A home-based intelligent monitoring and rehabilitation training system, method, terminal and medium for arteriovenous fistula

Abstract

This invention discloses a home-based intelligent monitoring and rehabilitation training system, method, terminal, and medium for arteriovenous fistulas (AVFs), relating to the field of smart medical technology. The system includes: home-based devices, a cloud server, and a software interactive management platform. The home-based devices include an AVF monitoring wristband and an intelligent arm exerciser. The AVF monitoring wristband collects AVF blood flow acoustic signature data and provides early warnings to the patient. The intelligent arm exerciser collects motion parameters and provides rehabilitation training and movement guidance functions for the patient. The cloud server processes the collected AVF blood flow acoustic signature data and motion parameters, performs thrombosis risk analysis, and generates a rehabilitation training plan. The software interactive management platform allows real-time viewing of the patient's AVF status, rehabilitation training plan, and exercise records. This invention fills the gap in integrated home-based AVF monitoring, early warning, rehabilitation training, and feedback, providing an intelligent solution for reducing AVF failure rate and extending AVF lifespan.

Description

Technical Field

[0001] This invention relates to the field of smart medical technology, and in particular to a home-based intelligent monitoring and rehabilitation training system, method, terminal and medium for arteriovenous fistula. Background Technology

[0002] Autologous arteriovenous fistulas (AVFs) are a lifeline for patients with end-stage renal disease undergoing hemodialysis. Through surgical anastomosis of the patient's own artery and vein, a high-flow vascular access is created to provide sufficient blood flow for extracorporeal circulation during hemodialysis. However, postoperative management of AVFs faces significant challenges: the failure rate after AVF surgery is relatively high, and thrombosis is the leading cause of fistula function loss, with vascular stenosis being a direct trigger for thrombosis. Furthermore, acute risks in home settings such as pressure during sleep or crushing by heavy objects can easily lead to sudden fistula occlusion.

[0003] The specific deficiencies in existing arteriovenous fistula management are as follows:

[0004] (1) There is a lack of home monitoring methods, the early warning is delayed and the cost is high. There are no home devices on the market that support real-time monitoring of fistula health, and acute risks such as nighttime compression cannot be warned in time.

[0005] (2) Poor rehabilitation training experience and low compliance: Traditional rehabilitation equipment such as grip balls and elastic balls only provide mechanical training, without quantitative feedback such as grip strength and wrist joint angle. Moreover, the interactive form is boring.

[0006] (3) Data fragmentation and lack of collaborative management loop: Monitoring data such as voiceprint and rehabilitation data such as training time and grip strength are not linked, and there is a lack of closed-loop management of "monitoring, early warning, rehabilitation training and feedback".

[0007] Therefore, existing technologies still have shortcomings. Summary of the Invention

[0008] The technical problem to be solved by this invention is to provide a home-based intelligent monitoring and rehabilitation training system, method, terminal, and medium for arteriovenous fistulas, addressing the aforementioned deficiencies of the prior art. The technical solution adopted by this invention is as follows:

[0009] In a first aspect, the present invention provides a home-based intelligent monitoring and rehabilitation training system for arteriovenous fistulas, the system comprising: home-based devices, a cloud server, and a software interactive management platform, wherein the home-based devices, the cloud server, and the software interactive management platform are interconnected in terms of data.

[0010] The home-based device includes an arteriovenous fistula (AVF) monitoring bracelet and an intelligent arm exerciser, which are interconnected. The AVF monitoring bracelet is used to collect the patient's AVF blood flow acoustic data and provide early warning prompts to the patient based on thrombosis risk information. The intelligent arm exerciser is used to collect the patient's movement parameters and provide the patient with rehabilitation training and movement guidance functions.

[0011] The cloud server is used to standardize the collected arteriovenous fistula blood flow acoustic print data and motion parameters, and to perform thrombosis risk analysis based on the arteriovenous fistula blood flow acoustic print data to generate a rehabilitation training program.

[0012] The software interaction management platform includes a software platform for patient terminals and a software platform for medical staff terminals, which are used to view the patient's arteriovenous fistula status, rehabilitation training plan, and exercise records in real time.

[0013] In one implementation, the arteriovenous fistula monitoring wristband includes:

[0014] A dual-channel voiceprint acquisition unit includes a main microphone and an auxiliary microphone. The main microphone is located inside the strap of the fistula monitoring wristband and is used to acquire the voiceprint data of the fistula blood flow. The auxiliary microphone is located outside the fistula monitoring wristband and is used to acquire ambient noise. The ambient noise is used to achieve differential noise reduction.

[0015] A graded early warning module, comprising a sound-emitting component, a multi-color indicator light, and a vibration component, wherein the graded early warning module is used to control one or more of the sound-emitting component, the multi-color indicator light, and the vibration component to provide early warning prompts according to the thrombosis risk information corresponding to the thrombosis early warning level.

[0016] A low-power control module is used to manage the power consumption of each module in the fistula monitoring wristband;

[0017] The data transmission module is used to transmit the blood flow acoustic data of the arteriovenous fistula to the cloud server.

[0018] In one implementation, the intelligent arm exerciser includes:

[0019] A six-axis motion sensor is used to capture the patient's wrist movements and collect the wrist joint angle during the movement.

[0020] A flexible stress sensor is used to record changes in the grip force of a patient's fingers, and the grip force changes can be converted into voltage signals through a bridge circuit.

[0021] The resistance level adjustment module provides multiple resistance levels to suit patients with different muscle strength levels.

[0022] The motion guidance module is used to provide prompts and guidance when patients' movements are not standardized.

[0023] The data recording module is used to record the duration of each patient's training session, the number of movements, and the rate of achieving the required level.

[0024] Secondly, embodiments of the present invention also provide a home-based intelligent monitoring and rehabilitation training method for arteriovenous fistulas, wherein the home-based intelligent monitoring and rehabilitation training method for arteriovenous fistulas is applied to any one of the above-described solutions of the home-based intelligent monitoring and rehabilitation training system for arteriovenous fistulas, and the method includes:

[0025] The patient's arteriovenous fistula (AVF) blood flow acoustic print data is collected in real time using an AVF monitoring wristband. The AVF blood flow acoustic print data is then standardized and analyzed for thrombosis risk to obtain thrombosis risk information.

[0026] The patient's movement parameters are collected by an intelligent arm exerciser, and the movement parameters are processed to obtain the patient's movement data.

[0027] Based on the thrombosis risk information, patient basic information, patient muscle strength level, duration of arteriovenous fistula use, and exercise data, a rehabilitation training plan is generated and transmitted to the patient terminal and medical staff terminal.

[0028] In one implementation, the arteriovenous fistula blood flow acoustic signature data is standardized, including:

[0029] Environmental noise in the arteriovenous fistula blood flow acoustic data is filtered out by adaptive differential noise reduction and Butterworth bandpass filtering;

[0030] Add an arteriovenous fistula duration label to the blood flow acoustic print data of the arteriovenous fistula after filtering out environmental noise.

[0031] In one implementation, thrombosis risk analysis is performed based on the processed arteriovenous fistula blood flow acoustic signature data to obtain thrombosis risk information, including:

[0032] Voiceprint features were extracted from the arteriovenous fistula blood flow voiceprint data after filtering out environmental noise.

[0033] The extracted voiceprint features are input into a support vector machine model for thrombosis risk analysis, and the thrombosis risk probability is output. The thrombosis risk probability is used as the thrombosis risk information.

[0034] In one implementation, the method further includes:

[0035] Based on the thrombosis risk information, a thrombosis warning level is determined, and the thrombosis warning level information is transmitted to the arteriovenous fistula monitoring wristband, the patient terminal, and the medical care terminal.

[0036] The fistula monitoring bracelet controls one or more of the following components—sound component, multi-color indicator light, and vibration component—to provide early warning based on the thrombosis warning level information, and provides synchronized early warning in conjunction with the patient terminal and medical staff terminal.

[0037] In one implementation, the method further includes:

[0038] After the patient performs the rehabilitation training program, the latest arteriovenous fistula blood flow voiceprint data uploaded by the arteriovenous fistula monitoring wristband and the latest exercise data uploaded by the smart arm exerciser are obtained.

[0039] The rehabilitation training program is updated based on the latest arteriovenous fistula blood flow voiceprint data and the latest exercise data.

[0040] Thirdly, embodiments of the present invention also provide a terminal, wherein the terminal includes a memory, a processor, and a home-based intelligent monitoring and rehabilitation training program for arteriovenous fistulas stored in the memory and executable on the processor. When the processor executes the home-based intelligent monitoring and rehabilitation training program for arteriovenous fistulas, it implements the steps of any of the above-mentioned solutions for the home-based intelligent monitoring and rehabilitation training method for arteriovenous fistulas.

[0041] Fourthly, embodiments of the present invention also provide a computer-readable storage medium, wherein the computer-readable storage medium stores a home-based intelligent monitoring and rehabilitation training program for arteriovenous fistulas, and the home-based intelligent monitoring and rehabilitation training program for arteriovenous fistulas implements the steps of the home-based intelligent monitoring and rehabilitation training method for arteriovenous fistulas as described in any of the above schemes on the computer-readable storage medium.

[0042] Beneficial Effects: Compared with existing technologies, this invention provides a home-based intelligent monitoring and rehabilitation training system for arteriovenous fistulas, comprising: home devices, a cloud server, and a software interactive management platform, all three of which are interconnected. The home devices include a fistula monitoring wristband and an intelligent arm exerciser, which are interconnected. The fistula monitoring wristband collects the patient's fistula blood flow acoustic signature data and provides early warnings based on thrombosis risk information. The intelligent arm exerciser collects the patient's movement parameters and provides rehabilitation training and movement guidance functions. The cloud server standardizes the collected fistula blood flow acoustic signature data and movement parameters, performs thrombosis risk analysis based on the data, and generates a rehabilitation training plan. The software interactive management platform includes a patient terminal software platform and a medical staff terminal software platform, used to view the patient's fistula status, rehabilitation training plan, and exercise records in real time.

[0043] This invention uses an arteriovenous fistula monitoring wristband to monitor blood flow and voiceprint data in the arteriovenous fistula in real time, which helps to provide timely warnings of thrombosis risk to patients. Furthermore, it uses an intelligent arm exerciser to collect and analyze patients' movement parameters, which helps to record training duration, number of movements, and achievement rate in real time, providing a basis for the generation and adjustment of subsequent rehabilitation training programs.

[0044] Furthermore, the software platform established by this invention for patient terminals and medical staff terminals not only allows patients to monitor the status of their fistulas in real time, but also enables medical staff to achieve remote and precise management. This fills the gap in the integration of home fistula monitoring, early warning, rehabilitation training, and feedback, providing an intelligent solution for reducing fistula failure rate and extending fistula lifespan, and combining clinical practicality with cost-effectiveness. Attached Figure Description

[0045] Figure 1 This is an architecture diagram of the home-based intelligent monitoring and rehabilitation training system for arteriovenous fistulas provided in an embodiment of the present invention.

[0046] Figure 2 This is a schematic diagram of the lithium battery module of the fistula monitoring bracelet in the home-based intelligent monitoring and rehabilitation training system for arteriovenous fistulas provided in an embodiment of the present invention.

[0047] Figure 3 This is a schematic diagram of the magnetic charging component of the fistula monitoring bracelet in the home-based intelligent monitoring and rehabilitation training system for arteriovenous fistulas provided in an embodiment of the present invention.

[0048] Figure 4 This is a flowchart illustrating a preferred embodiment of the home-based intelligent monitoring and rehabilitation training method for arteriovenous fistulas provided in this invention.

[0049] Figure 5 This is a schematic diagram illustrating the execution principle of the home-based intelligent monitoring and rehabilitation training method for arteriovenous fistulas provided in this embodiment of the invention.

[0050] Figure 6 A schematic diagram of a terminal provided in an embodiment of the present invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and effects of this invention clearer and more explicit, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0052] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content, operations, or steps, nor does it require execution in the described order. For example, some operations or steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0053] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0054] It should be understood that, in order to clearly describe the technical solutions of the embodiments of the present invention, the terms "first" and "second" are used in the embodiments of the present invention to distinguish identical or similar items with essentially the same function and effect. For example, "first control information" and "second control information" are only used to distinguish different control information and do not limit their order.

[0055] Those skilled in the art will understand that the words "first" and "second" do not limit the quantity or the order of execution, and that the words "first" and "second" do not necessarily imply that they are different.

[0056] It should also be understood that the term “and / or” as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0057] To address the problems existing in the prior art, this embodiment provides a home-based intelligent monitoring and rehabilitation training system for arteriovenous fistulas, such as... Figure 1 As shown, the system mainly includes home devices, a cloud server, and a software interaction management platform, forming a core architecture of "home device deployment - intelligent data analysis on the cloud server - intelligent management on the software interaction management platform." This constructs a closed-loop, remote, intelligent home-based arteriovenous fistula (AVF) monitoring and rehabilitation training system, specifically addressing issues such as delayed early warnings, lack of quantitative rehabilitation, and data fragmentation in existing AVF management. The home devices in this embodiment mainly include an AVF monitoring wristband and an intelligent arm exerciser. The AVF monitoring wristband collects the patient's AVF blood flow acoustic signature data and provides early warnings based on thrombosis risk information. The intelligent arm exerciser collects the patient's movement parameters and provides rehabilitation training and movement guidance functions. The AVF monitoring wristband and the intelligent arm exerciser can achieve data interconnection via Bluetooth and Wi-Fi dual-mode communication, adapting to the portability needs of home scenarios and facilitating stable interaction and data synchronization between the AVF monitoring wristband and the intelligent arm exerciser, thereby constructing a dual-device collaborative interaction mechanism.

[0058] Specifically, in combination Figure 1 As shown in the figure, the arteriovenous fistula monitoring wristband of this embodiment includes: a dual-channel voiceprint acquisition unit, a graded early warning module, a low-power control module, and a data transmission module, etc. Figure 1The main modules of the arteriovenous fistula (AVF) monitoring wristband are described. The dual-channel voiceprint acquisition unit includes a main microphone and an auxiliary microphone. The main microphone is located inside the wristband's strap and is used to acquire AVF blood flow voiceprint data. The auxiliary microphone is located outside the wristband and is used to acquire ambient noise. This ambient noise is used for differential noise reduction to ensure a signal-to-noise ratio of 65dB for the AVF blood flow voiceprint data, far exceeding the 30dB interference level of traditional single microphones. The graded warning module includes a sound-emitting component, a multi-color indicator light, and a vibration component. This module controls one or more of these components to provide warnings based on the thrombosis risk level. In practical applications, different components are activated for different warning levels. For example, a level one warning (low risk) triggers a slow flashing yellow indicator light combined with intermittent vibration. A level two warning (medium risk) triggers a fast flashing red indicator light combined with continuous vibration. If it is a Level 3 warning, indicating a high risk, a rapid flashing red indicator light combined with a strong vibration component can be triggered, along with an audible alert from the sound-emitting component to ensure no warnings are missed during nighttime sleep. The low-power control module in this embodiment manages the power consumption of each module in the fistula monitoring bracelet. For example, it monitors the power consumption when the dual-channel voiceprint acquisition unit is activated and enters sleep mode when the dual-channel voiceprint acquisition unit is idle. This low-power control module can also be paired with a lithium battery module and a magnetic charging component to extend battery life. The data transmission module in this embodiment transmits fistula blood flow voiceprint data to the cloud server. Specifically, the data transmission module can stably transmit fistula blood flow voiceprint data and motion parameters to the cloud server via a wireless communication antenna, covering common home scenarios. Furthermore, when transmitting to the cloud server, for patients less than one month post-fistula surgery, the data transmission automatically includes a fistula duration tag, allowing the cloud server to prioritize this type of data.

[0059] Specifically, in combination Figure 2 As shown, Figure 2 A structural schematic diagram of the lithium battery module is provided. The lithium battery module is located inside the outer casing 201 of the fistula monitoring wristband. For ease of demonstration, Figure 2 The diagram shown is a partial breakdown of the lithium battery module's structure. Figure 2 As can be seen, the lithium battery module includes a power button 202 and a lithium battery charging circuit board 203, providing battery life for the fistula monitoring bracelet. Combined with the monitoring circuit board 204, it can detect the power consumption of other modules, facilitating power management of those modules. Furthermore, the lithium battery module can also be paired with a magnetic charging component, such as... Figure 3 As shown, Figure 3This is a schematic diagram of the magnetic charging component inside the arteriovenous fistula monitoring bracelet. This magnetic charging component can be part of the lithium battery module, or it can be set up independently and connected to the lithium battery module. Figure 3 As can be seen, the magnetic charging component includes a magnetic charging interface base 301, which can be mounted on a sliding cover 302 and forms a sliding connection by cooperating with a sliding groove 303 on the lithium battery module. In one implementation, the magnetic charging component supports interface-free operation, avoiding interference from traditional interfaces on the wearer of the fistula monitoring bracelet. Furthermore, it provides a power-saving mode, such as automatically reducing the sampling frequency of the fistula monitoring bracelet between 10:00 PM and 6:00 AM the next day to achieve long battery life, balancing monitoring and ease of use.

[0060] Furthermore, the intelligent arm exerciser in this embodiment is an intelligent exercise device for training grip and arm strength. It serves as a carrier for collecting motion parameters, integrating both rehabilitation training and motion parameter collection functions. Specifically, it combines... Figure 1 As shown, the intelligent arm exerciser includes: a six-axis motion sensor, a flexible stress sensor, a resistance level adjustment module, a motion guidance module, and a data recording module. Figure 1The main modules of the intelligent arm exerciser are described. A six-axis motion sensor and a flexible stress sensor are built into the exerciser. The six-axis motion sensor captures the patient's wrist movements and collects the wrist joint angle during these movements. The collected wrist joint angle can be used to determine the correctness of the rehabilitation movements. The flexible stress sensor records the patient's finger flexion and extension, and changes in grip strength. These grip strength changes can be converted into voltage signals via a bridge circuit, quantifying muscle strength levels and providing objective data support for adjusting subsequent rehabilitation training programs. A resistance level adjustment module provides multiple resistance levels, specifically a knob that allows adjustment of the resistance level to suit patients with different muscle strength levels. The movement guidance module in this embodiment can include a built-in vibration component and indicator lights. Through the coordinated control of the vibration component and indicator lights, prompts and guidance are provided when the patient's movements are not correct. For example, when the collected wrist joint angle deviation is >5° or the grip strength is less than 80% of the target value, the vibration component emits a vibration prompt combined with a flashing red indicator light to remind the patient to perform the movements correctly. The data recording module records the duration of each training session, the number of movements, and the rate of achievement of correct movements, generating exercise data. This data is then synchronized to the arteriovenous fistula (AVF) monitoring wristband via Bluetooth and subsequently uploaded to a cloud server. Furthermore, in practical applications, this embodiment can associate the AVF duration tag from the collected AVF blood flow acoustic signature data with the exercise data, facilitating subsequent analysis of the impact of exercise on AVFs of different durations. This embodiment's intelligent arm exerciser innovatively integrates a six-axis motion sensor and a flexible stress sensor, achieving simultaneous quantification of movement accuracy and muscle strength levels. Its resistance level adjustment module can adapt to the low muscle strength needs of sarcopenic patients, filling a gap in rehabilitation equipment for this population.

[0061] Furthermore, the cloud server in this embodiment can also standardize the collected arteriovenous fistula (AVF) blood flow acoustic signature data and motion parameters. Specifically, when standardizing the AVF blood flow acoustic signature data, the cloud server first filters out environmental noise (such as talking sounds, television sounds, etc.) in the AVF blood flow acoustic signature data through adaptive differential denoising and Butterworth bandpass filtering. Next, the cloud server can add an AVF duration tag to the AVF blood flow acoustic signature data after filtering out environmental noise. In this way, the cloud server can analyze and process the AVF blood flow acoustic signature data according to the analysis rules corresponding to the AVF duration tag. Regarding the processing of motion parameters, not only can the patient's single training duration, number of movements, and movement attainment rate be summarized to obtain the patient's motion data, but the AVF duration tag of the collected AVF blood flow acoustic signature data can also be associated with the motion data.

[0062] Furthermore, the cloud server can also perform thrombosis risk analysis based on the arteriovenous fistula (AVF) blood flow voiceprint data and generate a rehabilitation training plan. In this embodiment, during the thrombosis risk analysis, voiceprint features are extracted from the AVF blood flow voiceprint data after filtering out environmental noise. The extracted voiceprint features are then input into a support vector machine model for thrombosis risk analysis, outputting a thrombosis risk probability, which is the thrombosis risk information. The voiceprint features extracted in this embodiment include: short-time energy, short-time average amplitude, short-time zero-crossing rate, root mean square of audio data, spectral centroid, spectral width, short-time autocorrelation function, power spectrum, and Mel-frequency cepstral coefficients of the AVF blood flow voiceprint data.

[0063] Specifically, short-time energy refers to the sum of squares of the arteriovenous fistula blood flow acoustic signature data over a period of time, and is calculated as shown in the following formula (1):

[0064] (1)

[0065] In formula (1), Short-term energy within a single analysis window (e.g., 500ms); For the first Signal amplitude of the acoustic fingerprint data of arteriovenous fistula blood flow at each sampling point; This is the sampling point index, with values ​​ranging from 0, 1, 2, ... (N=500, corresponding to 500 sampling points); The total number of sampling points in a single sampling is fixed at 500. By calculating the short-time energy, the energy of the arteriovenous fistula blood flow acoustic signature data over a period of time can be obtained, and the degree of blockage of the arteriovenous fistula can be analyzed based on the energy.

[0066] The short-term average amplitude refers to the average absolute value of the arteriovenous fistula blood flow acoustic signature data over a period of time, and is calculated as shown in the following formula (2):

[0067] (2)

[0068] In formula (2), The short-term average amplitude within a single analysis window; For the first Signal amplitude of the acoustic fingerprint data of arteriovenous fistula blood flow at each sampling point; This is the sampling point index, with values ​​ranging from 0, 1, 2, ... ; The total number of sampling points in a single sampling is fixed at 500. : No. The absolute value of the signal amplitude at each sampling point; This represents the sum of the absolute values ​​of the signals at all sampling points. Short-time average amplitude can also obtain the energy of the arteriovenous fistula blood flow acoustic signature data over a period of time. Unlike short-time energy, short-time average amplitude does not emphasize peak values, meaning it is less affected by peak data.

[0069] The short-time zero-crossing rate refers to the number of times the signal of the arteriovenous fistula blood flow acoustic signature crosses the zero point within a certain period of time. The calculation method is shown in formula (3):

[0070] (3)

[0071] In formula (3), Short-term zero-crossing rate within a single analysis window (times / window); For the first Signal amplitude of the acoustic fingerprint data of arteriovenous fistula blood flow at each sampling point; For the first Signal amplitude of the acoustic fingerprint data of arteriovenous fistula blood flow at each sampling point; This is the sampling point index, with values ​​ranging from 0, 1, 2, ... ; The total number of sampling points in a single sampling is fixed at 500. For symbolic functions, the rules are: ; The absolute value of the sign change between adjacent sampling points (2 when it changes, 0 when it remains unchanged); coefficient This represents the conversion of the number of sign changes into the number of zero crossings, meaning that every two sign changes correspond to one zero crossing. The short-term zero-crossing rate can also reflect the frequency of the arteriovenous fistula blood flow acoustic signature data, and can be used to distinguish between clear and dull sounds in the arteriovenous fistula.

[0072] The root mean square of the audio data refers to the square root of the average of the squared signals of the arteriovenous fistula blood flow acoustic signature data. The specific calculation method is shown in formula (4):

[0073] (4)

[0074] In formula (4), The root mean square of the signal within a single analysis window represents the effective amplitude of the arteriovenous fistula blood flow acoustic signature data; For the first Signal amplitude of the acoustic fingerprint data of arteriovenous fistula blood flow at each sampling point; This is the sampling point index, with values ​​ranging from 0, 1, 2, ... ; The total number of sampling points in a single sampling is fixed at 500. For the first The square of the signal amplitude at each sampling point; It represents the sum of the squares of the signals at all sampling points.

[0075] Variance refers to the dispersion of the arteriovenous fistula blood flow acoustic print data from the mean, and the calculation method is shown in formula (5).

[0076] (5)

[0077] In formula (5), It represents the variance of the signal within a single analysis window, reflecting the dynamic range of the signal. For the first Signal amplitude of the acoustic fingerprint data of arteriovenous fistula blood flow at each sampling point; This is the sampling point index, with values ​​ranging from 0, 1, 2, ... ; The total number of sampling points in a single sampling is fixed at 500. This is the average value of the signals from all sampled points within the window; For the first The deviation of the signal at each sampling point from the average value.

[0078] The spectral centroid can be used to describe the energy and frequency distribution of the arteriovenous fistula (AVF) blood flow acoustic signature data over a period of time. The calculation method is shown in formula (6). The spectral centroid represents the average point of the signal energy distribution of the AVF blood flow acoustic signature data. Generally speaking, the higher the frequency of the AVF blood flow acoustic signature data in patients with a high degree of AVF stenosis, the higher the spectral centroid will be.

[0079] (6)

[0080] In formula (6), The centroid of the spectrum of the signal within a single analysis window, in Hz, represents the average frequency of the energy distribution; This is a frequency point index, with values ​​ranging from 0, 1, 2, ... ; The total number of frequency points in the spectrum, determined by the number of sampling points. Decide, and Both are 500, corresponding to a frequency range of 0. 1000 Hz; For the first The linear frequency value at each frequency point is calculated as follows: ( =1000 Hz is the sampling rate). For the first The spectral amplitude at each frequency point can be obtained from the Fourier transform of the signal; It is a weighted sum of frequency and corresponding spectral amplitude, with the weight being the spectral amplitude.

[0081] The spectral width is used to describe the maximum width of the frequency of the arteriovenous fistula blood flow acoustic data over a period of time, and is calculated as shown in formula (7). Generally speaking, the higher the frequency of the arteriovenous fistula blood flow acoustic data in patients with a high degree of arteriovenous fistula stenosis, the greater the spectral width.

[0082] (7)

[0083] In formula (7), The spectral width of the signal within a single analysis window, in Hz, is used to characterize the frequency distribution range. For analyzing the window index; This is a frequency point index, with values ​​ranging from 0, 1, 2, ... ; The total number of frequency points in the spectrum. For the first Linear frequency values ​​at each frequency point; For the first The spectral centroid of a window; For the first spectral amplitude at each frequency point; For the first The square of the frequency deviation between each frequency point and the centroid of the spectrum.

[0084] The short-time autocorrelation function refers to the autocorrelation function of an audio signal, which is the convolution of the audio signal with its inverse signal. The calculation method is shown in formula (8). By calculating the autocorrelation function, the periodic information of the signal can be obtained, and then the fundamental period of the arteriovenous fistula blood flow acoustic signature data can be estimated.

[0085] (8)

[0086] In formula (8), For delay The short-time autocorrelation function value at time; The delay step is the number of steps, and its value ranges from 0, 1, 2, ... ; This is the sampling point index, with a value range of [value range missing]. arrive ; The total number of sampling points in a single sampling is fixed at 500. For the first Signal amplitude of the acoustic fingerprint data of arteriovenous fistula blood flow at each sampling point; For delay Step after step The signal amplitude of the arteriovenous fistula blood flow acoustic print data at each sampling point.

[0087] Power spectral density refers to the discrete Fourier transform of the short-time autocorrelation function of the signal, and the average value of the modulus is taken. The calculation method is shown in formula (9). Power spectral density represents the power of the per unit frequency band of the arteriovenous fistula blood flow acoustic signature data.

[0088] (9)

[0089] In formula (9), The power spectral density of the arteriovenous fistula blood flow acoustic data is expressed in W / Hz, representing the power per unit frequency band. This is the frequency value; It is a Discrete Fourier Transform (DFT) operator used to convert time-domain autocorrelation functions into frequency-domain functions; This is the short-time autocorrelation function of the signal; This is the delay step number; The total number of sampling points in a single sampling is fixed at 500.

[0090] Mel frequency cepstral coefficients are an audio feature extraction method based on the characteristics of human hearing. By simulating the human ear's perception of sounds at different frequencies, audio is analyzed from the perspective of the human ear. Based on cepstral analysis of sound features, the linear spectrum is converted to a nonlinear spectrum, and then further converted to the cepstral spectrum. The calculation method is shown in formula (10).

[0091] (10)

[0092] In formula (10), linear frequency The corresponding Mel frequency, measured in Mel, is used to simulate the nonlinear frequency scale of human hearing. The original linear frequency, with a value range of 0. 500Hz is the effective frequency range for arteriovenous fistula blood flow acoustic signature data; 2595 is the calibration coefficient for Mel frequency conversion, derived from experiments on human ear hearing characteristics; 700 is the mapping reference value between Mel frequency and linear frequency.

[0093] After extracting the aforementioned 10-dimensional voiceprint features, this embodiment inputs these features into a trained Support Vector Machine (SVM) model for thrombosis risk analysis. The SVM model uses a Gaussian kernel function, sets the parameter γ to 0.1, and the penalty coefficient C to 10, and optimizes the model parameters through 5-fold cross-validation. Based on the SVM model, the thrombosis risk probability can be output, which is the thrombosis risk information obtained from the analysis. Specifically, the analysis process of the 10-dimensional voiceprint features by the SVM model essentially transforms high-dimensional features into interpretable thrombosis risk probabilities through the logic of feature space mapping → optimal hyperplane construction → risk probability output. The specific process is as follows: The SVM model first standardizes and normalizes the aforementioned 10-dimensional voiceprint features to transform all voiceprint features into a uniform scale, ensuring that the contribution weights of the aforementioned 10-dimensional voiceprint features to the model are balanced. Next, a Gaussian kernel function is selected to map the original 10-dimensional voiceprint features to a higher-dimensional latent space, making the data linearly separable in the new space. Then, in the mapped high-dimensional space, an optimal hyperplane is found that maximizes the distance to the two nearest classes of samples. The larger the distance, the stronger the model's generalization ability. Next, the model parameters are optimized through 5-fold cross-validation to ensure that the model's predictive ability on the aforementioned 10 dimensions and features is stable and does not depend on specific training samples. Finally, the support vector machine model outputs a class label, which is the label indicating whether the patient has a risk of thrombosis, determined after analyzing and processing the input 10-dimensional voiceprint features. That is, the class label can be a normal label or an abnormal label. To quantify the risk of thrombosis, this embodiment needs to use Platt scaling to transform the class label of the support vector machine model into a probability in the [0, 1] interval, thereby obtaining the probability of thrombosis risk. The essence of Platt scaling is to train a simple sigmoid function (the core function of logistic regression) to map the classifier's decision value to a probability value in the [0, 1] interval. Thus, it can be seen that the above support vector machine can transform the abstract 10-dimensional voiceprint features into a clinically applicable probability of thrombosis risk.

[0094] After obtaining the probability of thrombosis risk, this embodiment can determine the thrombosis warning level by setting a warning threshold and comparing it with the probability of thrombosis risk. For example, the warning thresholds can be set to 0.3 and 0.7. If the probability of thrombosis risk is less than 0.3, the patient is considered normal and the warning mechanism will not be triggered. If the probability of thrombosis risk is within the range of [0.3, 0.5], the thrombosis warning level is determined to be a Level 1 warning; if the probability of thrombosis risk is within the range of [0.5, 0.7], the thrombosis warning level is determined to be a Level 2 warning; and if the probability of thrombosis risk is greater than 0.7, the thrombosis warning level is determined to be a Level 3 warning. Of course, the warning thresholds in this embodiment can be adjusted according to clinical needs, and this embodiment can greatly improve the timeliness and efficiency of thrombosis warnings.

[0095] Furthermore, after determining the thrombosis warning level information, the cloud server can transmit this information to the fistula monitoring bracelet, patient terminal, and medical staff terminal. Alternatively, it can transmit the information to the fistula monitoring bracelet, which then synchronizes it to the patient and medical staff terminals, achieving multi-terminal linkage between "device-cloud-patient". The fistula monitoring bracelet controls one or more of the following components—sound component, multi-color indicator light, and vibration component—to provide warnings based on the thrombosis warning level information, and provides synchronized warnings with the patient and medical staff terminals. For example, for a level 2 warning (mild risk), the fistula monitoring bracelet's yellow indicator light can flash slowly combined with intermittent vibration from the vibration component to provide a warning. This can also be combined with a pop-up notification on the patient terminal's app stating "mildly abnormal fistula voiceprint" and a simultaneous SMS notification to family members. For a level 2 warning (medium risk), the fistula monitoring bracelet's red indicator light can flash rapidly combined with continuous vibration from the vibration component to provide a warning. This can also be combined with a pop-up notification on the patient terminal's app and an SMS notification. If it is a Level 3 warning, which indicates a high risk, the red indicator light on the arteriovenous fistula monitoring bracelet can be triggered to flash rapidly, combined with a strong vibration from the vibration component, and then combined with a sound from the sound-emitting component to provide an alert. Similarly, it can be combined with the patient's terminal APP pop-up window and SMS reminder to ensure that no warning is missed during nighttime sleep.

[0096] Furthermore, the cloud server in this embodiment can also generate a rehabilitation training plan based on thrombosis risk information, patient basic information, patient muscle strength level, duration of arteriovenous fistula use, and exercise data, and transmit the rehabilitation training plan to the patient terminal and medical staff terminal. Specifically, the thrombosis risk information refers to the aforementioned thrombosis risk probability. Different thrombosis risk probabilities correspond to different degrees of arteriovenous fistula stenosis. For example, if the thrombosis risk probability is less than 0.3, the arteriovenous fistula is determined to be normal. If the thrombosis risk probability is in the range of [0.3, 0.5], the arteriovenous fistula stenosis is determined to be mild. If the thrombosis risk probability is in the range of [0.5, 0.7], the arteriovenous fistula stenosis is determined to be moderate. If the thrombosis risk probability is greater than 0.7, the arteriovenous fistula stenosis is determined to be severe. Patient basic information may include age, dialysis history, underlying diseases, etc. Inputting patient basic information allows the generated rehabilitation training plan to adjust the training parameters and intensity based on the patient's specific condition. The patient's muscle strength level can be determined based on their grip strength during training, specifically categorized into three levels: mild sarcopenia, moderate sarcopenia, and severe sarcopenia. Inputting the patient's muscle strength level allows for the setting of corresponding resistance levels in the generated rehabilitation training program. The duration of fistula use can be defined and staged based on post-operative time points. For example, one month post-surgery is the fistula maturation period; 1-3 months is the adaptation period; 3-6 months is the stable training period; 6-12 months is the reinforcement period; and over one year is the long-term maintenance period. This staged approach covers the entire post-operative period, from early surgery to long-term management, aiming to match the physiological characteristics and rehabilitation needs of patients at different post-operative stages and providing a crucial temporal reference for the dynamic adjustment of personalized rehabilitation training programs. Input exercise data can be the patient's data from the past 7 days, such as the average training achievement rate and the longest single training session.

[0097] Once the thrombosis risk information, patient basic information, patient muscle strength level, duration of arteriovenous fistula use, and exercise data are input into the rehabilitation plan generation module on the cloud server, a personalized rehabilitation training plan can be automatically generated. This plan includes core indicators such as corresponding resistance levels, exclusive training movements, appropriate exercise duration, and reasonable rest time. The rules for generating rehabilitation training plans are shown in Table 1 and Table 1 (continued).

[0098] Table 1

[0099]

[0100] Continued from Table 1

[0101]

[0102] The generated rehabilitation training plan is distributed to both the patient's and healthcare provider's terminals. Furthermore, after the patient completes the rehabilitation training plan, this embodiment can also acquire the latest fistula blood flow voiceprint data uploaded by the fistula monitoring wristband and the latest exercise data uploaded by the smart arm exerciser. Then, based on the latest fistula blood flow voiceprint data and the latest exercise data, the rehabilitation training plan is updated and adjusted. By gradually adjusting the rehabilitation training plan, it is ensured that it adapts to changes in the patient's condition.

[0103] Furthermore, the software interaction management platform in this embodiment includes a software platform for the patient terminal and a software platform for the medical staff terminal, which is used to view the patient's fistula status, rehabilitation training plan and exercise records in real time, and the data of the software interaction management platform can be synchronized to the cloud server or home devices.

[0104] Furthermore, combined Figure 1 As shown, the patient terminal software platform in this embodiment includes a data dashboard function, a home management guidance function, a gamified rehabilitation interaction function, and a personalized graded early warning push function. Specifically, the data dashboard function is used to view the fistula status, displaying waveforms of fistula blood flow voiceprint data, real-time grip strength values ​​and wrist joint angles collected by the arm smart exerciser. It also includes plotting line graphs of training duration over the past 7 / 30 days, bar graphs of target achievement rates, voiceprint energy entropy change curves, and generating weekly health reports, marking the fistula risk level and providing personalized recommendations. The home management guidance function includes a family monitoring assistance tool displayed by default on the software platform homepage. This tool includes a 3D animation of the palpation area within 5cm of the fistula anastomosis, a stethoscope usage tutorial, and an abnormal record button for family members to upload palpation or auscultation feedback and synchronize it to the terminal doctor's end. When the arteriovenous fistula has been in use for more than one year, the homepage of the patient terminal's software platform adds a self-monitoring check-in function, allowing patients to record daily palpation results (such as normal tremor or reduced tremor). The system will combine voiceprint data to generate a two-dimensional verification report, thereby reducing the risk of misjudgment for patients with cognitive impairment. The gamified rehabilitation interaction function maps the grip strength F collected by the intelligent arm exerciser to the game intensity P, specifically using the mapping method P=k1. F, where k1 is the calibration coefficient, and the wrist joint angle θ is mapped to the game height H, specifically H=k2. θ, where k2 is the dynamic gain. By developing games such as "Whack-a-Mole" and "Bird Flight" combined with mapped game intensity and height, players can complete levels and earn points through motion-sensing games. These points can be used to redeem health courses, solving the problems of "boring and low compliance" in traditional rehabilitation. In this embodiment, the personalized graded early warning push function of the patient terminal can adopt different push methods for patients with different conditions. For example, for patients with cognitive impairment: a level 1 warning will trigger a pop-up window in the APP to remind them of "mild abnormality of voiceprint in arteriovenous fistula" and simultaneously trigger a text message reminder on the family's end; a level 2 warning will trigger the above pop-up window + text message, and then pop up the contact information of the attending physician; while a level 3 warning will trigger the above pop-up window + text message, automatically dial the emergency contact number, and provide medical guidance. For normal patients: a level 1 warning will only trigger a pop-up window on the patient's end; a level 2 warning will trigger a text message to the family; and a level 3 warning will trigger a voice call only, reducing unnecessary interference to long-term stable patients.

[0105] Combination Figure 1 As shown, the software platform of the medical terminal in this embodiment includes data management and analysis functions and remote intervention functions. Specifically, the data management and analysis function supports screening patients by grouping them according to the level of thrombosis risk and viewing the patients' arteriovenous fistula status; it also includes viewing the patients' exercise records, such as changes in a single patient's voiceprint over the past 3 months and comparisons with exercise data uploaded from the smart arm exerciser, thereby generating a trend chart of arteriovenous fistula failure risk; and generating a chart to manage the occurrence of recent complications in patients. The remote intervention function of this embodiment can receive thrombosis warning information from patients and can send doctor replies or initiate video consultations with one click. Furthermore, the medical terminal also supports viewing rehabilitation training programs generated by the cloud server, allowing doctors to remotely adjust rehabilitation training programs and record patient feedback such as grip fatigue as reasons for modification, which are automatically synchronized to the patient's terminal and the smart arm exerciser after confirmation. In addition, it also supports remote adjustment of the wristband warning threshold to avoid false alarms.

[0106] This embodiment presents a home-based intelligent monitoring and rehabilitation training system for arteriovenous fistulas, integrating monitoring, early warning, rehabilitation training, and feedback. By establishing a standardized transmission protocol, it achieves real-time synchronization of fistula blood flow voiceprint data and motion data, enabling multi-terminal linkage of early warning signals. Furthermore, this embodiment innovatively incorporates gamified rehabilitation training design, mapping grip strength to game force parameters and wrist joint angle to game height parameters, addressing the monotony of traditional rehabilitation training. In addition, this embodiment constructs a multi-level early warning mechanism, achieving seamless integration from risk monitoring to patient-medical response.

[0107] Based on the above embodiments, the present invention also provides a home-based intelligent monitoring and rehabilitation training method for arteriovenous fistulas, which is implemented based on the home-based intelligent monitoring and rehabilitation training system for arteriovenous fistulas described in the above system embodiments. In specific applications, such as... Figure 4As shown, the home-based intelligent monitoring and rehabilitation training method for arteriovenous fistulas includes the following steps:

[0108] Step S100: Real-time acquisition of the patient's arteriovenous fistula blood flow acoustic print data based on the arteriovenous fistula monitoring wristband, standardization processing of the arteriovenous fistula blood flow acoustic print data, and thrombosis risk analysis based on the processed arteriovenous fistula blood flow acoustic print data to obtain thrombosis risk information.

[0109] Step S200: Collect the patient's motion parameters based on the intelligent arm exerciser, and process the motion parameters to obtain the patient's motion data;

[0110] Step S300: Based on the thrombosis risk information, patient basic information, patient muscle strength level, duration of arteriovenous fistula use, and exercise data, generate a rehabilitation training plan and transmit the rehabilitation training plan to the patient terminal and medical care terminal.

[0111] Combination Figure 5 As shown in the illustration, this embodiment uses a dual-channel voiceprint acquisition unit in the fistula monitoring wristband to collect fistula blood flow voiceprint data. The dual-channel voiceprint acquisition unit includes a main microphone and an auxiliary microphone. The main microphone is located inside the wristband's strap and is used to collect fistula blood flow voiceprint data. The auxiliary microphone is located outside the wristband and is used to collect ambient noise. This ambient noise is used for differential noise reduction to ensure that the signal-to-noise ratio of the fistula blood flow voiceprint data reaches 65dB, far exceeding the 30dB interference level of traditional single microphones. This solves the problem of noise interference in the home environment and fills a gap in domestic and international home fistula monitoring equipment.

[0112] This embodiment of the intelligent arm exerciser is an intelligent exercise device for training grip and arm strength. It serves as a carrier for collecting motion parameters, integrating both rehabilitation training and motion parameter acquisition functions. The six-axis motion sensor and flexible stress sensor of the intelligent arm exerciser can collect the wrist joint angle and grip force changes during patient movement. A bridge circuit converts these grip force changes into voltage signals, thus providing motion parameters. Furthermore, this embodiment can record the patient's single training duration, number of movements, and movement attainment rate through a data recording module, forming motion data. Next, the collected arteriovenous fistula blood flow acoustic signature data and motion data are associated and uploaded to a cloud server via a multi-mode gateway. This embodiment uses Bluetooth 5.0 or Wi-Fi dual-mode communication for multi-mode gateway transmission. In addition, in practical applications, this embodiment can associate the arteriovenous fistula duration tag of the collected arteriovenous fistula blood flow acoustic signature data with the motion data, facilitating subsequent analysis of the impact of exercise on arteriovenous fistulas of different durations.

[0113] Furthermore, cloud servers can employ support vector machine models for thrombosis risk analysis. In specific applications, this can be combined with... Figure 5As shown in the illustration, this embodiment extracts 10-dimensional voiceprint features from the arteriovenous fistula (AVF) blood flow voiceprint data after filtering out environmental noise. The extracted 10-dimensional voiceprint features are then input into a support vector machine (SVM) model for thrombosis risk analysis, outputting a thrombosis risk probability, which is the thrombosis risk information. The voiceprint features extracted in this embodiment include: short-time energy, short-time average amplitude, short-time zero-crossing rate, root mean square (RMS) of the audio data, spectral centroid, spectral width, short-time autocorrelation function, power spectrum, and Mel-frequency cepstral coefficients. The SVM model in this embodiment uses a Gaussian kernel function, sets the parameter γ to 0.1, and the penalty coefficient C to 10, and optimizes the model parameters through 5-fold cross-validation. Based on the SVM model, a thrombosis risk probability can be output, which is the analyzed thrombosis risk information.

[0114] After obtaining the probability of thrombosis risk, this embodiment can determine the thrombosis warning level by setting a warning threshold and comparing it with the probability of thrombosis risk. For example, the warning thresholds can be set to 0.3 and 0.7. If the probability of thrombosis risk is less than 0.3, the patient is considered normal and the warning mechanism will not be triggered. If the probability of thrombosis risk is within the range of [0.3, 0.5], the thrombosis warning level is determined to be a Level 1 warning; if the probability of thrombosis risk is within the range of [0.5, 0.7], the thrombosis warning level is determined to be a Level 2 warning; and if the probability of thrombosis risk is greater than 0.7, the thrombosis warning level is determined to be a Level 3 warning. Of course, the warning thresholds in this embodiment can be adjusted according to clinical needs, and this embodiment can greatly improve the timeliness and efficiency of thrombosis warnings.

[0115] Furthermore, after determining the thrombosis warning level information, the cloud server can transmit this information to the fistula monitoring bracelet, patient terminal, and medical staff terminal. Alternatively, it can transmit the information to the fistula monitoring bracelet, which then synchronizes it to the patient and medical staff terminals for thrombosis risk warning, achieving multi-terminal linkage between "device-cloud-patient". The fistula monitoring bracelet controls one or more of the following components—sound component, multi-color indicator light, and vibration component—to provide warnings based on the thrombosis warning level information, and provides synchronized warnings with the patient and medical staff terminals. For example, for a level 2 warning (mild risk), the fistula monitoring bracelet's yellow indicator light can flash slowly combined with intermittent vibration from the vibration component to provide a warning. This can also be combined with a pop-up notification on the patient terminal's app stating "mildly abnormal fistula voiceprint" and a simultaneous SMS notification to family members. For a level 2 warning (medium risk), the fistula monitoring bracelet's red indicator light can flash rapidly combined with continuous vibration from the vibration component to provide a warning. This can also be combined with a pop-up notification on the patient terminal's app and an SMS notification. If it is a Level 3 warning, which indicates a high risk, the red indicator light on the arteriovenous fistula monitoring bracelet can be triggered to flash rapidly, combined with a strong vibration from the vibration component, and then combined with a sound from the sound-emitting component to provide an alert. Similarly, it can be combined with the patient's terminal APP pop-up window and SMS reminder to ensure that no warning is missed during nighttime sleep.

[0116] Furthermore, combined Figure 5 As shown, this embodiment can input multi-dimensional data such as thrombosis risk information, patient basic information, patient muscle strength level, duration of arteriovenous fistula use, and exercise data to automatically generate a rehabilitation training plan, which is then transmitted to both the patient's and medical staff's terminals. The rehabilitation training plan generated in this embodiment includes core indicators such as corresponding resistance levels, specific training movements, appropriate exercise duration, and reasonable rest periods. By generating the above-mentioned rehabilitation training method, this embodiment helps achieve the synergistic goal of reducing thrombosis risk and improving muscle strength. Furthermore, the generated rehabilitation training plan is distributed to both the patient's and medical staff's terminals. Patients can perform rehabilitation training, engage in gamified interaction, and receive alert responses based on the rehabilitation training plan received on their patient's terminal. Medical staff can view patient data, adjust rehabilitation training plans and alert thresholds, and provide suggestions on their medical staff's terminals.

[0117] Furthermore, after the patient completes the rehabilitation training program, this embodiment can also acquire the latest fistula blood flow voiceprint data uploaded by the fistula monitoring wristband and the latest exercise data uploaded by the smart arm exerciser. Then, the latest fistula blood flow voiceprint data and the latest exercise data are transmitted back to the cloud server, which performs iterative analysis to update and adjust the rehabilitation training program. By gradually adjusting the rehabilitation training program, it is ensured that it adapts to changes in the patient's condition.

[0118] The home-based intelligent monitoring and rehabilitation training method for arteriovenous fistulas in this embodiment is based on the same principle as the various terminals and modules in the above system embodiment, and will not be elaborated further here.

[0119] Based on the above embodiments, the present invention also provides a terminal, the principle block diagram of which can be as follows: Figure 6 As shown. The terminal may include one or more processors 100 ( Figure 6 (Only one is shown in the image), memory 101, and computer program 102 stored in memory 101 and executable on one or more processors 100. For example, a home-based intelligent monitoring and rehabilitation training program for arteriovenous fistulas. When one or more processors 100 execute computer program 102, they can implement the various steps in the home-based intelligent monitoring and rehabilitation training method embodiment. Alternatively, when one or more processors 100 execute computer program 102, they can implement the functions of various modules / units in the home-based intelligent monitoring and rehabilitation training system embodiment, which is not limited here.

[0120] In one embodiment, the processor 100 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0121] In one embodiment, memory 101 may be an internal storage unit of an electronic device, such as a hard drive or RAM. Memory 101 may also be an external storage device of the electronic device, such as a plug-in hard drive, Smart Media Card (SMC), Secure Digital Card (SD), or Flash Card. Furthermore, memory 101 may include both internal and external storage units. Memory 101 is used to store computer programs and other programs and data required by the terminal. Memory 101 can also be used to temporarily store data that has been output or will be output.

[0122] Those skilled in the art will understand that Figure 6The block diagram shown is merely a partial structural diagram related to the present invention and does not constitute a limitation on the terminal to which the present invention is applied. A specific terminal may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0123] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), direct memory bus RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0124] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A home arteriovenous fistula intelligent monitoring and rehabilitation training system, characterized in that, The system comprises a home device, a cloud server and a software interaction management platform, and data interconnection is realized among the home device, the cloud server and the software interaction management platform; The home device comprises an internal fistula monitoring bracelet and an arm intelligent exerciser, and the internal fistula monitoring bracelet and the arm intelligent exerciser realize data interconnection; the internal fistula monitoring bracelet is used for collecting internal fistula blood flow voiceprint data of a patient and giving a prewarning prompt to the patient according to thrombus risk information; and the arm intelligent exerciser is used for collecting motion parameters of the patient and providing rehabilitation training function and action guiding function for the patient; The cloud server is used for standardizing the collected internal fistula blood flow voiceprint data and motion parameters, analyzing thrombus risk according to the internal fistula blood flow voiceprint data, and generating a rehabilitation training scheme; The software interaction management platform comprises a software platform of a patient terminal and a software platform of a medical care terminal, and is used for real-time viewing of internal fistula conditions, rehabilitation training schemes and exercise records of the patient; The arm intelligent exerciser comprises: a six-axis motion sensor, which is used for capturing wrist motion of the patient and collecting wrist joint angle when the patient is moving; a flexible stress sensor, which is used for recording grip strength change of fingers of the patient and converting the grip strength change into a voltage signal through a bridge circuit; a resistance gear adjusting module, which is used for providing multiple resistance gears to adapt to patients with different muscle strength levels; an action guiding module, which is used for prompting and guiding when the patient's motion is not standard; a data recording module, which is used for recording single training duration, action frequency and action compliance rate of the patient.

2. The intelligent monitoring and rehabilitation training system for home arteriovenous fistula according to claim 1, characterized in that, The internal fistula monitoring bracelet comprises: a dual-path voiceprint collecting unit, which comprises a main microphone and an auxiliary microphone, wherein the main microphone is arranged on the inner side of a watchband of the internal fistula monitoring bracelet and is used for collecting the internal fistula blood flow voiceprint data, and the auxiliary microphone is arranged on the outer side of the watchband of the internal fistula monitoring bracelet and is used for collecting environmental noise, which is used for realizing differential noise reduction; a hierarchical prewarning module, which comprises a sound emitting assembly, a multi-color indicating lamp and a vibration assembly, and is used for controlling one or more of the sound emitting assembly, the multi-color indicating lamp and the vibration assembly to give a prewarning prompt according to a thrombus prewarning level corresponding to thrombus risk information; a low-power consumption control module, which is used for managing power consumption of each module in the internal fistula monitoring bracelet; a data transmission module, which is used for transmitting the internal fistula blood flow voiceprint data to the cloud server.

3. A method for intelligent monitoring and rehabilitation training of a home arteriovenous fistula, characterized in that, The home arteriovenous internal fistula intelligent monitoring and rehabilitation training method is applied to the home arteriovenous internal fistula intelligent monitoring and rehabilitation training system of any one of claims 1-2, and the method comprises: collecting internal fistula blood flow voiceprint data of a patient in real time based on an internal fistula monitoring bracelet, standardizing the internal fistula blood flow voiceprint data, and analyzing thrombus risk based on the processed internal fistula blood flow voiceprint data to obtain thrombus risk information; collecting motion parameters of the patient based on an arm intelligent exerciser and processing the motion parameters to obtain motion data of the patient; and Generate a rehabilitation training scheme based on the thrombosis risk information, patient basic information, patient muscle strength level, internal fistula use time length, and motion data, and transmit the rehabilitation training scheme to the patient terminal and medical terminal.

4. The intelligent monitoring and rehabilitation training method for home arteriovenous fistula according to claim 3, characterized in that, The internal fistula blood flow acoustic fingerprint data is standardized, including: The environmental noise in the internal fistula blood flow acoustic fingerprint data is filtered out through adaptive differential noise reduction and Butterworth band-pass filtering; The internal fistula blood flow acoustic fingerprint data after filtering out the environmental noise is added with an internal fistula time length label.

5. The intelligent monitoring and rehabilitation training method for home arteriovenous fistula according to claim 4, characterized in that, Based on the processed internal fistula blood flow acoustic fingerprint data, thrombosis risk analysis is performed to obtain thrombosis risk information, including: The acoustic fingerprint features are extracted from the internal fistula blood flow acoustic fingerprint data after filtering out the environmental noise; The extracted acoustic fingerprint features are input into a support vector machine model for thrombosis risk analysis, and a thrombosis risk probability is output, which is taken as the thrombosis risk information.

6. The intelligent monitoring and rehabilitation training method for home arteriovenous fistula according to claim 5, characterized in that, The method further includes: Based on the thrombosis risk information, thrombosis early warning level information is determined, and the thrombosis early warning level information is transmitted to the internal fistula monitoring bracelet, patient terminal, and medical terminal; The internal fistula monitoring bracelet controls one or more of the sound emitting assembly, multi-color indicator light, and vibration assembly to give a warning prompt according to the thrombosis early warning level information, and synchronously gives a warning in combination with the patient terminal and medical terminal.

7. The intelligent monitoring and rehabilitation training method for home arteriovenous fistula according to claim 6, characterized in that, The method further includes: After the patient performs the rehabilitation training scheme, the latest internal fistula blood flow acoustic fingerprint data uploaded by the internal fistula monitoring bracelet and the latest motion data uploaded by the arm intelligent exerciser are obtained; Based on the latest internal fistula blood flow acoustic fingerprint data and the latest motion data, the rehabilitation training scheme is updated.

8. A terminal, characterized by comprising: The terminal includes a memory, a processor, and a home arteriovenous fistula intelligent monitoring and rehabilitation training program stored in the memory and executable on the processor, and when the processor executes the home arteriovenous fistula intelligent monitoring and rehabilitation training program, the steps of the home arteriovenous fistula intelligent monitoring and rehabilitation training method according to any one of claims 3-7 are implemented.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a home arteriovenous fistula intelligent monitoring and rehabilitation training program, and the home arteriovenous fistula intelligent monitoring and rehabilitation training program implements the steps of the home arteriovenous fistula intelligent monitoring and rehabilitation training method according to any one of claims 3-7 on the computer readable storage medium.

Images