High-speed blood flow doppler measurement method and system for miniaturized ultrasound catheter probes

Abstract

This invention discloses a high-speed blood flow Doppler measurement method and system for miniaturized ultrasound catheter probes. It employs a high-pulse repetition frequency complementary Golay-coded pulse sequence to replace continuous wave transmission, achieving a distance-dimensional information-discarding equivalent CW effect through extremely high PRF combined with range-gate coherent summation. Pulse compression processing gain is obtained through Golay BPSK encoding excitation. A pulse-by-pulse A / B alternating transmission strategy is used, with each physical pulse independently extracting Doppler samples using a matched filter corresponding to the codeword. Time-division multiplexing of pulse modes enables full-aperture transmission and reception, eliminating TX-RX acoustic leakage. This invention achieves a comprehensive signal-to-noise ratio gain of approximately 31 dB without exceeding the thermal load of traditional CW, restoring clinically usable CW equivalent Doppler spectra on ICE catheters, and can be implemented through firmware upgrades on existing PW Doppler hardware.

Description

Technical Field

[0001] This invention relates to a method and system for high-speed blood flow Doppler measurement using a miniaturized ultrasound catheter probe, which is particularly suitable for the application scenario of real-time spectral Doppler detection of high-speed blood flow using intracardiac ultrasound (ICE) phased array catheters in interventional cardiac surgery, and belongs to the field of medical ultrasound imaging technology. Background Technology

[0002] Doppler ultrasound is a core tool for assessing hemodynamics in clinical cardiac examinations. Based on the transmission method, Doppler ultrasound is mainly divided into two modes: pulsed wave (PW) Doppler and continuous wave (CW) Doppler. PW Doppler can precisely locate blood flow by setting a sampling volume at a specific depth, but it is limited by the nyquist limit of the pulse repetition frequency (PRF). When the blood flow velocity exceeds a certain threshold, frequency aliasing occurs, making accurate measurement impossible. CW Doppler, on the other hand, uses continuous transmission and reception, sacrificing distance resolution for aliasing-free high-speed blood flow measurement capabilities. It is irreplaceable in clinical scenarios assessing high-speed jets such as aortic stenosis, mitral regurgitation, and tricuspid regurgitation.

[0003] Intracardiac ultrasound (ICE) catheters are a rapidly developing interventional cardiac imaging technology in recent years. An ICE catheter encapsulates a miniature phased array probe (typically 64 elements, center frequency 5-10 MHz) within an 8F-10F (approximately 2.7-3.3 mm) catheter tip, which is then inserted into the heart chambers via the femoral vein. It provides real-time two-dimensional imaging, color Doppler, and spectral Doppler functions, and is used to guide interventional procedures such as atrial septal defect / patent foramen ovale closure, atrial fibrillation radiofrequency ablation, and left atrial appendage occlusion. Because the ICE catheter is located directly inside the heart chambers, it offers short imaging distances, high resolution, and does not require general anesthesia (unlike transesophageal echocardiography, TEE), leading to its rapid growth in clinical application in recent years.

[0004] However, ICE catheters face fundamental technical difficulties in achieving CW Doppler functionality, mainly in the following aspects:

[0005] 1. Severely limited transmission voltage In CW mode, the transmit duty cycle is 100% (continuous transmission). However, the heat dissipation capacity of the ICE catheter's miniature package structure is extremely limited. To prevent the temperature rise at the catheter tip from exceeding the safety threshold (typically no more than body temperature +2 degrees Celsius), the CW transmit voltage can only be set at an extremely low level (typically ±5V). In contrast, PW mode, due to its extremely low duty cycle (typically <1%), can use peak voltages of ±45V or even higher. This alone results in a CW transmit signal strength approximately 19dB lower than PW.

[0006] 2. Extremely poor acoustic isolation between transmitter and receiver Traditional CW Doppler probes require simultaneous transmission and reception, typically dividing the array into two subarrays: TX and RX. In conventional cardiac probes, due to their large physical size, the spacing between the TX and RX elements is sufficiently large, and with acoustic isolation design, an isolation level of 40-60 dB can be achieved. However, in ICE catheters, the 64 elements are densely packed within an effective aperture of only about 10 mm. The physical distance between the TX and RX subarrays is extremely close, resulting in severe acoustic and electrical coupling, and the actual isolation level is typically only 25-35 dB. A significant amount of TX leakage signal directly enters the receiving channel, creating an interference noise floor far exceeding the useful Doppler echo signal.

[0007] 3. Sensitivity loss due to half-aperture The CW mode splits the probe array into two, with only half of the array elements used for transmission and reception (e.g., 32+32). Compared to the full-aperture 64-element transmission and reception mode, the transmission sound field intensity and reception sensitivity both decrease by about 6dB, resulting in a total loss of about 12dB.

[0008] 4. Limitations of existing technology In the prior art, coded excitation technology has been widely used in B-mode ultrasound imaging to improve the signal-to-noise ratio (SNR), among which Golay complementary coding has attracted attention due to its theoretically perfect sidelobe cancellation characteristics. US Patent Publication No. 6487433B2 filed by GE discloses a method for using Golay coded excitation in cardiac ultrasound imaging, improving motion robustness through more than two transmissions, but this technique is aimed at B-mode imaging rather than Doppler mode. US Patent Publication No. 5564424A discloses a high PRF pulsed Doppler beamforming method that reduces distance ambiguity spurious signals by separating the TX / RX apertures, but does not involve coded excitation.

[0009] None of the aforementioned existing technologies have solved the fundamental problem of CW Doppler mode failure on miniaturized catheter probes (such as ICE catheters). In actual testing, PW Doppler measurements using an ICE phased array probe on a phantom can obtain very good signals, but the signal is completely invisible after switching to CW mode. If the ICE probe is replaced with a conventional cardiac phased array probe, the CW signal can be detected (although weaker than PW), indicating that the root cause of the problem lies in the special physical constraints of the ICE probe.

[0010] Therefore, how to achieve high-speed blood flow detection function equivalent to CW Doppler through signal processing schemes without changing the hardware design of ICE catheters is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0011] The technical problem to be solved by this invention is that the traditional CW mode results in undetectable signals due to low transmit voltage, poor transmit-receive isolation, and loss of half-aperture sensitivity.

[0012] To address the aforementioned technical problems, the first aspect of the present invention discloses a high-speed blood flow Doppler measurement method for a miniaturized ultrasonic catheter probe, comprising the following steps: Step 1: Based on the target maximum blood flow velocity and probe center frequency Determine the pulse repetition frequency To ensure that the Nyquist velocity is not lower than the target's maximum blood flow velocity, while maintaining the maximum unambiguous distance. Much smaller than the actual detection depth, making The maximum unambiguous distance generated when the Nyquist condition is satisfied is much smaller than the actual detection depth, resulting in multiple layers of distance ambiguity superposition. Step 2, generate a length of A Golay complementary coding pair (A, B) consisting of codeword A and codeword B is encoded on the same carrier to obtain the coded pulse A of codeword A and the coded pulse B of codeword B. The coded pulses of codeword A and codeword B are each composed of... It consists of several chips, each corresponding to one bit in codeword A and codeword B, containing several carrier cycles. The polarity of the chip is determined by the value of the corresponding bit in codeword A and codeword B. The chip polarity indicates whether to transmit a standard carrier or an inverted carrier. Step 3: Use all N array elements of the miniaturized catheter probe to perform beamforming and transmit coded pulse A and coded pulse B; After the transmission is completed, beamforming is performed using all N array elements of the miniaturized duct probe to receive the echo signal; In this invention, within each pulse repetition interval, the coded pulse is first transmitted with full aperture, and then the echo signal is received with full aperture. Step 4: Perform pulse-by-pulse matched filtering on the received echo signal to obtain a set of distance sampling points arranged along the distance dimension for each pulse repetition interval. , For the present Serial number Here is the distance to the sampling point index, where, , is the pulse repetition interval; Step 5: Collect the distance sampling points for each pulse repetition interval obtained in Step 4. Coherent summation is performed over the entire receiving window to obtain complex sample values ​​for each pulse repetition interval, so that the range dimension information is actively integrated and discarded, and the residual side lobes of the range dimension are averaged together, which is equivalent to realizing the full-depth signal integration of CW Doppler. Step 6: Arrange the complex sampled values ​​of the continuous pulse repetition interval in chronological order to form a physical... The Doppler sampling sequence is given by a sampling rate of [sample rate]. After filtering, the Doppler sampling sequence is segmented, windowed, and subjected to an N-point Fast Fourier Transform to obtain the Doppler power spectrum. The frequency axis span of the Doppler power spectrum is [specified value]. ; Step 7: Based on the Doppler power spectrum obtained in the previous step, perform spectrum display and clinical measurement. The Nyquist velocity is... The value is not halved due to the alternation of A and B.

[0013] Preferably, in step 1, the pulse repetition frequency With the target maximum blood flow velocity and the center frequency of the probe Between satisfy ,in, The speed of sound.

[0014] More preferably, the pulse repetition frequency The value range is from 50kHz to 200kHz, so that the maximum unambiguous distance does not exceed 15mm, while the detection depth is from 30mm to 100mm, and the distance ambiguity layer is not less than 3 layers.

[0015] Preferably, in step 2, codeword A and codeword B satisfy a complementary condition. ,in, For delay, and Let A and B be the autocorrelation functions of codewords A and B, respectively. Let be the impulse function.

[0016] More preferably, the length L of the Golay complementary coding pair is any one of 4, 8, 16 or 32, and each chip contains 1 to 4 carrier cycles.

[0017] Preferably, in step 3, when transmitting the encoded pulse A and the encoded pulse B, the encoded pulse A and the encoded pulse B are transmitted alternately, so that the Doppler sampling rate is equal to the physical... .

[0018] More preferably, the transmit peak voltage of full aperture time-division multiplexed transceiver. Based on the encoding duty cycle And the thermal budget constraint is determined to meet ,in, The transmit voltage is the traditional CW mode transmit peak voltage. It is at least 1.5 times the transmit voltage of a conventional CW.

[0019] More preferably, the pulse-by-pulse A / B alternation method ensures that each physical pulse independently contributes to a Doppler sample, and the Doppler sampling rate is equal to the physical pulse. The corresponding Nyquist speed is The FFT packet length M used is an integer between 32 and 512.

[0020] Preferably, for the echo signal received within each pulse repetition interval, step 4 includes the following steps: Step 401: After obtaining the focused receiving radio frequency signal through full aperture receiving beamforming, the baseband I / Q complex signal is obtained through IQ demodulation. Step 402: Select the corresponding matched filter vector as the matched filter coefficient according to the codeword used in the current pulse, perform cross-correlation operation between the baseband I / Q complex signal and the matched filter vector to achieve pulse compression, and obtain the set of distance sampling points y[n,k].

[0021] Preferably, in step 5, the receiving window covers all distance gates corresponding to all distance blur layers.

[0022] More preferably, the range gate coherent summation covers all range blur layers corresponding to the entire receiving window, and integrates and averages the residual sidelobes generated by the separate excitation of the range dimension by the Golay codeword in the summation, so that no further processing is required after the FFT. The complete coding processing gain can be obtained by synthesizing the spectrum of A / B blocks in the form of blocks. .

[0023] Preferably, after step 6 and in step 7, the method further includes: performing time averaging of the Doppler power spectra of multiple adjacent packets.

[0024] More preferably, the miniaturized ultrasound catheter probe is an intracardiac ultrasound (ICE) phased array catheter probe with 32 to 128 array elements, a center frequency of 5 to 10 MHz, and a catheter outer diameter not exceeding 12F.

[0025] The second aspect of the present invention discloses a high-speed blood flow Doppler measurement system for a miniaturized ultrasonic catheter probe, used to implement the above-mentioned high-speed blood flow Doppler measurement method, comprising: The encoded waveform generation and A / B alternation control module is used to generate encoded pulse A and encoded pulse B according to the pre-stored Golay complementary encoding pair (A,B), and to make encoded pulse A and encoded pulse B be transmitted alternately, so as to control the alternating use of codeword A and codeword B at the repetition interval of adjacent pulses according to the pulse-by-pulse A / B alternation strategy. The full-aperture beamforming transmitter module is used to delay and superimpose coded pulses A and B using all the array elements of a miniaturized duct probe to transmit coded pulses in a specified direction. A full-aperture beamforming receiver module is used to receive echo signals and perform receive beamforming using all elements of a miniaturized duct probe during the receive window of each pulse repetition interval. The adaptive matched filtering module is used to automatically switch the corresponding matched filter vector according to the codeword used in the current pulse repetition interval of the received signal after beamforming, and to perform pulse compression processing. For each pulse repetition interval, a set of range sampling points arranged along the range dimension is obtained. ; The range gate summation module is used to coherently sum all the range sampling points of the pulse compression output into a single complex Doppler sample value for each pulse repetition interval, thereby realizing the full-depth signal integration of the equivalent CW. The complex Doppler sample values ​​corresponding to all pulse repetition intervals constitute a continuous Doppler sampling sequence. The wall filtering and FFT module is used to sequentially perform high-pass filtering, windowing, and N-point FFT operations on a continuous Doppler sampling sequence to output the Doppler power spectrum. The spectrum display module is used to display the Doppler power spectrum in real time.

[0026] Preferably, it also includes a multi-packet averaging module for time averaging the Doppler power spectra of multiple adjacent packets.

[0027] The encoded waveform generation and alternation control module and the adaptive matched filter module are implemented in the FPGA or digital signal processor of the existing pulse wave Doppler ultrasound system through firmware upgrades, without the need for hardware modifications to the probe or analog front end.

[0028] More preferably, the high-speed blood flow Doppler measurement system is applied to the intracardiac ultrasound ICE catheter probe, and the peak transmission voltage of the full aperture beamforming transmission module is greater than ±8V and does not exceed ±15V, and the transmission duty cycle is 10% to 30%, so that the average thermal power consumption does not exceed 1.2 times the transmission power consumption of the traditional CW mode.

[0029] More preferably, the encoded waveform generation and alternation control module supports switching between multiple encoding lengths L, including L=4, 8, 16 and 32, and adaptively selects the optimal encoding length and transmit voltage combination according to the current thermal budget state, achieving a balance between SNR gain and thermal safety.

[0030] This invention provides a method and system for achieving equivalent CW Doppler high-speed blood flow measurement on miniaturized ultrasound catheter probes (especially ICE phased array catheters), which has the following advantages compared to existing technologies: (1) Completely eliminate TX-RX acoustic leakage: The full aperture time-division multiplexing of the transmitter and receiver is used, and the transmitter and receiver are strictly separated in time, which fundamentally eliminates the problem of the transmitted signal leakage to the receiver channel caused by the simultaneous operation of TX / RX in CW mode. This is especially critical for miniaturized probes such as ICE conduits, because the physical aperture of such probes is extremely small, and the TX-RX isolation in traditional CW mode is completely insufficient; (2) Doubling of transmit and receive aperture: Traditional CW divides N array elements into N / 2 transmit + N / 2 receive, while this invention uses all N array elements for transmit and receive, increasing the transmit and receive gain by about 6 dB each, for a total of about 12 dB; (3) Significantly increased transmit voltage: The low duty cycle (15-25%) of the pulse mode allows the peak voltage to be increased to more than twice that of the CW mode under the same thermal budget, resulting in an additional signal gain of 6-12dB; (4) Encoding processing gain: Golay BPSK encoding excitation is provided through pulse compression. The processing gain is 9dB when L=8 and 12dB when L=16. Its range sidelobe cancellation characteristic is naturally implied in the coding design. In the range gate summation (CW equivalent) scheme of the present invention, the range dimension is actively integrated and discarded, and the range sidelobe is averaged during the summation process. Therefore, there is no need to perform additional A / B spectrum complementary synthesis in the slow time dimension. (5) Maintain the Nyquist speed of the entire PRF: Use pulse-by-pulse A / B alternation instead of block alternation, each physical pulse independently generates a Doppler sample, and the Doppler sampling rate is equal to the physical PRF. It does not halve and is suitable for high-speed blood flow coverage at ±4m / s or higher in ICE scenarios. (6) Overall signal-to-noise ratio gain exceeds 30 dB: The sum of the above gains can theoretically improve the SNR by 30-31 dB, which is sufficient to restore the "completely undetectable" CW Doppler signal on the ICE probe to a clear clinically usable spectrum; (7) Thermal safety is guaranteed: Under the recommended parameter configuration, the average thermal power consumption of HPRF encoding mode does not exceed or is even lower than that of traditional CW mode, and will not increase the risk of temperature rise at the tip of the conduit; (8) Highly compatible with existing hardware: This invention can be implemented by firmware upgrade on the basis of existing pulse wave Doppler hardware links without the need to modify the probe or analog front end. Only the following are required: (a) add BPSK encoded waveform table and pulse-by-pulse A / B switching logic to the transmitter; (b) insert adaptive matched filter module to the receiver; (c) maintain the original PW Doppler wall filtering, windowing, FFT and averaging process on the slow time link.

[0031] (9) Good clinical equivalence: The output Doppler spectrum is completely consistent with the traditional CW Doppler in terms of format, refresh rate and clinical measurement method, without the need to change the doctor's operating habits and diagnostic process. Attached Figure Description

[0032] Figure 1 This is a schematic diagram of the overall process of the method of the present invention; Figure 2 This is a schematic diagram comparing the transmission and reception modes of traditional CW Doppler and the HPRF+Golay encoded Doppler of the present invention on an ICE probe, wherein (a) is the traditional CW Doppler mode and (b) is the HPRF+Golay encoded mode of the present invention. Figure 3 The diagram shows the autocorrelation characteristics of the Golay complementary coding pair, where (a) represents the autocorrelation of codeword A. (b) represents the autocorrelation of codeword B. (c) is a complementary synthesis ; Figure 4 The following is a schematic diagram comparing the transmitted waveforms, where (a) is the uncoded pulse waveform, (b) is the BPSK encoded waveform of Golay codeword A, and (c) is the BPSK encoded waveform of Golay codeword B. Figure 5 The diagram shows the pulse-by-pulse A / B alternating transmission timing (ABABAB…). In the diagram, (a) shows the pulse-by-pulse A / B alternating transmission timing, and (b) shows the BPSK encoding of the Golay-8 complementary pair (each chip contains two carrier cycles). The gray background in the diagram represents the codeword A pulse, and the shaded background represents the codeword B pulse. Figure 6The following are comparison diagrams of pulse compression effects: (a) shows the matched filter output of individual codewords A and B, and (b) shows the output after Golay complementary synthesis (only used to demonstrate the range sidelobe properties) compared with the sidelobe without coding output. Figure 7 The diagram shows the signal processing pipeline, where dashed boxes represent optional steps, and solid boxes with a gray background represent the core innovative steps of this invention. Figure 8 The simulation results are shown in the Doppler spectrum comparison diagram of the three modes, where (a) is the traditional CW Doppler spectrum, (b) is the pure HPRF Doppler spectrum, and (c) is the HPRF+Golay encoded Doppler spectrum. Figure 9 The simulation results are shown in the time-velocity spectrogram comparison chart for the three modes. Figure 10 A histogram showing the decomposition of SNR gain sources; Figure 11 A comparative analysis chart of thermal budgets; Figure 12 This is a functional block diagram of the system of the present invention, wherein the dashed boxes are optional modules, the solid boxes with a dark gray background are hardware modules, the solid boxes with a light gray background are firmware / FPGA modules, and the solid boxes with a white background are DSP modules. Detailed Implementation

[0033] The present invention will be further illustrated below with reference to specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Furthermore, it should be understood that after reading the teachings of this invention, those skilled in the art can make various alterations or modifications to the invention, and these equivalent forms also fall within the scope defined by the appended claims.

[0034] The first aspect of this invention discloses a high-speed blood flow Doppler measurement method for miniaturized ultrasound catheter probes, such as... Figure 1 As shown, it includes the following steps: Step S1: Parameter Determination. Based on the target maximum blood flow velocity. and probe center frequency Determine the pulse repetition frequency ,make satisfy ,in, The speed is the speed of sound, to ensure that Nyquist's velocity does not fall below the target's maximum blood flow velocity. Meanwhile, The choice makes the maximum unambiguous distance The distance is much smaller than the actual detection depth, resulting in multiple layers of distance ambiguity superposition, which provides the conditions for subsequent distance gate summation to achieve the distance information discarding effect of equivalent CW.

[0035] In one specific implementation of this invention, the parameters of the ICE catheter probe used are: 64 array elements, probe center frequency... The effective aperture is approximately 10 mm. The target clinical scenario is the detection of high-velocity blood flow within the cardiac chambers, mainly covering high-velocity blood flow in the ±4 m / s range (such as the main flow areas of partial aortic regurgitation and mitral / tricuspid regurgitation). For more extreme jet velocities (such as the main jet peak of 5-6 m / s in severe aortic stenosis), identifiable Nyquist folds are allowed.

[0036] Based on the Nyquist condition, the target maximum blood flow velocity is supported. Minimum required for:

[0037] In the formula, For the corresponding target maximum blood flow velocity The maximum Doppler shift caused.

[0038] Pick Corresponding pulse repetition interval At this point, the Nyquist velocity is: It covers the target blood flow velocity range.

[0039] The maximum unambiguous distance is When the detection depth is 50-80mm, the number of ambiguity layers is approximately 5-9, and the Doppler signal of each scatterer layer is in a... The layers are superimposed in the receiving window; subsequently, all range gates are coherently summed, that is, mathematically, the contributions of all blurred layers are merged into a scalar, which is equivalent to the effect of sacrificing range resolution to integrate full-depth echo in CW mode.

[0040] Step S2: Generate Golay complementary encoding pairs. The generated length is... A Golay complementary encoding pair (A, B) consisting of codeword A and codeword B, satisfying the complementarity condition. ,in, For delay, and Let A and B be the autocorrelation functions of codewords A and B, respectively. This is the impulse function. Binary Phase Shift Keying (BPSK) is used to encode the same carrier wave, obtaining the encoded pulses for codeword A and codeword B, such as... Figure 4 As shown. The encoding pulses of codeword A and codeword B are each generated by... The code consists of several chips, each corresponding to one bit in codeword A and codeword B, and contains several carrier cycles. The chip polarity is determined by the value of the corresponding bit in codeword A and codeword B, and the chip polarity indicates whether a standard carrier or an inverted carrier is transmitted. Specifically, the chip polarity is determined by +1 (standard carrier) or -1 (inverted carrier) of the corresponding bit in the Golay codeword.

[0041] In one specific implementation of this invention, a recursive method is used to generate the length. The Golay complementary encoding pair (A, B) can be generated from the seed. , Starting with recursive construction: ,

[0042] In the formula, and These are the first and second codes of codeword A and codeword B, respectively. Bit.

[0043] After three iterations, an 8-bit encoding pair was obtained: Code word A = [+1, +1, +1, -1, +1, +1, -1, +1]; Codeword B = [+1, +1, +1, -1, -1, -1, +1, -1].

[0044] Encoding the same carrier using binary phase shift keying (BPSK) yields the encoded pulses for codeword A and codeword B: when the chip polarity is +1, a standard carrier is transmitted. When the chip polarity is -1, transmit an inverted carrier wave. This is equivalent to a 180° phase flip. Each chip in the coded pulse contains two carrier cycles, and the chip duration is... The total typing time was Launch duty cycle .

[0045] Verifying complementarity: Calculation ,exist value In all The value is zero, such as Figure 3 As shown, the encoding pair satisfies the perfect complementarity condition.

[0046] Step S3: Full-aperture time-division multiplexing for transmission and reception. Beamforming is performed using all N array elements of the probe (e.g., N=64) to transmit coded pulses (the codeword selection for the coded pulses used during transmission is described in Step S4). After transmission, beamforming is performed using all N array elements to receive the echo signal. Because pulse mode is used instead of continuous wave mode, transmission and reception are strictly separated in time, eliminating the simultaneous operation of TX-RX and fundamentally eliminating the acoustic leakage problem between transmission and reception. Figure 2 The comparison demonstrates the difference between the half-aperture transceiver mode of traditional CW Doppler and the full-aperture time-division multiplexing transceiver mode of HPRF+Golay encoded Doppler of this invention. An ICE conduit probe (64 elements) is used. "1" indicates continuous TX, "2" indicates continuous RX, "3" indicates TX subarray (32 elements), and "4" indicates RX subarray (32 elements). Figure 2 The arrow in (a) indicates that TX leakage has occurred. "5" represents the TX pulse (full aperture), "6" represents the RX reception (full aperture), and "7" represents all 64 array elements (TX first, then RX, time division multiplexing). Figure 2 (b) in the diagram indicates no TX leakage. Figure 2 The arrow in (b) illustrates time separation: In traditional CW mode, the N array elements of the probe are divided into N / 2 transmitting elements and N / 2 receiving elements, which operate simultaneously, resulting in severe acoustic leakage between transmission and reception. In the present invention, all N array elements first transmit coded pulses with full aperture, and then receive the echo signal with full aperture after transmission. Transmission and reception are strictly separated in time, eliminating acoustic leakage and doubling both the transmitting and receiving apertures. Therefore, compared to the traditional CW Doppler half-aperture scheme that divides the N array elements into N / 2 transmitting + N / 2 receiving, the present invention utilizes all N array elements for both transmission and reception, increasing the transmission and reception gain by approximately 6dB each, for a total increase of approximately 12dB.

[0047] In one embodiment of the present invention, a specific implementation method is as follows: Transmit voltage determination: The voltage in the traditional CW mode is ±5V, the duty cycle is 100%, and the normalized average power is 5V. 2 ×1.0=25. In this embodiment of the invention, the duty cycle is 17.1%. If the peak voltage is ±10V, the normalized average power is 10. 2 ×0.171≈17.1, lower than CW mode, thermal safety is fully met. If the thermal budget allows, it can be further increased to ±12V (average power = 144×0.171≈24.6, still lower than 25 in CW mode) to obtain higher signal gain.

[0048] Taking a ±10V transmit voltage as an example, the signal-to-noise ratio gains compared to the traditional CW mode (±5V, 32-element half-aperture, 30dB isolation) are shown in the table below:

[0049] As shown in the table above, an overall SNR improvement of approximately 31 dB means a signal strength increase of more than 1,000 times, which is sufficient to restore a "completely undetectable" signal to a clinically usable high-quality spectrum.

[0050] Step S4: Pulse-by-pulse A / B alternating transmission. Adjacent pulses are alternately transmitted using two codewords from the Golay complement pair. Thus, all physical pulses participate in Doppler sampling, and the Doppler sampling rate is equal to the physical pulse rate. The reduction will not be halved.

[0051] In one embodiment of the present invention, a specific implementation method is as follows: odd-numbered pulses transmit codeword A, and even-numbered pulses transmit codeword B, alternating sequentially, with the specific timing sequence being: | ABABABABAB … |, as shown. Figure 5 As shown in the figure, "8" represents PRI=1 / PRF, and "9" represents "10" indicates the receiving window, "11" indicates "12" means , Figure 5 The time-corresponding matched filtering, as shown in (a) in the figure, is represented as conj(A) conj(B) conj(A) conj(B) conj(A) conj(B) ..., and the corresponding Doppler sampling is represented as... , Figure 5 (a) in the diagram indicates that all samples enter the same Doppler sequence. FFT (Doppler sampling rate = physical PRF, Nyquist sampling rate not halved), such as Figure 5 As shown in (b), the codeword sequence A can be represented as: A = [+1, +1, +1, -1, +1, +1, -1, +1], and the codeword sequence B can be represented as: B = [+1, +1, +1, -1, -1, -1, +1, -1]. The complementarity is: .

[0052] Step S5: Received signal pulse-by-pulse matched filtering (pulse compression). For each received signal... Echo signal within (pulse repetition interval): First, a focused radio frequency signal is obtained by full-aperture receiving beamforming; The baseband I / Q complex signal is then obtained through IQ demodulation. Then, according to the codeword used in the current pulse (codeword A or codeword B), the corresponding matched filter vector is selected—that is, the time-reversed conjugate of codeword A or codeword B, conj(A) or conj(B)—as the matched filter coefficient. The baseband I / Q complex signal is cross-correlated with the matched filter vector (equivalent to a sliding inner product operation) to achieve pulse compression. The output is a set of distance sampling points arranged along the distance dimension (fast time axis). ,in, For the present Serial number The compression effect is as follows: (Indexed by distance from sampling points) Figure 6 As shown.

[0053] Step S6: Obtain equivalent CW Doppler sampling by full-depth integration using a range gate. In this invention, the range gate refers to the spatial depth unit corresponding to each range sampling point in the pulse compression output; the range ambiguity layer refers to the layer formed by the range gate and the distance ambiguity layer. The extremely high depth makes the maximum unambiguous distance much smaller than the actual detection depth, causing echo signals from different depths to overlap in the same receiving window, resulting in the superposition of multiple depth signals (i.e., ambiguity).

[0054] The current output of step S5 All distance sampling points A coherent summation (i.e., full-depth integration) is performed over the entire receiving window (covering all range gates corresponding to all range blur layers) to obtain a complex sample value. As the current number indivual Doppler sampling. This full-depth integration operation actively discards range-resolved information and is mathematically equivalent to integrating the full-depth scatterer echo in CW mode. It should be noted that the sidelobe cancellation property of the Golay complementary pair essentially acts on the range dimension (i.e., (This is sidelobe cancellation in the fast time dimension), and since the distance dimension information has been integrated into a scalar in this step, there is no need to perform additional A / B complementary synthesis in the slow-time dimension.

[0055] In this embodiment of the invention, each pulse independently extracts a Doppler sample (using a matched filter corresponding to the codeword), and all samples constitute a physical... A rate-continuous Doppler time series. The frequency axis span of subsequent FFT processing is... Corresponding to Nyquist speed .exist , Below, the Nyquist velocity is approximately 4.11 m / s, without any reduction.

[0056] Step S7: Wall filtering and Doppler spectrum estimation. Since each Each step S6 generates a Doppler sample value. ,continuous The Doppler sample values ​​are arranged in chronological order, thus forming a physical... Doppler sampling sequence for sampling rate (Doppler sampling rate = physical) (The power spectrum is not halved due to A / B alternation). A high-pass wall filter is applied to the Doppler sampling sequence to remove low-frequency noise from tissue motion and the tube wall; then, the filtered sequence is segmented and windowed with a sliding packet of size M (e.g., M=128) and subjected to an N-point fast Fourier transform (FFT) to obtain the Doppler power spectrum.

[0057] In one embodiment of the invention, a specific implementation is as follows: Doppler packet length M and spectral update rate: M = 128 consecutive pulses are taken as one FFT packet, with a packet length of M·PRI = 128 × 12.5 μs = 1.6 ms, corresponding to a spectral update rate of approximately 625 Hz, which is sufficient for real-time clinical display. Verification of maximum blood flow displacement within the packet: 5 m / s blood flow displaces 5 × 1.6 × 10⁻⁶ pulses within 1.6 ms. -3 =8μm, much smaller than the length of ultrasound waves With respect to sampling volume size, the quasi-static assumption of slow-time correlation holds true.

[0058] The sidelobe cancellation property of Golay complementary pairs is mathematically expressed as follows: Its operating dimension is the range dimension (fast time), used to eliminate the range sidelobes of codewords A and B respectively. In the equivalent CW extraction scheme of this invention, the range gate signal of each pulse is summed and integrated into a scalar, and the range dimension information is actively discarded (equivalent CW effect), that is, the range sidelobes have been averaged onto the DC component during the summation process, and no longer need to be canceled by additional slow-time A / B synthesis. Pulse-by-pulse A / B alternation is the optimal structure in the equivalent CW mode: it obtains the complete range of each pulse. It achieves pulse compression gain while maintaining the Nyquist speed and highest spectral update rate of the full PRF.

[0059] Step S8: Multi-packet averaging (optional). A time-averaged Doppler power spectrum of multiple adjacent packets is performed to reduce estimation variance and improve the smoothness of the spectrum display.

[0060] Step S9: Spectrum Display and Clinical Measurement. The final Doppler power spectrum is displayed in real time as a time-velocity spectrogram. The frequency axis span of the N-point FFT is... The corresponding Nyquist speed is Maintain full It does not reduce the signal by half. The display method is completely consistent with the traditional CW Doppler spectrum, and it can perform standard clinical analyses such as peak velocity measurement and pressure step calculation.

[0061] In embodiments of the present invention, such as Figure 7 As shown, a complete received signal processing pipeline may specifically include the following steps: Step R1 – Full-aperture receiving beamforming: The echo signals from all 64 array elements are delayed and summed to form a focused receiving radio frequency signal (i.e., the beamformed radio frequency signal). The beamforming direction is consistent with the transmitting beam direction.

[0062] Step R2 – IQ Demodulation: Demodulate the focused received RF signal output from step R1 into a baseband I / Q complex signal.

[0063] Step R3 – Adaptive Matched Filtering (Pulse Compression): Based on the current... Code words used (odd number) Using codeword A, even number Using codeword B, select the corresponding matched filter vector—either the time-reversed conjugate of codeword A (conj(A)) or the time-reversed conjugate of codeword B (conj(B))—to construct a matched filter. Perform a cross-correlation operation (equivalent to a sliding inner product operation) between the baseband I / Q complex signal output from step R2 and this matched filter vector to achieve pulse compression. The length of the matched filter is... Number of sampling points corresponding to each chip ( (Approximately 16 sampling points). The output of pulse compression is a set of distance sampling points arranged along the distance dimension, denoted as... ,in, For the present Serial number This is the index for the distance sampling point.

[0064] Step R4 – Distance gate coherent summation (full depth integration): The current... indivual All distance sampling points of the pulse compression output in the inner step R3 Coherent summation results in a single complex value, i.e. ,in For the first indivual The Doppler sample value. This is equivalent to the full-depth integration of all scatterer signals along the entire beam path in CW mode, while also integrating and averaging the residual sidelobes in the range dimension.

[0065] Step R5 – Wall Filtering: The complex Doppler sampling sequence continuously output according to the physical PRF is processed. ( Apply a high-pass IIR or FIR wall filter with a cutoff frequency set at 100-300 Hz (corresponding to a blood flow velocity of approximately 1-3 cm / s) to remove low-frequency noise from vessel wall and tissue motion. After wall filtering, take M consecutive samples (e.g., M=128) to form a processing package. Then, it is sent to the subsequent FFT processing.

[0066] Step R6 – Windowing and FFT: Using M consecutive samples (M typically 128) as a packet, apply a Hanning window function, then perform an N-point Fast Fourier Transform (FFT, N ≥ M, typically N=256) to obtain the Doppler spectrum. ,in, Let be a window function. The Doppler power spectrum is... The frequency axis span is Maintain full The Doppler bandwidth is not halved.

[0067] Step R7 – Multi-packet averaging (optional): Perform time averaging of the power spectra of multiple adjacent packets. ,in, The number of packages participating in the average, For the index of the package ( ), For the first The Doppler power spectrum of each packet is calculated in step R6. Time averaging is used to reduce the variance of the spectrum estimation and improve the smoothness of the display. Since it has been demonstrated that the range dimension sidelobes have already been integrated and averaged in R4, it is not necessary to perform time averaging here. A / B block composition in the form of A / B.

[0068] Step R8 – Spectrum Display: Convert the frequency axis to the velocity axis ( It is displayed in real time in a standard time-speed scrolling spectrogram format.

[0069] In one specific implementation of this invention, due to the use of pulse transmission mode, the transmission duty cycle is... (in, (This refers to chip duration), typically 15-25%, far lower than CW's 100%. Therefore, under the constraint of the same average power consumption (i.e., the same thermal load), the pulse peak voltage... It can significantly improve and satisfy relationships ,in, This refers to the transmit voltage in CW mode. For example, when... When =17%, Can be increased to By increasing the voltage from ±5V in CW to ±10V or even higher in HPRF coding mode, a single phase can achieve a signal gain of 6-12dB.

[0070] To verify the effectiveness of the above technical solution, a complete system simulation model was built using MATLAB to simulate the Doppler detection performance under the following three modes with an ICE probe: (1) Conventional CW Doppler (±5V, 32+32 element half-aperture, TX-RX isolation 30 dB); (2) Pure HPRF Doppler (±10V, 64 elements, PRF = 80 kHz, no coding); (3) HPRF+Golay-8 encoded Doppler (±10V, full 64 elements, PRF=80kHz, pulse-by-pulse ABAB alternation).

[0071] The simulation included six blood flow scattering targets covering different depths (15–55 mm) and velocities (0.8–4.8 m / s, including forward and reverse directions), as well as three strong tissue clutter sources (amplitude 10–15 times that of the blood flow signal). Processing parameters: M = 128 pulses per packet, N_FFT = 256, Hanning windowing, and second-order Butterworth high-pass wall filter (cutoff normalization frequency 200 / (PRF / 2)).

[0072] Simulation results are as follows Figure 8 and Figure 9 As shown. In traditional CW mode, due to the extremely high TX leakage noise floor, almost no blood flow signal peaks are visible on the Doppler spectrum, consistent with the phenomenon of invisible CW signals in actual ICE probe tests. In pure HPRF mode, the signal can be detected, but the signal-to-noise ratio is limited. In HPRF+Golay encoding mode, clear Doppler spectral peaks can be observed at each target velocity, and the spectral purity and signal-to-noise ratio are significantly better than the previous two modes. The measured value of SNR gain matches the theoretical prediction (approximately 31dB) well. Since all three modes use the same physical PRF (or the slow time sampling rate of equivalent CW), a direct comparison on the same velocity axis yields clear conclusions.

[0073] like Figure 10 As shown, the sources of SNR gain of the HPRF+Golay encoded Doppler scheme of this invention compared to the traditional CW mode are decomposed. The figure shows the contributions of transmit voltage improvement (+6dB), transmit full aperture gain (+6dB), receive full aperture gain (+6dB), Golay pulse compression gain (+9dB), and elimination of TX leakage noise floor (+4dB) in the form of bar charts, with a total comprehensive SNR improvement of approximately +31dB, which is consistent with the simulation and measurement results.

[0074] like Figure 11 As shown, a comparative analysis of the thermal budgets of the traditional CW mode and the HPRF encoding mode of this invention under the same ICE catheter probe conditions was conducted. The traditional CW mode has a transmit voltage of ±5V, a duty cycle of 100%, and a normalized average power of 25. This invention uses a transmit voltage of ±10V, a duty cycle of approximately 17.1%, and a normalized average power of approximately 17.1, which is lower than the traditional CW mode. Even if the transmit voltage is further increased to ±12V, the normalized average power is approximately 24.6, still lower than the thermal load of the traditional CW mode. The comparison results in Figure 11 show that this invention achieves significant SNR gain while fully meeting thermal safety requirements.

[0075] The engineering implementation of this invention on the CM-70 ultrasound diagnostic platform can reuse most of the modules of the existing PW Doppler processing link. The main modifications include: (1) Transmitter: Add a Golay BPSK encoded waveform table and pulse-by-pulse A / B switching logic to the existing PW transmitter controller (FPGA implementation). If the existing transmitter supports cycle-by-cycle polarity control (most ultrasonic transmitters support this), only the waveform table content and A / B switching logic need to be modified, without any hardware changes.

[0076] (2) Receiver: After IQ demodulation and before Doppler processing, an adaptive matched filter is inserted to select the matched filter vector of the A or B codeword based on the current PRI parity. The filter coefficient length is only the number of sampling points corresponding to L chips (about 16 sampling points when L=8), and the computational load is extremely small, which can be implemented in the existing digital signal processing link.

[0077] (3) Doppler processing: The existing PW Doppler wall filtering, FFT and spectrum display modules are fully reused. Only the PRF parameter needs to be modified to the HPRF value (e.g., 80 kHz). No additional A / B Block spectrum synthesis step is required.

[0078] (4) User interface: Add "Encoded CW" or "HPRF-CW" options to the CW mode selection menu. The rest of the interaction methods are completely consistent with the traditional CW.

[0079] The core parameters of the solutions disclosed in the embodiments of this invention can be flexibly adjusted according to different application requirements: (1) Encoding length L: 4, 8, 16, 32, etc. can be selected. Longer codes provide higher encoding gain, but the duty cycle increases accordingly, which needs to be balanced under the hot budget constraint. For ICE applications, L=8 or L=16 is recommended.

[0080] (2) PRF: can be adjusted in the range of 50–200 kHz. A higher PRF provides a larger Nyquist velocity range, but increases the number of range blur layers (not a disadvantage for this invention, since all range gates are summed and integrated in R4).

[0081] (3) Packet length M: can be adjusted in the range of 32–512. A larger M provides higher velocity resolution, but the spectrum update rate is reduced.

[0082] (4) Encoding method extension: In addition to BPSK, QPSK (quad-phase encoding) can be used to further improve encoding flexibility, or Barker-Golay concatenated encoding can be used to obtain a longer equivalent code length and higher processing gain.

[0083] (5) Probe type expansion: Although the present invention is mainly applied to ICE catheters, it is also applicable to other miniaturized ultrasound probes (such as transesophageal ultrasound probes, intravascular ultrasound probes, transcranial Doppler probes, etc.) facing similar CW performance deficiencies in high-speed blood flow detection.

[0084] like Figure 12 As shown, a second aspect of this invention discloses a high-speed blood flow Doppler measurement system for a miniaturized ultrasound catheter probe, used to implement the above-mentioned high-speed blood flow Doppler measurement method, comprising: Encoded waveform generation and A / B alternation control module: Based on the pre-stored Golay complementary code pair (A,B) and configuration parameters, it generates BPSK modulated encoded pulses and controls the current waveform according to the pulse-by-pulse A / B alternation strategy (ABABAB…). The codewords used. This module can be implemented in the FPGA or transmitter controller of an existing ultrasound system.

[0085] Full-aperture beamforming transmitter module: Utilizing all elements of the ICE conduit probe, the encoded pulses are delayed and superimposed (beamforming) to transmit the encoded pulses in a specified direction. The transmit voltage is determined based on thermal budget constraints and the encoding duty cycle.

[0086] Full aperture beamforming receiver module: in each During the receiving window, all array elements are used to receive echo signals and perform receiving beamforming.

[0087] Adaptive matched filtering module: For the received signal after beamforming, based on the current... The codeword used (codeword A or codeword B) automatically switches to the corresponding matched filter vector for pulse compression processing. Obtain a set of distance sampling points arranged along the distance dimension (fast time axis). .

[0088] Distance gate summation module: For each The pulse compression output coherently sums all distance sampling points into a single complex Doppler sample value, achieving full-depth signal integration of the equivalent CW. The corresponding complex Doppler sample values ​​constitute a continuous Doppler sample sequence.

[0089] The wall filtering and FFT module sequentially performs high-pass filtering, windowing, and N-point FFT on the continuous Doppler sampling sequence, outputting the Doppler power spectrum. The frequency axis span of the FFT is... Maintain full Doppler bandwidth.

[0090] Multi-packet averaging module (optional): Performs time averaging of the Doppler power spectra of multiple adjacent packets.

[0091] Spectrum display module: Displays the Doppler power spectrum in real time in a clinically standard time-velocity spectrogram format.

[0092] The embodiments of the present invention have been described above. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of this disclosure. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. The scope of the present invention is defined by the appended claims and their equivalents. Various substitutions and modifications can be made by those skilled in the art without departing from the scope of the present invention, and all such substitutions and modifications should fall within the scope of this disclosure.

Claims

1. A method for high-speed blood flow Doppler measurement of a miniaturized ultrasonic catheter probe, characterized in that, Includes the following steps: Step 1: Based on the target maximum blood flow velocity and probe center frequency Determine the pulse repetition frequency To ensure that the Nyquist velocity is not lower than the target's maximum blood flow velocity, while maintaining the maximum unambiguous distance. Much smaller than the actual detection depth; Step 2, generate a length of A Golay complementary coding pair (A, B) consisting of codeword A and codeword B is encoded on the same carrier to obtain the coded pulse A of codeword A and the coded pulse B of codeword B. The coded pulses of codeword A and codeword B are each composed of... It consists of several chips, each corresponding to one bit in codeword A and codeword B, containing several carrier cycles. The polarity of the chip is determined by the value of the corresponding bit in codeword A and codeword B. The chip polarity indicates whether to transmit a standard carrier or an inverted carrier. Step 3: Use all N array elements of the miniaturized catheter probe to perform beamforming and transmit coded pulse A and coded pulse B; After the transmission is completed, beamforming is performed using all N array elements of the miniaturized duct probe to receive the echo signal; Step 4: Perform pulse-by-pulse matched filtering on the received echo signal to obtain a set of distance sampling points arranged along the distance dimension for each pulse repetition interval. , For the present Serial number Here is the distance to the sampling point index, where, , is the pulse repetition interval; Step 5: Collect the distance sampling points for each pulse repetition interval obtained in Step 4. Coherent summation is performed over the entire receiving window to obtain the complex sample value for each pulse repetition interval; Step 6: Arrange the complex sampled values ​​of the continuous pulse repetition interval in chronological order to form a physical... The Doppler sampling sequence is given as the sampling rate. After filtering the Doppler sampling sequence, it is segmented, windowed, and subjected to N-point fast Fourier transform to obtain the Doppler power spectrum. Step 7: Perform spectrum display and clinical measurement based on the Doppler power spectrum obtained in the previous step.

2. The high-speed blood flow Doppler measurement method for miniaturized ultrasonic catheter probes as described in claim 1, characterized in that, In step 1, the pulse repetition frequency With the target maximum blood flow velocity and the center frequency of the probe Between satisfy ,in, The speed of sound.

3. The high-speed blood flow Doppler measurement method for miniaturized ultrasonic catheter probes as described in claim 1, characterized in that, In step 2, codeword A and codeword B satisfy the complementary condition. ,in, For delay, and Let A and B be the autocorrelation functions of codeword A and codeword B, respectively. Let be the impulse function.

4. The high-speed blood flow Doppler measurement method for miniaturized ultrasonic catheter probes as described in claim 1, characterized in that, In step 3, when transmitting the encoded pulse A and the encoded pulse B, the encoded pulse A and the encoded pulse B are transmitted alternately.

5. The high-speed blood flow Doppler measurement method for a miniaturized ultrasonic catheter probe as described in claim 1, characterized in that, For the echo signal received within each pulse repetition interval, step 4 includes the following steps: Step 401: After obtaining the focused receiving radio frequency signal through full aperture receiving beamforming, the baseband I / Q complex signal is obtained through IQ demodulation. Step 402: Select the corresponding matched filter vector as the matched filter coefficient according to the codeword used in the current pulse, perform cross-correlation operation between the baseband I / Q complex signal and the matched filter vector to achieve pulse compression, and obtain the set of distance sampling points y[n,k].

6. The high-speed blood flow Doppler measurement method for a miniaturized ultrasonic catheter probe as described in claim 1, characterized in that, In step 5, the receiving window covers all distance gates corresponding to all distance blur layers.

7. The high-speed blood flow Doppler measurement method for a miniaturized ultrasonic catheter probe as described in claim 1, characterized in that, After step 6 and in step 7, the process further includes: performing time averaging of the Doppler power spectra of multiple adjacent packets.

8. A high-speed blood flow Doppler measurement system for miniaturized ultrasonic catheter probes, characterized in that, The method for implementing the high-speed blood flow Doppler measurement method according to any one of claims 1 to 7 includes: The encoded waveform generation and A / B alternation control module is used to generate encoded pulse A and encoded pulse B according to the pre-stored Golay complementary encoding pair (A,B), and to make encoded pulse A and encoded pulse B be transmitted alternately, so as to control the alternating use of codeword A and codeword B at the repetition interval of adjacent pulses according to the pulse-by-pulse A / B alternation strategy. The full-aperture beamforming transmitter module is used to delay and superimpose coded pulses A and B using all the array elements of a miniaturized duct probe to transmit coded pulses in a specified direction. A full-aperture beamforming receiver module is used to receive echo signals and perform receive beamforming using all elements of a miniaturized duct probe during the receive window of each pulse repetition interval. The adaptive matched filtering module is used to automatically switch the corresponding matched filter vector according to the codeword used in the current pulse repetition interval of the received signal after beamforming, and to perform pulse compression processing. For each pulse repetition interval, a set of range sampling points arranged along the range dimension is obtained. ; The range gate summation module is used to coherently sum all the range sampling points of the pulse compression output into a single complex Doppler sample value for each pulse repetition interval, thereby realizing the full-depth signal integration of the equivalent CW. The complex Doppler sample values ​​corresponding to all pulse repetition intervals constitute a continuous Doppler sampling sequence. The wall filtering and FFT module is used to sequentially perform high-pass filtering, windowing, and N-point FFT operations on a continuous Doppler sampling sequence to output the Doppler power spectrum. The spectrum display module is used to display the Doppler power spectrum in real time.

9. A high-speed blood flow Doppler measurement system for a miniaturized ultrasonic catheter probe as described in claim 8, characterized in that, It also includes a multi-packet averaging module, which is used to perform time averaging of the Doppler power spectrum of multiple adjacent packets.

Images