Ultrasonic three-dimensional volume rendering method and device

Abstract

The present invention relates to an ultrasonic three-dimensional volume rendering method, comprising the following steps: obtaining original volume data, preprocessing and reconstructing the volume data to obtain reconstructed volume data; obtaining an original opacity transfer function; setting a correction curve that varies with grayscale values, and fusing the correction curve to the original opacity transfer function to obtain a corrected opacity transfer function; and performing volume rendering based on the corrected opacity transfer function. The present invention corrects the opacity transfer function to improve its distribution effect, thereby improving the three-dimensional volume rendering effect.

Description

Technical Field

[0001] The present invention relates to the field of three-dimensional ultrasound technology, and in particular to an ultrasound three-dimensional volume rendering method, device and computer storage medium. Background Art

[0002] Traditional ultrasound imaging equipment can only provide two-dimensional medical images. Ultrasound doctors can only rely on experience to estimate the size, shape, location and geometric relationship of the lesion with its surrounding tissues from multiple two-dimensional images. They cannot obtain information about the inter-tissue and internal structures, which makes clinical diagnosis difficult for ultrasound doctors. With the development of medical imaging equipment and technology, medical image visualization plays an increasingly important role in clinical medicine.

[0003] At present, 3D ultrasound visualization imaging technology has become a mainstream ultrasound examination method in clinical practice. As an important component of 3D ultrasound visualization imaging system, 3D visualization volume rendering technology can not only intuitively and vividly display the overall surface of the imaging object, but also effectively reveal the internal structural information of 3D volume data. Therefore, it is very necessary to carry out 3D volume rendering research in the field of ultrasound.

[0004] In the existing ultrasound 3D volume rendering technology, the visualization algorithms involved mainly include ray casting algorithm (RayCasting), snowball (Splatting) algorithm and shear-warp algorithm. The ray casting algorithm has been widely used due to its high imaging quality, but its rendering speed is slower than other visualization algorithms. To this end, people have studied different volume rendering acceleration algorithms based on software and hardware, which has promoted the development of this technology. There are many acceleration algorithms for the ray casting algorithm, which can be summarized into two categories according to their algorithm ideas, namely, reducing the number of projected rays and reducing the number of sampling points in space jump. At the same time, with the improvement of hardware technology, GPU-based volume rendering acceleration and parallel volume rendering technology have made it more and more widely used in ultrasound 3D visualization imaging systems, providing assistance for ultrasound doctors' diagnosis.

[0005] Transfer function design is a key step in visual volume rendering. It is responsible for converting the data attributes in the volume data into certain optical attributes, such as color values ​​and opacity values, which directly affects the final volume rendering effect. Color is generally used to distinguish different substances in three-dimensional volume data. The RGB model is usually used, that is, the scalar value of the sampling point is mapped to an RGB three-dimensional vector through a transfer function. The visibility of the voxels in the volume data in the final visualization image depends on the value of the opacity transfer function. In order to highlight the tissue information of interest, a higher opacity is assigned, and structures of no interest are directly assigned zero or low opacity values. Therefore, designing a good opacity transfer function has a great impact on the results of ultrasound three-dimensional volume rendering, thereby affecting the clinical diagnosis of ultrasound doctors. Summary of the invention

[0006] In view of this, it is necessary to provide an ultrasonic three-dimensional volume rendering method, device and computer storage medium to solve the problem that the opacity transfer function distribution effect is poor and affects the three-dimensional volume rendering effect.

[0007] The present invention provides an ultrasonic three-dimensional volume rendering method, comprising the following steps:

[0008] Acquiring original volume data, preprocessing and reconstructing the volume data to obtain reconstructed volume data;

[0009] Get the original opacity transfer function;

[0010] Setting a correction curve that varies with grayscale values, and fusing the correction curve into the original opacity transfer function to obtain a corrected opacity transfer function;

[0011] Volume rendering is performed based on the corrected opacity transfer function.

[0012] Further, the original volume data is obtained, specifically:

[0013] The original volume data obtained by three-dimensional scanning according to the set probe scanning parameters is obtained.

[0014] Further, the volume data is preprocessed and reconstructed to obtain reconstructed volume data, specifically:

[0015] Performing filtering and smoothing processing on the volume data to obtain preprocessed volume data;

[0016] A reconstruction table required for volume reconstruction is calculated according to the set probe scanning parameters, and the preprocessed volume data is reconstructed by looking up the reconstruction table to obtain reconstructed volume data.

[0017] Furthermore, a correction curve that changes with the gray value is set, specifically:

[0018] f(x)=(1+λ) g(x) ;

[0019] Among them, f(x) is the correction curve function, x is the gray value, λ is the control parameter, λ∈(-1,1), which is used to control the amplitude of f(x), g(x) is the bias function, I H is the mean gray intensity corresponding to the structural organization in the volume data, I L is the mean gray intensity corresponding to uniform tissue in the volume data, a, b, c are constant parameters, e is a natural constant, and d is an empirical value;

[0020] The calibration curve is obtained by adjusting the d value and the λ value in the calibration curve function.

[0021] Furthermore, the correction curve is merged into the original opacity transfer function to obtain a corrected opacity transfer function, which is specifically:

[0022] The correction curve is multiplied by the opacity value corresponding to the original opacity transfer function at the same gray value to obtain a corrected opacity transfer function.

[0023] Furthermore, volume rendering is performed based on the corrected opacity transfer function, specifically:

[0024] Volume rendering is performed using a ray casting algorithm based on the corrected opacity transfer function.

[0025] Furthermore, a ray casting algorithm is used to perform volume rendering based on the corrected opacity transfer function, specifically:

[0026] Calculate the opacity value of each pixel based on the corrected opacity transfer function;

[0027] The ray casting algorithm is used to sample each pixel to obtain the color value and opacity value of each sampling point;

[0028] The color value and opacity value of each sampling point are synthesized according to the synthesis rule to obtain the color value of the corresponding pixel point on the screen;

[0029] Combining the color values ​​of each pixel on the screen, the overall volume rendering image is obtained.

[0030] Furthermore, it also includes:

[0031] Adjusting the correction curve according to the volume rendering result, and fusing the adjusted correction curve into the original opacity transfer function to obtain a re-calibrated opacity transfer function;

[0032] The volume rendering is re-performed based on the recalibrated opacity transfer function.

[0033] The present invention also provides an ultrasonic three-dimensional volume rendering device, comprising a processor and a memory, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the ultrasonic three-dimensional volume rendering method is implemented.

[0034] The present invention also provides a computer storage medium on which a computer program is stored. When the computer program is executed by a processor, the ultrasonic three-dimensional volume rendering method is implemented.

[0035] Beneficial effects: The present invention first performs a three-dimensional scan according to the probe scanning parameters to obtain the original volume data; pre-processes the original volume data and reconstructs the volume data to obtain the reconstructed volume data; obtains the original opacity transfer function; designs an opacity transfer function correction curve that changes with the pixel gray value, and merges the designed correction curve into the original opacity transfer function to obtain the final opacity transfer function; draws the three-dimensional volume data based on the corrected opacity transfer function and displays the final drawing effect on the output device. The present invention corrects the opacity transfer function and performs volume rendering based on the corrected opacity transfer function, thereby displaying the overall surface of the volume data and the internal structure information of the volume data, providing more help for ultrasound doctors to make quick and accurate diagnoses. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 A method flow chart of a first embodiment of an ultrasonic three-dimensional volume rendering method provided by the present invention;

[0037] Figure 2 A schematic diagram of a scanning method of a convex array fan scan in a first embodiment of the ultrasonic three-dimensional volume rendering method provided by the present invention;

[0038] Figure 3 An original opacity transfer function diagram of the first embodiment of the ultrasonic three-dimensional volume rendering method provided by the present invention;

[0039] Figure 4 A calibration curve diagram of the first embodiment of the ultrasonic three-dimensional volume rendering method provided by the present invention;

[0040] Figure 5 A schematic diagram of the fusion of the first correction curve and the original opacity transfer function of the first embodiment of the ultrasonic three-dimensional volume rendering method provided by the present invention;

[0041] Figure 6 A schematic diagram of the fusion of the second correction curve and the original opacity transfer function of the first embodiment of the ultrasonic three-dimensional volume rendering method provided by the present invention;

[0042] Figure 7This is a principle diagram of a ray casting algorithm of the first embodiment of the ultrasonic three-dimensional volume rendering method provided by the present invention. DETAILED DESCRIPTION

[0043] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.

[0044] Example 1

[0045] like Figure 1 As shown, embodiment 1 of the present invention provides an ultrasonic three-dimensional volume rendering method, comprising the following steps:

[0046] S1. Acquire original volume data, preprocess and reconstruct the volume data to obtain reconstructed volume data;

[0047] S2, obtaining the original opacity transfer function;

[0048] S3, setting a correction curve that changes with the gray value, and integrating the correction curve into the original opacity transfer function to obtain a corrected opacity transfer function;

[0049] S4. Perform volume rendering based on the corrected opacity transfer function.

[0050] This embodiment corrects the opacity transfer function, and performs volume rendering based on the corrected opacity transfer function, thereby displaying the overall surface of the volume data and the internal structural information of the volume data, providing more help for ultrasound doctors to make quick and accurate diagnoses. Specifically, firstly, a three-dimensional scan is performed according to the probe scanning parameters to obtain the original volume data; the original volume data is preprocessed and the reconstructed volume data is obtained through volume data reconstruction; the original opacity transfer function (OTF) is obtained; an opacity transfer function correction curve that changes with the pixel gray value is designed, and the designed correction curve is merged into the original opacity transfer function to obtain the final opacity transfer function; the three-dimensional volume data is drawn based on the corrected opacity transfer function and the final drawing effect is displayed on the output device.

[0051] By adopting the above-mentioned three-dimensional ultrasound imaging visualization volume rendering scheme, the reconstructed volume data can be drawn into a visible two-dimensional view to obtain the overall surface image of the volume data, and the internal structure information of the volume data can be observed therefrom, which is conducive to the analysis of the spatial structure and position morphological characteristics in the three-dimensional volume data. This embodiment designs a relatively good opacity transfer function, which is a good help for the volume rendering effect, thereby providing more help for ultrasound doctors to make quick and accurate diagnoses.

[0052] Preferably, the original volume data is obtained, specifically:

[0053] The original volume data obtained by three-dimensional scanning according to the set probe scanning parameters is obtained.

[0054] In this embodiment, a three-dimensional ultrasonic volume probe is used to perform three-dimensional scanning to obtain original volume data. The scanning method is a convex array fan scan. For details, see Figure 2 As shown, Figure 2 The number 1 represents the coronal section at the exact center of the reconstructed volume data, the number 2 represents the sagittal section at the exact center of the reconstructed volume data, and the number 3 represents the transverse section at the exact center of the reconstructed volume data. In addition, there are many ways to obtain the probe scanning parameters, which can be by reading preset values ​​or by receiving scanning parameters sent by other devices, which are not limited here. In addition, the scanning parameters specifically include scanning angle, sector scanning angle, scanning radius, sector scanning radius, imaging quality, etc., which are also not limited here.

[0055] Preferably, the volume data is preprocessed and reconstructed to obtain reconstructed volume data, specifically:

[0056] Performing filtering and smoothing processing on the volume data to obtain preprocessed volume data;

[0057] A reconstruction table required for volume reconstruction is calculated according to the set probe scanning parameters, and the preprocessed volume data is reconstructed by looking up the reconstruction table to obtain reconstructed volume data.

[0058] In this embodiment, before the original volume data is reconstructed twice, the original volume data is preprocessed, including filtering, smoothing and other operations, to improve the signal-to-noise ratio of the original volume data. In order to speed up the volume reconstruction process, the reconstruction table required for the two volume reconstructions is calculated and stored according to the preset scanning parameters, so that the original volume data can be reconstructed by looking up the table, thereby speeding up the reconstruction process. The reconstruction table specifically refers to the coordinate position pre-established according to the probe scanning parameters during the volume reconstruction process, and the coordinate position of the volume data is determined by looking up the table to obtain the reconstructed volume data.

[0059] In this embodiment, since the ray casting algorithm is used for volume rendering, it is necessary to use the optical properties such as opacity value and color value to be resampled and accumulated in the middle. The opacity value and color value are mapped by the pixel gray value, and finally the rendering effect is displayed. Therefore, the opacity transfer function is required as the gray mapping. The commonly used one-dimensional opacity transfer function is to map the gray value of the sampling point to the opacity value. In general, the one-dimensional opacity transfer function is based on the trapezoidal mapping relationship, the ramp mapping relationship, the triangle mapping relationship, etc. The mapping relationship of the original opacity transfer function selected in this embodiment is as follows Figure 3 The parabola shown.

[0060] Preferably, a correction curve that varies with the gray value is set, specifically:

[0061] f(x)=(1+λ) g(x) ;

[0062] Among them, f(x) is the correction curve function, x is the gray value, λ is the control parameter, λ∈(-1,1), which is used to control the amplitude of f(x), g(x) is the bias function, I H is the mean gray intensity corresponding to the structural organization in the volume data, I L is the mean gray intensity corresponding to uniform tissue in the volume data, a, b, c are constant parameters, e is a natural constant, and d is an empirical value;

[0063] The calibration curve is obtained by adjusting the d value and the λ value in the calibration curve function.

[0064] In this embodiment, in order to improve the original opacity transfer function, an opacity transfer function correction curve is constructed and merged into the original opacity transfer function to obtain a new opacity transfer function. Figure 4 As shown in the figure, the calibration curves under 10 sets of parameters (d value and λ value) are given. H ,I L are the mean gray intensity values ​​corresponding to the structural tissue and the mean gray intensity values ​​corresponding to the uniform tissue in the volume data, respectively. There are many methods for determining the structural tissue and the uniform tissue, which are not limited here. d is an empirical value, and in this embodiment, the value is 50. λ is used to control the amplitude of f(x), and the value range is (-1,1). Figure 4 The 10 groups of curves in the middle are the calibration curve result diagrams under different d values ​​and λ values. The values ​​of the two parameters d value and λ value can be adjusted and determined through multiple volume renderings to continuously enhance the volume rendering effect. Specifically, if the volume rendering effect is adjusted to be significantly better than the original rendering effect, mainly from the contrast effect, it means that the parameters are appropriate.

[0065] Preferably, the correction curve is merged into the original opacity transfer function to obtain a corrected opacity transfer function, specifically:

[0066] The correction curve is multiplied by the opacity value corresponding to the original opacity transfer function at the same gray value to obtain a corrected opacity transfer function.

[0067] The correction curve is multiplied by the gray value position corresponding to the original opacity transfer function to obtain the corrected opacity transfer function. Figure 5 and Figure 6The schematic diagrams of fusing and correcting the same original opacity transfer function with two different correction curves are given respectively. Figure 5 and Figure 6 The horizontal axis represents the grayscale intensity, and the vertical axis represents the opacity value corresponding to the grayscale intensity. Figure 5 In the figure: (a) is the original opacity transfer function curve, (c) is the first correction curve when d is 50 and λ is 0.75, (b) is the opacity transfer function after fusion correction, and (d) is a comparison diagram of the original opacity transfer function before correction and the opacity transfer function after correction. Figure 6 In the figure: (a) is the original opacity transfer function curve, (c) is the second correction curve when d is 50 and λ is -0.25, (b) is the opacity transfer function after fusion correction, and (d) is the comparison diagram of the original opacity transfer function before correction and the opacity transfer function after correction. In practice, the values ​​of d and λ need to be determined through multiple volume rendering experiments.

[0068] Preferably, volume rendering is performed based on the corrected opacity transfer function, specifically:

[0069] Volume rendering is performed using a ray casting algorithm based on the corrected opacity transfer function.

[0070] After the opacity transfer function is confirmed, any volume rendering algorithm in the prior art can be used for volume rendering. This embodiment uses a ray casting algorithm, which has a higher rendering quality than other algorithms.

[0071] Preferably, the volume rendering is performed using a ray casting algorithm based on the corrected opacity transfer function, specifically:

[0072] Calculate the opacity value of each pixel based on the corrected opacity transfer function;

[0073] The ray casting algorithm is used to sample each pixel to obtain the color value and opacity value of each sampling point;

[0074] The color value and opacity value of each sampling point are synthesized according to the synthesis rule to obtain the color value of the corresponding pixel point on the screen;

[0075] Combining the color values ​​of each pixel on the screen, the overall volume rendering image is obtained.

[0076] In this embodiment, the fused opacity transfer function is volume rendered using a ray casting algorithm. Figure 7This is a schematic diagram of the principle of the ray casting algorithm. The ray casting algorithm simulates light starting from the viewpoint, passing through the screen into the volume data, sampling the volume data, obtaining the color value and opacity value of each sampling point (given by the fused OTF mapping), and then synthesizing these sampling points according to certain synthesis rules to obtain the color value of the pixel point on the screen. When all the pixels on the screen are calculated, a complete volume rendering overall surface map is obtained. At present, there are mainly two different synthesis algorithms: synthesis from front to back and synthesis from back to front. This embodiment adopts the synthesis algorithm from front to back, and its synthesis formula is as follows:

[0077] c out *α out =c in *α in +c now *α now *(1-α in )

[0078] α out =α in +α now *(1-α in )

[0079] Among them, c now Indicates the color value of the current voxel, α now Indicates the opacity value of the current voxel; c in Represents the color accumulation value before the projection ray enters the voxel, α in Represents the accumulated opacity value before the ray is cast into the voxel. out Represents the color accumulation value after the projection ray enters the voxel, α out Represents the accumulated opacity value after the cast ray enters the voxel.

[0080] Preferably, it also includes:

[0081] Adjusting the correction curve according to the volume rendering result, and fusing the adjusted correction curve into the original opacity transfer function to obtain a re-calibrated opacity transfer function;

[0082] The volume rendering is re-performed based on the recalibrated opacity transfer function.

[0083] In this embodiment, after obtaining the three-dimensional volume rendering surface map, the parameters of the calibration curve (mainly d and λ) can be readjusted according to the rendering effect to obtain a new volume rendering result. By testing the parameters of the calibration curve, an opacity transfer function that meets the set conditions is obtained, based on which the spatial structure and position morphological characteristics of the three-dimensional volume data can be better observed, so as to clearly observe the desired internal structure information, thereby providing more help for ultrasound doctors to make a quick and accurate diagnosis.

[0084] After completing the volume rendering and obtaining the volume rendering surface map, the brightness, contrast, pseudo color, etc. of the volume rendering surface map are adjusted to further enhance the three-dimensional visualization effect and help ultrasound doctors make diagnoses.

[0085] Example 2

[0086] Embodiment 2 of the present invention provides an ultrasonic three-dimensional volume rendering device, including a processor and a memory, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the ultrasonic three-dimensional volume rendering method provided in embodiment 1 is implemented.

[0087] The ultrasonic 3D volume rendering device provided in the embodiment of the present invention is used to implement the ultrasonic 3D volume rendering method. Therefore, the ultrasonic 3D volume rendering device also has the technical effects of the ultrasonic 3D volume rendering method, which will not be described in detail here.

[0088] Example 3

[0089] Embodiment 3 of the present invention provides a computer storage medium on which a computer program is stored. When the computer program is executed by a processor, the ultrasonic three-dimensional volume rendering method provided in Embodiment 1 is implemented.

[0090] The computer storage medium provided in the embodiment of the present invention is used to implement the ultrasonic three-dimensional volume rendering method. Therefore, the computer storage medium also has the technical effects of the ultrasonic three-dimensional volume rendering method, which will not be described in detail here.

[0091] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. An ultrasonic three-dimensional volume rendering method, characterized in that: The following steps are involved: Acquiring original volume data, preprocessing and reconstructing the volume data to obtain reconstructed volume data; Get the original opacity transfer function; A correction curve that varies with grayscale values ​​is set, and the correction curve is merged into the original opacity transfer function to obtain a corrected opacity transfer function, wherein the correction curve that varies with grayscale values ​​is set as follows: f(x)=(1+λ) g(x) ; Where f(x) is the correction curve function, x is the gray value, λ is the control parameter, λ∈(-1,1), which is used to control the amplitude of f(x), and g(x) is the bias function. I H is the mean gray intensity corresponding to the structural organization in the volume data, I L is the mean gray intensity corresponding to uniform tissue in the volume data, a, b, c are constant parameters, e is a natural constant, and d is an empirical value; The calibration curve is obtained by adjusting the d value and the λ value in the calibration curve function; The correction curve is merged into the original opacity transfer function to obtain a corrected opacity transfer function, specifically: Multiplying the correction curve by the opacity value corresponding to the original opacity transfer function at the same grayscale value to obtain a corrected opacity transfer function; Volume rendering is performed based on the corrected opacity transfer function.

2. The ultrasonic three-dimensional volume rendering method according to claim 1, characterized in that: Get the original volume data, specifically: The original volume data obtained by three-dimensional scanning according to the set probe scanning parameters is obtained.

3. The ultrasonic three-dimensional volume rendering method according to claim 1, characterized in that: The volume data is preprocessed and reconstructed to obtain reconstructed volume data, specifically: Performing filtering and smoothing processing on the volume data to obtain preprocessed volume data; A reconstruction table required for volume reconstruction is calculated according to the set probe scanning parameters, and the preprocessed volume data is reconstructed by looking up the reconstruction table to obtain reconstructed volume data.

4. The ultrasonic three-dimensional volume rendering method according to claim 1, characterized in that: Volume rendering is performed based on the corrected opacity transfer function, specifically: Volume rendering is performed using a ray casting algorithm based on the corrected opacity transfer function.

5. The ultrasonic three-dimensional volume rendering method according to claim 1, characterized in that: Volume rendering is performed using a ray casting algorithm based on the corrected opacity transfer function, specifically: Calculate the opacity value of each pixel based on the corrected opacity transfer function; The ray casting algorithm is used to sample each pixel to obtain the color value and opacity value of each sampling point; The color value and opacity value of each sampling point are synthesized according to the synthesis rule to obtain the color value of the corresponding pixel point on the screen; Combining the color values ​​of each pixel on the screen, the overall volume rendering image is obtained.

6. The ultrasonic three-dimensional volume rendering method according to claim 1, characterized in that: Also includes: Adjusting the correction curve according to the volume rendering result, and fusing the adjusted correction curve into the original opacity transfer function to obtain a re-calibrated opacity transfer function; The volume rendering is re-performed based on the recalibrated opacity transfer function.

7. An ultrasonic three-dimensional volume rendering device, characterized in that: The method comprises a processor and a memory, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the ultrasonic three-dimensional volume rendering method according to any one of claims 1 to 6 is implemented.

8. A computer storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the ultrasonic three-dimensional volume rendering method according to any one of claims 1 to 6 is implemented.

Images