A method and system for processing urinary system pressure measurement signals based on biometric recognition

Abstract

This invention discloses a method and system for processing urinary system pressure measurement signals based on biometric recognition. The method includes: injecting a standard pressure source to simulate a known load; acquiring pressure signals from a calibrated pressure sensor and analyzing their spectral characteristics to construct a baseline noise model; extracting biometric features to establish an interference feature library; acquiring the patient's real-time pressure signal and filtering out environmental noise components based on the baseline noise model; matching the noise-free real-time pressure signal with the interference feature library, and applying targeted filtering enhancement to the matched interference features to obtain the filtered pressure signal. This invention solves the problems of noise interference and signal distortion caused by human factors in urinary system pressure measurement, improving the reliability of measurement results and the accuracy of diagnosis.

Description

Technical Field

[0001] This invention relates to the field of medical electronic signal processing technology, specifically to a method and system for processing urinary system pressure measurement signals based on biometric recognition. Background Technology

[0002] Urinary system pressure measurement is one of the core technologies in urodynamic testing, widely used in the diagnosis and assessment of functional urinary system disorders such as urinary incontinence, overactive bladder, and neurogenic bladder. Accurate acquisition of pressure signals such as intravesical pressure and urethral pressure is crucial for assessing bladder function, urethral sphincter status, and abnormal neural regulation. However, traditional urodynamic testing systems face numerous challenges in practical applications, primarily including system noise interference and signal distortion caused by human factors. These issues directly affect the reliability of measurement results and the accuracy of diagnosis.

[0003] Traditional urodynamic testing systems suffer from the following core problems: sensor drift, noise, and environmental electromagnetic interference cause signal baseline drift and noise superposition, masking true physiological pressure changes; instantaneous pressure fluctuations caused by patient behaviors such as coughing and changes in body position are highly similar to true bladder pressure changes, and traditional methods rely on fixed thresholds or simple filtering, lacking adaptability to individual differences and dynamic physiological states, easily leading to misjudgment or underjudgment of interference signals; furthermore, traditional urodynamic testing systems lack intelligent mechanisms and cannot dynamically adjust filtering parameters according to the patient's physiological state, resulting in inconsistent processing effects. Therefore, there is an urgent need for an innovative solution that can dynamically adapt to individual differences, intelligently filter interference, and ensure signal quality, in order to improve the accuracy and reliability of detection and provide high-quality data support for the precise diagnosis of urological diseases. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a signal processing method and system for urinary system pressure measurement based on biometric recognition, which addresses the shortcomings of the prior art. This method can solve the problems of noise interference and signal distortion caused by human factors in urinary system pressure measurement, thereby improving the reliability of measurement results and the accuracy of diagnosis.

[0005] To achieve the above objectives, according to one aspect of the present invention, a method for processing urinary system pressure measurement signals based on biometric recognition is provided, comprising:

[0006] A standard pressure source is injected to simulate a known load. The pressure signal from the calibrated pressure sensor is collected and its spectral characteristics are analyzed to construct a baseline noise model. The calibration process is periodically triggered, and the parameters of the baseline noise model are adjusted in conjunction with real-time environmental monitoring data to compensate for the sensitivity drift of the sensor. The pressure sensor is placed in the patient's bladder.

[0007] While the patient performs a standardized sequence of actions, pressure signals, inertial motion data, and real-time environmental monitoring data are simultaneously collected. The components of the pressure signals are decoupled, and interference events and their original feature parameters are extracted. These are then combined with the association rules of interference events and physiological states, as well as historical processing strategies, to form a hierarchical feature encryption storage system for interference features. Novel interference features are expanded through incremental learning algorithms to optimize the interference feature system. The interference events include physiological interference, behavioral interference, and environmental interference. Physiological interference includes the patient performing standardized actions, behavioral interference includes the patient's inertial motion, and environmental interference includes the real-time environmental monitoring data.

[0008] The patient's real-time pressure signal is acquired, and the environmental noise component of the real-time pressure signal is removed by filtering based on the baseline noise model. The real-time pressure signal after removing the environmental noise component is matched with the interference feature library, and targeted filtering enhancement is activated on the matched interference features to obtain the filtered pressure signal.

[0009] The above plan also includes:

[0010] Using signal fidelity and noise suppression ratio as reward functions, the environmental noise component of the real-time pressure signal is removed by filtering based on the baseline noise model, and the weights of the filtering parameters are dynamically adjusted by targeted filtering enhancement; wherein the signal fidelity includes waveform smoothness.

[0011] The above plan also includes:

[0012] The signal-to-noise ratio, waveform continuity index, and spectral purity of the filtered pressure signal are calculated in real time to quantify the signal quality level; when a pressure signal with a degraded signal quality level is detected, a graded response is triggered.

[0013] Users can use a visual interactive interface to label low-quality signal segments of the patient's real-time pressure signal, extract noise features of the low-quality signal segments, and incrementally update the interference feature library.

[0014] The signal quality level results are used to optimize the baseline noise model and the filtering;

[0015] Optimization reports are generated by tracking quality indicators over a long period.

[0016] In the above scheme, the calibration of the pressure sensor includes: placing the pressure sensor at a pressure-free reference point through a hardware zeroing program, and eliminating the initial offset using a multi-channel calibration circuit;

[0017] The periodic trigger calibration process, combined with real-time environmental monitoring data, adjusts the parameters of the baseline noise model, including:

[0018] The system periodically acquires the real-time environmental monitoring data. When the change in the real-time environmental monitoring data exceeds a threshold, the system automatically reloads the standard pressure source and performs the following operations:

[0019] Update the frequency domain attenuation coefficient of the baseline noise model and dynamically adjust the suppression weights of noise in different frequency bands;

[0020] Reconstruct the noise power spectral density (PSD) template and calibrate the noise energy distribution of each frequency band based on the spectral characteristics of the calibration signal;

[0021] Correct the temperature-sensitivity drift coefficient and fit the rate of change of sensor gain with temperature using temperature sensor data;

[0022] Optimize the vibration coupling gain matrix and establish the transfer function between environmental vibration frequency and pressure noise by combining inertial motion data;

[0023] Recalculate the baseline drift rate and quantify the sensor zero-point offset characteristics over time and temperature;

[0024] Adjust the non-Gaussian noise weights and update the robustness parameters of the baseline noise model based on the statistical characteristics of sudden interference events.

[0025] In the above scheme, the pressure-free reference points include the pubic symphysis region, the abdominal wall region (3-5cm below the umbilicus), the presacral region, the in vitro simulated tissue (silicone-hydrogel composite), and the zero point at the end of deep breathing;

[0026] The standard pressure sources include physiological saline, compressed carbon dioxide, the patient's natural bladder filling pressure, and an adjustable height water column.

[0027] The real-time environmental monitoring data includes temperature, electromagnetic field strength, and humidity;

[0028] The baseline noise model establishes a mapping relationship between noise components and signal amplitude by decomposing the frequency domain characteristics of environmental noise; wherein the frequency domain characteristics include power frequency interference and thermal noise.

[0029] In the above scheme, the standardized actions are physiological behaviors, including coughing and changing body position; the inertial movements include walking and climbing stairs.

[0030] The components of the pressure signal are decoupled using a time-frequency joint analysis method;

[0031] The original characteristic parameters of the interference event include the waveform characteristics of the physiological interference, the motion trajectory pattern of the behavioral interference, and the spectral fingerprint of the environmental interference.

[0032] The step of expanding novel interference features and optimizing the interference feature library through incremental learning algorithms includes: expanding novel interference features through incremental learning algorithms, adjusting matching thresholds in conjunction with real-time environmental monitoring data, and continuously optimizing the interference feature library by integrating clinical feedback data.

[0033] In the above scheme, the step of stripping the environmental noise component of the real-time pressure signal based on the baseline noise model includes:

[0034] In the frequency domain, an adaptive notch filter is designed to eliminate power frequency interference and equipment resonance peaks. In the time domain, a combination of moving average filtering and median filtering is used to suppress high-frequency random noise and sudden pulse interference.

[0035] The interference features in the real-time pressure signal after removing noise components are matched with the interference feature library, and targeted filtering enhancement is initiated for the matched interference features, including:

[0036] For specific frequency bands of the matched interference characteristics, the notch depth is dynamically increased and the filter window is narrowed to achieve precise elimination of local noise.

[0037] In the above scheme, the degraded pressure signal refers to the signal-to-noise ratio reduction rate, spectral purity shift, and waveform continuity index exceeding the corresponding thresholds;

[0038] The tiered response specifically includes distinguishing between primary and severe warnings based on the specific medical situation. Primary warnings will prompt medical staff to check the contact status of the equipment, while severe warnings will suspend monitoring and issue an audible and visual alarm.

[0039] In the above scheme, the low-quality signal segment includes the sensor detachment segment, the conduit blockage segment, the circuit saturation segment, and the wireless transmission packet loss segment.

[0040] According to another aspect of the present invention, a urinary system pressure measurement signal processing system based on biometric recognition is provided, comprising:

[0041] The baseline noise modeling module is used to inject a standard pressure source to simulate a known load, acquire pressure signals from calibrated pressure sensors and analyze their spectral characteristics to construct a baseline noise model; it periodically triggers the calibration process and adjusts the parameters of the baseline noise model in conjunction with real-time environmental monitoring data to compensate for the sensitivity drift of the sensor; wherein the pressure sensor is placed in the patient's bladder;

[0042] A biometric extraction module is used to simultaneously collect pressure signals, inertial motion data, and real-time environmental monitoring data while the patient performs a standardized action sequence. It decouples the components of the pressure signals, extracts interfering events and their original feature parameters, and combines these with association rules of interfering events and physiological states, along with historical processing strategies, to perform hierarchical feature encryption and storage, forming an interfering feature library. The module further expands the library with novel interfering features through an incremental learning algorithm, optimizing the interfering feature library. The interfering events include physiological interference, behavioral interference, and environmental interference. Physiological interference includes the patient performing standardized actions, behavioral interference includes the patient's inertial motion, and environmental interference includes the real-time environmental monitoring data.

[0043] An adaptive filtering algorithm module is used to acquire the patient's real-time pressure signal, and based on the baseline noise model, filter out the environmental noise component of the real-time pressure signal; match the real-time pressure signal after removing the environmental noise component with the interference feature library, and activate targeted filtering enhancement for the matched interference features to obtain the filtered pressure signal.

[0044] In summary, compared with the prior art, the above-described technical solutions conceived by this invention can achieve the following beneficial effects:

[0045] (1) This invention provides a signal processing method for urinary system pressure measurement based on biometric recognition. This method can solve the problems of noise interference and signal distortion caused by human factors in urinary system pressure measurement, and improve the reliability of measurement results and the accuracy of diagnosis. In application scenarios such as bladder pressure monitoring and urodynamic testing, this method helps to eliminate the influence of interference factors such as coughing and abdominal compression on doctors' diagnosis, improve the signal-to-noise ratio of the signals collected by the urodynamic experimental system, and provide high-quality data for the future machine intelligent diagnosis of urinary system diseases.

[0046] (2) In the method provided by this invention, by constructing an interference feature library, multi-dimensional feature decoupling of physiological activity artifacts, environmental noise and real pathological signals is achieved; a hierarchical adaptive filtering strategy is adopted to achieve significant interference suppression capability in the full dynamic range of bladder pressure signals, effectively eliminating baseline drift and waveform distortion caused by compound interference, and improving signal purity. Based on the establishment of the interference feature library, patient-specific interference patterns can be accurately identified, significantly improving the accuracy of personalized interference suppression; combined with the reinforcement learning mechanism, the filtering parameters can be dynamically adjusted in real time, adapting to changes in bladder filling status and signal feature migration under different physiological scenarios. Through the time-frequency joint analysis algorithm, the core morphological characteristics of key pathological features such as detrusor muscle contraction waveform are completely preserved in a strong interference environment; the multimodal signal mutual verification mechanism effectively ensures the measurement accuracy of diagnostic parameters such as urinary flow rate curve, avoiding the loss of key clinical information. Quality assessment achieves intelligent balance optimization between signal fidelity and noise suppression, and continuously expands the interference pattern recognition capability through a clinical feedback-driven incremental learning mechanism, significantly reducing the system misjudgment rate and improving processing efficiency. In urodynamic testing, this method effectively suppresses interference from postural changes and spontaneous pressure increases, significantly reducing the risk of clinical misdiagnosis. It provides high-confidence pressure-flow rate curve data support for the diagnosis of diseases such as neurogenic bladder, improving the consistency of diagnostic results. The environmental adaptive compensation algorithm maintains baseline stability under harsh conditions such as temperature fluctuations and electromagnetic interference, and the motion compensation model can effectively correct pressure artifacts caused by multi-degree-of-freedom body movements, ensuring signal reliability in complex environments. Attached Figure Description

[0047] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0048] Figure 1 This is a flowchart illustrating a method for processing urinary system pressure measurement signals based on biometric recognition, as described in Embodiment 1 of the present invention. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of this invention clearer, 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 and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0050] It should be understood that the sequence number of each step in the embodiment does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0051] Example 1

[0052] One embodiment of this application provides a method for processing urinary system pressure measurement signals based on biometric recognition. Please refer to [link to relevant documentation]. Figure 1 ,include:

[0053] S1, inject a standard pressure source to simulate a known load, collect the pressure signal of the calibrated pressure sensor and analyze its spectral characteristics to construct a baseline noise model; periodically trigger the calibration process and adjust the parameters of the baseline noise model in conjunction with real-time environmental monitoring data to compensate for the sensor's sensitivity drift; wherein the pressure sensor is placed in the patient's bladder.

[0054] Specifically, in this embodiment, it is understood that a highly reliable pressure measurement reference needs to be established to eliminate inherent sensor errors and environmental interference. In this embodiment, firstly, the sensor is placed at a pressure-free reference point (specifically the pubic symphysis region in this embodiment) through a hardware zeroing procedure, and initial offset is eliminated using a multi-channel calibration circuit to ensure the stability of the zero-point signal reference.

[0055] Subsequently, a standard pressure source (specifically physiological saline in this embodiment) is injected to simulate a known load. The sensor pressure signal is collected and its spectral characteristics are analyzed to construct a baseline noise model. This model establishes a mapping relationship between noise components and signal amplitude by decomposing the frequency domain characteristics of environmental noise (such as power frequency interference and thermal noise). To achieve long-term stability, a dynamic update mechanism is introduced: the calibration process is periodically triggered (specifically every 30 minutes in this embodiment), and the baseline noise model parameters are adjusted in conjunction with real-time environmental monitoring data (such as temperature, electromagnetic field strength, and humidity) to compensate for sensor sensitivity drift.

[0056] Specifically, in this embodiment, real-time environmental monitoring data is acquired periodically. When the change in the real-time environmental monitoring data exceeds a threshold, the standard pressure source is automatically reloaded and the following operations are performed:

[0057] Update the frequency domain attenuation coefficient of the baseline noise model and dynamically adjust the suppression weights of noise in different frequency bands;

[0058] Reconstruct the noise power spectral density (PSD) template and calibrate the noise energy distribution of each frequency band based on the spectral characteristics of the calibration signal;

[0059] Correct the temperature-sensitivity drift coefficient and fit the rate of change of sensor gain with temperature using temperature sensor data;

[0060] Optimize the vibration coupling gain matrix and establish the transfer function between environmental vibration frequency and pressure noise by combining inertial motion data;

[0061] Recalculate the baseline drift rate and quantify the sensor zero-point offset characteristics over time and temperature;

[0062] Adjust the non-Gaussian noise weights and update the robustness parameters of the baseline noise model based on the statistical characteristics of sudden disturbance events.

[0063] For example, when a significant change in ambient temperature exceeding 2°C is detected, the standard pressure source is automatically reloaded and the parameters of the baseline noise model are adjusted to ensure that the baseline noise model is adaptively corrected according to environmental conditions, providing a clean benchmark reference for subsequent signal processing.

[0064] S2, while the patient performs a standardized action sequence, simultaneously collects pressure signals, inertial motion data, and real-time environmental monitoring data; decouples the components of the pressure signal, extracts interfering events and their original feature parameters from the pressure signal, and combines the association rules of interfering events and physiological states with historical processing strategies to perform hierarchical feature encryption storage to form an interfering feature library; expands new interfering features through incremental learning algorithms to optimize the interfering feature library; the interfering events include physiological interference, behavioral interference, and environmental interference, where physiological interference includes the patient performing standardized actions, behavioral interference includes the patient's inertial motion, and environmental interference includes real-time environmental monitoring data.

[0065] Specifically, in this embodiment, the patient is first guided to perform a standardized sequence of actions (including but not limited to coughing, changing positions, and other physiological behaviors), while simultaneously collecting pressure signals, inertial motion data (including but not limited to walking, climbing stairs, and other movements), and real-time environmental monitoring data. Next, a time-frequency joint analysis method is used to decouple signal components, extracting waveform features of physiological interference, motion trajectory patterns of behavioral interference, and spectral fingerprints of environmental interference. Then, a hierarchical feature storage structure is established, encrypting and storing original feature parameters, association rules between interference events and physiological states, and historical processing strategies. Finally, a dynamic optimization mechanism is introduced, expanding novel interference features through incremental learning algorithms, adjusting matching thresholds based on real-time environmental monitoring data, and continuously optimizing the interference feature library by integrating clinical feedback data. In this embodiment, this step achieves end-to-end processing from multi-source data acquisition to intelligent feature modeling, ensuring the comprehensiveness and dynamic adaptability of interference identification.

[0066] S3: Acquire the patient's real-time pressure signal, and filter out the environmental noise component of the real-time pressure signal based on the baseline noise model; match the interference features in the real-time pressure signal after removing the environmental noise component with the interference feature library, and start targeted filtering enhancement for the matched interference features to obtain the filtered pressure signal.

[0067] Specifically, in this embodiment, firstly, environmental noise components are stripped based on a baseline noise model: an adaptive notch filter is designed in the frequency domain to eliminate power frequency interference and equipment resonance peaks; in the time domain, moving average filtering and median filtering are combined to suppress high-frequency random noise and sudden impulse interference, respectively. Subsequently, a biometric fingerprint matching mechanism is introduced: when the real-time signal matches interference features in the fingerprint database, targeted filtering enhancement is initiated. For example, for specific frequency bands of cough interference, the notch depth is dynamically increased and the filtering window is narrowed to achieve precise elimination of local noise. To adapt to complex physiological state changes, a reinforcement learning framework is integrated: using signal fidelity (e.g., waveform smoothness) and noise suppression ratio as reward functions, based on the baseline noise model, the environmental noise components of the real-time pressure signal are stripped through filtering, and the filtering parameter weights are dynamically adjusted through targeted filtering enhancement. For example, during urination, the low-frequency filtering intensity is automatically reduced to protect the detrusor muscle contraction signal, while broadband noise suppression is enhanced during the resting period. This step, through multimodal filtering collaboration and closed-loop parameter optimization, ensures a balance between signal integrity and noise suppression requirements in different scenarios.

[0068] In summary, this embodiment provides a signal processing method for urinary system pressure measurement based on biometric recognition. This method can solve the problems of noise interference and signal distortion caused by human factors in urinary system pressure measurement, thereby improving the reliability of measurement results and the accuracy of diagnosis.

[0069] Example 2

[0070] This application provides a method for processing urinary system pressure measurement signals based on biometric recognition. This method is basically the same as the method in Embodiment 1, except that the method in this embodiment further includes:

[0071] The signal-to-noise ratio, waveform continuity index, and spectral purity of the filtered pressure signal are calculated in real time to quantify the signal quality level. When a pressure signal with a degraded signal quality level is detected, a graded response is triggered.

[0072] In this embodiment, the signal-to-noise ratio (SNR), waveform continuity index, and spectral purity are first calculated in real time to quantify the signal quality level. When quality degradation is detected (such as SNR reduction rate, spectral purity shift, or waveform continuity index exceeding the corresponding threshold), a graded response is triggered: a primary warning and a severe warning are distinguished according to the specific medical situation. The primary warning prompts medical staff to check the equipment contact status, while the severe warning suspends monitoring and issues an audible and visual alarm.

[0073] The method in this embodiment also includes:

[0074] Users can use a visual interactive interface to label low-quality signal segments of the patient's real-time pressure signal, extract noise features of the low-quality signal segments, and incrementally update the interference feature library.

[0075] Signal quality level results are used to optimize the baseline noise model and filtering;

[0076] Optimization reports are generated by tracking quality indicators over a long period.

[0077] Specifically, in this embodiment, a visual interactive interface is provided, which allows users (specifically doctors in this embodiment) to manually label low-quality signal segments (sensor detachment intervals, catheter blockage intervals, circuit saturation intervals, and wireless transmission packet loss intervals), automatically extract the noise features of the intervals, and incrementally update them to the interference feature library.

[0078] In this embodiment, in order to form a closed-loop optimization, the signal quality level results are used to inversely optimize the parameters of the baseline noise model and the filter. For example, a continuous low signal-to-noise ratio triggers the calibration module to increase the dynamic update frequency, or drives the filter module to enhance the suppression strength of a specific frequency band.

[0079] In this embodiment, an optimization report is generated by tracking quality indicators over a long period of time. The optimization report can guide hardware design improvements (such as enhancing sensor electromagnetic shielding) and upgrading filtering algorithms and incremental learning algorithms.

[0080] Another aspect of this application provides a urinary system pressure measurement signal processing system based on biometric recognition, including:

[0081] The baseline noise modeling module is used to inject a standard pressure source to simulate a known load, acquire pressure signals from calibrated pressure sensors and analyze their spectral characteristics to construct a baseline noise model; it periodically triggers the calibration process and adjusts the parameters of the baseline noise model in conjunction with real-time environmental monitoring data to compensate for sensor sensitivity drift; the pressure sensor is placed in the patient's bladder;

[0082] The biometric extraction module is used to simultaneously collect pressure signals, inertial motion data, and real-time environmental monitoring data while the patient performs a standardized action sequence. It decouples the components of the pressure signal, extracts interfering events and their original feature parameters, and combines these with association rules of interfering events and physiological states, along with historical processing strategies, to create a hierarchical feature encryption storage system for interfering features. The module further expands the system with novel interfering features and optimizes the interfering feature library through incremental learning algorithms. Interfering events include physiological interference, behavioral interference, and environmental interference. Physiological interference includes the patient performing standardized actions, behavioral interference includes the patient's inertial motion, and environmental interference includes real-time environmental monitoring data.

[0083] An adaptive filtering algorithm module is used to acquire the patient's real-time pressure signal. Based on a baseline noise model, it filters out the environmental noise component of the real-time pressure signal. The real-time pressure signal after removing the environmental noise component is matched with an interference feature library. Targeted filtering enhancement is applied to the matched interference features to obtain the filtered pressure signal. It should be noted that, depending on the implementation needs, the various steps described in this application can be broken down into more steps, or two or more steps or parts of steps can be combined into new steps to achieve the purpose of this invention.

[0084] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for processing urinary system pressure measurement signals based on biometric recognition, characterized in that, include: A standard pressure source is injected to simulate a known load. Pressure signals from calibrated pressure sensors are collected and their spectral characteristics are analyzed to construct a baseline noise model. The calibration process is periodically triggered and the parameters of the baseline noise model are adjusted in conjunction with real-time environmental monitoring data to compensate for the sensitivity drift of the sensor; wherein the pressure sensor is placed in the patient's bladder; While the patient performs a standardized sequence of actions, pressure signals, inertial motion data, and real-time environmental monitoring data are simultaneously collected. The components of the pressure signals are decoupled, and interference events and their original feature parameters are extracted. These are then combined with the association rules of interference events and physiological states, as well as historical processing strategies, to form a hierarchical feature encryption storage system for interference features. Novel interference features are expanded through incremental learning algorithms to optimize the interference feature system. The interference events include physiological interference, behavioral interference, and environmental interference. Physiological interference includes the patient performing standardized actions, behavioral interference includes the patient's inertial motion, and environmental interference includes the real-time environmental monitoring data. The system acquires the patient's real-time stress signal and filters out the environmental noise component of the signal based on the baseline noise model. The real-time stress signal after environmental noise removal is then matched with the interference feature library. Targeted filtering enhancement is applied to the matched interference features to obtain the filtered stress signal. The signal-to-noise ratio, waveform continuity index, and spectral purity of the filtered stress signal are calculated in real-time to quantify the signal quality level. A graded response is triggered when a stress signal with a degraded quality level is detected. Users can annotate low-quality signal segments of the patient's real-time stress signal through a visual interactive interface, extract noise features from these low-quality signal segments, and incrementally update the interference feature library. The signal quality level results are used to optimize the baseline noise model and the filtering. An optimization report is generated by long-term tracking of quality indicators. The signal fidelity and noise suppression ratio are used as reward functions. Based on the baseline noise model, the environmental noise component of the real-time pressure signal is removed by filtering, and the weights of the filtering parameters are dynamically adjusted by the targeted filtering enhancement. The signal fidelity mentioned above includes waveform smoothness; The calibration of the pressure sensor includes: placing the pressure sensor at a pressure-free reference point through a hardware zeroing procedure, and eliminating the initial offset using a multi-channel calibration circuit; The periodic trigger calibration process, combined with real-time environmental monitoring data, adjusts the parameters of the baseline noise model, including: The system periodically acquires the real-time environmental monitoring data. When the change in the real-time environmental monitoring data exceeds a threshold, the system automatically reloads the standard pressure source and performs the following operations: Update the frequency domain attenuation coefficient of the baseline noise model and dynamically adjust the suppression weights of noise in different frequency bands; Reconstruct the noise power spectral density template and calibrate the noise energy distribution of each frequency band based on the spectral characteristics of the calibration signal; Correct the temperature-sensitivity drift coefficient and fit the rate of change of sensor gain with temperature using temperature sensor data; Optimize the vibration coupling gain matrix and establish the transfer function between environmental vibration frequency and pressure noise by combining inertial motion data; Recalculate the baseline drift rate and quantify the sensor zero-point shift characteristics over time and temperature; Adjust the non-Gaussian noise weights and update the robustness parameters of the baseline noise model based on the statistical characteristics of sudden interference events.

2. The method for processing urinary system pressure measurement signals based on biometric recognition according to claim 1, characterized in that, The pressure-free reference points include the pubic symphysis region, abdominal wall region, presacral region, in vitro simulated tissue, and the zero point at the end of deep breathing; The standard pressure sources include physiological saline, compressed carbon dioxide, the patient's natural bladder filling pressure, and an adjustable height water column. The real-time environmental monitoring data includes temperature, electromagnetic field strength, and humidity; The baseline noise model establishes a mapping relationship between noise components and signal amplitude by decomposing the frequency domain characteristics of environmental noise; wherein the frequency domain characteristics include power frequency interference and thermal noise.

3. The method for processing urinary system pressure measurement signals based on biometric recognition according to claim 1, characterized in that, The standardized movements are physiological behaviors, including coughing and changing body positions; the inertial movements include walking and climbing stairs. The components of the pressure signal are decoupled using a time-frequency joint analysis method; The original characteristic parameters of the interference event include the waveform characteristics of the physiological interference, the motion trajectory pattern of the behavioral interference, and the spectral fingerprint of the environmental interference. The step of expanding novel interference features and optimizing the interference feature library through incremental learning algorithms includes: expanding novel interference features through incremental learning algorithms, adjusting matching thresholds in conjunction with real-time environmental monitoring data, and continuously optimizing the interference feature library by integrating clinical feedback data.

4. The method for processing urinary system pressure measurement signals based on biometric recognition according to claim 1, characterized in that, The step of stripping the environmental noise component from the real-time pressure signal based on the baseline noise model includes: In the frequency domain, an adaptive notch filter is designed to eliminate power frequency interference and equipment resonance peaks. In the time domain, a combination of moving average filtering and median filtering is used to suppress high-frequency random noise and sudden pulse interference. The interference features in the real-time pressure signal after removing noise components are matched with the interference feature library, and targeted filtering enhancement is applied to the matched interference features, including: For specific frequency bands of the interference characteristics matched, the notch depth is dynamically increased and the filter window is narrowed to achieve local noise elimination.

5. The method for processing urinary system pressure measurement signals based on biometric recognition according to claim 1, characterized in that, The degraded pressure signal refers to the signal-to-noise ratio reduction rate, spectral purity shift, and waveform continuity index exceeding the corresponding threshold. The tiered response specifically includes distinguishing between primary and severe warnings based on the specific medical situation. Primary warnings will prompt medical staff to check the contact status of the equipment, while severe warnings will suspend monitoring and issue an audible and visual alarm.

6. The method for processing urinary system pressure measurement signals based on biometric recognition according to claim 1, characterized in that, The low-quality signal segment includes the sensor detachment segment, the conduit blockage segment, the circuit saturation segment, and the wireless transmission packet loss segment.

7. A urinary system pressure measurement signal processing system based on biometric recognition for implementing the method of claim 1, characterized in that, include: The baseline noise modeling module is used to inject a standard pressure source to simulate a known load, acquire pressure signals from calibrated pressure sensors and analyze their spectral characteristics to build a baseline noise model. The calibration process is periodically triggered and the parameters of the baseline noise model are adjusted in conjunction with real-time environmental monitoring data to compensate for the sensitivity drift of the sensor; wherein the pressure sensor is placed in the patient's bladder; A biometric extraction module is used to simultaneously collect pressure signals, inertial motion data, and real-time environmental monitoring data while the patient performs a standardized action sequence. It decouples the components of the pressure signals, extracts interfering events and their original feature parameters, and combines these with association rules of interfering events and physiological states, along with historical processing strategies, to perform hierarchical feature encryption and storage, forming an interfering feature library. The module further expands the library with novel interfering features through an incremental learning algorithm, optimizing the interfering feature library. The interfering events include physiological interference, behavioral interference, and environmental interference. Physiological interference includes the patient performing standardized actions, behavioral interference includes the patient's inertial motion, and environmental interference includes the real-time environmental monitoring data. An adaptive filtering algorithm module is used to acquire the patient's real-time pressure signal, and based on the baseline noise model, filter out the environmental noise component of the real-time pressure signal; match the real-time pressure signal after removing the environmental noise component with the interference feature library, and activate targeted filtering enhancement for the matched interference features to obtain the filtered pressure signal.

Images