A unipolar adaptive photoacoustic tomographic image reconstruction method

Abstract

The application belongs to the technical field of photoacoustic imaging, and discloses a unipolar adaptive photoacoustic tomographic image reconstruction method, which comprises the following steps: firstly, inserting a virtual array element between adjacent array elements of an ultrasonic array, and improving the spatial sampling rate through interpolation; secondly, reconstructing an original photoacoustic image based on a delay-and-sum algorithm; thirdly, performing time difference processing on the interpolated channel signals to obtain first-order time gradient data, and introducing a linear compensation factor to correct the negative gradient signals; fourthly, reconstructing a gradient photoacoustic image by using the compensated gradient signals; and finally, normalizing and then weighting and fusing the original image and the gradient image to generate a high-contrast unipolar photoacoustic image. The application compensates for negative signals through mathematical operation and fuses images, effectively solves the problems of low contrast and fuzzy edges of a traditional bipolar image, and avoids the loss of details caused by Hilbert transform. On the basis of retaining the characteristics of the sound pressure signal, the method significantly improves the image signal-to-noise ratio and edge sharpness.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of photoacoustic imaging, and particularly relates to a unipolar adaptive photoacoustic tomographic image reconstruction method. BACKGROUND

[0002] Photoacoustic Tomography (PAT) is a new non-invasive medical imaging technology, which realizes multi-dimensional analysis of biological tissues by combining the high-contrast advantage of optical imaging with the deep penetration ability of ultrasonic imaging. Its technical principle is based on the photoacoustic effect: when pulsed laser irradiates biological tissues, internal light-absorbing components (such as hemoglobin and melanin) absorb light energy and produce transient thermal expansion, exciting wide-band ultrasonic signals. After the array ultrasonic transducer captures these sound signals, through electric signal conversion and time delay correction processing, combined with the preset sound speed distribution model, the light-absorbing energy spatial distribution map of the internal tissues can be reconstructed.

[0003] The original channel data of photoacoustic tomography is approximately symmetrically distributed in the time domain, and the signal mean value is close to zero, so the traditional delay-and-sum (DAS) algorithm is used for reconstruction, which will generate a bipolar image. In this kind of image, the positive and negative sound pressure changes correspond to bright white and dark black pixels respectively, and the background area appears gray because the value is close to zero. Although the bipolar image completely retains the sound pressure polarity information, the background and weak signal gray scale are similar, and the low-contrast gray scale distribution seriously weakens the recognition of small blood vessels or deep tissues, which significantly limits the subsequent image processing and judgment.

[0004] In order to overcome the shortcomings of the bipolar image, Hilbert transform is often used in ultrasonic imaging to detect the envelope of the signal, and convert the bipolar image into a unipolar image. However, the envelope operation eliminates the contrast folding caused by the positive and negative polarity information, but inevitably leads to blurred edge details. The blood vessels and tissue boundaries in the unipolar image after Hilbert transform processing are shown as width expansion or edge blur, which makes it more difficult to accurately measure and quantitatively analyze the blood vessel diameter and tissue boundary, and also affects the image judgment. Therefore, how to process the positive and negative polarity signals while maintaining high contrast and clear edges has become a key challenge to improve the quality of photoacoustic tomography. SUMMARY

[0005] In order to solve the above problems existing in photoacoustic imaging, the application provides a unipolar adaptive photoacoustic tomographic image reconstruction method, which mainly processes the channel signals through mathematical operation, compensates for the negative signals, and converts the bipolar image into a unipolar image through image fusion, to obtain a high-contrast photoacoustic image.

[0006] To achieve the above object, the technical scheme adopted by the present application comprises the following steps:

[0007] A unipolar adaptive photoacoustic tomographic image reconstruction method comprises the following steps:

[0008] Step one, introduce a virtual element: the ultrasonic array contains multiple elements, a virtual element is inserted between any two adjacent elements, and the corresponding channel is an interpolation channel;

[0009] Step two, reconstruct the original photoacoustic image based on delay-and-sum: based on the interpolated channel data, the original photoacoustic image is reconstructed by using the delay-and-sum algorithm;

[0010] Step three, calculate the channel gradient data: the interpolated channel signal is differentially processed in the time dimension to obtain the first-order time gradient data of each channel signal;

[0011] Step four, calculate the gradient compensation data: for the signal with a negative gradient value, a compensation factor linearly related to time is introduced for weighting, the negative part weakened in the signal is corrected, and then the negative of the whole is taken to obtain the gradient compensation data;

[0012] Step five, reconstruct the gradient photoacoustic image based on delay-and-sum: the same delay-and-sum algorithm as step two is used, the input data is replaced by the compensated gradient signal, and a gradient photoacoustic image focusing on edge details and local changes is constructed;

[0013] Step six, fuse the images: the original photoacoustic image and the gradient photoacoustic image are normalized and weightedly fused to finally obtain a two-dimensional photoacoustic tomographic image with high contrast.

[0014] Further, the position of the virtual element in step one is set as the midpoint of the positions of the two adjacent elements, and the channel data is the average of the channel data corresponding to the two adjacent elements.

[0015] Further, in step two, it is assumed that the sound wave propagates along a straight line at a constant speed in the tissue, the propagation distance of each pixel point in the image to each element is calculated, and the speed of sound in the medium is divided to obtain the corresponding propagation time delay; then, the signal corresponding to the delay time point is extracted from each channel signal and superimposed to construct an original photoacoustic image reflecting the light absorption distribution inside the tissue.

[0016] Further, step one comprises the following specific steps:

[0017] Suppose the ultrasonic array contains elements, the position coordinates are , and the original channel data received is , , For sampling time, a virtual array element, interpolation array element, is inserted between any two adjacent array elements, and its position is set as the midpoint of the positions of the two adjacent array elements. The interpolated array element coordinates are expressed as:

[0018] ,

[0019] The interpolation channel data takes the average of the channel data corresponding to the two adjacent array elements. The interpolated channel data is expressed as:

[0020] .

[0021] Further, the original photoacoustic image of step two The calculation process is expressed as:

[0022] ,

[0023] wherein is the propagation speed of ultrasound in the medium.

[0024] Further, the first-order time gradient data of each channel signal of step three The calculation process is:

[0025] ,

[0026] wherein is the number of sampling points.

[0027] Further, the gradient compensation data of step four The calculation process is expressed as:

[0028] ,

[0029] wherein is the channel gradient data is the number of consecutive negative gradients in the row to the column , is an indicator function, which takes 1 when the condition is met, and 0 otherwise.

[0030] Further, the gradient photoacoustic image of step five is expressed as:

[0031] .

[0032] Further, the step six includes the following specific steps:

[0033] The original photoacoustic image and the gradient photoacoustic image are normalized respectively to transform the image values to -1 to 1, expressed as:

[0034] ,

[0035] wherein, represents a pixel point of the original photoacoustic image, is the whole original photoacoustic image; represents a pixel point of the gradient photoacoustic image, is the whole gradient photoacoustic image;

[0036] Then, weighted fusion is performed, and negative values are set to zero to obtain a unipolar image , which is represented as:

[0037] ,

[0038] wherein is a proportional factor.

[0039] The unipolar adaptive photoacoustic tomographic image reconstruction method has the following advantages:

[0040] The unipolar adaptive photoacoustic tomographic image reconstruction method can effectively enhance the details and contrast of the image while retaining the characteristics of the original sound pressure signal. Compared with the bipolar image, the technology significantly improves the signal-to-noise ratio of deep weak signals, reduces image artifacts, and maintains the high-contrast advantage of the unipolar image. On the basis of breaking through the limitations of traditional Hilbert envelope detection, the edge sharpness and detail features of the image are maintained. In addition, the expression method of the unipolar image is consistent with traditional medical imaging forms such as ultrasound imaging, has stronger readability and acceptance, and is helpful for more efficient image interpretation and identification. At the same time, the reconstruction algorithm used in the image processing process has a simple calculation structure and is easy to parallelize and accelerate, does not increase the overall computational complexity, and can realize real-time image reconstruction. BRIEF DESCRIPTION OF DRAWINGS

[0041] Figure 1 is a flow chart of the unipolar adaptive photoacoustic tomographic image reconstruction method of the present application;

[0042] Figure 2 is a schematic diagram of the original photoacoustic image (bipolar image) of the delay-and-sum reconstruction;

[0043] Figure 3 is a schematic diagram of the gradient photoacoustic image of the delay-and-sum reconstruction;

[0044] Figure 4 is a schematic diagram of the photoacoustic image (unipolar image) of the delay-and-sum reconstruction. DETAILED DESCRIPTION

[0045] In order to better understand the purpose, structure and function of the present application, a single polarity adaptive photoacoustic tomographic image reconstruction method is further described in detail below in combination with the drawings.

[0046] The single polarity adaptive photoacoustic tomographic image reconstruction method of the present application comprises the following steps:

[0047] Step one, introducing a virtual array element: the ultrasonic array contains a plurality of array elements, a virtual array element is inserted between any two adjacent array elements, the corresponding channel is an interpolation channel, the position is set as the midpoint of the positions of the two adjacent array elements, and the channel data is the average of the channel data corresponding to the two adjacent array elements. By introducing the virtual array element, the number of effective channels is increased, and the spatial sampling rate is improved;

[0048] Step two, reconstructing the original photoacoustic image based on delay and summation: assuming that the sound wave propagates along a straight line at a constant speed in the tissue, the propagation distance of each pixel point in the image to each array element is calculated, and the speed in the medium is divided to obtain the corresponding propagation time delay. Subsequently, the signal at the corresponding delay time point is extracted from each channel signal and superimposed, thereby enhancing the target signal and effectively suppressing the background noise, and an original photoacoustic image reflecting the light absorption distribution inside the tissue is constructed. The image retains the sound pressure polarity information, is a bipolar gray scale distribution map, and serves as the basis for subsequent enhancement processing;

[0049] Step three, calculating channel gradient data: the interpolated channel signal is differentially processed in the time dimension to obtain the first-order time gradient data of each channel signal;

[0050] Step four, calculating gradient compensation data: for the signal with a negative gradient value, a compensation factor linearly related to time is introduced for weighting to correct the weakened negative part in the signal, and then the whole is taken as negative to obtain the gradient compensation data;

[0051] Step five, reconstructing the gradient photoacoustic image based on delay and summation: the same delay and summation algorithm as in step two is adopted, but the input data is replaced by the compensated gradient signal, thereby constructing a gradient photoacoustic image focusing on edge details and local changes;

[0052] Step six, fusing the images: the original photoacoustic image and the gradient photoacoustic image are normalized and weightedly fused, and finally a two-dimensional photoacoustic tomographic image with high contrast is obtained.

[0053] Embodiment

[0054] As Figure 1 shown, the present application discloses a single polarity adaptive photoacoustic tomographic image reconstruction method, comprising the following steps:

[0055] Step one, introducing a virtual array element: assuming that the ultrasonic array contains array elements, and their position coordinates are ,The original channel data received is , , A virtual array element is inserted between any two adjacent array elements, namely the interpolation array element, and its position is set to the midpoint of the two adjacent array element positions. The interpolated array element coordinates are Expressed as:

[0056]

[0057] The interpolated channel data takes the average value of the corresponding channel data of two adjacent array elements. The interpolated channel data Expressed as:

[0058]

[0059] Step 2: Reconstruct the original photoacoustic image using delay summation based on the interpolated channel data , the calculation process is expressed as:

[0060]

[0061] in is the propagation speed of ultrasound in the medium. The original photoacoustic image obtained by delay-sum reconstruction is as follows: Figure 2 As shown in the figure, the yellow dotted line is the distribution position of the ultrasonic array elements.

[0062] Step 3: Calculate channel gradient data : Perform differential processing on the interpolated channel signal in the time dimension to obtain the first-order time gradient data of each channel signal , the calculation process is:

[0063]

[0064] in is the number of sampling points.

[0065] Step 4: Calculate gradient compensation data : For the continuous negative values ​​in the channel gradient data, multiply them by a factor that is linearly related to time, that is, dynamically amplify the continuous negative gradient value. The compensation intensity is linearly related to the number of times the continuous negative gradient appears. Finally, the overall data is negative. The calculation process is expressed as:

[0066]

[0067] in is the channel gradient data OK Middle to column the number of continuous negative gradients, is an indicator function, taking 1 if the condition is satisfied, otherwise 0.

[0068] Step five, reconstruction of the gradient photoacoustic image based on delay-and-sum: the same delay-and-sum algorithm as step two is adopted, but the input data is replaced by the compensated gradient signal , thus constructing a gradient photoacoustic image focusing on edge details and local changes , denoted as:

[0069]

[0070] The gradient photoacoustic image reconstructed based on delay-and-sum is shown in Figure 3 .

[0071] Step six, image fusion: the original photoacoustic image and the gradient photoacoustic image are normalized respectively to transform the image values to -1 to 1, denoted as:

[0072]

[0073] wherein, represents a pixel point of the original photoacoustic image, is the whole original photoacoustic image; represents a pixel point of the gradient photoacoustic image, is the whole gradient photoacoustic image;

[0074] weighted fusion is then performed, and the negative values are set to zero to obtain a unipolar image , denoted as:

[0075]

[0076] wherein is a proportional factor, generally taken as 1.

[0077] The unipolar image obtained after weighting is shown in Figure 4 , compared with Figure 2 the bipolar image, the unipolar image has higher contrast and fewer artifacts.

[0078] It can be understood that the present application is described through some embodiments, and those skilled in the art know that various changes or equivalent replacements can be made to the features and embodiments without departing from the spirit and scope of the present application. In addition, the features and embodiments can be modified to adapt to specific conditions and materials under the guidance of the present application without departing from the spirit and scope of the present application. Therefore, the present application is not limited by the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of the present application are within the scope of the present application.

Claims

1. A single-polarity adaptive photoacoustic tomography image reconstruction method, characterized in that: The steps include: Step 1: Introduce virtual array elements: The ultrasound array contains multiple array elements. A virtual array element is inserted between any two adjacent array elements, and the corresponding channel is the interpolation channel. Step 2: Reconstructing the original photoacoustic image based on delay-sum algorithm: Based on the interpolated channel data, the original photoacoustic image is reconstructed using the delay-sum algorithm; Original photoacoustic image The calculation process is expressed as: ， in is the propagation speed of ultrasound in the medium, N is the number of array elements, RF () represents the original channel data, Indicates the i array element x coordinate, Indicates the i array element z coordinate; Step 3: Calculate the channel gradient data: perform differential processing on the interpolated channel signal in the time dimension to obtain the first-order time gradient data of each channel signal; Step 4: Calculate the gradient compensation data: For signals with negative gradient values, introduce a compensation factor that is linearly related to time to weight them, correct the weakened negative part of the signal, and then negate the entire value to obtain the gradient compensation data; The gradient compensation data of step 4 The calculation process is expressed as: ， in is the channel gradient data OK Middle to column The number of consecutive negative gradients is, is an indicator function, which takes 1 when the condition is met and 0 otherwise; Step 5: Reconstructing a gradient photoacoustic image based on delayed summation: Using the same delayed summation algorithm as in step 2, the input data is replaced with a compensated gradient signal to construct a gradient photoacoustic image that focuses on edge details and local changes. Step 6: Image fusion: Normalize the original photoacoustic image and the gradient photoacoustic image separately and perform weighted fusion to finally obtain a high-contrast two-dimensional photoacoustic tomography image.

2. The single-polarity adaptive photoacoustic tomography image reconstruction method according to claim 1, characterized in that: The position of the virtual array element in step 1 is set to the midpoint of the positions of two adjacent array elements, and its channel data is the average value of the corresponding channel data of the two adjacent array elements.

3. The single-polarity adaptive photoacoustic tomography image reconstruction method according to claim 1, characterized in that: In step 2, assuming that acoustic waves propagate in a straight line at a constant speed in the tissue, the propagation distance from each pixel in the image to each array element is calculated and divided by the speed of sound in the medium to obtain the corresponding propagation time delay. Subsequently, the signal corresponding to the delay time point is extracted from each channel signal and superimposed to construct an original photoacoustic image reflecting the light absorption distribution within the tissue.

4. The single-polarity adaptive photoacoustic tomography image reconstruction method according to claim 1, characterized in that: The step 1 includes the following specific steps: Assume that the ultrasound array contains array elements, and their position coordinates are ,The original channel data received is , , For the sampling time, a virtual array element is inserted between any two adjacent array elements - the interpolation array element, and its position is set to the midpoint of the two adjacent array element positions. The interpolated array element coordinates Expressed as: ， The interpolated channel data takes the average value of the corresponding channel data of two adjacent array elements. The interpolated channel data Expressed as: 。 5. The single-polarity adaptive photoacoustic tomography image reconstruction method according to claim 1, characterized in that: The first-order time gradient data of each channel signal in step 3 The calculation process is: ， in is the number of sampling points, RF () represents the original channel data, m is the number of rows.

6. The single-polarity adaptive photoacoustic tomography image reconstruction method according to claim 1, characterized in that: The gradient photoacoustic image of step 5 , expressed as: 。 7. The single-polarity adaptive photoacoustic tomography image reconstruction method according to claim 1, characterized in that: The step six includes the following specific steps: The original photoacoustic image and the gradient photoacoustic image are normalized to transform the image values ​​to -1 to 1, which can be expressed as: ， in, represents the pixel of the original photoacoustic image, It is the entire original photoacoustic image; represents the pixel point of the gradient photoacoustic image, is the entire gradient photoacoustic image; Then perform weighted fusion and set the negative values ​​to zero to obtain a unipolar graph , expressed as: ， in is the scale factor.

Images