Automatic measurement method, system, medium and product for ghost image index of nuclear magnetic resonance equipment imaging

Abstract

The present invention discloses a method, system, medium, and product for automatically measuring ghost image indicators in nuclear magnetic resonance imaging equipment, including the following steps: obtaining an image file in DICOM format; performing pixel extraction and format conversion on the DICOM image file to obtain a second image in 8-bit format; automatically identifying regions of the second image to obtain signal areas, ghost areas, and noise areas; calculating statistical characteristics of the signal areas, ghost areas, and noise areas respectively; and calculating ghost image indicators based on the statistical characteristics of the signal areas, ghost areas, and noise areas. The present invention can automatically calculate ghost-related indicators, reduce manual intervention, and improve detection efficiency and consistency.

Description

Technical Field

[0001] The present invention relates to the field of nuclear magnetic resonance imaging technology, and in particular to an automatic measurement method, system, medium and product for ghost image indicators of nuclear magnetic resonance imaging. Background Art

[0002] MRI systems are prone to various artifacts during operation, with ghosting being a common and significant issue. Ghost artifacts are repetitive or blurry artifacts that appear in MRI images, typically occurring along specific directions (such as the phase encoding direction). Causes include gradient imbalance, patient motion, incomplete data acquisition, or system calibration errors. Ghosting not only degrades image aesthetics but also interferes with the accurate identification of true lesions or structures. Therefore, to ensure the accuracy and contrast of subsequent diagnostic images, ghosting is often assessed during pre-scan or routine quality assurance (QA) processes, typically by calculating ghosting metrics.

[0003] Existing practices mostly rely on manually selecting background areas and artifact areas to calculate ghost indicators. This is not only inefficient, but also easily affected by the operator's experience and subjective judgment, resulting in unstable detection results.

[0004] For example, Chinese patent application publication number CN101322648A proposes a method for automatically calculating ghost images in magnetic resonance imaging (MRI). This method first uses an iterative thresholding method to segment the image, dividing it into a foreground (signal region), a background (noise region), and a ghost region. Subsequently, by extracting pixel information from each region and combining it with statistical analysis methods, a ghost value index is calculated, enabling automated and objective assessment of ghost intensity. However, the iterative thresholding method is sensitive to noise. In particular, when the noise is strong, the foreground and background regions often become nested, resulting in a large number of isolated pixel regions. This can significantly bias the calculated results and affect the accuracy and stability of the ghost index. Summary of the Invention

[0005] In order to solve the technical problems existing in the background technology, the present invention proposes a method, system, medium and product for automatically measuring ghost image indicators of nuclear magnetic resonance equipment imaging.

[0006] The present invention proposes a method, system, medium and product for automatically measuring ghost image indicators of nuclear magnetic resonance imaging equipment, including:

[0007] Get image files in DICOM format;

[0008] Perform pixel extraction and format conversion on the DICOM format image file to obtain a second image in 8-bit format;

[0009] Automatically identifying regions of the second image to obtain signal regions, ghost regions, and noise regions;

[0010] Statistical characteristics of the signal area, ghost area and noise area are calculated respectively;

[0011] The ghost index is calculated based on the statistical characteristics of the signal area, ghost area and noise area.

[0012] Preferably, pixel extraction and format conversion are performed on the image file in DICOM format to obtain the second image in 8-bit format, specifically including:

[0013] The DICOM format image file is parsed and displayed, and the high-bit pixels are converted into a 16-bit first image and an 8-bit second image through the tag window width and window level of the image file.

[0014] Preferably, the conversion formula of the second image is expressed as: Where, Indicates the grayscale value displayed. Indicates the window width, Indicates the window level.

[0015] Preferably, performing automatic region recognition on the second image to obtain a signal region, a ghost region, and a noise region specifically includes:

[0016] The second image is binarized using the Otsu threshold method to obtain a first binarized image;

[0017] The Suzuki-Abe boundary tracking algorithm is used to track the boundaries of the first binary image, extract all the closed curves of the signal area, and select the closed curve of the signal area with the largest area as the outline of the signal area;

[0018] According to the outline of the signal area and the first binarized image, the noise rectangular areas in four directions are obtained;

[0019] The outline of the ghost area is obtained according to the outline of the signal area and the first image.

[0020] Preferably, according to the outline of the signal area and the first binarized image, four directions of noise rectangular areas are obtained, specifically including:

[0021] According to the outline of the signal area, the minimum enclosing rectangle of the signal area is calculated; the four corner points of the minimum enclosing rectangle of the signal area are extracted, and the four corner points of the entire first binary image are defined as reference corner points; the coordinates of the center point between each corner point of the minimum enclosing rectangle and the corresponding reference corner point are calculated respectively, and the coordinates of each center point are used as the center point of a noise rectangular area; the distances between the corner points of each minimum enclosing rectangle and the corresponding reference corner point in the x-axis and y-axis directions are calculated respectively, and half of them are taken as the width and height of the corresponding noise rectangular area, thereby obtaining noise rectangular areas in four directions.

[0022] Preferably, obtaining the outline of the ghost area according to the outline of the signal area and the first image specifically includes:

[0023] Calculating the centroid of the contour of the signal area and calculating the direction vector between each contour point and the centroid in the contour of the signal area, then amplifying all the direction vectors according to a preset ratio to obtain an intermediate signal area; mapping the intermediate signal area to a corresponding position in the first image to obtain an intermediate image block; removing the intermediate image block from the first image to obtain a ghost image to be identified area;

[0024] The ghost area to be identified is Gaussian filtered, and the Otsu threshold method is used to binarize the Gaussian filtered ghost area to obtain a second binary image; the second binary image is subjected to foreground erosion and then dilation to obtain the middle ghost area to be identified; the Suzuki-Abe boundary tracking algorithm is used to track the boundary of the middle ghost area to be identified, and all ghost area closed curves in the middle ghost area to be identified are extracted. The ghost area closed curve with the largest area is selected as the outline of the ghost area.

[0025] Preferably, the statistical features of the signal area, the ghost area, and the noise area are calculated separately, specifically including:

[0026] determining a signal image block corresponding to the signal area in the first image, and calculating a mean value of pixel values ​​of the signal image block;

[0027] Determine the noise image blocks corresponding to the noise rectangular area in the first image, and calculate the standard deviation of the pixel values ​​of each noise image block; calculate the noise level based on the standard deviation of the pixel values ​​of the four noise image blocks;

[0028] According to the outline of the ghost area, the centroid of the ghost area is calculated, and the centroid is used as the center point of the ghost area to calculate the mean of the ghost area.

[0029] In a second aspect, the present invention further provides an automatic measurement system for ghost image indicators of nuclear magnetic resonance imaging equipment, comprising:

[0030] Acquisition module, used to obtain image files in DICOM format;

[0031] A conversion module, used for performing pixel extraction and format conversion on the image file in DICOM format to obtain a second image in 8-bit format;

[0032] an identification module, configured to automatically identify regions of the second image file to obtain signal regions, ghost regions, and noise regions;

[0033] The calculation module is used to calculate the statistical characteristics of the signal area, the ghost area and the noise area respectively; and calculate the ghost index according to the statistical characteristics of the signal area, the ghost area and the noise area.

[0034] In a third aspect, the present invention further proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the automatic measurement method for ghost image indicators of nuclear magnetic resonance imaging equipment as described in any one of the first aspects.

[0035] In a fourth aspect, the present invention further proposes a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of the method for automatically measuring ghost image indicators of nuclear magnetic resonance imaging equipment as described in any one of the first aspects.

[0036] The present invention proposes a method, system, medium, and product for automatically measuring ghost image indicators in nuclear magnetic resonance imaging equipment. The method extracts pixels from a DICOM-formatted image file and converts its format to obtain an 8-bit second image. The method then automatically identifies regions in the second image to obtain signal, ghost, and noise regions. The statistical characteristics of the signal, ghost, and noise regions are calculated, respectively. The method then calculates ghost image indicators based on the statistical characteristics of the signal, ghost, and noise regions. The method automatically calculates ghost-related indicators, reduces manual intervention, and provides objective, standardized ghost image measurement results, significantly improving the measurement accuracy and consistency of nuclear magnetic resonance image quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 The figure is a flow chart of a method for automatically measuring ghost image indicators in magnetic resonance imaging equipment according to an embodiment of the present invention.

[0038] Figure 2 A schematic diagram of displaying an image file in DICOM format according to an embodiment of the present invention is shown.

[0039] Figure 3 FIG. 1 is a schematic diagram of a signal area according to an embodiment of the present invention.

[0040] Figure 4 FIG. 1 is a schematic diagram of a noise area according to an embodiment of the present invention.

[0041] Figure 5FIG. 1 is a schematic diagram of a ghost image to-be-identified area in an embodiment of the present invention.

[0042] Figure 6 FIG. 1 is a schematic diagram of a ghost area according to an embodiment of the present invention. DETAILED DESCRIPTION

[0043] It should be noted that, in the absence of conflict, the features of the embodiments and implementations of the present invention may be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0044] First, it's important to understand that ghosting generally refers to repetitive or blurry artifacts that appear in MRI images, typically occurring along specific directions (such as the phase encoding direction). Causes include gradient imbalance, patient motion, incomplete data acquisition, or system calibration errors. Ghosting not only detracts from the aesthetics of the image but also interferes with the accurate identification of true lesions or structures. The ghosting index is a numerical metric used to quantify the degree of ghosting in MRI images. It is typically measured by calculating the signal intensity ratio or difference between the main image area and the artifact area.

[0045] First, refer to Figure 1 The present invention proposes a method, system, medium and product for automatically measuring ghost image indicators of nuclear magnetic resonance imaging equipment, including:

[0046] Get image files in DICOM format;

[0047] Perform pixel extraction and format conversion on the DICOM format image file to obtain a second image in 8-bit format;

[0048] Automatically identifying regions of the second image to obtain signal regions, ghost regions, and noise regions;

[0049] Statistical characteristics of the signal area, ghost area and noise area are calculated respectively;

[0050] The ghost index is calculated based on the statistical characteristics of the signal area, ghost area and noise area.

[0051] The present invention extracts pixels and converts the format of a DICOM-formatted image file to obtain an 8-bit second image; automatically identifies regions of the 8-bit second image to obtain signal, ghost, and noise regions; calculates the statistical characteristics of the signal, ghost, and noise regions; and calculates ghost indices based on the statistical characteristics of the signal, ghost, and noise regions. The present invention can automatically calculate ghost-related indices, reduce manual intervention, and provide objective, standardized ghost measurement results to improve the measurement accuracy and consistency of MRI image quality.

[0052] In this embodiment, pixel extraction and format conversion are performed on the DICOM format image file to obtain the second image in 8-bit format, specifically including:

[0053] Parse and display DICOM format image files;

[0054] According to the displayed result, the high-bit pixels are converted into a first image in 16-bit format and a second image in 8-bit format through the tag window width and window level of the image file.

[0055] This embodiment is configured in this way to ensure the correct acquisition of DICOM image data.

[0056] Specifically, the image is converted into a 16-bit first image and an 8-bit second image. The 8-bit data is the result of window width and window level adjustment. The conversion method is shown in the following formula. The window width (WW) and window level (WL) are used to adjust the grayscale level, making the image clearer on the display device. 16-bit or 12-bit medical image data can be converted to 8-bit (0-255 grayscale values) for display.

[0057] Where, Indicates the grayscale value displayed. Indicates the window width, Indicates the window level.

[0058] The binarization methods of this embodiment include the average method, the bimodal method, and the OTSU threshold method.

[0059] In this embodiment, before automatically identifying the second image to obtain the signal area, the ghost area, and the noise area, the method further includes:

[0060] Perform Gaussian filtering on the second image to remove noise.

[0061] In this embodiment, the second image is automatically identified to obtain a signal area, a ghost area, and a noise area, specifically including:

[0062] The second image is binarized using the Otsu threshold method to obtain a first binarized image;

[0063] The Suzuki-Abe boundary tracking algorithm is used to track the boundaries of the first binary image, extract all closed curves, and select the curve with the largest area as the outline of the signal area;

[0064] According to the outline of the signal area and the first binarized image, the noise rectangular areas in four directions are obtained;

[0065] The outline of the ghost area is obtained according to the outline of the signal area and the first image.

[0066] This implementation utilizes the Otsu algorithm for foreground and background separation, combined with the Suzuki-Abe boundary tracking algorithm, to accurately identify signal and ghost areas. During quality assurance, MRI equipment scans water film phantoms to generate DICOM image files. These phantoms typically have clear outlines and sharp foreground-background contrast, rather than complex textures. The Otsu + Suzuki-Abe approach employed in this implementation is more suitable for this application scenario.

[0067] According to the outline of the signal area and the first binarized image, four noise rectangular areas are obtained, specifically including:

[0068] According to the outline of the signal area, the minimum enclosing rectangle of the signal area is calculated;

[0069] Extract the four corner points of the minimum circumscribed rectangle of the signal area, which are the upper left, upper right, lower left and lower right, and define the four corner points of the entire first binary image as reference corner points;

[0070] Calculate the coordinates of the center point between each corner point of the minimum bounding rectangle and the corresponding reference corner point, and use the coordinates of each center point as the center point of a noise rectangle area;

[0071] Calculate the distances between the corner points of each minimum bounding rectangle and the corresponding reference corner points in the x-axis and y-axis directions respectively, and take half of them as the width and height of the corresponding noise rectangular area to obtain the noise rectangular areas in four directions.

[0072] When extracting the noise area, this embodiment appropriately enlarges the signal area to reduce interference, thereby ensuring the accuracy of the identified area. Moreover, to optimize the selection of the noise area, this embodiment calculates the minimum circumscribed rectangle of the signal area and divides the noise area between its four corner points and the four corners of the image, thereby ensuring the rationality of the measurement.

[0073] The process of obtaining the outline of the ghost area according to the outline of the signal area and the first image specifically includes:

[0074] Calculating the centroid of the contour of the signal area, then calculating the direction vector between each contour point in the contour of the signal area and the centroid, and then amplifying all the direction vectors according to a preset ratio to obtain an intermediate signal area; mapping the intermediate signal area to a corresponding position in the first image to obtain an intermediate image block; removing the intermediate image block from the first image to obtain a ghost image to be identified area;

[0075] Performing Gaussian filtering on the ghost image to be identified area, and binarizing the ghost image to be identified area after Gaussian filtering using the Otsu threshold method to obtain a second binarized image;

[0076] Perform foreground erosion and dilation on the second binary image to obtain the middle ghost image identification area;

[0077] All ghost area closed curves in the middle ghost area to be identified are extracted, and the ghost area closed curve with the largest area is selected as the outline of the ghost area.

[0078] When extracting ghost areas, this embodiment appropriately enlarges the signal area to reduce interference and ensure the accuracy of the identified area. In the process of extracting ghost areas, an opening operation (morphological processing) is used to remove the influence of isolated pixel areas, thereby improving the accuracy of the ghost areas and making the calculated ghost indicators closer to their actual sizes. At the same time, the main structure of the ghost area is retained to improve the stability, accuracy and robustness of the detection.

[0079] The preset ratio in this embodiment is 1.1.

[0080] In this embodiment, the statistical features of the signal area, the ghost area, and the noise area are calculated respectively, specifically including:

[0081] Determine a signal image block corresponding to the signal area in the first image, and calculate a mean value S of the signal image block;

[0082] Determine the noise image blocks corresponding to the noise rectangular area in the first image, and calculate the standard deviation of each noise image block; , calculate the noise level ;

[0083] According to the outline of the ghost area, the center of mass of the ghost area is calculated, and the center of mass is used as the center point of the ghost area to calculate the mean of the ghost area. .

[0084] The statistical characteristics of this embodiment include mean and standard deviation.

[0085] In this embodiment, the ghosting index includes a ghost-to-signal ratio, a ghost-to-noise ratio, and a signal-to-noise ratio;

[0086] Among them, the ratio of ghost to signal Where, represents the mean value of the ghost area, represents the mean of the signal;

[0087] Ghost to noise ratio Where, represents the mean value of the ghost area, represents the mean of the noise; Where, represents the signal-to-noise ratio, represents the mean of the signal; represents the mean of the noise.

[0088] In a specific implementation, based on the identified signal area, noise area and ghost area, the mean and standard deviation of each area are calculated to quantify the intensity of the ghost.

[0089] In summary, the present invention can automatically calculate ghost-related indicators, reduce manual intervention, improve detection efficiency and consistency, and adapt to complex imaging environments. Even in the case of low signal-to-noise ratio or poor image quality, it can still stably and accurately identify signal areas and ghost areas.

[0090] In a second aspect, the present invention further provides an automatic measurement system for ghost image indicators of nuclear magnetic resonance imaging equipment, comprising:

[0091] Acquisition module, used to obtain image files in DICOM format;

[0092] A conversion module, used for performing pixel extraction and format conversion on the image file in DICOM format to obtain a second image in 8-bit format;

[0093] an identification module, configured to automatically identify regions of the second image file to obtain signal regions, ghost regions, and noise regions;

[0094] The calculation module is used to calculate the statistical characteristics of the signal area, the ghost area and the noise area respectively; and calculate the ghost index according to the statistical characteristics of the signal area, the ghost area and the noise area.

[0095] In this embodiment, a display module is also included;

[0096] During the format conversion process, the display module is used to parse and display the DICOM format image file; the DICOM format image file is parsed and displayed; the conversion module is used to determine whether the DICOM format image file is high-bit data greater than 8 bits; if so, the DICOM format image file is pixel extracted to obtain a first image, and the first image is converted into a second image in 8-bit format.

[0097] The automatic region recognition process of the recognition module in this embodiment includes:

[0098] The second image is binarized using the Otsu threshold method to obtain a first binarized image;

[0099] The Suzuki-Abe boundary tracking algorithm is used to track the boundaries of the first binary image, extract all the closed curves of the signal area, and select the closed curve of the signal area with the largest area as the outline of the signal area;

[0100] According to the outline of the signal area and the first binarized image, the noise rectangular areas in four directions are obtained;

[0101] The outline of the ghost area is obtained according to the outline of the signal area and the first image.

[0102] According to the outline of the signal area and the first binarized image, four noise rectangular areas are obtained, specifically including:

[0103] According to the outline of the signal area, the minimum enclosing rectangle of the signal area is calculated; the four corner points of the minimum enclosing rectangle of the signal area are extracted, and the four corner points of the entire first binary image are defined as reference corner points; the coordinates of the center point between each corner point of the minimum enclosing rectangle and the corresponding reference corner point are calculated respectively, and the coordinates of each center point are used as the center point of a noise rectangular area; the distances between the corner points of each minimum enclosing rectangle and the corresponding reference corner point in the x-axis and y-axis directions are calculated respectively, and half of them are taken as the width and height of the corresponding noise rectangular area, thereby obtaining noise rectangular areas in four directions.

[0104] The process of obtaining the outline of the ghost area according to the outline of the signal area and the first image specifically includes:

[0105] Calculating the centroid of the contour of the signal area and calculating the direction vector between each contour point and the centroid in the contour of the signal area, then amplifying all the direction vectors according to a preset ratio to obtain an intermediate signal area; mapping the intermediate signal area to a corresponding position in the first image to obtain an intermediate image block; removing the intermediate image block from the first image to obtain a ghost image to be identified area;

[0106] The ghost area to be identified is Gaussian filtered, and the Otsu threshold method is used to binarize the Gaussian filtered ghost area to obtain a second binary image; the second binary image is subjected to foreground erosion and then dilation to obtain the middle ghost area to be identified; the Suzuki-Abe boundary tracking algorithm is used to track the boundary of the middle ghost area to be identified, and all ghost area closed curves in the middle ghost area to be identified are extracted. The ghost area closed curve with the largest area is selected as the outline of the ghost area.

[0107] In a third aspect, the present invention further proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the automatic measurement method for ghost image indicators of nuclear magnetic resonance imaging equipment as described in any one of the first aspects.

[0108] In a fourth aspect, the present invention further proposes a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of the method for automatically measuring ghost image indicators of nuclear magnetic resonance imaging equipment as described in any one of the first aspects.

[0109] The present invention will be described below with reference to specific embodiments.

[0110] Example 1

[0111] The present invention proposes an automatic measurement method for ghost image indicators of nuclear magnetic resonance imaging equipment, comprising:

[0112] Step 1: Obtain the DICOM format image file obtained by scanning the phantom with a nuclear magnetic resonance device;

[0113] Step 2: Display the DICOM format image file, such as Figure 2 As shown;

[0114] Step 3: Convert the DICOM image file into a 16-bit first image and an 8-bit second image. The 8-bit second image is the result of window width and window level adjustment. The conversion method is shown in the following formula. The window width (WW) and window level (WL) are used to adjust the grayscale level to make the image clearer on the display device. 16-bit or 12-bit medical image data can be converted to 8-bit (0-255 grayscale values) for display. Where, Indicates the grayscale value displayed. Indicates the window width, Indicates window level;

[0115] Step 4: Noise filter the second image by Gaussian filtering, using Gaussian kernel processing;

[0116] Step 5: Use the Otsu threshold method to separate the foreground and background of the filtered image in step 4 to generate a first binary image;

[0117] Step 6: Use the Suzuki-Abe algorithm to track the boundary of the signal area on the first binary image obtained in step 5, obtain all closed curves, and select the one with the largest area as the signal area, as shown in Figure 3 As shown; and calculate the mean value S of the 16-bit image in the signal area;

[0118] Step 7: Based on the signal area calculated in step 6, its minimum bounding rectangle is obtained, and its four corner points are obtained, namely the upper left, upper right, lower left, and lower right. At the same time, the four corner points of the entire image are defined as references. Subsequently, by traversing the corner points of each bounding rectangle and the corresponding corner points of the image, the center point of the two is calculated as the center point of the noise rectangular area; then, the distance between the two points in the x-axis and y-axis directions is calculated, and half of it is taken as the width and height of the noise rectangular area, and the noise rectangular areas in four directions are obtained, as shown in the figure. Figure 4 ; Then calculate the standard deviation of the 16-bit image in the noise area , then the noise level is obtained ;

[0119] Step 8: To avoid the influence of the signal detector on the ghost area recognition, the signal area outline in step 6 is appropriately enlarged. Specifically, the centroid of the signal area outline is calculated, and then the direction vector between each contour point and the centroid is calculated. The signal area is then enlarged by the vector at a ratio of 1.1. The enlarged signal area is then removed from the 16-bit image to obtain the ghost area to be recognized.

[0120] Step 9: Filter the ghost image to be identified area obtained in step 8, and use the Otsu threshold method to separate the foreground and background to generate a second binary image, such as Figure 5 As shown in the figure, the foreground is the ghost area. To avoid the influence of isolated pixel areas, a morphological opening operation is performed on the second binary image. Specifically, the foreground erosion and then dilation are performed on the image to remove small area noise and retain the main structure of the ghost area.

[0121] Step 10: For all the ghost area closed curves obtained in step 9, select the one with the largest area as the ghost area, calculate its centroid, and use the centroid as the center point of the final ghost area. The length and width of the ghost area are 5 respectively. Calculate the mean of the ghost area. ,like Figure 6 As shown;

[0122] Step 11: Calculate the final ghost index for the statistics calculated in steps 6, 7, and 9, and get the ratio of ghost to signal , get the ratio of ghost to noise , calculate the signal-to-noise ratio .

[0123] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A method for automatically measuring ghost image indicators of nuclear magnetic resonance imaging equipment, characterized in that: include: Get image files in DICOM format; Parse and display the DICOM format image file, and convert it from high-bit pixels to 16-bit format for the first image and 8-bit format for the second image according to the tag window width and window level of the image file; Automatically identifying regions of the second image to obtain signal regions, ghost regions, and noise regions; Statistical characteristics of the signal area, ghost area and noise area are calculated respectively; Calculate the ghost index based on the statistical characteristics of the signal area, ghost area and noise area; The second image is automatically identified to obtain a signal area, a ghost area, and a noise area, specifically including: The second image is binarized using the Otsu threshold method to obtain a first binarized image; The Suzuki-Abe boundary tracking algorithm is used to track the boundaries of the first binary image, extract all the closed curves of the signal area, and select the closed curve of the signal area with the largest area as the outline of the signal area; According to the outline of the signal area and the first binarized image, the noise rectangular areas in four directions are obtained; Obtaining an outline of the ghost area according to an outline of the signal area and the first image; The process of obtaining the outline of the ghost area according to the outline of the signal area and the first image specifically includes: Calculating the centroid of the contour of the signal area and calculating the direction vector between each contour point and the centroid in the contour of the signal area, then amplifying all the direction vectors according to a preset ratio to obtain an intermediate signal area; mapping the intermediate signal area to a corresponding position in the first image to obtain an intermediate image block; removing the intermediate image block from the first image to obtain a ghost image to be identified area; wherein the preset ratio is 1.1; The ghost area to be identified is Gaussian filtered, and the Otsu threshold method is used to binarize the Gaussian filtered ghost area to obtain a second binary image; the second binary image is subjected to foreground erosion and then dilation to obtain the middle ghost area to be identified; the Suzuki-Abe boundary tracking algorithm is used to track the boundary of the middle ghost area to be identified, and all ghost area closed curves in the middle ghost area to be identified are extracted. The ghost area closed curve with the largest area is selected as the outline of the ghost area.

2. The automatic measurement method for ghost image index of nuclear magnetic resonance imaging equipment according to claim 1, characterized in that: The transformation formula of the second image is expressed as: ； Where, Indicates the grayscale value displayed. Indicates the window width, Indicates the window level.

3. The automatic measurement method for ghost image index of nuclear magnetic resonance imaging equipment according to claim 1, characterized in that: According to the outline of the signal area and the first binarized image, the noise rectangular areas in four directions are obtained, including: According to the outline of the signal area, the minimum enclosing rectangle of the signal area is calculated; the four corner points of the minimum enclosing rectangle of the signal area are extracted, and the four corner points of the entire first binary image are defined as reference corner points; the coordinates of the center point between each corner point of the minimum enclosing rectangle and the corresponding reference corner point are calculated respectively, and the coordinates of each center point are used as the center point of a noise rectangular area; the distances between the corner points of each minimum enclosing rectangle and the corresponding reference corner point in the x-axis and y-axis directions are calculated respectively, and half of them are taken as the width and height of the corresponding noise rectangular area, thereby obtaining noise rectangular areas in four directions.

4. The automatic measurement method for ghost image index of nuclear magnetic resonance imaging equipment according to claim 1, characterized in that: The statistical characteristics of the signal area, ghost area and noise area are calculated separately, including: determining a signal image block corresponding to the signal area in the first image, and calculating a mean value of pixel values ​​of the signal image block; Determine the noise image blocks corresponding to the noise rectangular area in the first image, and calculate the standard deviation of the pixel values ​​of each noise image block; calculate the noise level according to the mean of the standard deviations of the pixel values ​​of the four noise image blocks; According to the outline of the ghost area, the centroid of the ghost area is calculated, and the centroid is used as the center point of the ghost area to calculate the mean of the ghost area.

5. An automatic measurement system for ghost image indicators of nuclear magnetic resonance imaging equipment, characterized in that: include: Acquisition module, used to obtain image files in DICOM format; A conversion module is used to parse and display the image file in DICOM format, and convert the high-bit pixel into a first image in 16-bit format and a second image in 8-bit format according to the tag window width and window level of the image file; an identification module, configured to automatically identify regions of the second image to obtain signal regions, ghost regions, and noise regions; A calculation module, used to calculate the statistical characteristics of the signal area, the ghost area and the noise area respectively; Calculate the ghost index based on the statistical characteristics of the signal area, ghost area and noise area; The automatic region recognition process of the recognition module includes: The second image is binarized using the Otsu threshold method to obtain a first binarized image; The Suzuki-Abe boundary tracking algorithm is used to track the boundaries of the first binary image, extract all the closed curves of the signal area, and select the closed curve of the signal area with the largest area as the outline of the signal area; According to the outline of the signal area and the first binarized image, the noise rectangular areas in four directions are obtained; Obtaining an outline of the ghost area according to an outline of the signal area and the first image; The process of obtaining the outline of the ghost area according to the outline of the signal area and the first image specifically includes: Calculating the centroid of the contour of the signal area and calculating the direction vector between each contour point and the centroid in the contour of the signal area, then amplifying all the direction vectors according to a preset ratio to obtain an intermediate signal area; mapping the intermediate signal area to a corresponding position in the first image to obtain an intermediate image block; removing the intermediate image block from the first image to obtain a ghost image to be identified area; wherein the preset ratio is 1.1; The ghost area to be identified is Gaussian filtered, and the Otsu threshold method is used to binarize the Gaussian filtered ghost area to obtain a second binary image; the second binary image is subjected to foreground erosion and then dilation to obtain the middle ghost area to be identified; the Suzuki-Abe boundary tracking algorithm is used to track the boundary of the middle ghost area to be identified, and all ghost area closed curves in the middle ghost area to be identified are extracted. The ghost area closed curve with the largest area is selected as the outline of the ghost area.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for automatically measuring ghost image indicators of nuclear magnetic resonance imaging equipment described in any one of claims 1 to 4 are implemented.

7. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method for automatically measuring ghost image indicators of nuclear magnetic resonance imaging equipment described in any one of claims 1 to 4 are implemented.

Images