Medical imaging gating system based on non-contact multi-channel vibration signal compound judgment

Abstract

The application discloses a medical imaging gating system based on non-contact multi-channel vibration signal composite judgment. Combining light projection and image acquisition technology, the mechanical vibration signal of a measured person's heart and the mechanical signal of the carotid artery pulse are synchronously collected through a non-contact method, and a heart beat trigger point is extracted from the mechanical vibration signal. Based on the composite judgment strategy of double signal timing alignment and delay correction, the heart beat trigger points from different mechanical vibration signals are judged for timing consistency. When the time difference of the heart beat trigger points from different mechanical signals in the same heart cycle is within a preset physiological delay window, the average value is taken as the heart beat trigger time of the current heart cycle. Otherwise, the time of the heart beat trigger point of the signal with the best signal-to-noise ratio is selected as the heart beat trigger time. The method can generate a high-precision and high-robustness gating synchronous trigger signal in real time, and drive medical imaging devices such as MRI and CT to realize accurate synchronous acquisition.

Description

Technical Field

[0001] This invention belongs to the field of surgical identification technology, and relates to the synchronization of biomedical engineering and medical imaging gating, specifically to a medical imaging gating system based on non-contact multi-channel vibration signal composite decision. Background Technology

[0002] With the continuous development of medical imaging technology, equipment such as cardiac magnetic resonance imaging (MRI), computed tomography (CT), and ultrasound have played a crucial role in the clinical diagnosis and treatment of cardiovascular diseases. These high-resolution imaging techniques place increasingly higher demands on the accuracy of simultaneous acquisition and motion gating. Because patients inevitably experience physiological activities such as heartbeats, blood flow pulsations, and respiratory movements during examinations, motion artifacts are easily generated, significantly affecting image quality.

[0003] Currently, the most commonly used gating method in clinical practice is synchronous triggering based on electrocardiogram (ECG). ECG can acquire cardiac electrical activity signals with high temporal resolution, providing periodic gating references for imaging equipment. However, in practical applications, this contact-based signal acquisition method of ECG has many drawbacks, such as electromagnetic interference, cumbersome operation, and easy signal loss.

[0004] Some studies have proposed using non-contact mechanical vibration signals as gating signals to address the problem of poor signal quality in electrocardiogram (ECG) and other signal acquisition processes, which is negatively impacted by factors such as adhesion quality, body surface environment, and patient movement, affecting sensor sensitivity and stability. Specifically, cardiac mechanical vibration signals (SCG) can be acquired non-contactly using sensors or laser speckle techniques to capture the mechanical vibrations in the chest caused by heartbeats, offering advantages such as being contactless, electrode-free, and having short delays. Carotid artery pulse mechanical signals can be acquired using laser speckle, mechanical oscillators, or accelerometers, directly reflecting the mechanical response of large artery pulse waves, exhibiting smooth waveforms, distinct peak characteristics, and relatively low susceptibility to local interference. This provides a solution for non-contact medical imaging gating systems.

[0005] However, cardiac mechanical vibration signals are easily affected by body position, respiration, and skin surface condition, resulting in unstable waveforms and limited anti-interference capabilities. Furthermore, carotid pulse mechanical signals, compared to in-situ cardiac events, exhibit a certain physiological delay, limiting their ability to accurately synchronize the cardiac cycle. Therefore, in practical applications, while non-contact signal-based gating systems can improve comfort and reduce electromagnetic shielding requirements, they still suffer from limited anti-interference capabilities, easily leading to misjudgment, loss, or timing abnormalities of gating trigger points, thus affecting the synchronization quality of medical imaging. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention proposes a medical imaging gating system based on non-contact multi-channel vibration signal composite decision-making. It synchronously acquires cardiac mechanical vibration signals and carotid artery mechanical signals through two optical channels. Based on a composite decision-making strategy of dual-signal timing alignment and delay correction, it effectively suppresses single-path signal noise and abnormal interference, and generates high-precision, robust gating synchronization trigger signals in real time to drive MRI, CT and other medical imaging equipment to achieve precise synchronous acquisition.

[0007] In the first aspect, this application provides a medical imaging gating system based on non-contact multi-channel vibration signal composite decision, including a multi-channel image acquisition module, a signal preprocessing module, a cardiac cycle feature extraction module, a composite gating decision module, and a gating signal output module.

[0008] The multi-channel image acquisition module includes a structured light projection module, a single-point laser illumination module, a high-speed imaging module, and a time-base control module. The structured light projection module projects a pattern composed of coded units with spatially encoded information onto the surface of the subject's chest cavity. The single-point laser illumination module illuminates a single point of laser light onto the carotid artery region of the same subject. The time-base control module controls two high-speed industrial cameras in the high-speed imaging module to acquire image sequences of the chest cavity surface and the carotid artery region simultaneously, ensuring that the acquisition times of the two sequences are consistent.

[0009] The signal preprocessing module uses motion estimation methods to collect information on the displacement changes of patterns caused by vibration from the image sequence, which serves as the original waveforms of the cardiac mechanical vibration signal and the carotid pulse mechanical signal.

[0010] The cardiac cycle feature extraction module first extracts the main vibration signal from the original waveform output by the signal preprocessing module, and then filters the main vibration signal to extract the cardiac trigger point from the filtered cardiac mechanical vibration signal and carotid pulse mechanical signal, respectively.

[0011] The composite gating decision module automatically identifies and aligns the cardiac trigger point output from the cardiac cycle feature extraction module through a preset physiological delay window and signal quality adaptive criteria, and outputs a gating synchronization signal.

[0012] The gating signal output module provides an output interface to output the time series composed of the cardiac trigger time t* obtained by the composite gating decision module to the imaging equipment in real time, providing a gating signal reference for medical imaging or monitoring equipment, and realizing a fully automatic, anti-interference, and high-precision cardiac synchronous gating system.

[0013] Secondly, this application provides a method for generating medical imaging gated signals based on non-contact multi-channel vibration signal composite decision, specifically including the following steps:

[0014] Step 1: Project a pattern composed of coded units with spatial coded information onto the surface of the subject's chest cavity, while simultaneously irradiating a single point laser onto the carotid artery region.

[0015] Step 2: Using two cameras, acquire image sequences of the subject's chest cavity surface and carotid artery region at the same clock.

[0016] Step 3: Using motion estimation methods, collect the displacement change information of the pattern from the image sequences of the chest cavity surface and the carotid artery region, respectively, as the original waveforms of the cardiac mechanical vibration signal and the carotid pulse mechanical signal.

[0017] Step 4: Using a filtering algorithm, extract the cardiac trigger points from the filtered cardiac mechanical vibration signal and carotid pulse mechanical signal, respectively.

[0018] Step 5: Set the physiological delay window. This refers to the time t between the cardiac mechanical vibration signal and the carotid pulse mechanical signal's cardiac trigger point within the same cardiac cycle. SCG t CA When the time difference is within the physiological delay window, half of the time difference is compared with the cardiac trigger point t of the cardiac mechanical vibration signal in the next cardiac cycle. SCG The times are added together and used as the cardiac trigger time t* for the next cardiac cycle. Otherwise, the signal-to-noise ratio (SNR) of the cardiac mechanical vibration signal and the carotid pulse mechanical signal are calculated separately, and the time of the cardiac trigger point of the signal with the best SNR is selected as the cardiac trigger time t*.

[0019] Step 6: Use the cardiac trigger time t* generated in Step 5 as the gating signal for medical imaging to control the medical imaging equipment.

[0020] The present invention has the following beneficial effects:

[0021] 1. For the first time, a composite fusion decision based on synchronously acquired cardiac mechanical vibration signals and carotid pulse mechanical signals is proposed, which makes full use of the temporal complementarity between the cardiac mechanical source and the blood flow response of the large artery to achieve high-precision and robust detection of the cardiac cycle trigger point.

[0022] 2. To address the physiological delay differences between dual signals, a timing alignment and window matching algorithm is proposed. A multi-strategy adaptive decision process is designed, including consistency composite, primary-secondary switching, weighted averaging, and anomaly compensation. This enables dynamic delay calibration and accurate output of composite gating trigger points, significantly improving the robustness and reliability of gating trigger signals. It is particularly suitable for high-end medical imaging fields such as MRI and CT, and especially for precise synchronous gating in environments with strong magnetic fields and high electromagnetic interference.

[0023] 3. It fully leverages the complementary advantages of cardiac mechanical vibration signals and carotid pulse mechanical signals, effectively suppressing single-path signal noise and abnormal interference, significantly improving the accuracy of gating triggering and the system's adaptability. The entire solution is completely non-contact and electrode-free, with advantages such as strong anti-electromagnetic interference and high timing accuracy, and can be widely applied in the fields of clinical medical image gating and synchronous detection of vital signs. Attached Figure Description

[0024] Figure 1 This is a block diagram of a medical imaging gating system based on non-contact multi-channel vibration signal composite decision-making.

[0025] Figure 2 Block diagram of the multi-channel image acquisition module;

[0026] Figure 3 A schematic diagram of speckle patterns acquired by a high-speed imaging module in the carotid artery region;

[0027] Figure 4 Example of coding unit localization for a thoracic cavity surface image;

[0028] Figure 5 A static example of a speckle pattern formed by laser reflection in the carotid artery region;

[0029] Figure 6 The original waveforms and filtered waveforms are for cardiac mechanical vibration signals and carotid pulse mechanical signals. Detailed Implementation

[0030] The present invention will be further explained below with reference to the accompanying drawings;

[0031] Example 1

[0032] like Figure 1 As shown, this embodiment provides a medical imaging gating system based on non-contact multi-channel vibration signal composite decision, including a multi-channel image acquisition module, a signal preprocessing module, a cardiac cycle feature extraction module, a composite gating decision module, and a gating signal output module.

[0033] like Figure 2 As shown, the multi-channel image acquisition module includes a structured light projection module, a single-point laser illumination module, a high-speed imaging module, and a time-base control module.

[0034] The structured light projection module projects a pattern with spatially encoded information, such as a dotted or striped pattern, composed of coded units onto the surface of the subject's chest cavity. When the heartbeat causes minute vibrations on the chest cavity surface, this pattern shifts slightly over time. The single-point laser irradiation module irradiates the carotid artery region of the same subject with a single-point laser. When the carotid artery pulsation causes minute vibrations on the skin surface, the laser reflection forms a dynamic speckle pattern. The time-base control module controls two high-speed industrial cameras in the high-speed imaging module to acquire image sequences of the chest cavity surface and the carotid artery region simultaneously, ensuring that the image acquisition time in both sequences is consistent, thereby acquiring synchronized multi-channel mechanical vibration signals in subsequent processes.

[0035] The high-speed industrial cameras have a frame rate of ≥200fps. One camera is positioned directly facing the chest cavity to capture image sequences of the subtle vibrations of the chest wall caused by the heartbeat, based on a structured light projection pattern. The other camera captures image sequences of a single-point laser reflecting light spots or speckles from the carotid artery region, capturing details of the carotid artery's mechanical fluctuations, such as... Figure 3 As shown.

[0036] The signal preprocessing module is used to extract the original waveforms of cardiac mechanical vibration signals and carotid pulse mechanical signals from the image sequences acquired by the multi-channel image acquisition module.

[0037] For an image sequence of the thoracic cavity surface, the coding units in a frame of the image are first located using image processing algorithms, such as... Figure 4 As shown, the movement trajectory of each coding unit on the time axis is then acquired from the image sequence. Signal-to-noise ratio analysis is performed on the movement trajectory of each coding unit, and the unit with the highest motion energy is selected as the primary analysis target. Then, using motion estimation algorithms such as optical flow or template matching, the displacement change sequence on the time axis is extracted from the movement trajectory of the primary analysis target, generating the original waveform of the cardiac mechanical vibration signal, including both horizontal and vertical directions.

[0038] For image sequences of the carotid artery region, such as Figure 5 As shown, firstly, the Region of Interest (ROI) of each frame of the image is selected. Then, optical flow is applied to the ROI to acquire the displacement change signal of the laser irradiation point on the time axis from the image sequence. Finally, energy analysis or periodic evaluation methods are used to select the direction or component with the highest signal-to-noise ratio as the original waveform of the carotid pulse mechanical signal.

[0039] The cardiac cycle feature extraction module is used to extract the cardiac trigger point sequence from the raw waveform output by the signal preprocessing module.

[0040] For the original waveform of the cardiac mechanical vibration signal, the channel with the best signal-to-noise ratio (SNR) is first selected as the main vibration signal through SNR evaluation. Then, the main vibration signal undergoes detrending processing, employing a Butterworth bandpass filter (0.5–5 Hz) to suppress non-target frequency noise and enhance cardiac-related components. Finally, extreme point detection is performed on the filtered signal within a sliding time window to output the cardiac trigger point sequence of the cardiac mechanical vibration signal.

[0041] The original waveform of the carotid pulse mechanical signal is subjected to Butterworth bandpass filtering (0.5–5 Hz), followed by multi-stage processing including detrending, wavelet denoising, and normalization to suppress baseline drift and high-frequency noise. Finally, a sliding window is used for overall signal inversion, and the first derivative extremum method is used to detect the signal periodic peak, outputting the cardiac trigger point sequence of the carotid pulse mechanical signal.

[0042] As an optional embodiment, the bandpass filtered signal is further subjected to wavelet decomposition reconstruction, absolute value conversion, signal envelope analysis and moving average smoothing to improve signal morphological stability and feature clarity.

[0043] The composite gating decision module performs a time-series comparison of the cardiac mechanical vibration signal and the carotid pulse mechanical signal output by the cardiac cycle feature extraction module, and outputs the cardiac trigger time by combining a preset physiological delay window and signal quality adaptive criteria. Specifically:

[0044] Step 1: Initialize the physiological delay window [τ] min ,τ max [80 ms, 200 ms]. For the most recent M valid cardiac cycles, after removing outliers exceeding μ±2σ, a robust mean is calculated. With robust standard deviation Then, the starting position τ of the physiological delay window... min and the final position τ max Perform adaptive updates:

[0045] ;

[0046] ;

[0047] Where μ and σ represent the mean and standard deviation of M valid cardiac cycles, respectively, and M is 15 in this embodiment. k = 1.0 to 1.5 represents the adjustment coefficient. The clip() function is used to limit the value of the variable to a defined range.

[0048] Step 2: In each cardiac cycle, calculate the time difference between the cardiac trigger point of the cardiac mechanical vibration signal and the cardiac trigger point of the carotid pulse mechanical signal. ;when ∈[τ min ,τ max If the timing of the heartbeat trigger points from multiple channels is considered to be consistent, proceed to step 3; otherwise, if the time difference is considered to be excessive, proceed to step 4 to perform a primary / secondary switching or anomaly handling process.

[0049] Step 3, due to the cardiac trigger point time t of the carotid pulse mechanical signal CA There is a certain delay compared to the actual cardiac time. In order to meet the accuracy and real-time requirements of medical imaging equipment for gating signals, half of the time difference calculated from the current cardiac cycle is compared with the cardiac trigger point t of the cardiac mechanical vibration signal in the next cardiac cycle. SCGp The sum of these times is used as the trigger time t* for the next heartbeat cycle. p

[0050]

[0051] Step 4: Define the method for calculating signal quality Q as follows:

[0052] Q=α·z(SNR)+β·z(r)+γ·z(s)

[0053] Where z() represents the normalization function, SNR represents the signal-to-noise ratio, r represents the Pearson correlation coefficient between the signal waveform in the current cardiac cycle and the signal waveforms in the most recent K cardiac cycles, and s represents the selectable interval stability of the signal. α, β, and γ represent proportionality coefficients. In this embodiment, K=5, α=0.5, β=0.4, and γ=0.1.

[0054] Set the signal reliability quality threshold to Q. valid =0.5, the quality threshold for unreliable signals is Q. poor =0.3.

[0055] When the time difference between the cardiac trigger point and the physiological delay window exceeds the Q-factor, if Q exists... SCG Or Q CA More than Q valid If the condition is met, proceed to s4.1 for primary / secondary switching; otherwise, proceed to s4.2 for exception handling.

[0056] s4.1, the time t of the cardiac trigger point of the cardiac mechanical vibration signal within the current cardiac cycle. SCG As the trigger moment for attraction, t*. If in N consecutive... s Within a period, Q SCG <Q poor Then, the time t of the cardiac trigger point of the carotid pulse mechanical signal within the current cardiac cycle is used. CA As the trigger moment for attraction, t*. If in N consecutive... r Within a period, QSCG ≥Q poor Then, the time t for restoring the cardiac trigger point based on the cardiac mechanical vibration signal within the current cardiac cycle is restored. SCG The moment t* is the trigger for falling in love.

[0057] To avoid jitter, set the T in the contact. h No second switching is allowed within a single cycle.

[0058] In this embodiment, N is set s =2、N r =3、T h =3.

[0059] s4.2. Using the mean of the previous cycle, trend extrapolation, or Kalman filtering, predict and compensate for the time interval between two R waves, i.e., the ventricular depolarization interval RR, and output the cardiac trigger time t*.

[0060] The gating signal output module provides an interface with the imaging equipment. The time series composed of the cardiac trigger time t* obtained by the composite gating decision module is output to the imaging equipment in real time, providing gating signal references for medical imaging or monitoring equipment such as MRI, CT, ultrasound, and EEG.

[0061] As a preferred embodiment, a minimum trigger interval T is set. ref =250ms, meaning the time interval between any two heartbeat trigger moments t* should be greater than the minimum trigger interval T. ref This is to suppress repeated triggering within the same cycle.

[0062] Example 2

[0063] This embodiment provides a method for generating medical imaging gated signals based on non-contact multi-channel vibration signal composite decision-making, specifically including the following steps:

[0064] Step 1: Project a pattern composed of coded units with spatial coded information onto the surface of the subject's chest cavity, while simultaneously irradiating a single point laser onto the carotid artery region.

[0065] Step 2: Using two high-speed industrial cameras, acquire image sequences of the subject's chest cavity surface and carotid artery region at the same clock speed.

[0066] Step 3: Using motion estimation methods, such as optical flow or template matching, collect the displacement change information of the pattern from the image sequences of the chest cavity surface and the carotid artery region, respectively, as the original waveforms of the cardiac mechanical vibration signal and the carotid pulse mechanical signal.

[0067] Step 4: Using a filtering algorithm, extract the cardiac trigger points from the cardiac mechanical vibration signal and the carotid pulse mechanical signal, respectively. Figure 6As shown.

[0068] Step 5: Set the physiological delay window. This refers to the time t between the cardiac mechanical vibration signal and the carotid pulse mechanical signal's cardiac trigger point within the same cardiac cycle. SCG t CA When the time difference is within the physiological delay window, half of the time difference is compared with the cardiac trigger point t of the cardiac mechanical vibration signal in the next cardiac cycle. SCG The times are added together and used as the cardiac trigger time t* for the next cardiac cycle. Otherwise, the signal-to-noise ratio (SNR) of the cardiac mechanical vibration signal and the carotid pulse mechanical signal are calculated separately, and the time of the cardiac trigger point of the signal with the best SNR is selected as the cardiac trigger time t*.

[0069] Step 6: Use the cardiac trigger time t* generated in Step 5 as the gating signal for medical imaging to control the medical imaging equipment.

Claims

1. A medical imaging gating system based on non-contact multi-channel vibration signal composite decision, characterized in that: By combining light projection and image acquisition technology, the mechanical vibration signals of the heart and the mechanical signals of the carotid pulse of the subject are acquired simultaneously in a non-contact manner, and the cardiac trigger point is extracted from the mechanical vibration signal. Timing consistency is determined for cardiac trigger points from different mechanical vibration signals; when the time t of the cardiac mechanical vibration signal and the carotid pulse mechanical signal within the same cardiac cycle is consistent... SCG t CA The time difference is within a preset physiological delay window. Half of the time difference is then compared with the cardiac trigger point t of the cardiac mechanical vibration signal in the next cardiac cycle. SCG The times are added together and used as the cardiac trigger time t* for the next cardiac cycle; otherwise, the signal-to-noise ratio of the cardiac mechanical vibration signal and the carotid pulse mechanical signal are calculated separately, and the time of the cardiac trigger point of the signal with the best signal-to-noise ratio is selected as the cardiac trigger time t*.

2. The medical imaging gating system based on non-contact multi-channel vibration signal composite decision as described in claim 1, characterized in that: The medical imaging gating system includes a multi-channel image acquisition module, a signal preprocessing module, a cardiac cycle feature extraction module, a composite gating decision module, and a gating signal output module. The multi-channel image acquisition module is used to project patterns and single-point lasers with spatial coding information onto the chest cavity surface and carotid artery region of the subject, respectively. Then, the camera acquires image sequences of the chest cavity surface and carotid artery region under the same clock control. The signal preprocessing module uses motion estimation methods to collect pixel displacement change information from the image sequence, which serves as the original waveform of the cardiac mechanical vibration signal and the original waveform of the carotid pulse mechanical signal. The cardiac cycle feature extraction module is used to extract cardiac trigger points from the original waveforms of cardiac mechanical vibration signals and carotid pulse mechanical signals. The composite gating decision module automatically identifies and aligns the cardiac trigger point output from the cardiac cycle feature extraction module through a preset physiological delay window and signal quality adaptive criteria, and outputs the cardiac trigger time. The gating signal output module provides an output interface to output the time series composed of cardiac trigger times obtained by the composite gating decision module to the imaging equipment in real time, providing a gating signal reference for medical imaging or monitoring equipment.

3. The medical imaging gating system based on non-contact multi-channel vibration signal composite decision as described in claim 2, characterized in that: For the image sequence of the chest cavity surface, the coding units in a frame of the image are first located by the image processing algorithm. Then, the movement trajectory of each coding unit on the time axis is collected from the image sequence. The signal-to-noise ratio of the movement trajectory of each coding unit is analyzed, and the one with the largest motion energy is selected as the main analysis target. Then, the displacement change sequence on the time axis of the main analysis target is extracted from its movement trajectory by the motion estimation algorithm to generate the original waveform of the cardiac mechanical vibration signal.

4. The medical imaging gating system based on non-contact multi-channel vibration signal composite decision as described in claim 2, characterized in that: For the image sequence of the carotid artery region, the ROI region of each frame is first selected, and then optical flow is applied to the ROI region to collect the displacement change signal of the laser irradiation point on the time axis from the image sequence. Finally, energy analysis or periodic evaluation methods are used to select the direction or component with the highest signal-to-noise ratio as the original waveform of the carotid pulse mechanical signal.

5. The medical imaging gating system based on non-contact multi-channel vibration signal composite decision as described in claim 2, characterized in that: For the original waveform of the cardiac mechanical vibration signal, the channel with the best signal-to-noise ratio is first selected as the main vibration signal through signal-to-noise ratio evaluation. Then, the main vibration signal is detrended by using a Butterworth bandpass filter of 0.5~5Hz to suppress non-target frequency noise and enhance cardiac-related components. Finally, the extreme point detection is performed on the filtered signal within a sliding time window to output the cardiac trigger point sequence of the cardiac mechanical vibration signal. The original waveform of the carotid pulse mechanical signal is subjected to Butterworth bandpass filtering of 0.5~5Hz, followed by multi-stage processing of detrending, wavelet denoising, and normalization to suppress baseline drift and high-frequency noise. Finally, the sliding window and first derivative extremum method are used to detect the peak value of the signal period and output the cardiac trigger point sequence of the carotid pulse mechanical signal.

6. The medical imaging gating system based on non-contact multi-channel vibration signal composite decision as described in claim 5, characterized in that: The bandpass filtered signal is further subjected to wavelet decomposition reconstruction, absolute value conversion, signal envelope analysis and moving average smoothing, and then cardiac trigger point detection is performed.

7. The medical imaging gating system based on non-contact multi-channel vibration signal composite decision as described in claim 1, characterized in that: The timing consistency of cardiac trigger points from different mechanical vibration signals is determined by the following method: initializing the physiological delay window [τ] min ,τ max [80 ms, 200 ms]; For the most recent M valid cardiac cycles, after removing outliers exceeding μ±2σ, a robust mean is calculated. With robust standard deviation Then, the starting position τ of the physiological delay window... min and the final position τ max Perform adaptive updates: ； ； Where μ and σ represent the mean and standard deviation of M valid cardiac cycles, respectively; k represents the adjustment coefficient; the clip() function is used to limit the value of the variable to a defined range; In each cardiac cycle, calculate the time difference between the cardiac trigger point of the cardiac mechanical vibration signal and the cardiac trigger point of the carotid pulse mechanical signal. ;when ∈[τ min ,τ max At that time, it was assumed that the timing of the cardiac trigger points from different mechanical vibration signals was consistent.

8. The medical imaging gating system based on non-contact multi-channel vibration signal composite decision as described in claim 1, characterized in that: Set the signal reliability quality threshold Q valid and unreliable quality threshold Q poor Q valid >Q poor ; When the time difference between the cardiac trigger point and the physiological delay window exceeds the Q-factor, if Q exists... SCG Or Q CA More than Q valid If the condition is met, a primary / secondary switch will be performed; otherwise, exception handling will be performed. The method for calculating signal quality Q is as follows: Q=α·z(SNR)+β·z(r)+γ·z(s); Where z() represents the standardization function, SNR represents the signal-to-noise ratio, r represents the Pearson correlation coefficient between the signal waveform in the current cardiac cycle and the signal waveform in the most recent K cardiac cycles, s represents the selectable interval stability of the signal, and α, β, and γ represent proportionality coefficients.

9. The medical imaging gating system based on non-contact multi-channel vibration signal composite decision as described in claim 8, characterized in that: The primary / secondary switching is: if in N consecutive... s Within a period, Q SCG <Q poor Then, the time t of the cardiac trigger point of the carotid pulse mechanical signal within the current cardiac cycle is used. CA As the trigger moment for attraction, t*; if in consecutive N... r Within a period, Q SCG ≥Q poor Then, the time t of the cardiac trigger point of the cardiac mechanical vibration signal within the current cardiac cycle is used. SCG As the trigger moment of attraction t*; and, in consecutive T... h No second switching is allowed within a single cycle; The anomaly handling is as follows: using the mean of the previous cycle, trend extrapolation, or Kalman filtering, the time interval between two R waves, i.e., the ventricular depolarization interval RR, is predicted and compensated, and the cardiac trigger time t* is output.

10. A method for generating medical imaging gating signals based on non-contact multi-channel vibration signal composite decision, characterized in that: Specifically, the following steps are included: Step 1: Project a pattern composed of coded units with spatial coded information onto the surface of the subject's chest cavity, while simultaneously irradiating a single point of laser onto the carotid artery region; Step 2: Using two cameras, acquire image sequences of the subject's chest cavity surface and carotid artery region at the same clock speed; Step 3: Using motion estimation methods, collect the displacement change information of the pattern from the image sequences of the chest cavity surface and the carotid artery region, respectively, as the original waveforms of the cardiac mechanical vibration signal and the carotid pulse mechanical signal; Step 4: Extract the cardiac trigger points from the waveforms of the cardiac mechanical vibration signal and the carotid pulse mechanical signal, respectively, using a filtering algorithm. Step 5: Set the physiological delay window; the time t between the cardiac mechanical vibration signal and the carotid pulse mechanical signal trigger point within the same cardiac cycle. SCG t CA When the time difference is within the physiological delay window, half of the time difference is compared with the cardiac trigger point t of the cardiac mechanical vibration signal in the next cardiac cycle. SCG The times are added together and used as the cardiac trigger time t* for the next cardiac cycle; otherwise, the signal-to-noise ratio of the cardiac mechanical vibration signal and the carotid pulse mechanical signal are calculated separately, and the time of the cardiac trigger point of the signal with the best signal-to-noise ratio is selected as the cardiac trigger time t*. Step 6: Use the cardiac trigger time t* generated in Step 5 as the gating signal for medical imaging to control the medical imaging equipment.

Images