Dual-angle photoacoustic sensing system

Abstract

The application discloses a dual-angle photoacoustic sensing system, comprising a laser for irradiating a measured tissue, characterized in that further comprising: two ultrasonic probes, wherein one ultrasonic probe detects a photoacoustic signal PA1 from directly above the measured tissue, and the other ultrasonic probe detects a photoacoustic signal PA2 from the side of the measured tissue; a proportional coefficient calculation module for calculating the signal proportional coefficient of the photoacoustic signal PA1 and the photoacoustic signal PA2; a depth prediction module: a function relationship between the signal proportional coefficient and the depth information is established in the depth prediction module, and after the real-time obtained signal proportional coefficient is input into the depth prediction module, the corresponding depth information is output by the depth prediction module based on the function relationship. The application places one ultrasonic sensor directly above the measured tissue and on the side of the measured tissue respectively, and proposes a matching signal processing method to extract the characteristic information of the measured tissue, so that more clinical useful information can be extracted under the premise of basically not increasing the cost.

Description

Technical Field

[0001] This invention relates to a photoacoustic sensing system, specifically a photoacoustic sensing system in the depth direction. Background Technology

[0002] Photoacoustic effect is a physical process in which nanosecond pulsed laser light is irradiated onto biological tissue. The tissue absorbs some of the laser energy, resulting in a momentary increase in temperature and thermal expansion and contraction, which further excites pulsed ultrasound waves. By selecting different wavelengths of laser light to irradiate the tissue, the light absorption characteristics of different types of biological tissues can be distinguished. Furthermore, by scanning with an ultrasound probe, two-dimensional imaging results of the light absorption distribution can be obtained.

[0003] However, due to limitations such as a limited field of view, the sensing and imaging effects in the depth direction are often insufficient. For example, when multiple blood vessels overlap in the longitudinal direction, detection from directly above often only reveals the most superficial vessel. Similarly, when imaging melanoma, detection from directly above typically only obtains surface signals and cannot reflect more complex structural information in the depth direction, such as the depth of invasion. While higher-quality photoacoustic 3D images can be obtained through precise 2D scanning or the application of high-density ultrasound array sensors, this also introduces problems such as excessively long overall detection times or high system costs. Summary of the Invention

[0004] The technical problem this invention aims to solve is that traditional photoacoustic sensing methods based on a single ultrasonic probe cannot effectively perform quantitative analysis and evaluation of the depth direction because there is no reference signal.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is to provide a dual-angle photoacoustic sensing system, including a laser for irradiating the tissue under test, characterized in that it further includes:

[0006] Two ultrasound probes are used. One ultrasound probe detects the photoacoustic signal PA1 from directly above the tissue being tested, and the other ultrasound probe detects the photoacoustic signal PA2 from the side of the tissue being tested.

[0007] The proportionality coefficient calculation module is used to calculate the signal proportionality coefficient between photoacoustic signal PA1 and photoacoustic signal PA2.

[0008] Depth Prediction Module: The depth prediction module establishes a functional relationship between the signal scaling factor and the depth information. After the real-time signal scaling factor is input into the depth prediction module, the depth prediction module outputs the corresponding depth information based on the functional relationship.

[0009] Preferably, it further includes a distance adjustment mechanism for adjusting the distance between the two ultrasonic probes, and making the distance between the two ultrasonic probes continuously adjustable.

[0010] Preferably, the signal scaling factor is a time-domain scaling factor, or a frequency-domain scaling factor, or a combination of both, wherein:

[0011] The time-domain scaling factor is calculated using the following method:

[0012] The effective time-domain signals of photoacoustic signals PA1 and PA2 are extracted respectively;

[0013] Peak detection was performed on the truncated photoacoustic signals PA1 and PA2 respectively to obtain the time-domain maximum values ​​of photoacoustic signals PA1 and PA2, peak(PA1) and peak(PA2).

[0014] The time-domain scaling factor is obtained by performing scaling operations on the time-domain maximum values ​​peak(PA1) and peak(PA2);

[0015] The frequency domain scaling factor is calculated using the following method:

[0016] The effective signals of photoacoustic signals PA1 and PA2 are extracted respectively;

[0017] Fourier transforms were performed on the truncated photoacoustic signals PA1 and PA2 respectively to obtain their information in the frequency domain;

[0018] In the frequency domain, the frequency information of photoacoustic signals PA1 and PA2 is extracted and then scaled to obtain the frequency domain scaling factor.

[0019] Preferably, the frequency information is the amplitude at the peak frequency point or the amplitude at any frequency point.

[0020] Preferably, when the signal scaling factor is a time-domain scaling factor, a functional relationship between the time-domain scaling factor and the depth information is established in the depth prediction module, wherein the time-domain scaling factor is the independent variable and the depth information is the dependent variable. After the real-time obtained time-domain scaling factor is input into the depth prediction module, the depth prediction module outputs the corresponding depth information based on the functional relationship.

[0021] When the signal scaling factor is a frequency domain scaling factor, a functional relationship between the frequency domain scaling factor and the depth information is established in the depth prediction module. The frequency domain scaling factor is the independent variable and the depth information is the dependent variable. After the frequency domain scaling factor obtained in real time is input into the depth prediction module, the depth prediction module outputs the corresponding depth information based on the functional relationship.

[0022] When the signal scaling factor is a time-domain scaling factor and a frequency-domain scaling factor, the depth prediction module simultaneously establishes a functional relationship between the time-domain scaling factor and the depth information, and a functional relationship between the frequency-domain scaling factor and the depth information. The frequency-domain scaling factor and the time-domain scaling factor are independent variables, and the depth information is the dependent variable. After the real-time obtained time-domain scaling factor and frequency-domain scaling factor are input into the depth prediction module, the depth prediction module obtains the corresponding depth information 1 and depth information 2 based on functional relationship 1 and functional relationship 2, respectively. Then, it combines depth information 1 and depth information 2 to output the final depth information.

[0023] Preferably, when the signal scaling factor is a time-domain scaling factor and a frequency-domain scaling factor, the depth prediction module uses the average of the first depth information and the second depth information as the final depth information.

[0024] Preferably, the functional relationship between the time-domain scaling factor and the depth information, as well as the functional relationship between the frequency-domain scaling factor and the depth information, are obtained through data fitting methods.

[0025] This invention places an ultrasound sensor on the top and side of the tissue being tested, and proposes a matching signal processing method to extract the characteristic information of the tissue being tested. This allows for the extraction of more clinically useful information, such as predicting subcutaneous vascularity, tumor invasion depth, and real-time monitoring of treatment effects during laser therapy, without significantly increasing costs. Attached Figure Description

[0026] Figure 1 This diagram illustrates the placement of the ultrasonic probe in the invention.

[0027] Figure 2 The calculation method of the time-domain scaling factor is illustrated;

[0028] Figure 3 The calculation method for the frequency domain scaling factor is illustrated.

[0029] Figure 4 This illustrates the implementation method of the depth prediction module in this embodiment;

[0030] Figure 5 The diagram illustrates the time-domain photoacoustic signals corresponding to the two ultrasonic probes. Detailed Implementation

[0031] 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.

[0032] This embodiment discloses a dual-angle photoacoustic sensing system, comprising:

[0033] A laser used to irradiate the tissue being tested.

[0034] Two ultrasound probes: such as Figure 1 As shown in (a) and (b), when the geometric features of the tissue under test differ in the depth direction (e.g., there are more blood vessels in the longitudinal direction), the traditional photoacoustic sensing method based on a single ultrasound probe cannot effectively perform quantitative analysis and evaluation because there is no reference signal. To solve this problem, in this invention, two ultrasound probes are used to simultaneously detect photoacoustic signals PA1 and PA2 from directly above and from the side of the tissue under test, respectively.

[0035] The distance adjustment mechanism is used to adjust the distance between the two ultrasonic probes, and this distance is continuously adjustable. We propose that the distance between the two probes should be continuously adjustable. This is because if the distance is too close, the photoacoustic signals acquired by the two probes will have high similarity, reducing the accuracy of feature extraction. Conversely, if the side ultrasonic probe is too far from the top probe, the photoacoustic signal will travel too far, resulting in severe sound wave attenuation and a reduced signal-to-noise ratio. Therefore, there should be an optimal distance between the two ultrasonic probes, and this optimal distance varies depending on the application scenario. The distance can be adjusted manually or by a motor.

[0036] The scaling factor calculation module is used to calculate the time-domain scaling factor of photoacoustic signals PA1 and PA2, or to calculate the frequency-domain scaling factor of photoacoustic signals PA1 and PA2, or to calculate both the time-domain and frequency-domain scaling factors of photoacoustic signals PA1 and PA2.

[0037] The calculation method for the time-domain scaling factor is as follows: Figure 2 As shown: First, the effective time-domain signals of photoacoustic signals PA1 and PA2 are truncated to remove coupling and external interference. Then, peak detection is performed on PA1 and PA2 to obtain their time-domain maximum values, peak(PA1) and peak(PA2). Finally, the time-domain maximum values, peak(PA1) and peak(PA2), of the two photoacoustic signals are scaled to obtain the time-domain scaling coefficient Rt.

[0038] The calculation method for the frequency domain scaling factor is as follows: Figure 3As shown: First, the effective signals of photoacoustic signals PA1 and PA2 are extracted respectively. Then, Fourier transforms are performed on photoacoustic signals PA1 and PA2 respectively to obtain their information in the frequency domain. In the frequency domain, we can extract the peak frequency points of each of the two photoacoustic signals, as well as the amplitude at specific frequency points. Based on the frequency information extracted from the two photoacoustic signals, further scaling operations are performed to obtain the frequency domain scaling coefficient Rf. FFT(·) performs a Fourier transform on the time-domain signal to obtain the frequency-domain information of the signal, and peakf(·) extracts the peak values ​​from the frequency-domain information, such as the peak frequency points and their amplitudes or the amplitude values ​​at any frequency point.

[0039] The depth prediction module is used to predict the corresponding depth information based on the time-domain scaling factor and / or frequency-domain scaling factor calculated by the scaling factor calculation module.

[0040] If the scaling factor calculation module obtains a time-domain scaling factor, a functional relationship between the time-domain scaling factor and depth information is established in the depth prediction module, where the time-domain scaling factor is the independent variable and the depth information is the dependent variable. If the scaling factor calculation module obtains a frequency-domain scaling factor, a functional relationship between the frequency-domain scaling factor and depth information is established in the depth prediction module, where the frequency-domain scaling factor is the independent variable and the depth information is the dependent variable. If the scaling factor calculation module obtains both time-domain and frequency-domain scaling factors, a functional relationship between both the time-domain and frequency-domain scaling factors is established in the depth prediction module, where the time-domain and frequency-domain scaling factors are the independent variables and the depth information is the dependent variable.

[0041] In this embodiment, the scaling factor calculation module obtains both the time-domain scaling factor and the frequency-domain scaling factor. Therefore, the depth prediction module simultaneously establishes the functional relationship between the time-domain scaling factor and the depth information, as well as the functional relationship between the frequency-domain scaling factor and the depth information.

[0042] like Figure 4 As shown, a data fitting method is used to establish the functional relationship between the time-domain scaling factor and depth information, and the functional relationship between the frequency-domain scaling factor and depth information: First, we obtain multiple sets of samples with known immersion depths and depth data. We then perform photoacoustic signal acquisition on the samples using dual probes and obtain the time-domain and frequency-domain scaling factors through analysis. At this point, we can combine the known standard depth information of the samples to perform data fitting on the depth and scaling factors, obtaining two fitting functions in the time and frequency domains, namely: Fitting function 1 f1(x) = a + bx + cx 2 +… is used to describe the functional relationship between depth information and temporal scaling coefficients; the fitting function is f2(x) = α + βx + γx 2+… is used to describe the functional relationship between depth information and frequency domain scaling coefficients, where the independent variables are the time domain scaling coefficient and the frequency domain scaling coefficient, and the function value is the immersion depth. The values ​​of coefficients a, b, c, α, β, γ, etc. are obtained.

[0043] In this embodiment: After inputting the real-time obtained temporal scaling factor into the depth prediction module, depth information one can be obtained through fitting function one; after inputting the real-time obtained frequency domain scaling factor into the depth prediction module, depth information two can be obtained through fitting function two. The depth prediction module then calculates the final depth information based on depth information one and depth information two. In this embodiment, the depth prediction module uses the average of depth information one and depth information two as the final depth information.

[0044] It should be noted that, depending on the actual needs, those skilled in the art can choose either the time-domain scaling factor or the frequency-domain scaling factor to predict the structural information of subcutaneous tissue, i.e., depth information (such as vascular richness, tumor invasion depth, and laser treatment effect), or they can combine the time-domain scaling factor and the frequency-domain scaling factor for prediction.

[0045] The aforementioned dual-angle photoacoustic sensing system can effectively overcome the problem of quantitative analysis that cannot be performed due to the lack of a reference signal in single-probe photoacoustic sensing, and provides more and more accurate quantitative tissue analysis and assessment, such as subcutaneous vascular richness and pigment infiltration depth.

[0046] To verify the effectiveness of the aforementioned dual-angle photoacoustic sensing system, an agar phantom was fabricated to simulate mole tissue for phantom experiments:

[0047] 1) Using agar powder, water, and black ink as basic raw materials and controlling their proportions, two types of biomimetic bodies with different melanin concentrations were produced.

[0048] 2) Based on the actual thickness of moles in clinical practice, the prosthetic body was made into circular pieces with a diameter of about one centimeter and four different thicknesses, in the order of thin, relatively thin, relatively thick, and thick.

[0049] 3) The absorption coefficient of melanin was obtained by consulting relevant materials, and 1064nm was selected as the excitation wavelength based on the absorption and penetration capabilities of laser in tissue.

[0050] 4) Signal excitation was performed on each phantom, and signal acquisition was conducted at two angles: directly above and to the side. Experimental signals at different concentrations and thicknesses were obtained, such as... Figure 5 As shown, (a) represents the signal corresponding to low concentration - thin and relatively thin, and (b) represents the signal corresponding to high concentration - relatively thick and thick.

[0051] 5) Draw the time-domain waveforms and power spectrum distributions of each group of signals, and perform inter-group comparisons and calculate key parameters.

[0052] Here, we use time-domain signal analysis to compare the peak-to-peak values ​​of two time-domain signals collected by probes at different distances from the same sample, and obtain the Rt value distribution as shown in the table below (where the signal collected by the closer probe corresponds to PA1 in the formula):

[0053]

[0054] As can be seen from the table above, the Rt values ​​vary considerably for different concentrations. At the same concentration, the ratio decreases as the sample thickness gradually increases, which is consistent with our theoretical analysis.

Claims

1. A dual-angle photoacoustic sensing system, comprising a laser for irradiating a tissue under test, characterized in that, Also includes: Two ultrasound probes are used. One ultrasound probe detects the photoacoustic signal PA1 from directly above the tissue being tested, and the other ultrasound probe detects the photoacoustic signal PA2 from the side of the tissue being tested. The proportionality coefficient calculation module is used to calculate the signal proportionality coefficient between photoacoustic signal PA1 and photoacoustic signal PA2. Depth Prediction Module: The depth prediction module establishes a functional relationship between the signal scaling factor and the depth information. After the real-time signal scaling factor is input into the depth prediction module, the depth prediction module outputs the corresponding depth information based on the functional relationship. The signal scaling factor is a time-domain scaling factor, a frequency-domain scaling factor, or a combination of both, wherein: The time-domain scaling factor is calculated using the following method: The effective time-domain signals of photoacoustic signals PA1 and PA2 are extracted respectively; Peak detection was performed on the truncated photoacoustic signals PA1 and PA2 respectively to obtain the time-domain maximum values ​​of photoacoustic signals PA1 and PA2, peak(PA1) and peak(PA2). The time-domain scaling factor is obtained by performing scaling operations on the time-domain maximum values ​​peak(PA1) and peak(PA2); The frequency domain scaling factor is calculated using the following method: The effective signals of photoacoustic signals PA1 and PA2 are extracted respectively; Fourier transforms were performed on the truncated photoacoustic signals PA1 and PA2 respectively to obtain their information in the frequency domain; In the frequency domain, the frequency information of photoacoustic signals PA1 and PA2 is extracted and then scaled to obtain the frequency domain scaling factor. When the signal scaling factor is a time-domain scaling factor, a functional relationship between the time-domain scaling factor and the depth information is established in the depth prediction module, wherein the time-domain scaling factor is the independent variable and the depth information is the dependent variable. After the real-time obtained time-domain scaling factor is input into the depth prediction module, the depth prediction module outputs the corresponding depth information based on the functional relationship. When the signal scaling factor is a frequency domain scaling factor, a functional relationship between the frequency domain scaling factor and the depth information is established in the depth prediction module. The frequency domain scaling factor is the independent variable and the depth information is the dependent variable. After the frequency domain scaling factor obtained in real time is input into the depth prediction module, the depth prediction module outputs the corresponding depth information based on the functional relationship. When the signal scaling factor is a time-domain scaling factor and a frequency-domain scaling factor, the depth prediction module simultaneously establishes a functional relationship between the time-domain scaling factor and the depth information, and a functional relationship between the frequency-domain scaling factor and the depth information. The frequency-domain scaling factor and the time-domain scaling factor are independent variables, and the depth information is the dependent variable. After the real-time obtained time-domain scaling factor and frequency-domain scaling factor are input into the depth prediction module, the depth prediction module obtains the corresponding depth information 1 and depth information 2 based on functional relationship 1 and functional relationship 2, respectively. Then, it combines depth information 1 and depth information 2 to output the final depth information.

2. The dual-angle photoacoustic sensing system as described in claim 1, characterized in that, It also includes a distance adjustment mechanism for adjusting the distance between the two ultrasonic probes, making the distance between the two ultrasonic probes continuously adjustable.

3. The dual-angle photoacoustic sensing system as described in claim 1, characterized in that, The frequency information refers to the amplitude at the peak frequency point or the amplitude at any frequency point.

4. The dual-angle photoacoustic sensing system as described in claim 1, characterized in that, When the signal scaling factor is a time-domain scaling factor and a frequency-domain scaling factor, the depth prediction module uses the average of the first depth information and the second depth information as the final depth information.

5. A dual-angle photoacoustic sensing system as described in claim 1, characterized in that, The functional relationships between the time-domain scaling factor and the depth information, as well as the functional relationships between the frequency-domain scaling factor and the depth information, were obtained through data fitting methods.

Images