A scanning method, system and storage medium for medical imaging equipment

By using binocular cameras and multi-angle position sensors to calculate three-dimensional human data in medical imaging equipment, the problem of error under the guidance of monocular cameras is solved, and a more accurate scanning and simplified operation process is achieved.

CN118542642BActive Publication Date: 2024-11-22ANALOGIC MEDICAL EQUIP (SHANGHAI) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410476433.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-04-19
Publication Date
2024-11-22
Estimated Expiration
2044-04-19

AI Technical Summary

Technical Problem

When using a monocular camera in existing medical imaging equipment to guide patients to adjust their positions, the two-dimensional image display causes physician errors, affecting the accuracy of the scan results.

Method used

Using binocular cameras and multi-angle position sensors, three-dimensional human body contour and position data are calculated through fitting algorithms, combined with the spatial data of the imaging equipment to generate scanning protocols, optimize scanning parameters, and realize one-stop human body positioning and scanning parameters optimization.

Benefits of technology

It reduces the error of physician guidance, improves the accuracy of scanning results, simplifies the operation process, and improves operation efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118542642B_ABST
    Figure CN118542642B_ABST
Patent Text Reader

Abstract

The present application relates to the field of medical imaging technology, and discloses a scanning method, system and storage medium of a medical imaging device, the method comprising: acquiring first image data and second image data; calculating first human body graphic data; identifying human body contour and generating three-dimensional human body contour data; acquiring first position data and second position data; calculating first human body position data; identifying human body features and generating human body feature position data; correcting the coordinate content of three-dimensional human body contour data based on human body feature position data; extracting scanning part data; generating a scanning area according to the scanning part data and the spatial data of the imaging device; generating a scanning protocol based on the confirmation information generated corresponding to the scanning area, and sending the scanning protocol to the imaging device; the imaging device performs scanning in response to the scanning protocol; reducing the error of the physician's guidance and improving the accuracy of the scanning result.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of medical imaging, and in particular to a scanning method, system and storage medium of a medical imaging device. Background Art

[0002] In the prior art, a monocular camera is added to the scanning room. The monocular camera captures images in the scanning room, and the doctor guides the patient to adjust the position or posture according to the images in the scanned image.

[0003] However, the existing monocular camera display solution can only simply present the image in the scanning room. The image is a two-dimensional picture. When doctors look at the two-dimensional picture for guidance, errors will occur, and these errors will affect the accuracy of the scanning results. Summary of the invention

[0004] In order to reduce the errors in physician guidance and improve the accuracy of scanning results, the present application provides a scanning method, system and storage medium of a medical imaging device.

[0005] In a first aspect, the present application provides a scanning method of a medical imaging device, which adopts the following technical solution:

[0006] A scanning method of a medical imaging device comprises the following steps:

[0007] Acquire first image data and second image data based on a first binocular camera and a second binocular camera beside a scanning bed;

[0008] Calculate first human body graphic data using a preset first fitting algorithm according to the first image data and the second image data;

[0009] Recognizing a human body contour from the first human body graphic data to generate three-dimensional human body contour data;

[0010] Acquiring first position data and second position data based on a first position sensor and a second position sensor beside a scanning bed;

[0011] Calculate first human body position data using a preset second fitting algorithm according to the first position data and the second position data;

[0012] Identifying human features from the first human position data to generate human feature position data;

[0013] Correcting the coordinate content of the three-dimensional human body contour data based on the human body feature position data;

[0014] Extracting scanned part data from the latest three-dimensional human body contour data;

[0015] Generate a scanning area according to the scanning part data and the spatial data of the imaging device;

[0016] generating a scanning protocol based on the confirmation information generated corresponding to the scanning area, and sending the scanning protocol to an imaging device;

[0017] The imaging device performs scanning in response to the scanning protocol.

[0018] By adopting the above technical solution, an external binocular camera is added to the traditional medical imaging equipment, and the three-dimensional human body contour data is identified from multiple angles using a first fitting algorithm based on the first image data and the second image data. The first position data and the second position data output by the multi-angle position sensor are then combined, and the first human body position data is calculated using a second fitting algorithm. The three-dimensional human body contour data is corrected to achieve multi-angle data fusion and form complete three-dimensional human body data. The three-dimensional human body data is then combined with the scanning workflow of the imaging equipment, and a scanning protocol is generated based on the confirmation information to achieve the specified scanning area and optimize the scanning parameters, so as to achieve one-stop human body positioning and scanning parameter optimization of the medical imaging equipment, reduce the error of physician guidance, and improve the accuracy of the scanning results.

[0019] Optionally, the first fitting algorithm comprises the following steps:

[0020] Extracting first contour data from the first image data, and extracting second contour data from the second image data; the first contour data includes first human body data and first device contour data, and the second contour data includes second human body data and second device contour data;

[0021] Calculating first deformation distribution data according to the first device contour data, adjusting the first human body data according to the first deformation distribution data, matching first template data from a preset first human body template set according to the adjusted first human body data, and calculating first fitting human body data according to the adjusted first human body data and the first template data;

[0022] Calculating second deformation distribution data according to the second device contour data, adjusting the second human body data according to the second deformation distribution data, matching second template data from a preset second human body template set according to the adjusted second human body data, and calculating second fitting human body data according to the adjusted second human body data and the second template data;

[0023] The first human body graphic data is calculated according to the first fitting human body data and the second fitting human body data by taking pixel values ​​of pixels at corresponding positions according to a weighted average algorithm.

[0024] By adopting the above technical solution, human body data and device contour data are extracted from the image data. The device contour data is relatively stable. The deformation of the human body data in the image data is obtained based on the deformation of the device contour data in the image data, and then the fitted human body data close to the actual one is calculated. The fitted human body data at two different angles are combined and calculated, which is conducive to calculating the first human body graphic data that is more in line with the actual one.

[0025] Optionally, the first fitting algorithm further comprises the following steps:

[0026] Calculating a difference between pixel values ​​of corresponding pixels in the first fitting human body data and the second fitting human body data;

[0027] Calculating the sum of all said differences;

[0028] The contrast of the first human body graphic data is adjusted according to the sum value, wherein the greater the sum value is, the greater the contrast is, and the smaller the sum value is, the smaller the contrast is.

[0029] By adopting the above technical solution, the greater the feature gap in the fitted human body data at the two angles, the greater the sum value, and it is necessary to highlight the contour features in the first human body graphic data by increasing the contrast; the smaller the feature gap, the smaller the sum value, and less contrast can be added to avoid losing some less obvious features; if the contour positions in the fitted human body data at the two angles are misaligned or offset in detail, the calculated sum value will not be too large, reducing the impact of the misalignment and offset on the contrast.

[0030] Optionally, the first fitting algorithm further comprises the following steps:

[0031] Calculating a difference between pixel values ​​of corresponding pixels in the first fitting human body data and the second fitting human body data;

[0032] Calculate the sum of the absolute values ​​of all the differences;

[0033] The weight of the first template data and the weight of the second template data in the weighted average algorithm are adjusted according to the sum value. The larger the sum value is, the smaller the weight of the first template data and the weight of the second template data are. The smaller the sum value is, the larger the weight of the first template data and the weight of the second template data are.

[0034] By adopting the above technical solution, by calculating the sum of absolute values, the feature difference in the fitted human body data of the two angles can be reflected, and the detailed misalignment or offset of the contour position in the fitted human body data of the two angles can be magnified. It is more likely that misalignment or offset will occur between the template data and the actual contour data, so that the weight of the template data can be adjusted according to the actual situation to make the calculated data more realistic.

[0035] Optionally, the first position sensor and the second position sensor are ultrasonic sensors arranged in a linear array; or, the first position sensor and the second position sensor are laser sensors having visible light spots and arranged in a linear array, and the visible light spots of the linear array are projected on the side of the human body;

[0036] The second fitting algorithm comprises the following steps:

[0037] extracting first boundary data from the first position data, and extracting second boundary data from the second position data;

[0038] Matching first edge data from a preset first boundary template set according to the first boundary data, and calculating first fitting boundary data according to the first boundary data and the first edge data;

[0039] Matching second edge data from a preset second boundary template set according to the second boundary data, and calculating second fitting boundary data according to the second boundary data and the second edge data;

[0040] The first human body position data is calculated according to the first fitting boundary data and the second fitting boundary data through the position values ​​of corresponding points according to a weighted average algorithm.

[0041] By adopting the above technical solution, position data is obtained through a position sensor, boundary data is extracted from the position data, and then fitting boundary data close to the actual one is fitted according to the boundary data at two different angles. The fitting human body data at two different angles are combined and calculated, which is conducive to calculating the first human body position data that is more in line with the actual one.

[0042] Optionally, the second fitting algorithm further comprises the following steps:

[0043] Calculate the difference between the position values ​​of the corresponding points in the first fitting boundary data and the second fitting boundary data as the position difference;

[0044] Calculate the sum of the absolute values ​​of all the position differences as the position sum value;

[0045] The weight of the first edge data and the weight of the second edge data in the weighted average algorithm are adjusted according to the position and value; the larger the sum value is, the smaller the weight of the first edge data and the weight of the second edge data are, and the smaller the sum value is, the larger the weight of the first edge data and the weight of the second edge data are.

[0046] By adopting the above technical solution, by calculating the sum of absolute values, the feature difference in the fitting position data of the two angles can be reflected, and the detailed misalignment or offset of the contour position in the fitting position data of the two angles can be magnified. It is more likely that misalignment or offset will occur between the template data and the actual contour data, so that the weight of the edge data can be adjusted according to the actual situation to make the calculated data more realistic.

[0047] Optionally, the method further comprises the following steps:

[0048] The first device contour data and the second device contour data are visible light spots of a linear array on the side of a human body. The deformation variables of the distances between multiple groups of adjacent visible light spots in the image data are calculated, and corresponding deformation distribution data are obtained by matching multiple deformation variables from a preset distribution data set.

[0049] By adopting the above technical solution, the linear array of laser sensors can measure the distance to obtain boundary data, and can also be used as the device contour data captured by the binocular camera for correction or calibration of image data.

[0050] Optionally, the first binocular camera and the second binocular camera are located on both sides of the scanning bed, and a third binocular camera facing the human body is also arranged beside the scanning bed;

[0051] The method further comprises the steps of:

[0052] Acquire third image data based on the third binocular camera;

[0053] Identifying two-dimensional human body contour data from the third image data;

[0054] Correcting the first human body position data according to the two-dimensional human body contour data;

[0055] Correction methods include:

[0056] Calculating a first deviation value between the two-dimensional human body contour data and preset standard two-dimensional human body contour data;

[0057] Calculating a second deviation value between the first human body position data and a preset standard human body position;

[0058] Calculating a first comprehensive deviation value using a weighted average algorithm according to the first deviation value and the second deviation value;

[0059] The first comprehensive deviation value is used as a gain value to correct the first human body position data.

[0060] By adopting the above technical solution, adding a third binocular camera can correct the human body position data through the third image data, which is beneficial to improve the accuracy of the human body position data.

[0061] Optionally, a third position sensor is further provided next to the third binocular camera;

[0062] The method further comprises the steps of:

[0063] Acquire third position data based on the third position sensor;

[0064] identifying two-dimensional contour data according to the third position data;

[0065] Correcting data of a contour portion of the three-dimensional human body contour data that is not facing the first binocular camera and the second binocular camera according to the two-dimensional contour data;

[0066] Correction methods include:

[0067] Calculating a third deviation value between the two-dimensional contour data and preset standard two-dimensional contour data;

[0068] Calculating a fourth deviation value between the three-dimensional human body contour data and a preset standard human body contour;

[0069] Calculating a second comprehensive deviation value using a weighted average algorithm according to the third deviation value and the fourth deviation value;

[0070] The second comprehensive deviation value is used as a gain value to correct the data of the contour part of the three-dimensional human body contour data that is not facing the first binocular camera and the second binocular camera.

[0071] By adopting the above technical solution, adding a third position sensor can correct the three-dimensional human body contour data through the third position data, which is beneficial to improving the accuracy of the three-dimensional human body contour data.

[0072] In a second aspect, the present application provides a scanning system for a medical imaging device, which adopts the following technical solution:

[0073] A scanning system for medical imaging equipment comprises a processor, wherein a program of any one of the above-mentioned scanning methods for medical imaging equipment is running on the processor.

[0074] To summarize, the present application includes at least one of the following beneficial technical effects: an external binocular camera is added to the traditional medical imaging equipment, and a first fitting algorithm is used to identify three-dimensional human body contour data from multiple angles based on the first image data and the second image data, and then the first position data and the second position data output by the multi-angle position sensor are combined, and the first human body position data is calculated using a second fitting algorithm, and the three-dimensional human body contour data is corrected to achieve multi-angle data fusion and form complete human body three-dimensional volume data; then the human body three-dimensional volume data is combined with the scanning workflow of the imaging equipment, and a scanning protocol is generated according to the confirmation information to achieve the specified scanning area and optimize the scanning parameters, so as to achieve one-stop human body positioning and scanning parameter optimization of the medical imaging equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 It is a schematic diagram of the overall method flow of the scanning method of the medical imaging device in the embodiment of the present application.

[0076] Figure 2 It is a hardware schematic diagram corresponding to the scanning method of the medical imaging device in the embodiment of the present application.

[0077] Figure 3 It is a schematic flow chart of a method for calculating first human body graphic data in an embodiment of the present application.

[0078] Figure 4 It is a flow chart of a method for adjusting the contrast of first human body graphic data in an embodiment of the present application.

[0079] Figure 5 It is a flowchart of a method for adjusting the weight of template data in an embodiment of the present application.

[0080] Figure 6 It is a graphical diagram corresponding to the fitted human body data in the embodiment of the present application (the chest thickness is not shown).

[0081] Figure 7 It is a hardware schematic diagram containing a position sensor in an embodiment of the present application.

[0082] Figure 8 It is a flowchart of a method for calculating first human body position data in an embodiment of the present application.

[0083] Fig. 9 It is a flowchart of a method for adjusting the weight of edge data in an embodiment of the present application.

[0084] Fig.10 It is a schematic flow chart of a method for correcting first human body position data in an embodiment of the present application.

[0085] Fig.11It is a schematic flow chart of a method for correcting data of a contour portion of a three-dimensional human body contour data that is not facing the first binocular camera and the second binocular camera in an embodiment of the present application. DETAILED DESCRIPTION

[0086] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the attached Figure 1-11 Shown in.

[0087] In the description of this specification, the description with reference to the terms "certain embodiments", "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiments or examples are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0088] The present application embodiment discloses a scanning method of a medical imaging device, referring to Figure 1 and Figure 2 , based on the first binocular camera and the second binocular camera next to the scanning bed. The first binocular camera and the second binocular camera are located on both sides of the length direction of the scanning bed, with shooting directions opposite to each other and facing the human body on the scanning bed. The height corresponds to the height range of the human body in the scanning room. The first binocular camera captures one side image of the human body, and the second binocular camera captures another side image of the human body.

[0089] like Figure 1 As shown, the method comprises the following steps:

[0090] Based on the first binocular camera and the second binocular camera beside the scanning bed, the first image data and the second image data are obtained. The first image data contains a side image of a human body, an image of the scanning bed, and an image of the scanning bedside device. The image data contains the image deformation of the camera lens, and the straight edge of the scanning bed or the straight edge or straight line feature of the scanning bedside device can be used to correct the image deformation.

[0091] The first human body graphic data is calculated using a preset first fitting algorithm according to the first image data and the second image data. Figure 3 As shown, the first fitting algorithm includes the following steps:

[0092] The first contour data is extracted from the first image data, and the second contour data is extracted from the second image data; the first contour data includes the first human body data and the first device contour data, and the second contour data includes the second human body data and the second device contour data. Among them, the first image data and the second image data both contain grayscale value images or color value images of the human body image, the scanning bed image and the scanning bedside device image. In this embodiment, a grayscale value image is used, and the grayscale value of the pixel is 0-255. The grayscale value of the pixel is 0, which is displayed as black, and the grayscale value of the pixel is 255, which is displayed as white; if the color value image is used, it is divided into primary color images showing the three primary colors, and is processed separately in different primary color images. In the grayscale value image, the grayscale value of the human body image is lower than that of the scanning bed, and the grayscale value of the scanning bed image is lower than that of the scanning bedside device image. The contours of the human body, the scanning bed or the scanning bedside device can be calculated by the existing contour recognition algorithm, and then the corresponding relationship between the contours of the human body, the scanning bed or the scanning bedside device can be identified by the existing template matching algorithm. In addition, the partial shape of the human body may block the scanning bed, so the data of the outline of the equipment beside the scanning bed is used as the equipment outline data.

[0093] The first deformation distribution data is calculated according to the first device contour data, the first human body data is adjusted according to the first deformation distribution data, the first template data is matched from the preset first human body template set according to the adjusted first human body data, and the first fitting human body data is calculated according to the adjusted first human body data and the first template data. In this embodiment, the deformation degree of the straight edge or straight line feature of the scanning bedside device in the image is first calculated, for example, the curvature of the straight edge of the scanning bedside device in the image is calculated, and the corresponding first deformation distribution data is matched or calculated by table lookup according to the size of the curvature. The first deformation distribution data can be matrix data, and the matrix format of the matrix data is adapted to the matrix format of the human body data. After the matrix data and the first human body data are multiplied, the imaging of the first human body data is stretched. The stretched first human body data is half of the human body data. According to this half of the data, the first template data is matched or adapted to the corresponding first template data from the first human body template set by table lookup. The first template data is the data of the other half of the human body, which is half of the standard model data, and then the first human body data and the first template data are pieced together to obtain the first fitting human body data.

[0094] The second deformation distribution data is calculated according to the second device contour data, the second human body data is adjusted according to the second deformation distribution data, the second template data is matched from a preset second human body template set according to the adjusted second human body data, and the second fitting human body data is calculated according to the adjusted second human body data and the second template data. The calculation process of the second fitting human body data is the same as the calculation process of the first fitting human body data, and will not be repeated in this article.

[0095] like Figure 6 As shown, the first human body graphic data is calculated according to the weighted average algorithm by taking the pixel values ​​of the pixels at the corresponding positions according to the first fitting human body data and the second fitting human body data. For example, the first fitting human body data contains the real-time data on the left side of the human body and the template data on the right side of the human body, and the second fitting human body data contains the real-time data on the right side of the human body and the template data on the left side of the human body. Human body data and device contour data are extracted from the image data. The device contour data is relatively stable. The deformation of the human body data in the image data is obtained according to the deformation of the device contour data in the image data, and then the fitting human body data close to the actual is calculated. The fitting human body data of two different angles are combined and calculated, which is conducive to calculating the first human body graphic data that is more in line with the actual situation.

[0096] like Figure 4 As shown, in order to improve the accuracy of the subsequent data calculation results, the difference between the pixel values ​​of the corresponding pixels in the first fitting human body data and the second fitting human body data is calculated; the sum of all the differences is calculated; the contrast of the first human body graphic data is adjusted according to the sum, and the larger the sum, the greater the contrast, and the smaller the sum, the smaller the contrast. The larger the feature gap in the fitting human body data at the two angles, the larger the sum, and it is necessary to increase the contrast to highlight the contour features in the first human body graphic data; the smaller the feature gap, the smaller the sum, and the less contrast can be added to avoid losing some less obvious features; if the contour position in the fitting human body data at the two angles is misaligned or offset in detail, the calculated sum will not be too large, reducing the impact of the misalignment and offset on the contrast. For example, if the first fitting human body data and the second fitting human body data are misaligned or offset in detail, there are pixels A and pixel B in the first fitting human body data, and there are corresponding pixels a and pixel b in the second fitting human body data; the value of pixel A is 100, the value of pixel a is 30, the value of pixel B is 40, and the value of pixel b is 110, then Aa=70, Bb=-70, and the sum of the differences between the two is 0. The sum of the differences can reduce the impact of the misalignment and offset on the contrast.

[0097] like Figure 5 and Figure 6As shown, in addition, the difference between the pixel values ​​of the corresponding pixels in the first fitting human body data and the second fitting human body data can be calculated; the sum of the absolute values ​​of all the differences can be calculated; the weight of the first template data and the weight of the second template data in the weighted average algorithm are adjusted according to the sum value, and the larger the sum value, the smaller the weight of the first template data and the weight of the second template data, and the smaller the sum value, the larger the weight of the first template data and the weight of the second template data. By calculating the sum of the absolute values, the feature gap in the fitting human body data of the two angles can be reflected, and the detailed misalignment or offset of the contour position in the fitting human body data of the two angles can be magnified, and the template data and the actual contour data are more likely to be misaligned or offset, so that the weight of the template data can be adjusted according to the actual situation, so that the calculated data is more in line with the actual situation. For example, if the first fitting human body data and the second fitting human body data are misaligned or offset in detail, then there are pixels A and pixel B in the first fitting human body data, and there are corresponding pixels a and pixel b in the second fitting human body data; the value of pixel A is 100, the value of pixel a is 30, the value of pixel B is 40, and the value of pixel b is 110, then Aa=70, Bb=-70, the sum of the absolute values ​​of the difference between the two is 140, the weight of the first template data is 0.3 and the weight of the second template data is 0.3. If the value of pixel A is 70, the value of pixel a is 30, the value of pixel B is 40, and the value of pixel b is 80, then Aa=40, Bb=-40, the sum of the absolute values ​​of the difference between the two is 80, the weight of the first template data is 0.5 and the weight of the second template data is 0.5. It is further supplemented that pixel A can be calculated and pixel B can also be calculated; or, pixel A can be calculated and pixel B can be obtained from a set.

[0098] A human body contour is identified from the first human body graphic data to generate three-dimensional human body contour data.

[0099] like Figure 7 and Figure 8 As shown, based on the first position sensor and the second position sensor beside the scanning bed, the first position data and the second position data are obtained. The first position sensor and the second position sensor are ultrasonic sensors arranged in a linear array; or, the first position sensor and the second position sensor are laser sensors with visible light spots and arranged in a linear array, and the visible light spots of the linear array are projected on the side of the human body. The first device contour data and the second device contour data are visible light spots of the linear array on the side of the human body, and the deformation variables of the distances between multiple groups of adjacent visible light spots in the image data are calculated, and the corresponding deformation distribution data are obtained by matching the multiple deformation variables from the preset distribution data set. The linear array of laser sensors can measure the distance to obtain boundary data, and can also be used as device contour data captured by the binocular camera for correction or calibration of image data.

[0100] The first human body position data is calculated using a preset second fitting algorithm according to the first position data and the second position data. The second fitting algorithm includes the following steps: extracting the first boundary data from the first position data, and extracting the second boundary data from the second position data. According to the first boundary data, the first edge data is matched from the preset first boundary template set, and the first fitting boundary data is calculated according to the first boundary data and the first edge data. The first position data can be a series of distance values ​​of points, and the outer edge contour of a side of the human body and the outer edge contour of other objects can be fitted by connecting lines through the distance values ​​of these points, and then the outer edge contour of a side of the human body is calculated by matching or looking up the table to obtain the first boundary data. In the graphic dimension, the shape of the connecting line is the outer edge contour shape of a side of the human body. At this time, if the outer edge contour data of the other side of the human body is missing, the first edge data corresponding to the contour shape represented by the content of the first boundary data can be calculated by matching or looking up the table, and the first boundary data and the first edge data can be merged to obtain the first fitting boundary data.

[0101] According to the second boundary data, the second edge data is matched from the preset second boundary template set, and the second fitting boundary data is calculated according to the second boundary data and the second edge data. The first human body position data is calculated according to the weighted average algorithm through the position values ​​of the corresponding points based on the first fitting boundary data and the second fitting boundary data. The calculation process of the second fitting boundary data is the same as the calculation process of the first fitting boundary data, which will not be repeated in this article. The position data is obtained through the position sensor, the boundary data is extracted from the position data, and then the fitting boundary data close to the actual one is fitted according to the boundary data of two different angles, and the fitting human body data of two different angles are combined and calculated, which is conducive to calculating the first human body position data that is more in line with the actual one.

[0102] like Fig. 9 As shown, in order to adjust the weight of the edge data according to the actual situation and make the calculated data more realistic, the second fitting algorithm also includes the following steps: calculating the difference between the position values ​​of the corresponding points in the first fitting boundary data and the second fitting boundary data as the position difference, and calculating the sum of the absolute values ​​of all position differences as the position sum. The weight of the first edge data and the weight of the second edge data in the weighted average algorithm are adjusted according to the position sum. The larger the sum value, the smaller the weight of the first edge data and the weight of the second edge data, and the smaller the sum value, the larger the weight of the first edge data and the weight of the second edge data.

[0103] For example, if the first fitting boundary data and the second fitting boundary data are misaligned or offset in detail, then there are points C and D in the first fitting boundary data, and there are corresponding points c and d in the second fitting boundary data; the distance value of point C is 50, the distance value of point c is 45, the distance value of point D is 55, and the distance value of point d is 49, then Cc=5, Dd=-4, the sum of the absolute values ​​of the difference between the two is 9, the weight of the first edge data is 0.4 and the weight of the second edge data is 0.4. If the distance value of point C is 55, the distance value of point c is 45, the distance value of point D is 40, and the distance value of point d is 50, then Aa=10, Bb=-5, the sum of the absolute values ​​of the difference between the two is 15, the weight of the first edge data is 0.39 and the weight of the second edge data is 0.39. By calculating the sum of the absolute values, the feature difference in the fitting position data of the two angles can be reflected, and the detailed misalignment or offset of the contour position in the fitting position data of the two angles can be magnified, and the misalignment or offset is more likely to occur between the template data and the actual contour data. It is further supplemented that point C can be calculated, and point D can also be calculated; or, point C can be calculated, and point D can be obtained from the set.

[0104] Human body features are identified from the first human body position data to generate human body feature position data. The identification method may be a template matching method, an AI algorithm based on big data, or a convolutional neural network algorithm in the prior art. The coordinate content of the three-dimensional human body contour data is corrected based on the human body feature position data. The correction method may be a weighted average algorithm, which performs a weighted average calculation on the data corresponding to the human body part in the three-dimensional human body contour data based on the human body feature position data. For example, the position of the human hand, the human body feature position data is 120, and the corresponding coordinate position is 20. The data corresponding to the hand position in the three-dimensional human body contour data is 110, and the corresponding coordinate position is 18. The weighted value of the human body feature position data is 0.4, and the weighted value of the three-dimensional human body contour data is 0.6. Then 120*0.4+110*0.6=114, and 114 is used to replace the data corresponding to the hand position in the three-dimensional human body contour data. The coordinate position of the corresponding hand position in the updated three-dimensional human body contour data is 19.

[0105] like Figure 7 As shown, in the above embodiment, a third binocular camera and a third position sensor may be added, the third binocular camera is located above the scanning bed and facing the human body, and the third position sensor is arranged beside the third binocular camera.

[0106] like Fig.10 and Fig.11As shown, based on the third binocular camera, the third image data is obtained, the two-dimensional human body contour data is identified from the third image data, and the first human body position data is corrected according to the two-dimensional human body contour data. The correction method includes: calculating the first deviation value between the two-dimensional human body contour data and the preset standard two-dimensional human body contour data, and calculating the second deviation value between the first human body position data and the preset standard human body position. The first comprehensive deviation value is calculated using a weighted average algorithm based on the first deviation value and the second deviation value. The weighted average algorithm here can adopt the weighted average algorithm mentioned above in the text. The first comprehensive deviation value is used as a gain value to correct the first human body position data. The correction method here is the same as the method for correcting three-dimensional human body contour data. Adding a third binocular camera can correct the human body position data through the third image data, which is conducive to improving the accuracy of the human body position data.

[0107] Based on the third position sensor, the third position data is obtained, the two-dimensional contour data is identified according to the third position data, and the data of the contour part that is not facing the first binocular camera and the second binocular camera in the three-dimensional human body contour data is corrected according to the two-dimensional contour data. The correction method includes: calculating the third deviation value between the two-dimensional contour data and the preset standard two-dimensional contour data, and calculating the fourth deviation value between the three-dimensional human body contour data and the preset standard human body contour. The second comprehensive deviation value is calculated using a weighted average algorithm based on the third deviation value and the fourth deviation value, and the weighted average algorithm here can adopt the weighted average algorithm mentioned above in the text. The second comprehensive deviation value is used as a gain value to correct the data of the contour part that is not facing the first binocular camera and the second binocular camera in the three-dimensional human body contour data. The correction method here is the same as the method for correcting the three-dimensional human body contour data. Adding a third position sensor can correct the three-dimensional human body contour data through the third position data, which is conducive to improving the accuracy of the three-dimensional human body contour data.

[0108] The scanned part data is extracted from the latest 3D human body contour data. The 3D human body contour data includes the edge contour data of the human body. The corresponding regional contour can be extracted from the 3D human body contour data according to the preset chest template, abdomen template, head template and other templates, and then the regional contour is displayed on the 3D human body contour. The data corresponding to the displayed part is the scanned part data.

[0109] The scanning area is generated according to the scanning part data and the spatial data of the imaging device. The scanning part data is only data that exists relative to the three-dimensional human body contour data. The scanning device cannot directly analyze useful data from the scanning part data, so the relative position data of the scanning part data and the three-dimensional human body contour data are calculated, and then the scanning area data corresponding to the spatial data is calculated according to the relative position relationship between the three-dimensional human body contour data and the spatial data of the imaging device. The scanning device is an imaging device, and the scanning device can analyze the scanning area to be scanned according to the scanning area data.

[0110] Based on the confirmation information generated corresponding to the scan area, a scan protocol is generated and sent to the imaging device. In practice, the scan area is displayed to the physician through a touch-sensitive display screen away from the scanning device. The physician selects the scan area to be scanned through a touch action. After the selection is completed, corresponding confirmation information is generated. The confirmation information can be recognized and analyzed by the scanning device. An executable scan protocol is generated based on the analysis result, and then the scan protocol is sent to the scanning device in the scanning room. The scanning device or imaging device scans in response to the scan protocol.

[0111] An external binocular camera is added to the traditional medical imaging equipment. The binocular cameras with multiple angles are connected to the scanning control center of the imaging equipment via an Ethernet switch. The binocular camera is used to collect video data; the scanning control center of the imaging equipment calculates real-time calculations and modeling, and provides physicians with scanning area selection and recommended scanning parameters to form a complete scanning protocol and scanning workflow. The binocular camera is connected to the network switch via Ethernet, and PoE is used to realize a single interface for power supply and communication. The scanning control center of the imaging equipment is also connected to the network switch via Ethernet to realize the collection and exchange of camera video data. The scanning control center of the imaging equipment is responsible for processing the three-dimensional patient body data, combining the daily scanning control flow to control the scanning frame and scanning bed to complete the scanning task.

[0112] The multi-angle patient volume data acquisition is carried out in real time. The acquired multi-angle volume data needs to be first corrected, and the multi-angle human body data is aligned with the central coordinates of the imaging device to form a unified coordinate system, and then the data is fused to form complete volume data. In the scanning control system, it is combined with the scanning workflow to directly display the patient's real-time three-dimensional volume data when the patient is positioned, and the scanning protocol is recommended for efficient patient scanning.

[0113] According to the first image data and the second image data, the first fitting algorithm is used to identify the three-dimensional human body contour data from multiple angles, and then the first position data and the second position data output by the multi-angle position sensor are combined, and the first human body position data is calculated using the second fitting algorithm, and the three-dimensional human body contour data is corrected to achieve multi-angle data fusion and form complete human body three-dimensional volume data; then the human body three-dimensional volume data is combined with the scanning workflow of the imaging device, and a scanning protocol is generated according to the confirmation information to achieve the specified scanning area and optimize the scanning parameters, so as to achieve one-stop human body positioning and scanning parameter optimization of the medical imaging device, reduce the error of the doctor's guidance, and improve the accuracy of the scanning results. It greatly simplifies the daily operations of doctors, and can automatically give scanning conditions and scanning parameters according to the patient's body data, further improving the operation efficiency and reducing the complexity of the operation.

[0114] An embodiment of the present application further discloses a scanning system for a medical imaging device, including a processor, wherein a program for any one of the above-mentioned scanning methods for the medical imaging device is running in the processor.

[0115] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A scanning method of a medical imaging device, characterized in that: The steps include: Acquire first image data and second image data based on a first binocular camera and a second binocular camera beside a scanning bed; Calculate first human body graphic data using a preset first fitting algorithm according to the first image data and the second image data; Recognizing a human body contour from the first human body graphic data to generate three-dimensional human body contour data; Acquiring first position data and second position data based on a first position sensor and a second position sensor beside a scanning bed; Calculate first human body position data using a preset second fitting algorithm according to the first position data and the second position data; Identifying human features from the first human position data to generate human feature position data; Correcting the coordinate content of the three-dimensional human body contour data based on the human body feature position data; Extracting scanned part data from the latest three-dimensional human body contour data; Generate a scanning area according to the scanning part data and the spatial data of the imaging device; generating a scanning protocol based on the confirmation information generated corresponding to the scanning area, and sending the scanning protocol to an imaging device; The imaging device performs scanning in response to the scanning protocol.

2. The scanning method of the medical imaging device according to claim 1, characterized in that: The first fitting algorithm comprises the following steps: Extracting first contour data from the first image data, and extracting second contour data from the second image data; the first contour data includes first human body data and first device contour data, and the second contour data includes second human body data and second device contour data; Calculating first deformation distribution data according to the first device contour data, adjusting the first human body data according to the first deformation distribution data, matching first template data from a preset first human body template set according to the adjusted first human body data, and calculating first fitting human body data according to the adjusted first human body data and the first template data; Calculating second deformation distribution data according to the second device contour data, adjusting the second human body data according to the second deformation distribution data, matching second template data from a preset second human body template set according to the adjusted second human body data, and calculating second fitting human body data according to the adjusted second human body data and the second template data; The first human body graphic data is calculated according to the first fitting human body data and the second fitting human body data by taking pixel values ​​of pixels at corresponding positions according to a weighted average algorithm.

3. The scanning method of medical imaging equipment according to claim 2, characterized in that: The first fitting algorithm also includes the following steps: Calculating a difference between pixel values ​​of corresponding pixels in the first fitting human body data and the second fitting human body data; Calculating the sum of all said differences; The contrast of the first human body graphic data is adjusted according to the sum value, wherein the greater the sum value is, the greater the contrast is, and the smaller the sum value is, the smaller the contrast is.

4. The scanning method of the medical imaging device according to claim 2, characterized in that: The first fitting algorithm also includes the following steps: Calculating a difference between pixel values ​​of corresponding pixels in the first fitting human body data and the second fitting human body data; Calculate the sum of the absolute values ​​of all the differences; The weight of the first template data and the weight of the second template data in the weighted average algorithm are adjusted according to the sum value. The larger the sum value is, the smaller the weight of the first template data and the weight of the second template data are. The smaller the sum value is, the larger the weight of the first template data and the weight of the second template data are.

5. The scanning method of medical imaging equipment according to claim 2, characterized in that: The first position sensor and the second position sensor are ultrasonic sensors arranged in a linear array; or, the first position sensor and the second position sensor are laser sensors having a visible light spot and arranged in a linear array, and the visible light spot of the linear array is projected on the side of the human body; The second fitting algorithm comprises the following steps: extracting first boundary data from the first position data, and extracting second boundary data from the second position data; Matching first edge data from a preset first boundary template set according to the first boundary data, and calculating first fitting boundary data according to the first boundary data and the first edge data; Matching second edge data from a preset second boundary template set according to the second boundary data, and calculating second fitting boundary data according to the second boundary data and the second edge data; The first human body position data is calculated according to the first fitting boundary data and the second fitting boundary data through the position values ​​of corresponding points according to a weighted average algorithm.

6. The scanning method of the medical imaging device according to claim 5, characterized in that: The second fitting algorithm also includes the following steps: Calculate the difference between the position values ​​of the corresponding points in the first fitting boundary data and the second fitting boundary data as the position difference; Calculate the sum of the absolute values ​​of all the position differences as the position sum value; adjusting the weight of the first edge data and the weight of the second edge data in the weighted average algorithm according to the position and value; The larger the sum value is, the smaller the weight of the first edge data and the weight of the second edge data are; and the smaller the sum value is, the larger the weight of the first edge data and the weight of the second edge data are.

7. The scanning method of medical imaging equipment according to claim 5, characterized in that: The method further comprises the steps of: The first device contour data and the second device contour data are visible light spots of a linear array on the side of a human body. The deformation variables of the distances between multiple groups of adjacent visible light spots in the image data are calculated, and corresponding deformation distribution data are obtained by matching multiple deformation variables from a preset distribution data set.

8. The scanning method of medical imaging equipment according to claim 1, characterized in that: The first binocular camera and the second binocular camera are located on both sides of the scanning bed, and a third binocular camera facing the human body is also arranged beside the scanning bed; The method further comprises the steps of: Acquire third image data based on the third binocular camera; Identifying two-dimensional human body contour data from the third image data; Correcting the first human body position data according to the two-dimensional human body contour data; Correction methods include: Calculating a first deviation value between the two-dimensional human body contour data and preset standard two-dimensional human body contour data; Calculating a second deviation value between the first human body position data and a preset standard human body position; Calculating a first comprehensive deviation value using a weighted average algorithm according to the first deviation value and the second deviation value; The first comprehensive deviation value is used as a gain value to correct the first human body position data.

9. The scanning method of medical imaging equipment according to claim 8, characterized in that: A third position sensor is also arranged next to the third binocular camera; The method further comprises the steps of: Acquire third position data based on the third position sensor; correcting the third image data according to the third position data; identifying two-dimensional contour data from the third position data; Correcting data of a contour portion of the three-dimensional human body contour data that is not facing the first binocular camera and the second binocular camera according to the two-dimensional contour data; Correction methods include: Calculating a third deviation value between the two-dimensional contour data and preset standard two-dimensional contour data; Calculating a fourth deviation value between the three-dimensional human body contour data and a preset standard human body contour; Calculating a second comprehensive deviation value using a weighted average algorithm according to the third deviation value and the fourth deviation value; The second comprehensive deviation value is used as a gain value to correct the data of the contour part of the three-dimensional human body contour data that is not facing the first binocular camera and the second binocular camera.

10. A scanning system for medical imaging equipment, characterized in that: It comprises a processor, in which a program of a scanning method of a medical imaging device as described in any one of claims 1 to 9 is run.