Three-dimensional positioning method and system for ultrasonic medical diagnosis position

Abstract

The invention discloses a three-dimensional positioning method and system for an ultrasonic medical diagnosis position. Ultrasonic images are obtained through different ultrasonic positions and ultrasonic angles. And detecting the features of each ultrasonic image to obtain an ultrasonic feature map containing the information of the detected object. And converting the gray value of the ultrasonic image into the distance from the detection object. And constructing a three-dimensional position matrix which can represent the distance by using positions and comprises ultrasonic features. A three-dimensional position matrix is used to detect features of diagonals, features of a plurality of ultrasonic positions and features of a plurality of ultrasonic angles, as values for determining whether a detection diagnosis position is accurate. And the features of the diagonals are fused when detecting the diagnostic position.

Description

Technical Field

[0001] The present invention relates to the field of ultrasonic medical technology. Specifically, it relates to a method and system for three-dimensional positioning of ultrasonic medical diagnosis positions. Background Art

[0002] Ultrasonic examination is a medical imaging diagnosis technology based on ultrasonic waves, which can visualize human tissues to detect lesions in human tissues. When using an ultrasonic probe for ultrasonic examination, generally only two-dimensional ultrasonic images can be obtained. Screening for human lesions through two-dimensional ultrasonic images mainly depends on the experience and techniques of ultrasonic doctors. For two-dimensional ultrasonic images, if the ultrasonic examination fails to achieve the expected effect due to factors such as lack of experience of ultrasonic doctors. Therefore, three-dimensional reconstruction of two-dimensional ultrasonic images can provide more intuitive, rich, and comprehensive diagnostic information for the ultrasonic examination site, improving the disease diagnosis level of ultrasonic examination. Having spatial three-dimensional pose information for two-dimensional ultrasonic images is a necessary condition for realizing three-dimensional reconstruction of two-dimensional ultrasonic images.

[0003] Currently, the three-dimensional positioning of ultrasonic probes mainly relies on multiple external sensor devices for positioning, and the positioning sensor devices are expensive and bulky. Or the three-dimensional positioning method for converting two-dimensional ultrasonic images into three-dimensional ultrasonic images is adopted, and the method in the conversion process and the secondary detection of the detection results are not carried out, resulting in inaccurate discrimination. Summary of the Invention

[0004] The purpose of the present invention is to provide a method and system for three-dimensional positioning of ultrasonic medical diagnosis positions to solve the above problems existing in the prior art.

[0005] In a first aspect, an embodiment of the present invention provides a method for three-dimensional positioning of ultrasonic medical diagnosis positions, including:

[0006] Obtaining a plurality of ultrasonic plane images, corresponding ultrasonic positions, and ultrasonic angles; the ultrasonic position is the position of the ultrasonic device; the ultrasonic angle represents the angle between the ultrasonic device and the detection object; the plurality of ultrasonic plane images represent ultrasonic images of a detection object at multiple angles and multiple positions;

[0007] Detecting the features of the ultrasonic plane images to obtain an ultrasonic feature map; the ultrasonic feature map represents the correlation between the gray value of the ultrasonic plane image and the length from the ultrasonic device; a plurality of ultrasonic feature maps are obtained corresponding to the plurality of ultrasonic plane images;

[0008] Based on the ultrasonic position and ultrasonic angle, constructing the plurality of ultrasonic feature maps three-dimensionally to obtain a three-dimensional ultrasonic feature matrix;

[0009] Based on a three-dimensional ultrasonic feature matrix, detect the features at multiple ultrasonic positions and multiple ultrasonic angles to obtain a diagnostic position; the diagnostic position represents the three-dimensional position of the detection object.

[0010] Optionally, the method of constructing a three-dimensional ultrasonic feature matrix based on the ultrasonic positions and ultrasonic angles includes:

[0011] Detect the correlation relationship in the ultrasonic feature map based on the ultrasonic feature map to obtain a three-dimensional position matrix;

[0012] Multiple three-dimensional position matrices are obtained corresponding to multiple ultrasonic feature maps;

[0013] Based on the multiple three-dimensional position matrices, perform mapping at different ultrasonic positions and ultrasonic angles to obtain a three-dimensional ultrasonic feature matrix.

[0014] Optionally, the method of detecting the features at multiple ultrasonic positions and multiple ultrasonic angles based on the three-dimensional ultrasonic feature matrix to obtain a diagnostic position includes:

[0015] Obtain a 3*3*3 first three-dimensional convolutional kernel;

[0016] Obtain 4 diagonals in the three-dimensional ultrasonic feature matrix;

[0017] With a step size of 1, after matching the center point of the first three-dimensional convolutional kernel with the positions on the diagonal, perform convolution on the three-dimensional ultrasonic feature matrix to obtain diagonal convolution features;

[0018] 4 diagonal convolution features are obtained corresponding to 4 diagonals;

[0019] Based on the 4 diagonal convolution features and the three-dimensional ultrasonic feature matrix, obtain a diagnostic position.

[0020] Optionally, the method of obtaining a diagnostic position based on the 4 diagonal convolution features and the three-dimensional ultrasonic feature matrix includes:

[0021] Based on the 4 diagonal convolution features, obtain an accurate value of the diagnostic position;

[0022] If the accurate value of the diagnostic position is greater than the diagnostic threshold, perform convolution based on the diagonal convolution features and the three-dimensional ultrasonic feature matrix to obtain a second feature, and obtain a diagnostic position.

[0023] Optionally, the method of obtaining an accurate value of the diagnostic position based on the 4 diagonal convolution features includes:

[0024] Fuse the 4 diagonal convolution features to obtain a first fusion feature;

[0025] Obtain a first ultrasonic plane image and a second ultrasonic plane image; the first ultrasonic plane image and the second ultrasonic plane image are two different ultrasonic plane images with an included angle difference less than 90 degrees;

[0026] Take the points at the overlapping positions in the first ultrasonic plane image and the second ultrasonic plane image as labeled detection points;

[0027] Input the first fusion feature into the first neural network to obtain predicted detection points;

[0028] Take the shortest distance between the labeled detection points and the predicted detection points as the accurate value of the diagnosis position.

[0029] Optionally, if the accurate value of the diagnosis position is greater than the diagnosis threshold, based on the diagonal convolution feature and the three-dimensional ultrasonic feature matrix, perform convolution to obtain a second feature and obtain the diagnosis position, including:

[0030] Obtain a 3*3*3 second three-dimensional convolution kernel;

[0031] With a stride of 1, perform convolution of the second three-dimensional convolution kernel on the three-dimensional ultrasonic feature matrix to obtain ultrasonic convolution features;

[0032] Extract the diagonal values of the ultrasonic convolution features to obtain a plurality of second diagonal convolution features; one diagonal corresponds to one diagonal convolution feature and one second diagonal convolution feature;

[0033] Fuse the diagonal convolution feature with the corresponding second diagonal convolution feature to obtain a fused diagonal feature;

[0034] Replace the corresponding diagonal values in the ultrasonic convolution features with the fused diagonal features to obtain fused ultrasonic convolution features;

[0035] Input the fused ultrasonic convolution features into a three-dimensional convolution network to obtain the diagnosis position.

[0036] Optionally, based on the ultrasonic feature map, detecting the correlation relationship in the ultrasonic feature map to obtain a three-dimensional position matrix includes:

[0037] Input the ultrasonic feature map into a distance detection network to detect the distance between each pixel point of each ultrasonic plane image and the detection object, and obtain a detection distance matrix;

[0038] Construct an initial three-dimensional matrix; the number of rows of the initial three-dimensional matrix corresponds to the length of the ultrasonic feature map; the number of columns of the initial three-dimensional matrix corresponds to the width of the ultrasonic feature map; the number of pages of the initial three-dimensional matrix is a fixed value; the fixed value is less than the minimum difference in the detection distance matrix;

[0039] The initial value of the initial three-dimensional matrix is 0;

[0040] Divide the difference between the maximum value and the minimum value in the detection distance matrix by a fixed value to obtain the page distance length; the page distance length represents the range of the actual distances included in the page represented by 1.

[0041] Divide the difference between the value in the detection distance matrix and the minimum value by the page distance length to obtain the detection page number; a position in the detection distance matrix corresponds to a detection page number.

[0042] Find the position corresponding to the ultrasonic feature map in the initial three-dimensional matrix, and set the value corresponding to the detection page number to the value at the position corresponding to the ultrasonic feature map to obtain a three-dimensional position matrix.

[0043] Optionally, mapping based on the multiple three-dimensional position matrices at different ultrasonic positions and ultrasonic angles to obtain a three-dimensional ultrasonic feature matrix includes:

[0044] Construct a position coordinate; the position coordinate is a three-dimensional matrix.

[0045] According to the ultrasonic position and ultrasonic angle, map the values in the three-dimensional position matrix into the position coordinate to obtain multiple mapped three-dimensional positions.

[0046] Divide the multiple mapped three-dimensional positions into a position three-dimensional ultrasonic feature matrix according to the axes parallel to the x-axis, y-axis, and z-axis of the position coordinate.

[0047] Optionally, detecting the features of the detected ultrasonic plane image to obtain an ultrasonic feature map includes:

[0048] Obtain a 2×2 two-dimensional convolution kernel.

[0049] Convolve the two-dimensional convolution kernel on the ultrasonic plane image with a step size of 1 to obtain ultrasonic plane features.

[0050] In a second aspect, an embodiment of the present invention provides a three-dimensional positioning system for ultrasonic medical diagnosis positions, including:

[0051] An acquisition module, configured to acquire a plurality of ultrasonic plane images, corresponding ultrasonic positions, and ultrasonic angles; the ultrasonic position is the position of an ultrasonic device; the ultrasonic angle represents the angle between the ultrasonic device and a detection object; the plurality of ultrasonic plane images represent ultrasonic images of a detection object at multiple angles and multiple positions.

[0052] A feature extraction module, configured to detect the features of the ultrasonic plane image to obtain an ultrasonic feature map; the ultrasonic feature map represents the correlation between the gray value of the ultrasonic plane image and the length from the ultrasonic device; a plurality of ultrasonic plane images correspondingly obtain a plurality of ultrasonic feature maps.

[0053] A three-dimensional module for constructing a three-dimensional structure from multiple ultrasonic feature maps based on the ultrasonic position and ultrasonic angle to obtain a three-dimensional ultrasonic feature matrix;

[0054] A detection module for detecting features at multiple ultrasonic positions and multiple ultrasonic angles based on the three-dimensional ultrasonic feature matrix to obtain a diagnostic position; the diagnostic position represents the three-dimensional position of the detection object.

[0055] Compared with the prior art, the embodiments of the present invention achieve the following beneficial effects:

[0056] The embodiments of the present invention also provide a method and a system for three-dimensional positioning of an ultrasonic medical diagnosis position.

[0057] In the present invention, ultrasonic images at different ultrasonic positions and ultrasonic angles are used. The features of each ultrasonic image are detected to obtain ultrasonic feature maps containing information of the detection object. The gray values of the ultrasonic images are converted into distances from the detection object. A three-dimensional position matrix that can represent the distances with positions and contains ultrasonic features is constructed. It is achieved that the distances from the detection object and the features of the detection object reflected by the ultrasonic waves are represented by the feature three-dimensional position matrix. Thus, the features of the diagonal are detected using the three-dimensional position matrix, and the predicted values of two points that are different on the two-dimensional ultrasonic plane images respectively but the same on the three-dimensional position coordinates are obtained as the values for judging whether the detected diagnostic position is accurate. And when detecting the diagnostic position, the features of the diagonal are fused, so that the obtained diagnostic position is more accurate, and thus the three-dimensional positioning is more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 is a flowchart of a method for three-dimensional positioning of an ultrasonic medical diagnosis position provided by an embodiment of the present invention.

[0059] Figure 2 is a convolution schematic diagram of a first three-dimensional convolution kernel in a method for three-dimensional positioning of an ultrasonic medical diagnosis position provided by an embodiment of the present invention.

[0060] Figure 3 is a schematic diagram of position coordinates in a method for three-dimensional positioning of an ultrasonic medical diagnosis position provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0061] The present invention will be described in detail below with reference to the accompanying drawings.

[0062] Embodiment 1

[0063] As Figure 1 shown, an embodiment of the present invention provides a method for three-dimensional positioning of an ultrasonic medical diagnosis position, and the method includes:

[0064] S101: Obtain multiple ultrasonic plane images, corresponding ultrasonic positions, and ultrasonic angles; the ultrasonic position is the position of the ultrasonic device; the ultrasonic angle represents the angle between the ultrasonic device and the detection object; the multiple ultrasonic plane images represent ultrasonic images of a detection object at multiple angles and multiple positions.

[0065] Among them, the ultrasonic device in this embodiment is a handheld ultrasonic device, and the detection object is a human body that has undergone ultrasonic detection in history.

[0066] Among them, in this embodiment, the angle between the two planes formed by the plane emitting ultrasonic waves in the ultrasonic device and the front of the detection object is used as the ultrasonic angle.

[0067] Among them, in this embodiment, the position of the ultrasonic device on the geographical position coordinates is used as the ultrasonic position. The relationship between the time length and the actual distance can be known from the ultrasonic ranging formula.

[0068] Among them, the ultrasonic plane image is an image reflected back when ultrasonic waves encounter an obstacle.

[0069] S102: Detect the features of the ultrasonic plane image to obtain an ultrasonic feature map; the ultrasonic feature map represents the correlation between the gray value of the ultrasonic plane image and the length from the ultrasonic device; multiple ultrasonic plane images correspond to obtain multiple ultrasonic feature maps.

[0070] S103: Based on the ultrasonic position and ultrasonic angle, construct the multiple ultrasonic feature maps into three dimensions to obtain a three-dimensional ultrasonic feature matrix;

[0071] S104: Based on the three-dimensional ultrasonic feature matrix, detect the features at multiple ultrasonic positions and multiple ultrasonic angles to obtain a diagnosis position; the diagnosis position represents the three-dimensional position of the detection object.

[0072] Optionally, the step of constructing the multiple ultrasonic feature maps into three dimensions based on the ultrasonic position and ultrasonic angle to obtain a three-dimensional ultrasonic feature matrix includes:

[0073] Based on the ultrasonic feature map, detect the correlation in the ultrasonic feature map to obtain a three-dimensional position matrix.

[0074] Among them, the three-dimensional position matrix is used to extract the feature of the distance in the ultrasonic plane image.

[0075] Multiple ultrasonic feature maps correspond to obtain multiple three-dimensional position matrices.

[0076] Based on the multiple three-dimensional position matrices, perform mapping at different ultrasonic positions and ultrasonic angles to obtain a three-dimensional ultrasonic feature matrix.

[0077] Optionally, detecting features at multiple ultrasound positions and multiple ultrasound angles based on the three-dimensional ultrasound feature matrix to obtain a diagnostic position includes:

[0078] Obtain a 3×3×3 first three-dimensional convolution kernel.

[0079] Obtain 4 diagonals in the three-dimensional ultrasound feature matrix;

[0080] With a step size of 1, after matching the center point of the first three-dimensional convolution kernel with the positions on the diagonal, perform convolution on the three-dimensional ultrasound feature matrix to obtain diagonal convolution features.

[0081] Among them, the convolution of the first three-dimensional convolution kernel is as Figure 2 shown.

[0082] 4 diagonals correspond to obtaining 4 diagonal convolution features;

[0083] Based on the 4 diagonal convolution features and the three-dimensional ultrasound feature matrix, obtain the diagnostic position.

[0084] Optionally, obtaining the diagnostic position based on the 4 diagonal convolution features and the three-dimensional ultrasound feature matrix includes:

[0085] Based on the 4 diagonal convolution features, obtain an accurate diagnostic position value; the accurate diagnostic position value represents the accuracy of diagnostic position detection;

[0086] If the accurate diagnostic position value is greater than the diagnostic threshold, based on the diagonal convolution features and the three-dimensional ultrasound feature matrix, perform convolution to obtain a second feature and obtain the diagnostic position.

[0087] Among them, in this embodiment, the diagnostic threshold is 0.9.

[0088] Optionally, obtaining the accurate diagnostic position value based on the 4 diagonal convolution features includes:

[0089] Fuse the 4 diagonal convolution features to obtain a first fusion feature.

[0090] Among them, in this embodiment, fusion is performed by averaging corresponding positions;

[0091] Obtain a first ultrasound plane image and a second ultrasound plane image; the first ultrasound plane image and the second ultrasound plane image are two different ultrasound plane images with an included angle less than 90 degrees.

[0092] Use the points at the overlapping positions of the first ultrasound plane image and the second ultrasound plane image as labeled detection points;

[0093] Input the first fusion feature into a first neural network to obtain predicted detection points.

[0094] Among them, the first neural network is a (Fully Connected Neural Network, FCNN). The first neural network has only one output neuron.

[0095] Take the shortest distance between the labeled detection point and the predicted detection point as the accurate diagnosis position value.

[0096] Among them, in this embodiment, the Euclidean distance is used to calculate the shortest distance.

[0097] Optionally, if the accurate diagnosis position value is greater than the diagnosis threshold, based on the diagonal convolution feature and the three-dimensional ultrasound feature matrix, perform convolution to obtain a second feature and obtain the diagnosis position, including:

[0098] Obtain a 3×3×3 second three-dimensional convolution kernel;

[0099] With a stride of 1, perform convolution of the second three-dimensional convolution kernel on the three-dimensional ultrasound feature matrix to obtain an ultrasound convolution feature; the ultrasound convolution feature represents the feature of the three-dimensional position image corresponding to the constructed three-dimensional ultrasound feature matrix.

[0100] Among them, three-dimensional convolutional neural network (3D Convolutional Neural Networks, CNN) is used for convolution.

[0101] Extract the diagonal values of the ultrasound convolution feature to obtain a plurality of second diagonal convolution features; one diagonal corresponds to one diagonal convolution feature corresponding to one second diagonal convolution feature;

[0102] Fuse the diagonal convolution feature with the corresponding second diagonal convolution feature to obtain a fused diagonal feature.

[0103] Among them, in this embodiment, the method of averaging the corresponding positions is used for fusion.

[0104] Replace the diagonal values in the ultrasound convolution feature with the fused diagonal feature to obtain a fused ultrasound convolution feature;

[0105] Input the fused ultrasound convolution feature into a three-dimensional convolutional network to obtain the diagnosis position.

[0106] Among them, the third convolutional network is a three-dimensional convolutional neural network (3D Convolutional Neural Networks, CNN).

[0107] Among them, in this embodiment, the diagnosis position is the smallest target box containing the detection object, including the center point, length, width, and height of the detection object.

[0108] Optionally, detecting the internal depth based on the ultrasonic feature map to obtain a three-dimensional position matrix includes:

[0109] Inputting the ultrasonic feature map into a distance detection network to detect the distance between each pixel point of each ultrasonic plane image and the detection object, thereby obtaining a detection distance matrix.

[0110] Among them, the distance detection network is a deconvolutional convolutional neural network (Convolutional Neural Networks, CNN).

[0111] Construct an initial three-dimensional matrix; the number of rows of the initial three-dimensional matrix corresponds to the length of the ultrasonic feature map; the number of columns of the initial three-dimensional matrix corresponds to the width of the ultrasonic feature map; the number of pages of the initial three-dimensional matrix is a fixed value; the fixed value is less than the minimum difference in the detection distance matrix.

[0112] Among them, the fixed value represents setting the initial three-dimensional matrix with a fixed number of pages according to the method that one page only contains one value in the detection distance matrix.

[0113] The initial value of the initial three-dimensional matrix is 0.

[0114] Among them, the initial value of 0 represents the distance between each position of the unknown detection object and the ultrasonic device.

[0115] Subtract the minimum value from the maximum value in the detection distance matrix and divide the result by the fixed value to obtain the page distance length; the page distance length represents the range of the actual distance included in a page with a value of 1.

[0116] Among them, if the fixed value is 10, the minimum detected value is 4 cm, the maximum value is 6 cm, and the page distance length is 2 mm.

[0117] Subtract the minimum value from the value in the detection distance matrix and divide the result by the page distance length to obtain the detection page number; one position in the detection distance matrix corresponds to one detection page number.

[0118] Among them, in this embodiment, the subscript of the number of pages of the initial three-dimensional matrix starts from 0.

[0119] Among them, if the value in the detection distance matrix is 4.3 cm, then (4.3 - 4) / 0.2 = 1, and set the page number corresponding to the subscript 1 in the initial three-dimensional matrix as the detection page number.

[0120] Find the position corresponding to the ultrasonic feature map in the initial three-dimensional matrix, and set the value corresponding to the detection page number as the value at the position corresponding to the ultrasonic feature map, thereby obtaining a three-dimensional position matrix.

[0121] Among them, for behavior 1 with a value of 4.3 cm in the detection distance matrix and column 2, find the position in the initial three-dimensional matrix where the row is 1, the column is 2, and the page number is 1 and set it to 1.

[0122] Optionally, based on the multiple three-dimensional position matrices, mapping is performed at different ultrasonic positions and ultrasonic angles to obtain a three-dimensional ultrasonic feature matrix, including:

[0123] Construct position coordinates.

[0124] Among them, in this embodiment, the position coordinates represent the coordinates of the position of the detection object relative to the actual geographical location. The position coordinates are as Figure 3 shown.

[0125] According to the x-axis, y-axis, and z-axis parallel to the position coordinates, divide the multiple mapped three-dimensional position segmentation position three-dimensional ultrasonic feature matrix.

[0126] Convert the three-dimensional position point cloud into a three-dimensional ultrasonic feature matrix.

[0127] Optionally, detecting the features of the detected ultrasonic plane image to obtain an ultrasonic feature map, including:

[0128] Obtain a 2×2 two-dimensional convolution kernel;

[0129] With a step size of 1, perform convolution of the two-dimensional convolution kernel on the ultrasonic plane image to obtain ultrasonic plane features.

[0130] Among them, convolution is performed using a Convolutional Neural Networks (CNN).

[0131] Embodiment 2

[0132] Based on the above method for three-dimensional positioning of ultrasonic medical diagnosis positions, an embodiment of the present invention further provides a three-dimensional positioning system for ultrasonic medical diagnosis positions. The system includes an acquisition module, a feature extraction module, a three-dimensional module & a detection module.

[0133] The acquisition module is used to acquire multiple ultrasonic plane images, corresponding ultrasonic positions, and ultrasonic angles; the ultrasonic position is the position of the ultrasonic device; the ultrasonic angle represents the angle between the ultrasonic device and the detection object; the multiple ultrasonic plane images represent ultrasonic images of a detection object at multiple angles and multiple positions.

[0134] The feature extraction module is used to detect the features of the ultrasonic plane image to obtain an ultrasonic feature map; the ultrasonic feature map represents the correlation between the gray value of the ultrasonic plane image and the length from the ultrasonic device; multiple ultrasonic plane images correspond to obtain multiple ultrasonic feature maps.

[0135] A three-dimensional module for constructing a three-dimensional structure from multiple ultrasonic feature maps based on the ultrasonic position and ultrasonic angle to obtain a three-dimensional ultrasonic feature matrix.

[0136] A detection module for detecting features at multiple ultrasonic positions and multiple ultrasonic angles based on the three-dimensional ultrasonic feature matrix to obtain a diagnosis position; the diagnosis position represents the three-dimensional position of the detection object.

[0137] The algorithms and displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used in conjunction with the teachings provided herein. The structure required to construct such a system will be apparent from the above description. Additionally, the present invention is not directed to any particular programming language. It should be understood that the teachings of the present invention can be implemented in various programming languages, and the description of a particular language above is for the purpose of disclosing the best mode of the present invention.

[0138] In the specification provided herein, a large number of specific details are set forth. However, it can be understood that embodiments of the present invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0139] Each component embodiment of the present invention can be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art should understand that a microprocessor or a digital signal processor (DSP) can be used in practice to implement some or all of the functions of some or all of the components of the device according to the embodiments of the present invention. The present invention can also be implemented as a device or device program (e.g., a computer program and a computer program product) for performing part or all of the methods described herein. Such a program implementing the present invention can be stored on a computer-readable medium, or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, or provided on a carrier signal, or in any other form.

Claims

1. A method for three-dimensional positioning of ultrasonic medical diagnosis position, characterized in that: include: Acquire multiple ultrasonic plane images and corresponding ultrasonic positions and ultrasonic angles; the ultrasonic position is the position of the ultrasonic device; the ultrasonic angle represents the angle between the ultrasonic device and the detection object; the multiple ultrasonic plane images represent ultrasonic images of a detection object at multiple angles and multiple positions; Detecting the characteristics of the ultrasonic plane image to obtain an ultrasonic characteristic graph; the ultrasonic characteristic graph represents the correlation between the grayscale value of the ultrasonic plane image and the distance from the ultrasonic device; and obtaining multiple ultrasonic characteristic graphs corresponding to multiple ultrasonic plane images; Based on the ultrasound position and the ultrasound angle, a plurality of ultrasound characteristic graphs are constructed into three dimensions to obtain a three-dimensional ultrasound characteristic matrix; Based on the three-dimensional ultrasound feature matrix, the features of multiple ultrasound positions and multiple ultrasound angles are detected to obtain a diagnosis position; the diagnosis position represents the three-dimensional position of the detection object.

2. The method for three-dimensional positioning of ultrasonic medical diagnosis position according to claim 1, characterized in that: The method constructs a three-dimensional ultrasonic feature map based on the ultrasonic position and the ultrasonic angle to obtain a three-dimensional ultrasonic feature matrix, including: Based on the ultrasonic characteristic graph, detecting the correlation relationship in the ultrasonic characteristic graph to obtain a three-dimensional position matrix; Multiple three-dimensional position matrices are obtained corresponding to multiple ultrasonic characteristic images; Based on the multiple three-dimensional position matrices, mapping is performed at different ultrasound positions and ultrasound angles to obtain a three-dimensional ultrasound feature matrix.

3. The method for three-dimensional positioning of ultrasonic medical diagnosis position according to claim 1, characterized in that: The method of detecting the features of multiple ultrasound positions and multiple ultrasound angles based on the three-dimensional ultrasound feature matrix to obtain the diagnosis position includes: Get the first three-dimensional convolution kernel of 3*3*3; Acquire four diagonal lines in the three-dimensional ultrasound feature matrix; With a step size of 1, after matching the center point of the first three-dimensional convolution kernel with the position on the diagonal line, convolution is performed on the three-dimensional ultrasound feature matrix to obtain a diagonal convolution feature; The 4 diagonal lines correspond to 4 diagonal convolution features; Based on the four diagonal convolution features and the three-dimensional ultrasound feature matrix, the diagnostic position is obtained.

4. The method for three-dimensional positioning of ultrasonic medical diagnosis position according to claim 3, characterized in that: The diagnostic position is obtained based on the four diagonal convolution features and the three-dimensional ultrasound feature matrix, including: Based on the four diagonal convolution features, the accurate value of the diagnosis position is obtained; If the diagnostic position accuracy value is greater than the diagnostic threshold, convolution is performed based on the diagonal convolution feature and the three-dimensional ultrasound feature matrix to obtain a second feature and obtain the diagnostic position.

5. The method for three-dimensional positioning of ultrasonic medical diagnosis position according to claim 4, characterized in that: Based on the four diagonal convolution features, the accurate value of the diagnosis position is obtained, including: The four diagonal convolution features are fused to obtain the first fused feature; Acquire a first ultrasonic plane image and a second ultrasonic plane image; the first ultrasonic plane image and the second ultrasonic plane image are two different ultrasonic plane images with an angle difference of less than 90 degrees; Using points at overlapping positions in the first ultrasonic plane image and the second ultrasonic plane image as marked detection points; Inputting the first fusion feature into the first neural network to obtain a predicted detection point; The shortest distance between the marked detection point and the predicted detection point is taken as the diagnostic position accuracy value.

6. The method for three-dimensional positioning of ultrasonic medical diagnosis position according to claim 4, characterized in that: If the diagnostic position accuracy value is greater than the diagnostic threshold, convolution is performed based on the diagonal convolution feature and the three-dimensional ultrasound feature matrix to obtain a second feature and obtain the diagnostic position, including: Get the second three-dimensional convolution kernel of 3*3*3; Convolving the second three-dimensional convolution kernel on the three-dimensional ultrasonic feature matrix with a step size of 1 to obtain ultrasonic convolution features; Extract the diagonal value of the ultrasonic convolution feature to obtain multiple second diagonal convolution features; one diagonal line corresponds to one diagonal convolution feature and one second diagonal convolution feature; The diagonal convolution feature is fused with the corresponding second diagonal convolution feature to obtain a fused diagonal feature; The fused diagonal features are used to replace the corresponding diagonal values ​​in the ultrasonic convolution features to obtain the fused ultrasonic convolution features; The fused ultrasonic convolution features are input into a three-dimensional convolutional network to obtain a diagnosis position.

7. The method for three-dimensional positioning of ultrasonic medical diagnosis position according to claim 2, characterized in that: The detecting the correlation relationship in the ultrasonic characteristic graph based on the ultrasonic characteristic graph to obtain a three-dimensional position matrix includes: According to inputting the ultrasonic characteristic graph into the distance detection network, the distance between each pixel point of each ultrasonic plane image and the detection object is detected to obtain a detection distance matrix; Constructing an initial three-dimensional matrix; the number of rows of the initial three-dimensional matrix corresponds to the length of the ultrasonic characteristic graph; the number of columns of the initial three-dimensional matrix corresponds to the width of the ultrasonic characteristic graph; the number of pages of the initial three-dimensional matrix is ​​a fixed value; the fixed value is less than the minimum difference in the detection distance matrix; The initial value of the initial three-dimensional matrix is ​​0; The difference between the maximum value and the minimum value in the detection distance matrix is ​​divided by a fixed value to obtain a page distance length; the page distance length represents the range of actual distances contained in a page of 1; The difference between the value in the detection distance matrix and the minimum value is divided by the page distance length to obtain the number of detection pages; one position in the detection distance matrix corresponds to one detection page number; The position corresponding to the ultrasonic characteristic graph is found in the initial three-dimensional matrix, and the value corresponding to the number of detected pages is set as the value of the position corresponding to the ultrasonic characteristic graph to obtain a three-dimensional position matrix.

8. The method for three-dimensional positioning of ultrasonic medical diagnosis position according to claim 2, characterized in that: The mapping is performed based on the multiple three-dimensional position matrices at different ultrasound positions and ultrasound angles to obtain a three-dimensional ultrasound feature matrix, including: Constructing position coordinates; the position coordinates are a three-dimensional matrix; According to the ultrasound position and the ultrasound angle, the values ​​in the three-dimensional position matrix are mapped in the position coordinates to obtain a plurality of mapped three-dimensional positions; According to the x-axis, y-axis and z-axis parallel to the position coordinates, the plurality of mapped three-dimensional positions are segmented into a three-dimensional ultrasonic feature matrix.

9. The method for three-dimensional positioning of ultrasonic medical diagnosis position according to claim 1, characterized in that: The detecting the characteristics of the ultrasonic plane image to obtain the ultrasonic characteristic graph includes: Get a 2*2 two-dimensional convolution kernel; The two-dimensional convolution kernel is convolved on the ultrasound plane image with a step size of 1 to obtain ultrasound plane features.

10. A three-dimensional positioning system for ultrasonic medical diagnosis, characterized in that: include: An acquisition module, used to acquire multiple ultrasonic plane images, and corresponding ultrasonic positions and ultrasonic angles; the ultrasonic position is the position of the ultrasonic device; the ultrasonic angle represents the angle between the ultrasonic device and the detection object; the multiple ultrasonic plane images represent ultrasonic images of a detection object at multiple angles and multiple positions; A feature extraction module is used to detect the features of the ultrasonic plane image and obtain an ultrasonic feature map; the ultrasonic feature map represents the correlation between the gray value of the ultrasonic plane image and the distance from the ultrasonic device; multiple ultrasonic feature maps are obtained corresponding to multiple ultrasonic plane images; A three-dimensional module, used to construct three dimensions of the plurality of ultrasonic feature maps based on the ultrasonic position and the ultrasonic angle to obtain a three-dimensional ultrasonic feature matrix; The detection module is used to detect the features of multiple ultrasound positions and multiple ultrasound angles based on a three-dimensional ultrasound feature matrix to obtain a diagnosis position; the diagnosis position represents the three-dimensional position of the detection object.

Images