation and wireless transmission, the signal processing device can accurately monitor and evaluate the patient's cardiac function in real time, effectively improving the accuracy and efficiency of signal processing, and overcoming the problems of large signal noise, insufficient imaging quality and lack of evaluation accuracy in the prior art.

[0108] The specific implementation method is:

[0109] Step 1: Preparation of flexible ultrasound patch

[0110] Brief description: First, the wearable flexible ultrasound patch is made, including the preparation and integration of the flexible substrate, acoustic matching layer, piezoelectric array sensor, etc. The flexible ultrasound patch is the core of the entire system, and its preparation is the basis for the subsequent function realization.

[0111] 1. Preparation of flexible substrate: A thin layer of PDMS (polydimethylsiloxane) with soft and sticky properties is used as the flexible substrate material. The preparation method is to mix the PDMS main agent and the curing agent in a mass ratio of 1:7.5, refrigerate for 12 hours, and then heat in a 45-degree oven for 15 hours to obtain a highly sticky film. This flexible substrate has sufficient softness and adhesion to be attached to the patient's skin surface, improving wearing comfort.

[0112] 2. Fabrication of piezoelectric array sensors: Multiple sets of directional linear arrays are prepared on piezoelectric ceramics by cutting. The specific operation is to first use a diamond cutting machine to cut the ceramics at equal intervals to form deep grooves, then fill them with epoxy resin, and polish them after curing to obtain the final array. Piezoelectric ceramics have good piezoelectric properties and are used to convert electrical signals into ultrasonic signals.

[0113] 3. Integration of acoustic matching layer: The double-layer acoustic matching layer is integrated with the piezoelectric array. The matching layer provides the necessary acoustic impedance gradient to ensure that the sound energy is smoothly transmitted to the human tissue. The matching layer optimizes its acoustic impedance parameters through the KLM model, so that the performance of the sensor can be optimized under flexible conditions.

[0114] Step 2: Sensor installation and electrode design for phased array control

[0115] Brief description: Integrate the flexible sensor with the electrode that can realize phased array control, and complete the sensor installation and electrode design. The main purpose of this step is to achieve independent control and data acquisition of each sub-array element, providing a basis for subsequent signal acquisition and monitoring.

[0116] 1. Phased array electrode design: A combination of bottom-layer common-pole electrodes and upper-layer separated collection electrodes is used. The bottom-layer electrodes are used for common grounding shared by multiple sub-array elements, which simplifies the electrode design and reduces the complexity of the structure. The upper-layer electrodes independently receive the signals of each sub-array element to achieve precise phased array control, thereby completing signal synthesis and decomposition by controlling the phase, amplitude and start-up time of each sub-array element.

[0117] 2. Array sensor installation: Integrate the manufactured piezoelectric array sensor onto a flexible substrate and connect the wireless transmission module via flexible printed circuit board (FPCB) technology to ensure the integrity of the signal transmission path. After installation, the system needs to be calibrated to ensure that the sensor fits the human body surface well.

[0118] Step 3: Signal acquisition and preprocessing

[0119] Brief description: After the sensor is installed, signal acquisition and preprocessing begins, including ultrasonic signal acquisition, filtering and noise processing. This step provides high-quality preliminary data for subsequent cross-section reconstruction and result analysis.

[0120] 1. Signal acquisition: Using a flexible ultrasound patch, scan the heart area based on the phased B-mode mode, and simultaneously collect ultrasound data in different directions through multiple sets of directional linear arrays to obtain two-dimensional cross-sectional imaging of the heart, ensuring that different areas of the heart are covered. Sufficient image data is obtained through multiple scans.

[0121] 2. Signal preprocessing: Use adaptive wavelet transform filtering (to remove noise at multiple resolutions) and SAR-BM3-D joint filtering (to improve image clarity) to process the two-dimensional ultrasound image. The purpose of this step is to eliminate noise and interference to improve imaging quality and signal accuracy, and provide a reliable data basis for subsequent image reconstruction and data analysis.

[0122] Step 4: Cross-sectional reconstruction and 3D volume calculation

[0123] Brief description: Based on the acquired and preprocessed data, cross-sectional reconstruction and left ventricular volume calculation are performed. This step provides a basis for subsequent cardiac function assessment by reconstructing the three-dimensional structure of the heart.

[0124] 1. Cross-sectional reconstruction: Use the ray casting method and the Ray Casting algorithm to perform three-dimensional reconstruction on the preprocessed two-dimensional cross-sectional data, rearrange the grayscale information, and perform filtering and assignment operations to generate a clearer three-dimensional heart image. This step ensures that the reconstructed image depicts the geometric morphology of the heart more comprehensively.

[0125] 2. Three-dimensional volume calculation: Based on the improved Simpson method, the three-dimensional volume calculation of the left ventricle is performed. This method divides the left ventricle into multiple cylinders through two-dimensional imaging data of multiple orthogonal sections, and calculates the volume of each cylinder to obtain the total volume. Compared with the traditional single-section calculation method, the multi-section technology significantly improves the accuracy of the description of the geometric morphology, especially the volume estimation in the case of left ventricular distortion.

[0126] Step 5: Result Analysis and Application of Machine Learning Model

[0127] Brief description: The reconstructed cardiac images and volume calculation results are analyzed using a machine learning model. This step aims to quantitatively evaluate cardiac function and provide real-time monitoring data for the patient's condition.

[0128] 1. Cardiac cavity contour recognition: Use the Canny algorithm to perform contour recognition on the cardiac cavity in the 3D image to extract the boundary information of the cardiac cavity, especially the boundary of the left ventricle in the systolic and diastolic states. This contour information provides the basis for the subsequent ejection fraction calculation.

[0129] 2. Machine learning analysis: Use Gaussian Naive Bayes (GNB) or decision tree models to analyze the extracted cardiac chamber contours and ejection fraction data. Through repeated learning and verification with other ejection fraction calculation methods, the machine learning model can continuously improve the accuracy of cardiac function assessment and provide personalized diagnosis and monitoring recommendations based on the patient's historical data.

[0130] Step 6: Wireless transmission and monitoring alarm system

[0131] Brief description: The analyzed data is sent to the remote receiving end through the wireless transmission module, and the patient's abnormal physiological signals are alarmed through the monitoring alarm system. The purpose of this step is to achieve real-time data transmission and immediate response to abnormal conditions.

[0132] 1. Integration of wireless transmission module: The wireless signal transmission module is integrated with the flexible ultrasonic patch, and Wi-Fi 6 technology is used to achieve real-time data transmission with low power consumption and high anti-interference. The transmission module includes signal acquisition, analog-to-digital conversion, data processing and other units to ensure the stability and reliability of data from acquisition to transmission.

[0133] 2. Monitoring and alarm system: Through the monitoring system integrating acquisition, processing and alarm functions, the physiological signals of the heart are monitored in real time. Once an abnormal ejection fraction or heart rate is detected, the alarm system will sound an alarm through a speaker to remind medical staff to deal with it in time. At the same time, the monitoring system has the function of storing and managing the patient's physiological data, which is convenient for subsequent disease tracking and diagnosis.

[0134] Step 7: Device packaging and clinical application

[0135] Brief description: Finally, the device is packaged as a whole to ensure that it is suitable for clinical use. This step ensures that the device has sufficient stability and safety in actual application.

[0136] 1. Overall waterproof and antibacterial packaging: The wireless module is embedded in the flexible substrate of the patch using a flexible printed circuit board (FPCB), and the device is waterproof and antibacterial packaged to ensure that it can work stably for a long time in a complex clinical environment. This packaging design makes the device suitable for long-term wear by critical care patients and reduces the risk of infection.

[0137] 2. Clinical suitability test: After the device is packaged, clinical suitability test is required to verify the fit and comfort of the device in different patients, as well as the reliability and accuracy of data collection. The test ensures that the device can meet the real-time monitoring needs in critical care and provide effective physiological data support for patients.

[0138] Embodiment 8: This embodiment further limits the multi-section configuration wearable ultrasound patch detection alarm system for CICU intensive care provided in Embodiment 7, and also includes a wireless signal transmission module, including:

[0139] A signal acquisition module, connected to the ultrasonic patch, for collecting ultrasonic signals;

[0140] An analog-to-digital conversion unit, used to convert analog ultrasonic signals into digital signals, providing a basis for subsequent processing;

[0141] A data processing unit, used for pre-processing the collected digital signals, including compression and encoding;

[0142] A wireless communication module, used for transmitting the pre-processed digital signal to a receiving device;

[0143] Anti-interference and calibration modules reduce external interference and ensure signal stability through hardware and software optimization;

[0144] Low noise amplifier, used to amplify ultrasound signals.

[0145] Embodiment 9: This embodiment provides a computer storage medium for storing a computer program. When the computer program is read by a computer, the computer executes the method provided in embodiment 6.

[0146] Embodiment 10: This embodiment provides a computer, including a processor and a storage medium. When the processor reads a computer program stored in the storage medium, the computer executes the method provided in embodiment 6.

[0147] Implementation Method XI: Combination Figure 1-7This embodiment is described in detail by using specific examples to further describe the technical solution provided above. Specifically:

[0148] The sensor consists of Figure 1 As shown, it is composed of multiple sets of directional linear arrays, and the direction of its detection cross section is represented by a red thin line in the figure. The specific three-dimensional size of the piezoelectric ceramic and the scale and direction of the array should be determined according to the needs and are all within the protection scope of this embodiment.

[0149] Figure 1 The white part in the middle is a polymer adhesive material, usually a thin layer of sticky PDMS. It is produced by adding PDMS curing agent at a ratio of 1:7.5 of PDMS main agent mass ratio. After the evenly stirred PDMS mixture is refrigerated in a refrigerator for 12 hours, and then heated in an oven at 45 degrees for 15 hours, a highly sticky PDMS film can be obtained.

[0150] Figure 1 The green part in the middle uses a double-layer acoustic matching layer. The matching layer provides the necessary acoustic impedance gradient to allow the acoustic energy of the transducer to pass smoothly through the body tissue and allow the reflected sound waves (returned echoes) to return to the transducer for detection. In this work, two matching layers are added to the array sensor. Using the KLM model, the acoustic impedance Z1, Z2 and thickness h1, h2 of the matching layer can be estimated:

[0151]

[0152] Where Zp is the acoustic impedance information of the piezoelectric ceramic layer, Zh is the acoustic impedance value of human tissue (usually 1.5 MRayls). λ is the wavelength of ultrasound in the matching layer. At the same time, N is an empirical parameter obtained from the experiment, usually taking a value of 1.1 to 1.14.

[0153] The electrode design described in this embodiment (such as Figure 1 The design of this patent has a unique advantage and can send each sub-array element to achieve phased array control. First, the design of the bottom common electrode enables multiple sub-array elements to share a common electrode, thereby reducing the number and complexity of electrodes. The design of this patent not only simplifies the electrode structure, but also reduces manufacturing costs and improves the reliability and stability of the acoustic system. Secondly, the design of the upper separate collection electrode enables each sub-array element to receive and send signals independently. The design of this patent can realize phased array control, that is, by controlling the phase, amplitude and start-up time of each sub-array element, the synthesis and decomposition of the array signal can be achieved. This control method can be used to expand the monitoring field of view.

[0154] The bottom common electrode helps to reduce crosstalk. Figure 1 The left side shows that the common ground electrode is on the backing side, which introduces the parasitic capacitance crosstalk of the human body. Figure 1 The right side shows the test results of the experiment of the present invention on the matching side, and its stability and reliability test results.

[0155] Figure 2 The medium-dark gray part is the backing layer of the sensor, which is used to prevent the echo and ringing effect of the backward-emitted sound waves from returning to the sensor. In order to obtain a larger sensor bandwidth, by comprehensively considering the bandwidth and signal amplitude, the acoustic impedance design of the backing layer can be selected from 7.5 to 10 MRayls, and nano-scale metal oxide particles can be mixed in each acoustic layer to improve its impedance performance and antibacterial properties (already protected by part of our patent ZL 2023 1 1454624.6).

[0156] A wireless communication module and power supply module for a wearable ultrasound patch used in critical care. This module is stable, reliable, and has low power consumption to ensure long-term use. In addition, the module also has anti-interference capabilities to prevent electromagnetic interference from affecting data transmission. Since the wearable ultrasound patch needs to monitor the patient's physiological parameters for a long time, the power supply module has a high energy density to ensure sufficient endurance. Finally, the integrated package is completed and placed in a sterile package for easy access at any time.

[0157] The figure shows a double cross section, but in reality this patent can cover more cross sections.

[0158] The analysis flow diagram of the signal processing of the wearable ultrasound patch is shown in the figure. Figure 3 As shown in the figure, the process can be roughly divided into four steps: system calibration, data processing, cross-section reconstruction, and result analysis. First of all, system calibration is the basis of the entire process, which involves establishing a Cartesian coordinate system based on the center point of the flexible sensor. The establishment of this coordinate system provides a basic basis for subsequent calculation and analysis, ensuring the accuracy and reliability of the data.

[0159] In the preliminary data processing process, the heart area is first scanned using a phased B-mode scan under known curvature conditions. This scanning method can obtain several sets of cross-sectional imaging. Then, the two-dimensional result graphics are clarified using signal processing methods including but not limited to adaptive wavelet transform filtering and SAR-BM3-D. The purpose of this step is to remove noise and interference in the image and improve the clarity and accuracy of the image.

[0160] In the cross-sectional reconstruction process, the ray casting method and the Ray Casting algorithm are first used to perform rearrangement filtering and assignment operations on the grayscale information of the initially formed three-dimensional image. These algorithms can further optimize the image quality and improve the image clarity and contrast. Finally, the contours of each heart chamber are identified using methods including but not limited to the Canny algorithm. The purpose of this step is to extract important cardiac ejection parameters and provide the necessary data basis for subsequent calculations.

[0161] In the process of result analysis, machine learning can be completed using methods such as Gaussian Naive Bayes GNB or decision trees. These machine learning algorithms can complete the calculation of the extracted contour information and repeatedly learn and verify with other ejection fraction calculation methods, so as to accurately evaluate various cardiac parameters. These analysis results have important reference value for the diagnosis and treatment of diseases.

[0162] Adaptive wavelet transform filtering: effectively removes noise of different scales through multi-resolution analysis and retains key details of cardiac images. SAR-BM3-D: Under the joint filtering of spatial domain and frequency domain, the denoising effect is better while retaining image details to the maximum extent. Ray Casting Method and Ray Casting Algorithm: used for three-dimensional image reconstruction and grayscale information processing, providing higher resolution reconstruction effect, especially suitable for dynamic monitoring of cardiac morphology.

[0163] The functions and calculation methods of wearable ultrasound patches are as follows Figure 6 As shown, the ultrasound patch uses a modified Simpson method and can use multi-section echocardiography, four cross sections, two groups of apical four-chamber sections and two groups of apical two-chamber sections.

[0164] The mathematical basis of the Simpson method for calculating the left ventricular volume is based on an important geometric principle: no matter what the shape of an object is, as long as it is evenly cut into multiple equal parts, the total volume of the object is equal to the sum of the volumes of each equal section. This principle provides a solid mathematical foundation for the Simpson method, making it widely used in calculating the left ventricular volume. Clinical ultrasound usually uses the simplified Simpson method, taking two orthogonal sections at the apex (apical four-chamber and apical two-chamber), dividing the left ventricle into 20 equally divided cylinders along the long axis of the left ventricle, and the sum of the volumes of each cylinder is the left ventricular volume.

[0165] The figure shows a double cross section, but in reality this patent can cover more cross sections.

[0166] like Figure 4 The traditional single orthogonal plane Simpson method only uses a simple estimate of the volume of each layer to perform superposition.

[0167] The improvement of the traditional single orthogonal plane Simpson method in this embodiment is specifically: introducing multiple orthogonal sections to form a three-dimensional volume algorithm, thereby effectively improving the calculation accuracy (such as Figure 5 ). Unlike the traditional single-orthogonal plane Simpson method, the present application is not only a simple repetition of these existing methods, but also achieves a leap forward improvement in volume estimation through a unique multi-orthogonal plane technology. In addition, the present application also adopts an innovative transducer arrangement, which significantly improves the accuracy of volume estimation and reaches a clinical test accuracy level close to that of invasive operations such as the Swan-Ganz catheter method.

[0168] Depend on Figure 6 The short-axis section of the papillary muscle shown in the upper middle right shows that the heart cross section is not a complete ellipse. Figure 3 In the right figure below, two sets of cross sections are used to fully represent the accuracy of the cross-sectional area calculation of this layer. This calculation method corrects the distorted geometry of the left ventricle as much as possible and completes the description of the left ventricle as completely as possible. In addition, for devices with multiple cross sections, a virtual center line can also be established to ensure the consistency of the center line in three-dimensional space as much as possible.

[0169] The monitoring and alarm system of the wearable ultrasound patch is a system that integrates the functions of acquisition, processing, analysis and alarm. Its basic function is to monitor the physiological signals of the human body in real time and generate alarm signals for abnormal physiological signals, and process and analyze these signals to provide useful information.

[0170] The basic functions and processing flow of the monitoring and alarm system of the wearable ultrasound patch are as follows: Figure 7 As shown:

[0171] Acquisition system of multi-section wearable ultrasound patch for CICU intensive care: The main power supply system consists of AC / DC mains conversion system and voltage stabilization system. The regulated power supply is used as system power supply and standard reference voltage value input to the array sensor acquisition system. These sensors are responsible for collecting multiple signals. Front-end core control unit: Under a unified clock, the acquisition and transmission system is distributed and collected at high speed. This means that all sensors collect data at the same time to ensure the real-time and accuracy of the data. Mid-end preprocessing unit: As a hardware adaptive noise reduction system, it processes noise and crosstalk to provide basic support for the next step of signal recognition. This step is very important for improving signal quality and reducing noise interference. Back-end signal processing system: Proposes information that can be read. This information can include physiological parameters such as heart rate, blood pressure, ejection fraction, etc., and completes relevant warning preprocessing. Interactive interface control system: After integration, basic operations can be performed on these data, such as date coding and creating or deleting. This allows users to easily view and analyze data. Warning part: When the piezoresistive sensor is changed and different from the standard value, the voltage signal output by the sensor is collected in the amplification module. The voltage value can be set by programming. When it exceeds the set value, the speaker starts to alarm. This can remind users to pay attention to changes in their physical condition in a timely manner. Peripheral control devices and storage units: basic control and storage of physiological and motion signals for subsequent detailed analysis with the help of doctors. These data can be used to diagnose diseases, evaluate treatment effects, etc.

[0172] Wireless signal transmission unit of a multi-section configuration wearable ultrasound patch for CICU intensive care: In order to enable the wearable ultrasound patch to achieve efficient and stable wireless signal transmission to meet the needs of real-time monitoring in intensive care, this embodiment specially designs a low-power, high-anti-interference wireless signal transmission module. The wireless signal transmission module includes the following parts: Signal acquisition module: connected to the ultrasound patch sensor to collect ultrasound signals. Analog-to-digital conversion unit (ADC): converts analog ultrasound signals into digital signals to provide a basis for subsequent processing. Data processing unit (MCU): pre-processes the collected digital signals (such as compression and encoding). Wireless communication module (Wi-Fi 6): transmits the pre-processed digital signals to the receiving device. Anti-interference and calibration module: through hardware and software optimization, reduce external interference and ensure signal stability.

[0173] Among them, the analog-to-digital conversion unit (ADC) can select AD7768 (24-bit high-precision ADC) or other devices that can support 24-bit or higher resolution, can accurately capture the slight changes of ultrasonic signals, and have low power consumption characteristics, which are suitable for long-term operation of wearable devices. The data processing unit (MCU) can select the STM32L4 series (low-power MCU) or other processors with a main frequency of 80MHz to meet the needs of signal preprocessing and compression algorithms. The wireless communication module can select ESP32-C6 or other dual-mode communications that support Wi-Fi 6 and BLE, which are suitable for large data transmission and complex scenarios. The antenna part can select Taoglas's flexible FPC antenna series because it adapts to the curved structure of wearable devices. The low-noise amplifier (LNA) can select LMP7721 or other amplifiers that can provide high gain, which is suitable for amplifying ultrasonic signals and ensuring the signal-to-noise ratio (SNR). The anti-interference shielding material uses a copper-plated aluminum foil shielding layer, which is suitable for the packaging design of flexible devices due to its easy integration.

[0174] Integration with Ultrasonic Patch When integrating the wireless signal transmission module with the flexible ultrasonic patch, an integrated packaging design is adopted to maintain the miniaturization and flexibility of the module:

[0175] Step 1: Connect the wireless module to the ultrasonic sensor circuit board to ensure the integrity of the signal path.

[0176] Step 2: Use flexible printed circuit board (FPCB) technology to embed the wireless module into the flexible substrate of the patch.

[0177] Step 3: Complete the overall waterproof and antibacterial packaging to ensure that the equipment is suitable for long-term wear and complex environments by critically ill patients.

[0178] This embodiment uses flexible miniaturized manufacturing technology to successfully realize the wearable manufacturing of multi-section linear array probes. The application of this technology makes the ultrasonic sensor array elements richer, thus having certain self-compensation or error correction functions. Through the further development of ultrasonic phased array technology, the array probe can be deflected at a certain angle in the vertical direction, thereby improving the field of view to a level of plus or minus 30 degrees. In intensive care, this wearable multi-section linear array probe and monitoring system can monitor the patient's vital signs in a long-term and stable manner. Due to its miniaturized and flexible characteristics, it can be easily attached to the patient without restricting the patient's activities. At the same time, its self-compensation or error correction function can also effectively reduce errors and improve the accuracy and reliability of monitoring.

[0179] In this embodiment, the orthogonal multi-section configuration is one of the core innovations, and this design is directly related to the accurate presentation and calculation of ejection fraction. The traditional single-section imaging method is difficult to fully present the geometric shape of the heart, especially in the volume measurement of complex three-dimensional structures such as the left ventricular region. The following are the technical advantages of the orthogonal multi-section configuration in ejection fraction measurement:

[0180] 1 Accurate depiction of the complete ventricular volume The orthogonal multi-section configuration provides a more comprehensive and accurate basis for the volume measurement of the left ventricle by obtaining cross-sectional images on two mutually perpendicular planes, the apical four-chamber and the two-chamber. Based on the calculation principle of the Simpson method, this configuration divides the left ventricle into multiple cylindrical units and accumulates the volume of each unit to obtain the total volume of the left ventricle. Compared with single-plane images, orthogonal multi-sections can more accurately capture the geometric complexity of the ventricle, especially in the case of ventricular distortion, providing a true depiction of the morphology of the heart chamber.

[0181] 2. Improved accuracy of ejection fraction calculation. Ejection fraction is an important indicator for evaluating cardiac function, and its accuracy depends on the accurate measurement of left ventricular volume in systolic and diastolic states. Through the orthogonal multi-section configuration, this embodiment can obtain images of the apical four chambers and two chambers on different planes, respectively, reducing shape distortion in a single plane, making the calculation of end-systolic volume and end-diastolic volume more accurate. This directly improves the accuracy of ejection fraction calculation and helps provide reliable data support for clinical diagnosis.

[0182] 3. Ability to correct the distortion of the left ventricle The geometric shape of the left ventricle is often not a regular ellipse, and it will be obviously distorted, especially in pathological conditions. Orthogonal multi-section imaging can capture the structural information of the left ventricle in different directions by imaging in two perpendicular planes, making up for the morphological deviation under a single perspective. This configuration allows the overall morphology of the ventricle to be accurately presented, thereby correcting the volume measurement error caused by morphological distortion in traditional methods.

[0183] 4. Guarantee of 3D centerline consistency The multi-section configuration can also ensure the spatial consistency of data in different image planes by constructing a virtual 3D centerline. This consistency is crucial in ejection fraction calculation, and can effectively reduce data deviations between planes and further improve the calculation accuracy of ejection fraction.

[0184] 5. Multi-section data fusion improves signal-to-noise ratio

[0185] Through orthogonal multi-section imaging, this embodiment realizes the fusion processing of multi-view data. This configuration can obtain cardiac images from different angles, and can compensate for noise effects in the process of data fusion, thereby further improving the signal-to-noise ratio. This multi-section data fusion provides higher accuracy and reliability when extracting parameters such as cardiac morphology and ejection fraction.

[0186] The preparation method of the polymer adhesive material is as follows:

[0187] Polymer adhesive materials usually use a thin layer of PDMS with viscosity. The production method is to add PDMS curing agent in a ratio of 1:7.5 of PDMS main agent mass ratio. After the PDMS mixture is evenly stirred, it is refrigerated in a refrigerator for 12 hours, and then heated in an oven at 45 degrees for 15 hours to obtain a highly viscous PDMS film.

[0188] Piezoelectric wire array production method:

[0189] Cutting is a commonly used method of operation, which is simple to operate and has significant effects. When performing the cutting method, a key equipment, a diamond cutting machine, is required. The steps of the cutting method are as follows:

[0190] I. Place the complete piezoelectric ceramic on the cutting platform of the diamond cutting machine. Then, the cutting tool is controlled to make equidistant deep grooves on the piezoelectric ceramic. The purpose of this step is to form a certain number of deep grooves on the piezoelectric ceramic to prepare for the subsequent filling operation.

[0191] II. Rotate the cutting platform 90 degrees. The purpose of this step is to cut the same equidistant deep grooves on the other side of the piezoelectric ceramic. Through this operation, deep grooves can be formed on both sides of the piezoelectric ceramic, thereby increasing its structural strength and stability.

[0192] III. Fill the deep groove with epoxy resin. Epoxy resin is a commonly used packaging material with excellent mechanical properties and corrosion resistance. After filling the epoxy resin, it needs to be cured at room temperature. The cured epoxy resin will be tightly combined with the piezoelectric ceramic to form a solid structure. Finally, the piezoelectric ceramic base is removed by grinding to obtain the final cutting result.

[0193] The technical solution provided by the present invention is further described in detail above through several specific implementation modes in order to highlight the advantages and benefits of the technical solution provided by the present invention. However, the several specific implementation modes described above are not intended to be used as limitations on the present invention. Any reasonable modification and improvement of the present invention, combination of implementation modes and equivalent substitution within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A multi-section wearable ultrasound patch for CICU intensive care, characterized in that: The patch comprises: A flexible substrate for attaching to the patient's skin surface; A multi-section linear array ultrasonic sensor for acquiring two-dimensional ultrasonic images of multiple sections; The electrode capable of realizing the phased array control function comprises sub-array elements arranged in a double-layer array, and a bottom common-pole electrode and an upper separated collection electrode, wherein the bottom common-pole electrode is used for common grounding of the sub-array elements, and the upper separated collection electrode is used for independently receiving the signal of each sub-array element; Acoustic matching layer, used to provide acoustic impedance gradient.

2. The multi-section wearable ultrasound patch for CICU intensive care according to claim 1, characterized in that: The flexible substrate, the multi-section linear array ultrasonic sensor, the phased array control electrode and the double-layer acoustic matching layer are tightly fitted in sequence.

3. The multi-section wearable ultrasound patch for CICU intensive care according to claim 1, characterized in that: The flexible substrate is made of a flexible material having adhesiveness.

4. The multi-section wearable ultrasound patch for CICU intensive care according to claim 1, characterized in that: The multi-section linear array ultrasonic sensor comprises at least two groups of linear arrays in different directions.

5. The multi-section wearable ultrasound patch for CICU intensive care according to claim 1, characterized in that: The acoustic matching layer is a double-layer acoustic matching layer.

6. A signal processing method for a multi-section wearable ultrasound patch for CICU intensive care, characterized in that: The method is applied to the patch according to claim 1, comprising: The steps of collecting and denoising the patient's heart scan data; A step of reconstructing three-dimensional graphics on the scan data; A step of identifying the outline of the three-dimensional figure; The step of analyzing the contour according to a preset analysis method.

7. A multi-section wearable ultrasound patch detection alarm system for CICU intensive care, characterized in that: The system is implemented based on the patch according to claim 1, and includes: The main power supply system, including the AC / DC mains conversion system and the voltage stabilization system, is used as the standard reference voltage value input to the array sensor acquisition system; The arrayed sensor acquisition system is used to collect multi-channel signals; A front-end core control unit performs distributed acquisition of the arrayed sensor acquisition system under a unified clock; The mid-end pre-processing unit, as an adaptive noise reduction system, pre-processes the collected signals; A back-end signal processing system, used to process the pre-processed signals and generate information containing physiological parameters; When the signal detected by the piezoresistive sensor in the arrayed sensor acquisition system is different from the standard value, a reminder signal is output.

8. The multi-section wearable ultrasonic patch detection alarm system for CICU intensive care according to claim 7, characterized in that: Also includes a wireless signal transmission module, including: A signal acquisition module, connected to the ultrasonic patch, for collecting ultrasonic signals; An analog-to-digital conversion unit, used to convert analog ultrasonic signals into digital signals, providing a basis for subsequent processing; A data processing unit, used for pre-processing the collected digital signals, including compression and encoding; A wireless communication module, used for transmitting the pre-processed digital signal to a receiving device; Anti-interference and calibration modules reduce external interference and ensure signal stability through hardware and software optimization; Low noise amplifier, used to amplify ultrasound signals.

9. A computer storage medium for storing a computer program, characterized in that: When the computer program is read by a computer, the computer executes the method according to claim 6.

10. A computer, comprising a processor and a storage medium, characterized in that: When the processor reads the computer program stored in the storage medium, the computer executes the method of claim 6 .

Images