A system and method for enhancing ultrasound imaging of the spermatic vein based on double-wall filtering and adaptive doppler signal processing

Abstract

The present application belongs to the technical field of medical ultrasonic signal processing, and specifically relates to a spermatic vein ultrasonic imaging enhancement system and method based on double-wall filtering and adaptive Doppler signal processing. The system comprises a dynamic reconfigurable analog front end, a dual-channel heterogeneous data processing unit and a closed-loop feedback controller. The method extracts fluid dynamics characteristics in real time through an auxiliary path, intelligently identifies resting state and action mutation state. At the action mutation moment, the physical layer automatically switches the analog filtering parameters to prevent signal saturation, and the digital layer adaptively increases the FIR filter weight to eliminate the ringing artifact; in the resting state, the high-gain analog configuration is maintained and the IIR filter is dominated to capture weak signals. With the delay buffer mechanism, accurate timing alignment of control instructions and imaging data is realized. The present application effectively solves the contradiction between low flow rate detection sensitivity and strong tissue motion interference suppression, eliminates the flicker artifact generated by breath holding action, and significantly improves the diagnosis accuracy of weak reflux.

Description

Technical Field

[0001] This invention belongs to the field of medical ultrasound signal processing technology, specifically a spermatic vein ultrasound imaging enhancement system and method based on dual-wall filtering and adaptive Doppler signal processing. Background Technology

[0002] Varicocele is a common cause of male infertility, and its ultrasound diagnosis hinges on detecting weak reflux signals within the veins. However, clinical detection faces significant technical challenges, primarily manifested in the contradiction between extremely low flow rates and strong interference. Weak signal: The reflux velocity of the spermatic vein is extremely low (usually less than 2 cm / s), and its Doppler frequency shift is often submerged in system noise and low-frequency signals of vascular wall pulsation, requiring extremely high detection sensitivity.

[0003] Dynamic range conflict: To induce reflux, doctors often ask patients to perform the Valsalva maneuver (holding their breath to increase abdominal pressure). At the moment of initiation of this maneuver, the intense contraction of the abdominal muscles generates a large-amplitude tissue movement signal (Clutter).

[0004] Limitations of existing technology: Traditional ultrasound systems typically use a high-pass filter with fixed parameters in the analog front end (AFE). If the cutoff frequency is set too low to preserve weak blood flow, the strong interference generated by the Valsalva maneuver will directly cause the analog-to-digital converter (ADC) to clip, rendering all subsequent digital signal processing ineffective.

[0005] Existing digital wall filters typically have single parameters and slow adjustment. While higher-order IIR filters can remove low frequencies, they can produce severe "ringing effects" during signal abrupt changes, resulting in persistent flickering artifacts in the image and seriously interfering with doctors' judgment. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a spermatic vein ultrasound imaging enhancement system and method based on dual-wall filtering and adaptive Doppler signal processing. It has the advantages of preventing analog end saturation, distinguishing between weak blood flow and strong tissue interference in real time, and seamlessly switching processing strategies between the two.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a spermatic vein ultrasound imaging enhancement system based on dual-wall filtering and adaptive Doppler signal processing, the system comprising: The reconfigurable analog front-end module, as the first layer of filtering unit, is used to receive the electrical signal converted by the ultrasound probe. This module includes an RC network with topology switching capability, which is used to perform high-pass filtering with variable cutoff frequency in the analog domain, and adjust the hardware gain and filtering parameters in real time according to the received feedback instructions to output an unsaturated analog RF signal. An analog-to-digital converter module is used to convert the analog radio frequency signal into a raw digital sequence; The dual-path data splitting module is used to copy the original digital sequence into two parallel data paths: one is the main signal imaging path that maintains the original high sampling rate, and the other is a low-rate control path that has undergone multi-stage downsampling processing. The fluid dynamics state observation module, located in the low-rate control path, is used to extract time-frequency features from the downsampled baseband signal and identify the current fluid state, which includes at least the resting microfluidic state, the tissue motion disturbance state, and the Valsalva action mutation state. An adaptive digital filtering execution module, serving as a second-layer filtering unit, is located in the main signal imaging path. This module contains parallel filter banks of different types, and dynamically selects the corresponding filtering strategy to perform secondary fine filtering on the main signal based on the identification results of the hydrodynamic state observation module. The closed-loop feedback controller generates two sets of coordinated control signals based on the state results output by the fluid dynamics state observation module: the first set of signals is sent in reverse to the reconfigurable analog front-end module to physically adjust the cutoff frequency of the first wall filter to prevent signal saturation; the second set of signals is sent to the adaptive digital filter execution module to adjust the weight coefficient of the second wall filter to separate the blood flow signal.

[0008] Preferably, the reconfigurable analog front-end module includes: an active filter circuit composed of operational amplifiers; An array of electronic switches connected in parallel to the key nodes of the filter circuit; and a precision resistor network connected to the electronic switch array; The closed-loop feedback controller sends high and low level commands to the electronic switch array to change the resistance combination value of the connected circuit, thereby physically switching the cutoff frequency of the analog high-pass filter between a low-frequency threshold and a high-frequency threshold within milliseconds. The low-frequency threshold is used to allow weak blood flow signals to pass through, while the high-frequency threshold is used to block low-frequency high-energy interference generated by large-amplitude tissue movement.

[0009] Preferably, the feature extraction logic of the fluid dynamics state observation module is as follows: Perform a sliding window short-time Fourier transform on the signal in the low-rate domain to obtain the time-frequency distribution matrix; Calculate the ratio of low-frequency energy to full-frequency energy. If the ratio exceeds a preset interference threshold, it is determined to be a tissue motion interference state. Calculate the autocorrelation coefficient of the spectrum of adjacent time windows. If the correlation coefficient is higher than the preset continuity threshold and there is a unidirectional Doppler frequency shift, it is determined to be a resting microfluidic state. Monitor the instantaneous rate of change of energy across the entire frequency band. If the rate of change exceeds the mutation threshold, it is determined to be a Valsalva action mutation state.

[0010] Preferably, the adaptive digital filtering execution module includes: An infinite impulse response (IIR) filter unit is configured with a steep transition band to maximize the preservation of low-velocity blood flow signals near zero frequency in a resting microfluidic state. A finite impulse response (FIR) filter unit, configured to have linear phase characteristics, is used to suppress broadband transient noise without producing phase distortion under Valsalva action abrupt states; A weighted fusion unit is used to connect the outputs of an IIR filter unit and an FIR filter unit. The closed-loop feedback controller outputs a dynamic weighting factor based on the current fluid state, and controls the weighted fusion unit to perform linear cross-fade-in and fade-out between the two filter outputs, thereby achieving seamless and smooth switching of the filtering mode.

[0011] Preferably, the dual-path data splitting module further includes: A delay buffer unit is connected in series in the main signal imaging path; The delay duration of the delay buffer unit is configured to be equal to the computation time required for the fluid dynamics state observation module to perform feature extraction and state determination; The delay buffer unit ensures that the main signal data frame arriving at the adaptive digital filter execution module is precisely aligned with the control command generated by the closed-loop feedback controller on the time axis, thus ensuring that the adjustment of filter parameters is synchronized with the actual state changes of the signal.

[0012] Preferably, the control strategy of the closed-loop feedback controller specifically includes: When the fluid dynamics state observation module predicts an impending Valsalva action abrupt change, it prioritizes triggering the first set of control signals to increase the cutoff frequency of the reconfigurable simulation front-end module and reduce the front-end gain in order to reserve dynamic range. After confirming that the signal is not saturated and has entered a stable backflow phase, the second set of control signals is triggered to reduce the cutoff frequency of the adaptive digital filter execution module and increase the output weight of the IIR filter unit in order to capture the low-speed backflow signal.

[0013] Preferably, the system also includes a color flow imaging unit: Used to receive the pure blood flow signal output by the adaptive digital filtering execution module; Interpolate and upsample the signal to restore it to the imaging rate; Blood flow velocity, variance, and energy parameters were calculated using an autocorrelation algorithm. Based on the fluid state identified by the fluid dynamics state observation module, color coding is performed only within the time window determined to be effective blood flow, and color display is suppressed within the time window of tissue motion interference state, thereby eliminating flicker artifacts.

[0014] A method for ultrasound imaging of the spermatic vein based on dual-wall filtering and adaptive Doppler signal processing, characterized by the following steps: S1: The ultrasonic echo signal is first-stage analog wall filtering and amplification using a reconfigurable analog front-end module, and then converted into a digital sequence; S2: Split the digital sequence into high-sampling-rate main path data and low-rate control data that has been downsampled; S3: Extract fluid dynamic features from low-rate control data and use state machine logic to determine whether the current signal type is weak blood flow, strong tissue artifact, or instantaneous action transition. S4: Generate closed-loop feedback instructions based on the judgment results; S5: According to the feedback instruction, reverse the adjustment of the hardware filtering parameters of the analog front end in step S1 to prevent saturation, and at the same time, forward adjust the type weight and cutoff frequency of the second digital wall filter in the main path. S6: Perform parameter estimation and color imaging on the main path signal after double filtering.

[0015] Preferably, the parameter adjustment strategy in step S5 includes: When the spermatic vein is detected to be in a resting state, the cutoff frequency of the first analog wall filter is controlled to the range of 20Hz-50Hz, and the second digital wall filter is controlled to mainly adopt the high-order IIR mode. Upon detecting the start of the Valsalva action, the cutoff frequency of the first analog wall filter is instantly increased to the 100Hz-200Hz range, and the second digital wall filter is controlled to primarily adopt the linear phase FIR mode.

[0016] Preferably, the extraction of fluid dynamic features in step S3 does not rely on full sampling rate data, but is based on baseband data with a downsampling rate of 1 / 10 to 1 / 100 of the original sampling rate, in order to reduce the computational load of real-time processing.

[0017] Compared with existing technologies, this invention provides a spermatic vein ultrasound imaging enhancement system and method based on dual-wall filtering and adaptive Doppler signal processing, which has the following beneficial effects: Breaking through dynamic range limitations: By using digital domain decision-making to inversely control the analog domain circuit, strong interference is attenuated by physical means before it reaches the ADC, fundamentally solving the signal loss problem caused by ADC saturation.

[0018] Balancing sensitivity and immunity: The dual-filter architecture allows the system to reach extremely low cutoff frequencies during the resting period (preserving weak blood flow) and instantly switch to high cutoff frequencies during the action period (filtering out artifacts), significantly improving diagnostic accuracy.

[0019] Image artifact elimination: A hybrid filtering strategy is adopted, which utilizes the linear phase characteristics of the FIR filter to process abrupt signals, thereby eliminating the ringing flicker caused by traditional IIR filtering.

[0020] Computationally efficient: It adopts a dual-time-scale architecture, and complex feature analysis is completed in the low-rate domain, which reduces the system's computing power requirements and is easy to implement in engineering. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the system architecture of the present invention; Figure 2 This is a flowchart illustrating the implementation of the method of the present invention. Detailed Implementation

[0022] To better understand the purpose, structure, and function of this invention, the following detailed description, in conjunction with the accompanying drawings, provides an overview of the spermatic vein ultrasound imaging enhancement system and method based on dual-wall filtering and adaptive Doppler signal processing.

[0023] Please see Figure 1-2 The operation of this system is a closed-loop dynamic feedback process, which can be divided into the following six key steps: Step S1: Dynamically reconfigurable analog signal conditioning (physical layer first-wall filtering) Initialization Configuration: After system startup, the closed-loop feedback controller defaults to setting the reconfigurable analog front-end to "high-sensitivity monitoring mode." At this time, the controller sends a command to the electronic switch array in the analog circuit to connect a high-resistance resistor network. This operation physically sets the cutoff frequency of the analog high-pass filter to a lower value (e.g., 20 Hz) while simultaneously setting the gain of the preamplifier to a higher level (e.g., 40 decibels) to ensure the capture of weak blood flow signals.

[0024] Signal preprocessing: After the ultrasound probe receives the echo signal, it first passes through the analog circuit configured above. Before the signal enters the analog-to-digital converter, this circuit filters out extremely low-frequency body sway signals and amplifies them through a physical resistor-capacitor network.

[0025] Anti-saturation mechanism: Once the system predicts an impending strong interference in subsequent steps, the feedback controller immediately sends a "strong interference warning" command. This command triggers an electronic switch, disconnecting the high-resistance path and switching to a low-resistance resistor. This physical switch instantly raises the cutoff frequency of the analog filter (e.g., to 150 Hz) within microseconds, while simultaneously reducing the amplifier gain. This "hard adjustment" ensures that the voltage swing output to the analog-to-digital converter remains within the linear range, fundamentally preventing clipping distortion caused by excessively strong signals.

[0026] Step S2: Dual-channel heterogeneous data offloading and timing alignment Digital conversion: Analog-to-digital converters convert analog signals into raw digital sequences at high sampling rates (e.g., 40 MHz).

[0027] Data splitting: The system copies the numerical sequence into two copies, which are then sent to two parallel processing paths: Imaging Main Path: To ensure image quality, this path maintains a high sampling rate and integrity of the original signal. Data is directly stored in a first-in, first-out (FIFO) delay buffer. The storage depth of this buffer is precisely designed so that its delay duration is exactly equal to the time required for analysis and calculation by the control path. This means that when this data flows out of the buffer, the control path has just calculated the control command for this data, achieving precise alignment of "data and command" on the time axis.

[0028] Decision auxiliary path: To improve real-time performance, data in this path enters a multi-stage downsampling filter. The data volume is significantly compressed (e.g., to one percent of the original data volume), generating a low-rate baseband signal. This enables the system to quickly perform complex feature analysis on a low-power processor.

[0029] Step S3: Extraction of hydrodynamic fingerprint features In the decision auxiliary pathway, the processor performs sliding window analysis on the low-rate baseband signal to extract the following three key fluid features (full text description): Low-frequency energy percentage: This refers to the proportion of low-frequency energy in the total energy. An extremely high percentage usually indicates significant tissue movement interference.

[0030] Signal autocorrelation: Calculates the similarity between the current signal and the signal at the previous time step. Blood flow signals typically have high similarity, while broadband noise or abrupt interference has extremely low similarity.

[0031] Directional consistency index: Compare positive and negative frequency energy. If the energy in one direction is absolutely dominant, it is determined to be unidirectional blood flow; if the energy in both directions is balanced, it may be tissue vibration.

[0032] Step S4: State Machine Decision and Parameter Generation Based on the above characteristics, the system's intelligent state machine logic determines the current fluid state and generates a corresponding control strategy: Resting-state decision: When a signal exhibits low energy, high similarity, and unidirectional characteristics, it is determined to be in a resting state. At this point, a "high-sensitivity" control command is generated.

[0033] Action abrupt change state determination: When a sudden burst of energy across the entire frequency band is detected and the signal similarity drops sharply (such as at the moment of starting a breath-holding action), it is determined to be an action abrupt change state. At this time, a "strong suppression" control command is generated, including triggering the anti-saturation switching of the analog front end.

[0034] Parameter generation: The state machine not only outputs state labels, but also generates specific digital filter parameters in real time, including the weighting factor of the hybrid filter (which determines whether it is biased towards steady-state filtering or transient filtering) and the cutoff frequency value of the digital filter.

[0035] Step S5: Adaptive hybrid filtering is performed (second-wall filtering). The main data flowing out of the delay buffer then enters the adaptive digital filtering execution module. This module contains two sets of filters operating in parallel: IIR filter banks: have extremely steep frequency response characteristics, suitable for preserving weak signals in steady state.

[0036] FIR filter banks: have linear phase characteristics and do not produce ringing artifacts when processing abrupt signals.

[0037] Dynamic fusion: The system weights and superimposes the outputs of the two filters according to the weighting factors generated in step S4.

[0038] In the resting state, the weights are fully biased towards the IIR filter to maximize the extraction of low-velocity blood flow.

[0039] During the action abrupt state, the weights are quickly and smoothly transferred to the FIR filter, utilizing its linear phase characteristics to eliminate flicker artifacts while removing broadband noise.

[0040] Step S6: Artifact-free imaging reconstruction The purified signal, after double filtering, is sent to the color imaging unit. To further enhance the user experience, the system has added display control logic: when the state machine determines that it is in a "sudden change state" or "strong interference state", it will automatically temporarily turn off the gain of the color display to prevent messy color noise from appearing in the screen background; after the state returns to stability, the color blood flow display will be automatically restored.

[0041] Case Background: A male patient suspected of having early-stage varicocele underwent an ultrasound examination. To confirm the diagnosis, the doctor needed to perform the Valsalva maneuver (breath-holding to increase abdominal pressure) to observe whether there was blood reflux in the vein.

[0042] Full-process dynamic response analysis: Phase 1: Quiet Breathing Phase (Baseline Monitoring) Physiological phenomenon: When the patient lies flat and relaxes, the blood flow in the spermatic vein is extremely slow, or even close to stagnation.

[0043] System behavior: The system identified the current signal energy as weak and stable.

[0044] Analog end: Maintain high gain and low cutoff frequency to collect as weak echoes as possible.

[0045] Digital end: High-order IIR filtering mode is adopted, and the cutoff frequency of the digital filter is further reduced to prevent the accidental removal of extremely low-speed blood flow signals.

[0046] Results: The screen clearly displayed intermittent, weak spontaneous venous blood flow signals, and the image background was clean, without any black hole effect caused by excessive filtering.

[0047] Phase Two: The initial moment of breath-holding (strong interference impact) Physiological phenomenon: When the doctor gives the instruction, "Please hold your breath!", the patient's abdominal muscles contract violently. At this moment, the muscles and blood vessels in the groin area undergo significant displacement, generating extremely strong low-frequency motion signals (scintillation artifacts), whose intensity far exceeds that of blood flow signals.

[0048] System behavior: The decision auxiliary pathway detects that the energy burst rate exceeds the threshold within milliseconds and immediately identifies it as an "action mutation".

[0049] On the analog side: Before the strong interference signal reaches the saturation point of the analog-to-digital converter, the feedback controller triggers a physical switch to instantly raise the analog filter cutoff frequency and reduce the gain. This "emergency braking" operation protects the original signal waveform from being clipped.

[0050] On the digital end: the hybrid weighting quickly switches to FIR filtering mode, utilizing its linear phase characteristics to process abrupt signal changes, avoiding the "ringing" phenomenon common in traditional equipment. Simultaneously, the back-end imaging unit temporarily disables the color display.

[0051] Result: In this situation, the screen of a traditional device would typically display a chaotic burst of color flickering, obscuring the location of the blood vessels. However, the screen background of this system remained in grayscale mode, very clean, without any color noise interference.

[0052] Phase 3: Reflux Duration (Critical Period for Diagnosis) Physiological phenomenon: The patient maintains a breath-holding state, and the increased abdominal pressure causes the venous valves to fail, resulting in blood backflow. At this point, tissue movement ceases, and the velocity of the refluxed blood flow gradually increases and then slowly decreases.

[0053] System behavior: The system detected that the tissue motion interference had disappeared, and was replaced by a blood flow signal with a clear direction.

[0054] Simulation end: Simulation parameters are reverted to standard state.

[0055] Digital side: The weights gradually and smoothly transition back to the IIR-dominant mode to finely depict the blood flow profile.

[0056] Result: A red reverse blood flow jet appeared precisely on the screen. Because the previous strong interference did not cause circuit saturation, the system was able to completely record the first time point of reflux onset and accurately measure the duration of reflux (a core indicator for judging the severity of the lesion), effectively avoiding missed diagnoses.

[0057] As can be seen from the above cases, this system successfully resolved the contradiction between "low flow rate detection" and "strong interference suppression" by combining analog and digital technologies and coordinating time and frequency domains, thus significantly improving the accuracy of clinical diagnosis. Example

[0058] Detailed Implementation Steps Step S1: Dynamically reconfigurable analog signal conditioning (physical wall filtering) Hardware configuration: When the system starts up, the closed-loop feedback controller will reconfigure the analog front-end (AFE) and initialize it to "high sensitivity mode". At this time, the control command closes the high and low resistance paths in the electronic switch array, setting the cutoff frequency (f_{c_analog}) of the analog amplifier to 20Hz and the preamplifier gain (LNAGain) to 40dB.

[0059] Signal acquisition: An ultrasound probe, for example a 75MHz linear probe, receives echo signals from the spermatic vein region. This analog signal is first passed through a fault-tolerant network configured as described above to eliminate extremely low-frequency body sway signals and is then amplified.

[0060] Anti-saturation mechanism: If strong interference is detected in subsequent steps, the alarm will be triggered and a command will be sent to disconnect the high-frequency channel and connect a low-resistance power supply. Instantly, f_canalog will increase to 150Hz and the gain will decrease to 20dB. This process is completed in the analog domain, ensuring that the voltage swing of the output to the digital-to-digital converter (ADC) remains online within the range of 2Vpp, physically eliminating digital distortion. Step S2: Dual-channel heterogeneous data splitting. Digitization: The ADC converts the analog signal into a raw digital sequence x[n] with a sampling rate of 40 MHz.

[0061] Flow splitting logic: Path A (Main Imaging Path): Maintains the integrity of the original signal. Signal x[n] is directly stored in a FIFO (First In First Out) delay buffer. The buffer depth D is set to 512 sampling points (corresponding to approximately 12.8 μs, plus a processing delay of approximately 5 ms). This delay is strictly equal to the computation time of Path B, and is used to achieve timing alignment between control commands and signal data.

[0062] Path B (Decision Auxiliary Path): Signal x[n] enters a three-stage cascaded integrator-comb (CIC) decimation filter, with a downsampling factor set to 40, outputting a baseband signal with a sampling rate of 1MHz. This reduced the amount of data by 97.5%, significantly decreasing the computational burden on real-time analysis.

[0063] Step S3: Extraction of hydrodynamic fingerprint features In path B, a 64-point sliding window FFT (Fast Fourier Transform) is performed on the baseband signal to calculate the following eigenvectors: Low-frequency energy ratio (CR): Calculates the ratio of energy in the 0-100Hz frequency band to the energy across the entire frequency band. If it is greater than 0.85, it is marked as "potential tissue movement".

[0064] Autocorrelation coefficient (p): Calculates the spectral correlation between adjacent time windows. If it is greater than 0.9, it is marked as "stable signal"; if it is less than 0.4, it is marked as "random noise or sudden interference".

[0065] Doppler asymmetry (AI): compares positive and negative frequency energies. If IE+-E_1 / Etotal>0.6, it is judged as "unidirectional blood flow".

[0066] Step S4: State Machine Decision and Parameter Generation Based on the eigenvectors of S3, a finite state machine (FSM) determines the fluid state at the current time t and generates control parameters: State determination logic: Resting state: CR < 0.85 and p > 0.9 and AI > 0.6.

[0067] Transient state: a sudden increase in energy across the entire frequency band (EnergyRate>Threshold) and p<0.4 (typically the instant at the start of the Valsalva action).

[0068] Strong artifacts: CR>0.9 and p>0.6 (typically respiratory motion).

[0069] Parameter generation: Parameter generation: α (Adjustable resistor switch) Digital filter mixing weight α (0-1, first-generation IIR, 0 represents full FIR) Generate the digital cutoff frequency fc_digital.

[0070] Step S5: Adaptive hybrid filtering execution (main path processing) The main path data x[nD] read from the delay buffer coincides in time with the control parameters generated by S4. The adaptive filter module performs the following operations: Parallel filtering: Data is simultaneously fed into a fourth-order IIR high-pass filter (output yIIR) and a 64th-order linear-phase FIR filter (output yFIR).

[0071] The linear phase characteristics of FIR are used to eliminate ringing artifacts.

[0072] Dynamic fusion: The final output signal Yout = α·yIIR + (1-α)·yFIR In the resting state, α=1. The higher-order properties of IIR are utilized to deeply explore weak blood flow.

[0073] In the action abrupt state, α linearly drops to 0 within 5ms, and the ringing artifact is eliminated by utilizing the linear phase characteristics of FIR.

[0074] Step S6: Artifact-free imaging reconstruction The clean signal Yout is color encoded. If the state machine determines that it is in an "action change state" or "strong artifact state", the back-end image processor will automatically set the color gain to zero temporarily to prevent the screen from displaying "color mosaic" flickering; after the state returns to the "resting state" or "return state", the color display will be automatically restored.

[0075] II. Explanation of Typical Clinical Application Cases Case Background: The patient is a 30-year-old male, suspected of having subclinical varicocele on the left side (not palpable on routine physical examination, but visible on ultrasound). The doctor performed the Valsalva maneuver (breath-holding to increase abdominal pressure) induction test.

[0076] Timeline full process analysis: Phase T1: Resting baseline period (0s-2s) Physiological phenomenon: When the patient breathes calmly, the blood flow in the spermatic vein is extremely slow (about 1-2 cm / s), or even stops.

[0077] System response: State determination: Identified as "resting microfluidic state".

[0078] Simulation operation: Maintain high gain, simulation cutoff frequency 20Hz.

[0079] Digital action: Hybrid weight α=1.0 (full IIR), digital cutoff frequency down to 30Hz.

[0080] Image results: The screen clearly displays the intermittent, weak blood flow signals within the vein, without excessive filtering of black holes. Compared to traditional equipment, more details of low-velocity blood flow can be seen at this point.

[0081] Phase T2: The instant the action begins (2s-2.5s) Physiological phenomenon: When the doctor instructs, "Please inhale and hold your breath!", the patient's abdominal muscles contract violently. The groin area tissues undergo significant displacement, generating an extremely strong low-frequency Doppler signal (Clutter), with an intensity exceeding that of the blood flow signal by more than 60 dB.

[0082] System response: State determination: The state machine detects an energy burst and a sudden drop in correlation within 2ms and determines it as an "action mutation state".

[0083] Simulated Action: Immediately triggers the anti-saturation command, physically switching the simulated cutoff frequency to 150Hz with a gain attenuation of 20dB. Key Effect: At this point, the ADC input voltage is not saturated, preserving the original waveform information.

[0084] Digital action: The mixed weight α quickly becomes 0 (full FIR), the digital cutoff frequency is increased to 200Hz, and broadband rejection is enabled.

[0085] Image results: Traditional devices would display a chaotic, flickering array of colors on the screen, obscuring the blood vessels. This system, however, automatically suppresses color display, maintaining a clean grayscale B mode background without any interference noise.

[0086] Phase T3: Duration of backflow (2.5s-5s) Physiological phenomenon: The patient holds their breath, and the increased abdominal pressure causes the venous valves to fail, resulting in backflow of blood. At this time, tissue movement stops, and the velocity of the refluxed blood flow gradually increases to a peak (about 15 cm / s) and then slowly decreases.

[0087] System response: Status determination: Tissue movement disappears, and a reverse unidirectional blood flow signal is detected, which is determined to be "continuous reflux state".

[0088] Simulation action: Simulation cutoff frequency is reset to 50Hz (balancing interference immunity and retention), gain is restored.

[0089] Digital action: The weight α is gradually restored to 0.8 (IIR dominant, low-frequency reflux is preserved), and the digital cutoff frequency is dynamically adjusted to follow the average blood flow velocity.

[0090] Image results: A red reversed blood flow jet appears precisely on the screen with clear boundaries, and the onset time and duration of the regurgitation are recorded in detail (this is a key indicator for diagnosing the severity of VC).

[0091] It is understood that the present invention has been described through some embodiments, and those skilled in the art will recognize that various changes or equivalent substitutions can be made to these features and embodiments without departing from the spirit and scope of the invention. Furthermore, under the teachings of the present invention, these features and embodiments can be modified to adapt to specific situations and materials without departing from the spirit and scope of the invention. Therefore, the present invention is not limited to the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application are within the protection scope of the present invention.

Claims

1. A spermatic vein ultrasound imaging enhancement system based on dual-wall filtering and adaptive Doppler signal processing, characterized in that, The system includes: The reconfigurable analog front-end module, as the first layer of filtering unit, is used to receive the electrical signal converted by the ultrasound probe. This module includes an RC network with topology switching capability, which is used to perform high-pass filtering with variable cutoff frequency in the analog domain, and adjust the hardware gain and filtering parameters in real time according to the received feedback instructions to output an unsaturated analog RF signal. An analog-to-digital converter module is used to convert the analog radio frequency signal into a raw digital sequence; The dual-path data splitting module is used to copy the original digital sequence into two parallel data paths: one is the main signal imaging path that maintains the original high sampling rate, and the other is a low-rate control path that has undergone multi-stage downsampling processing. The fluid dynamics state observation module, located in the low-rate control path, is used to extract time-frequency features from the downsampled baseband signal and identify the current fluid state, which includes at least the resting microfluidic state, the tissue motion disturbance state, and the Valsalva action mutation state. An adaptive digital filtering execution module, serving as a second-layer filtering unit, is located in the main signal imaging path. This module contains parallel filter banks of different types, and dynamically selects the corresponding filtering strategy to perform secondary fine filtering on the main signal based on the identification results of the hydrodynamic state observation module. The closed-loop feedback controller generates two sets of coordinated control signals based on the state results output by the fluid dynamics state observation module: the first set of signals is sent in reverse to the reconfigurable analog front-end module to physically adjust the cutoff frequency of the first wall filter to prevent signal saturation; the second set of signals is sent to the adaptive digital filter execution module to adjust the weight coefficient of the second wall filter to separate the blood flow signal.

2. The spermatic vein ultrasound imaging enhancement system based on dual-wall filtering and adaptive Doppler signal processing according to claim 1, characterized in that, The reconfigurable analog front-end module includes: an active filter circuit composed of operational amplifiers; An array of electronic switches connected in parallel to the key nodes of the filter circuit; and a precision resistor network connected to the electronic switch array; The closed-loop feedback controller sends high and low level commands to the electronic switch array to change the resistance combination value of the connected circuit, thereby physically switching the cutoff frequency of the analog high-pass filter between a low-frequency threshold and a high-frequency threshold within milliseconds. The low-frequency threshold is used to allow weak blood flow signals to pass through, while the high-frequency threshold is used to block low-frequency high-energy interference generated by large-amplitude tissue movement.

3. The spermatic vein ultrasound imaging enhancement system based on dual-wall filtering and adaptive Doppler signal processing according to claim 1, characterized in that, The feature extraction logic of the fluid dynamics state observation module is as follows: Perform a sliding window short-time Fourier transform on the signal in the low-rate domain to obtain the time-frequency distribution matrix; Calculate the ratio of low-frequency energy to full-frequency energy. If the ratio exceeds a preset interference threshold, it is determined to be a tissue motion interference state. Calculate the autocorrelation coefficient of the spectrum of adjacent time windows. If the correlation coefficient is higher than the preset continuity threshold and there is a unidirectional Doppler frequency shift, it is determined to be a resting microfluidic state. Monitor the instantaneous rate of change of energy across the entire frequency band. If the rate of change exceeds the mutation threshold, it is determined to be a Valsalva action mutation state.

4. The spermatic vein ultrasound imaging enhancement system based on dual-wall filtering and adaptive Doppler signal processing according to claim 1, characterized in that, The adaptive digital filtering execution module includes: An infinite impulse response (IIR) filter unit is configured with a steep transition band to maximize the preservation of low-velocity blood flow signals near zero frequency in a resting microfluidic state. A finite impulse response (FIR) filter unit, configured to have linear phase characteristics, is used to suppress broadband transient noise without producing phase distortion under Valsalva action abrupt states; A weighted fusion unit is used to connect the outputs of an IIR filter unit and an FIR filter unit. The closed-loop feedback controller outputs a dynamic weighting factor based on the current fluid state, and controls the weighted fusion unit to perform linear cross-fade-in and fade-out between the two filter outputs, thereby achieving seamless and smooth switching of the filtering mode.

5. The spermatic vein ultrasound imaging enhancement system based on dual-wall filtering and adaptive Doppler signal processing according to claim 1, characterized in that, The dual-path data splitting module also includes: A delay buffer unit is connected in series in the main signal imaging path; The delay duration of the delay buffer unit is configured to be equal to the computation time required for the fluid dynamics state observation module to perform feature extraction and state determination; The delay buffer unit ensures that the main signal data frame arriving at the adaptive digital filter execution module is precisely aligned with the control command generated by the closed-loop feedback controller on the time axis, thus ensuring that the adjustment of filter parameters is synchronized with the actual state changes of the signal.

6. The spermatic vein ultrasound imaging enhancement system based on dual-wall filtering and adaptive Doppler signal processing according to claim 1, characterized in that, The control strategy of the closed-loop feedback controller specifically includes: When the fluid dynamics state observation module predicts an impending Valsalva action abrupt change, it prioritizes triggering the first set of control signals to increase the cutoff frequency of the reconfigurable simulation front-end module and reduce the front-end gain in order to reserve dynamic range. After confirming that the signal is not saturated and has entered a stable backflow phase, the second set of control signals is triggered to reduce the cutoff frequency of the adaptive digital filter execution module and increase the output weight of the IIR filter unit in order to capture the low-speed backflow signal.

7. The spermatic vein ultrasound imaging enhancement system based on dual-wall filtering and adaptive Doppler signal processing according to claim 1, characterized in that, The system also includes a color flow imaging unit: Used to receive the pure blood flow signal output by the adaptive digital filtering execution module; The signal is upsampled and interpolated to restore it to the imaging rate; Blood flow velocity, variance, and energy parameters were calculated using an autocorrelation algorithm. Based on the fluid state identified by the fluid dynamics state observation module, color coding is performed only within the time window determined to be effective blood flow, and color display is suppressed within the time window of tissue motion interference state, thereby eliminating flicker artifacts.

8. A spermatic vein ultrasound imaging method based on dual-wall filtering and adaptive Doppler signal processing, characterized in that, The method includes the following steps: S1: The ultrasonic echo signal is first filtered and amplified using a reconfigurable analog front-end module and then converted into a digital sequence; S2: Split the digital sequence into high-sampling-rate main path data and low-rate control data that has been downsampled; S3: Extract fluid dynamic features from low-rate control data and use state machine logic to determine whether the current signal type is weak blood flow, strong tissue artifact, or instantaneous action transition. S4: Generate closed-loop feedback instructions based on the judgment results; S5: According to the feedback instruction, reverse the adjustment of the hardware filtering parameters of the analog front end in step S1 to prevent saturation, and at the same time, forward adjust the type weight and cutoff frequency of the second digital wall filter in the main path. S6: Perform parameter estimation and color imaging on the main path signal after double filtering.

9. The spermatic vein ultrasound imaging method based on dual-wall filtering and adaptive Doppler signal processing according to claim 8, characterized in that, The parameter adjustment strategy in step S5 includes: When the spermatic vein is detected to be in a resting state, the cutoff frequency of the first analog wall filter is controlled to the range of 20Hz-50Hz, and the second digital wall filter is controlled to mainly adopt the high-order IIR mode. Upon detecting the start of the Valsalva action, the cutoff frequency of the first analog wall filter is instantly increased to the 100Hz-200Hz range, and the second digital wall filter is controlled to primarily adopt the linear phase FIR mode.

10. The spermatic vein ultrasound imaging method based on dual-wall filtering and adaptive Doppler signal processing according to claim 8, characterized in that, The extraction of fluid dynamics features in step S3 does not rely on full sampling rate data, but is based on baseband data with a downsampling rate of 1 / 10 to 1 / 100 of the original sampling rate to reduce the computational load of real-time processing.

Images