Medical image processing method, device, computer equipment and readable storage medium

Abstract

The present application relates to a medical image processing method, apparatus, computer equipment, and storage medium. The method comprises: acquiring a medical image, performing segmentation processing on a reference frame in the medical image to obtain a mask of a target area in the reference frame, performing registration processing on a background area image of a floating frame in the medical image based on the reference frame and the mask to obtain an updated floating frame, performing motion correction processing on the background area in the updated floating frame to obtain a motion-corrected image corresponding to the floating frame, and performing motion analysis on the target area image in the motion-corrected image. This method can directly process a real medical image to determine a motion-corrected image corresponding to a target organ in a floating frame without resorting to other information to achieve motion correction, thereby ensuring that the motion of the target organ in the motion-corrected image is consistent with the actual motion state of the target organ, thereby improving the accuracy of the acquired respiratory motion-corrected image.

Description

Technical Field

[0001] The present application relates to the field of medical technology, and in particular to a medical image processing method, apparatus, computer equipment, and readable storage medium. Background Art

[0002] In clinical practice, the temporal dimension of cine images is used to assess the motion of certain organs. However, human organ motion includes cardiac, respiratory, and intestinal movements, and these movements can be coupled. Therefore, to analyze the motion of the target organ and avoid affecting the respiratory motion of other organs, the respiratory motion of these other organs must be corrected to protect the respiratory motion of the target organ.

[0003] Traditional techniques use motion models such as rotation, translation, shearing, and temporal or spatial angles to correct for respiratory motion in other organs, thereby protecting them. However, this traditional approach can lead to inaccurate images from respiratory motion correction. Summary of the Invention

[0004] Based on this, it is necessary to provide a medical image processing method, apparatus, computer equipment and readable storage medium to address the above technical problems.

[0005] A medical image processing method, comprising:

[0006] Acquiring a medical image, the medical image comprising a reference frame and at least one floating frame;

[0007] Perform segmentation processing on the reference frame to obtain a mask of the target area in the reference frame;

[0008] Based on the reference frame and the mask, the background area in the floating frame is registered to obtain an updated floating frame, where the floating frame includes the target area and the background area;

[0009] A motion correction process is performed on the background area in the updated floating frame to obtain a motion corrected image corresponding to the floating frame, and a motion analysis is performed on the target area image in the motion corrected image.

[0010] In one embodiment, for each floating frame, a background region in the floating frame is registered based on a reference frame and a mask to obtain an updated floating frame, including:

[0011] determining an initial driving force based on the reference frame and the floating frame;

[0012] determining a target driving force based on the initial driving force and the mask;

[0013] The target driving force and the floating frame are processed to determine an updated floating frame.

[0014] In one embodiment, determining the initial driving force based on the reference frame and the floating frame includes:

[0015] Obtaining correlation coefficients between pixels in the reference frame and pixels in the floating frame;

[0016] The initial driving force is obtained by performing negative gradient operation on the floating frame through the correlation coefficient.

[0017] In one embodiment, obtaining a correlation coefficient between a pixel in a reference frame and a pixel in a floating frame includes:

[0018] Calculating the variance of the reference frame and the variance of the floating frame according to the pixel values ​​of the pixel points in the reference frame and the pixel values ​​of the pixel points in the floating frame, and calculating the covariance between the reference frame and the floating frame;

[0019] Correlation parameters between each pixel point in the reference frame and the pixel point in the floating frame are determined through the covariance between the reference frame and the floating frame, the variance of the reference frame, and the variance of the floating frame.

[0020] In one embodiment, determining a target driving force based on an initial driving force and a mask includes:

[0021] Perform subtraction operation using a preset standard matrix and mask to obtain a subtraction result;

[0022] The initial driving force and the subtraction result are multiplied to obtain the target driving force.

[0023] In one embodiment, processing the target driving force and the floating frame to determine an updated floating frame includes:

[0024] Regularize the target driving force to determine the deformation field update amount of the floating frame;

[0025] The initial deformation field of the floating frame is updated by the deformation field update amount to obtain an updated deformation field;

[0026] Based on the updated deformation field, pixel values ​​in the floating frame are resampled to obtain an updated floating frame.

[0027] In one embodiment, performing motion correction processing on the background area in the updated floating frame to obtain a motion-corrected image includes: one or more iterative operations;

[0028] At least one iteration operation includes:

[0029] Determine the updated floating frame as the current floating frame;

[0030] Based on the reference frame and the mask, the background area in the current floating frame is registered to obtain a motion-corrected image of the current floating frame;

[0031] The current floating frame is updated according to the motion correction image of the current floating frame; the convergence condition of the iterative operation is that the number of iterations reaches a preset number threshold or the accuracy of related parameters reaches a preset accuracy.

[0032] A medical image processing device, comprising:

[0033] An image acquisition module, configured to acquire a medical image, the medical image comprising a reference frame and at least one floating frame;

[0034] A segmentation module is used to segment the reference frame to obtain a mask of the target area in the reference frame;

[0035] a registration processing module, configured to perform registration processing on the background area in the floating frame with reference to the frame and the mask, to obtain an updated floating frame, wherein the floating frame includes the target area and the background area;

[0036] The motion correction module is used to perform motion correction processing on the background area in the updated floating frame to obtain a motion correction image corresponding to the floating frame, and perform motion analysis on the target area image in the motion correction image.

[0037] A computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are performed:

[0038] Acquire a medical image, the medical image including a reference frame and at least one floating frame, wherein the reference frame and the floating frame respectively correspond to different physiological phases of a scan of the imaging subject;

[0039] Acquire a mask of a target region, where the target region corresponds to a region of interest in the reference frame or the floating frame;

[0040] performing registration processing on at least a portion of the floating frame based on the reference frame and the mask to obtain an updated floating frame;

[0041] A motion correction process is performed on at least a portion of the updated floating frame to obtain a motion-corrected image corresponding to the floating frame; the corresponding area of ​​the region of interest in the motion-corrected image remains consistent with the corresponding area in the floating frame.

[0042] A readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0043] Acquiring a medical image, the medical image comprising a reference frame and at least one floating frame;

[0044] Perform segmentation processing on the reference frame to obtain a mask of the target area in the reference frame;

[0045] Based on the reference frame and the mask, the background area in the floating frame is registered to obtain an updated floating frame, where the floating frame includes the target area and the background area;

[0046] A motion correction process is performed on the background area in the updated floating frame to obtain a motion corrected image corresponding to the floating frame, and a motion analysis is performed on the target area image in the motion corrected image.

[0047] The above-mentioned medical image processing method, device, computer equipment and readable storage medium, the computer equipment can obtain a medical image, segment the reference frame in the medical image, obtain a mask of the target area in the reference frame, based on the reference frame and the mask, align the background area image of the floating frame in the medical image to obtain an updated floating frame, perform motion correction on the background area in the updated floating frame to obtain a motion correction image corresponding to the floating frame, and perform motion analysis on the target area image in the motion correction image; this method can directly process the real medical image to determine the motion correction image corresponding to the target organ in the floating frame, without the need to use other information to achieve motion correction, so that the movement of the target organ in the motion correction image can be consistent with the real movement state of the target organ of the imaging object, thereby improving the accuracy of the acquired respiratory motion correction image. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A diagram showing an application environment of a medical image processing method in one embodiment;

[0049] Figure 2 is a flowchart of a medical image processing method in one embodiment;

[0050] Figure 3 1 is a flow chart of a method for performing registration processing on a background area image in a floating frame in one embodiment;

[0051] Figure 4 A schematic flow chart of a method for determining an initial driving force based on a reference frame and a floating frame in one embodiment;

[0052] Figure 5 FIG1 is a flow chart of a method for determining a target driving force based on an initial driving force and a mask in one embodiment;

[0053] Figure 6 is a flow chart of a method for processing a target driving force and a floating frame to obtain an updated floating frame in another embodiment;

[0054] Figure 7A floating frame including a heart region in another embodiment;

[0055] Figure 8 In another embodiment, Figure 8 Corresponding to the image with motion offset;

[0056] Figure 9 In another embodiment, Figure 9 Achieve motion-corrected images;

[0057] Figure 10 In another embodiment, Figure 9 The corresponding deformed image;

[0058] Figure 11 is a structural block diagram of a medical image processing device in one embodiment;

[0059] Figure 12 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0060] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0061] The medical image processing method provided in this application can be applied to Figure 1 The magnetic resonance scanning device and computer device shown. Optionally, the magnetic resonance scanning device and the computer device can be communicatively connected, and the connection method can be Bluetooth, mobile network, WiFi, etc. Optionally, the computer device can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers and portable wearable devices. It can also be implemented using an independent server or a server cluster composed of multiple servers. This embodiment does not limit the specific form of the computer device. Optionally, the magnetic resonance scanning device can scan the target area corresponding to the imaging and collect medical data, and then send the medical data to the computer device, which reconstructs the medical data to obtain a reference frame and a floating frame. Optionally, the reference frame and the floating frame can be at least one of an X-ray image, a radionuclide image, an ultrasound image, a magnetic resonance image, etc. The specific process of the medical image processing method will be specifically introduced in the following embodiments.

[0062] In order to analyze the pathological condition of the target part or target organ of the imaging object, it is necessary to protect the images corresponding to other parts or other organs in the medical image of the imaging object, further analyze the images of the target part or target organ in the medical image, and determine the movement of the target organ. In this way, the analysis results can be consistent with the analysis results when the other parts or other organs of the imaging object are not moving, thereby improving the accuracy of the analysis results and helping doctors to perform pathological analysis of the imaging object. Based on this, this embodiment provides a medical image processing method, such as Figure 2 The figure shows a flow chart of the medical image processing method, which is applied to Figure 1 The computer device in the example is used to illustrate the process, including the following steps:

[0063] S100: Acquire a medical image, where the medical image includes a reference frame and at least one floating frame.

[0064] Specifically, the medical image can be an image of the imaging object acquired in real time, or can be an image of the imaging object acquired in advance and stored locally or in the cloud. In this embodiment, the reference frame can be used as a benchmark image to perform motion correction processing on each floating frame.

[0065] It should be noted that the reference frame and the floating frame are medical images of the same imaging object corresponding to different physiological phases during one scanning process; the reference frame can be a phase with relatively stable physiological movement.

[0066] S200 , performing segmentation processing on the reference frame to obtain a mask of the target area in the reference frame.

[0067] Specifically, the reference frame includes images of the target region and a background region excluding the target region. The target region may be a region of interest (ROI) of the imaged object, corresponding to a tissue or organ of interest, i.e., the target tissue or organ. The background region may be a non-region of interest of the imaged object. Alternatively, the reference frame may be a combined image of the target region and the background region. The reference frame may be of any size, as long as the reference frame includes the target region image.

[0068] Optionally, if it is currently necessary to analyze the target part or target organ of the imaging object to obtain a pathological state, the reference frame can be a medical image including the target part and other parts around the target part, or a medical image including the target organ and other parts around the target organ.

[0069] It is understandable that the computer device can use a threshold-based, region-based, edge-based and / or neural network-based segmentation method to perform segmentation processing on the target area in the reference frame to obtain a segmented image and a mask of the target area in the segmented image. Among them, the parameters involved in the segmentation method used can be set according to the target area actually segmented. In this embodiment, a neural network-based segmentation method can be used to implement the segmentation processing; specifically, the neural network model used can be a self-organizing map neural network segmentation model, a graph neural network segmentation model and / or a genetic neural network segmentation model, etc. In this embodiment, a fully convolutional neural network segmentation model can be used to implement the segmentation processing, such as a Vnet segmentation model.

[0070] It should be noted that the mask of the target area in the segmented image can be a mask of the target area in the medical image. Optionally, the above-mentioned segmented image can be an image corresponding to the target part of the imaging object, or an image corresponding to the target organ of the imaging object. Optionally, the reference frame can be a large-area regional image including the target part or target organ, and correspondingly, the segmented image can be a small-area regional image including the target part or target organ. Optionally, the size of the segmented image can be the same as the size of the mask of the target area. The mask of the target area can be a pixel value of 1 for the pixel points corresponding to the target part or target organ to be protected, and a pixel value of 0 for the pixel points corresponding to other organs to be corrected.

[0071] In this embodiment, the target region is a target organ region. The target organ may be an organ with localized motion, such as the heart, intestines, uterus, lungs, or breast of the imaging subject. Since magnetic resonance cine imaging technology can rapidly image moving organs of the imaging subject, this embodiment uses the example of magnetic resonance cine images as both the reference frame and the floating frame to illustrate the specific process of the medical image processing method.

[0072] S300 , performing registration processing on the background area image in the floating frame based on the reference frame and the mask to obtain an updated floating frame; the floating frame includes images of the target area and the background area except the target area.

[0073] Specifically, the images processed for registration can be locally deformed images. Alternatively, registration can refer to the spatial alignment of images acquired on the same or different subjects using the same or different modalities. This is widely used in radiotherapy planning, surgical guidance, imaging motion calibration, disease diagnosis, image segmentation, and treatment efficacy tracking.

[0074] It should be noted that since the movement of the target organ can cause local deformation in the corresponding floating frame, the computer device can perform registration processing on the background area image in the floating frame based on the reference frame and the mask of the target area in the reference frame to obtain an updated floating frame. Optionally, the floating frame can be one or more medical images acquired in real time that include the target area and background areas outside the target area. Optionally, the computer device can perform registration processing on the background area image in the floating frame based on the reference frame and the mask of the target area in the reference frame.

[0075] It is understandable that the registration process can be understood as alignment and matching, and can also be understood as the process of performing translation, conversion, rotation, scaling, flipping, calculation, etc. on the image to be registered. Optionally, the image to be registered can be a floating frame.

[0076] Among them, the above-mentioned translation processing can be understood as the process of moving part or all of the pixel points in the image to be registered horizontally (x-axis) or vertically (y-axis) according to a given translation amount; the above-mentioned conversion processing can be understood as the process of converting part or all of the pixel points in the image to be registered and the corresponding coordinate points; the above-mentioned rotation processing can be understood as the process of rotating part or all of the pixel points in the image to be registered by a preset angle with the origin as the center of the circle; the above-mentioned scaling processing can be understood as the process of scaling all the pixel points in the image to be registered by a preset multiple in the horizontal (x-axis) or vertical (y-axis) direction; the above-mentioned flipping processing can be understood as the process of flipping part or all of the pixel points in the image to be registered with the preset axis as the axis of symmetry; the above-mentioned operation processing can be understood as the process of performing addition, subtraction, multiplication, division, exponential and / or logarithmic operations on part or all of the pixel points in the image to be registered.

[0077] S400 , based on the updated floating frame, performing motion correction processing on the background area image in the floating frame to obtain a motion corrected image, and performing motion analysis on the target area image in the motion corrected image.

[0078] Specifically, the aforementioned motion correction processing can be understood as translation, translation, rotation, scaling, flipping, and calculation processing. The computer device can perform translation, translation, rotation, scaling, flipping, and calculation processing on the background area image in the floating frame based on the updated floating frame to obtain a motion-corrected image. Furthermore, the computer device can perform motion analysis on the target area image in the motion-corrected image to assess the motion of the target organ of the imaging subject.

[0079] For example, when the target organ is the heart, the computer device can further perform motion analysis on the cardiac region image to evaluate the cardiac morphology and function of the imaged subject. This motion analysis can include ejection fraction calculation, cardiac stress analysis, etc. If the target organ is another organ, the purpose of the motion analysis is similar.

[0080] It should be noted that the motion-corrected image may be an image obtained by motion-correcting the background region image in the floating frame, but not the target region image. Alternatively, the target region image in the motion-corrected image and the target region image in the floating frame may be identical, but the background region image in the motion-corrected image and the background region image in the floating frame may be different. In this embodiment, the floating frame may be understood as the image to be corrected.

[0081] In the above-mentioned medical image processing method, a computer device can acquire a medical image, perform segmentation processing on a reference frame in the medical image, obtain a mask of a target area in the reference frame, perform registration processing on a background area image of a floating frame in the medical image based on the reference frame and the mask, obtain an updated floating frame, perform motion correction processing on the background area image in the floating frame based on the updated floating frame to obtain a motion-corrected image, and perform motion analysis on the target area image in the motion-corrected image; the method can protect the respiratory motion of the target organ image in the floating frame by the respiratory motion of other organ images in the floating frame, so that the respiratory motion of the target organ image in the floating frame can remain original, providing accurate basic information for the next step of analyzing the motion of the target organ; at the same time, the method can obtain the floating frame without affecting the free breathing of the imaging subject. The motion correction image corresponding to the target organ in the moving frame will not cause breathing difficulties for the imaging subject. A series of processing can be directly performed on the reference frame and the floating frame in the free breathing state of the imaging subject to determine the motion correction image corresponding to the target organ in the floating frame, thereby improving the universal applicability of the respiratory motion correction method; in addition, the method can directly process the real medical image to determine the motion correction image corresponding to the target organ in the floating frame, and does not need to use other information to realize motion correction, so that the movement of the target organ in the motion correction image can be consistent with the real movement state of the target organ of the imaging subject, thereby improving the accuracy of the acquired respiratory motion correction image; in addition, the method can also perform motion analysis on the target area image in the motion correction image to evaluate the movement of the target organ of the imaging subject.

[0082] In order to perform motion correction processing on the floating frame, it is necessary to first perform registration processing on the floating frame. Therefore, in one embodiment, if Figure 3As shown, for each floating frame, the step of performing registration processing on the background area image in the floating frame based on the reference frame and the mask in S200 to obtain an updated floating frame can be achieved by the following steps:

[0083] S210 : Determine an initial driving force based on a reference frame and a floating frame.

[0084] Specifically, the computer device may perform a first operation on the pixel values ​​of some or all of the pixels in the reference frame and the floating frame to obtain the initial driving force f1.

[0085] It should be noted that the first operation can be an operation such as addition, subtraction, multiplication, division, exponential, square root, square sum, or logarithm. Optionally, the initial driving force f1 can be understood as the deformation of the target organ and other organs in the reference frame and the floating frame. Optionally, the size of the initial driving force f1, the size of the reference frame, and the size of the floating frame can all be the same.

[0086] S220 : Determine the target driving force based on the initial driving force and the mask.

[0087] Specifically, the computer device may perform a second operation on the initial driving force f1 and the mask of the target area in the reference frame to obtain the target driving force f2. Optionally, the second operation may be the same as or different from the first operation.

[0088] S230: Process the target driving force and the floating frame to obtain an updated floating frame.

[0089] It should be noted that the computer device may perform a third operation, comparison, convolution, and / or data conversion on the target driving force and the floating frame to obtain an updated floating frame. The third operation may be the same as or different from the second and first operation processes.

[0090] The above-mentioned medical image processing method can determine the initial driving force based on the reference frame and the floating frame, determine the target driving force based on the initial driving force and the mask, process the target driving force and the floating frame, and obtain an updated floating frame; this method can protect the target organ image in the floating frame through the mask of the target area to improve the accuracy of the results of motion analysis of the target area image in the floating frame.

[0091] In one embodiment, Figure 4 As shown, the step of determining the initial driving force based on the reference frame and the floating frame in S210 may include:

[0092] S211 . Obtain correlation coefficients between pixels in the reference frame and pixels in the floating frame.

[0093] Specifically, this embodiment can perform motion correction on one or more floating frames. Therefore, for each floating frame, the computer device can use a mean square error method, a mutual information method, a correlation coefficient method, or a local cross-correlation coefficient method to perform operations on the reference frame and the floating frame, respectively, to obtain correlation coefficients between pixels in the reference frame and pixels in the floating frame. Optionally, the operations can be addition, subtraction, multiplication, division, exponential, square root, square sum, and / or logarithm operations.

[0094] The above correlation coefficient can be understood as the similarity between each pixel in the reference frame and the corresponding pixel in the floating frame.

[0095] The following describes a process for calculating correlation coefficients between pixels in a reference frame and pixels in a floating frame using a local cross-correlation coefficient method. In one embodiment, the step of obtaining correlation coefficients between pixels in the reference frame and pixels in the floating frame in S211 may specifically include: calculating the variance of the reference frame and the variance of the floating frame based on pixel values ​​in the reference frame and pixel values ​​in the floating frame, and calculating the covariance between the reference frame and the floating frame; and determining correlation parameters between each pixel in the reference frame and the pixel in the floating frame using the covariance between the reference frame and the floating frame, the variance of the reference frame, and the variance of the floating frame.

[0096] Among them, the computer device can perform calculations on the covariance between the reference frame and the floating frame, the variance of the reference frame, and the variance of the floating frame to obtain the relevant parameters lcc between each pixel point in the reference frame and the pixel point in the floating frame; the calculation processing can also be addition, subtraction, multiplication, division, exponential, square root, square sum / or logarithm and other calculation processing.

[0097] In this embodiment, if the reference frame is frame 1 and the floating frame is explained as frame 2, the computer device can calculate the variance of the pixel values ​​of all pixels in frame 1 and frame 2, which are v1 and v2 respectively, and calculate the covariance Cov between frame 1 and frame 2. 1,2 , and then calculate the correlation parameter lcc between each pixel in the reference frame and the pixel in the floating frame 1,2 =Cov 2 1,2 / (v1*v2). For each floating frame, the above method can be used to calculate the relevant parameters between the reference frame and the floating frame.

[0098] S212 , performing a negative gradient operation on the floating frame using the correlation coefficient to obtain an initial driving force.

[0099] Specifically, the computer device may calculate the negative gradient directional derivative (negative gradient operation) of the corresponding floating frame through the related parameters between the reference frame and each floating frame, and obtain the initial driving force f1 corresponding to each floating frame.

[0100] The above-mentioned medical image processing method can obtain the correlation coefficient between the pixel points in the reference frame and the pixel points in the floating frame, perform a negative gradient operation on the floating frame through the correlation coefficient, obtain the initial driving force, and further determine the target driving force through the initial driving force, so as to perform motion correction on the floating frame through the target driving force, thereby improving the accuracy of the motion correction result.

[0101] In one embodiment, Figure 5 As shown, the step of determining the target driving force based on the initial driving force and the mask in S220 may include:

[0102] S221 . Perform a subtraction operation on the preset standard matrix and the mask to obtain a subtraction operation result.

[0103] Specifically, the preset standard matrix can be a constant matrix, the size of which is the same as the size of the mask of the target area in the reference frame. Optionally, all values ​​in the standard matrix can be the same or different. Each value in the standard matrix can be any value, but in this embodiment, all values ​​in the standard matrix are 1.

[0104] It should be noted that the computer device can use a preset standard matrix to perform a subtraction with the mask of the target area in the reference frame to obtain a subtraction result. Optionally, the subtraction result can be in matrix form, and the size of the subtraction result can also be the same as the size of the mask of the target area in the reference frame.

[0105] S222: Multiply the initial driving force and the subtraction result to obtain the target driving force.

[0106] It should be noted that the computer device may multiply the initial driving force and the value at the position corresponding to the subtraction result to obtain the target driving force. Optionally, the target driving force may be equal to the initial driving force. The value at the corresponding position of the target driving force may be equal to the product of the initial driving force and the value at the position corresponding to the subtraction result.

[0107] The above-mentioned medical image processing method can determine the target driving force based on the initial driving force, and further perform motion correction on the floating frame by using the target driving force, so that the method can improve the accuracy of the motion correction result.

[0108] The following describes how to obtain the updated floating frame. In one embodiment, Figure 6 As shown, the step of processing the target driving force and the floating frame to obtain an updated floating frame in S230 may specifically include:

[0109] S231 : Regularize the target driving force to determine the deformation field update amount of the floating frame.

[0110] Specifically, the computer device can perform regularization processing on the target driving force f2 to obtain the floating frame's deformation field update u. Alternatively, the floating frame's deformation field update u can be understood as the change in the target organ's deformation field s during motion. In this embodiment, the size of the floating frame's deformation field update u is the same as the size of the floating frame.

[0111] It is understandable that the above regularization processing can be an addition operation, a subtraction operation, a multiplication operation, a division operation, an exponential operation and / or a logarithmic operation. However, in this embodiment, the above regularization processing is a convolution operation. Specifically, the computer device can perform a convolution operation on the target driving force f2 and the convolution kernel k to obtain the deformation field update amount u of the floating frame. The deformation field update amount u of the floating frame is equal to Optionally, the convolution kernel k may be a transposed convolution kernel, a separable convolution kernel, a dilated convolution kernel, etc. However, in this embodiment, the convolution kernel k is a Gaussian convolution kernel.

[0112] S232: Update the initial deformation field of the floating frame using the deformation field update amount to obtain an updated deformation field.

[0113] Specifically, the computer device can perform an arithmetic operation on the deformation field update amount u and the initial deformation field s1 of the floating frame to implement the update process and obtain the updated deformation field s2. In this embodiment, the deformation field update amount u and the initial deformation field s1 of the floating frame are added to obtain the updated deformation field s2, i.e., the updated deformation field s2 is equal to s1 + u. Optionally, the size of the initial deformation field s1 of the floating frame is the same as the size of the updated deformation field s2.

[0114] S233 : Resample the pixel values ​​in the floating frame based on the updated deformation field to obtain an updated floating frame.

[0115] In this embodiment, the size of the updated deformation field s2 is the same as the size of the floating frame. Optionally, the computer device can perform a fourth operation on the updated deformation field s2 and the pixel values ​​in the floating frame to complete resampling, thereby obtaining an updated floating frame. Optionally, the fourth operation may be the same as or different from the third, second, and first operations described above. However, in this embodiment, the fourth operation is an addition operation.

[0116] Among them, the computer device can perform addition operations on each numerical value in the updated deformation field s2 and the pixel value at the corresponding position in the floating frame to obtain the sum corresponding to each position in the floating frame, and combine the sums at all positions according to the corresponding positions to obtain the updated floating frame.

[0117] The above-mentioned medical image processing method can process the target driving force and the floating frame to obtain an updated floating frame, thereby realizing the registration of the floating frame to further improve the accuracy of the results of motion analysis of the target area image in the floating frame; and the method can also eliminate the overfitting processing of the target driving force through regularization processing, thereby improving the accuracy of the motion correction results.

[0118] In order to protect the target organ image in the floating frame and accurately perform motion correction on the background area image in the floating frame, in one embodiment, the step of performing motion correction on the background area in the updated floating frame in S300 may include one or more iterative operations:

[0119] At least one iterative operation includes: determining the updated floating frame as the current floating frame; performing registration processing on the background area in the current floating frame based on the reference frame and the mask to obtain a motion-corrected image of the current floating frame; updating the current floating frame according to the motion-corrected image of the current floating frame; and the convergence condition of the iterative operation is that the number of iterations reaches a preset number threshold or the accuracy of relevant parameters reaches a preset accuracy.

[0120] It should be noted that the computer device can determine the acquired updated floating frame as the current floating frame, and perform alignment processing on the background area in the current floating frame to obtain the updated current floating frame. If the current number of iterations reaches a preset threshold or the accuracy of the current relevant parameters reaches a preset accuracy, the updated current floating frame will be determined as the motion corrected image of the current floating frame.

[0121] It is understood that if the current number of iterations does not reach a preset threshold and the accuracy of the current relevant parameters does not reach a preset accuracy, the motion-corrected image of the current floating frame can be determined as the current floating frame. The step of registering the background area in the current floating frame based on the reference frame and the mask is continued for the current floating frame until the current number of iterations reaches the preset threshold or the accuracy of the current relevant parameters reaches a preset accuracy. The iterative operation ends and the updated current floating frame is determined as the motion-corrected image of the current floating frame. The motion-corrected image is an image that protects the target area in the current floating frame and performs motion correction on the background area in the current floating frame.

[0122] In this embodiment, the iterative operation can be understood as repeatedly executing the above step S300 for the current floating frame until the number of iterations is equal to a preset iteration threshold or the precision of the obtained relevant parameters is equal to a preset precision, and the iterative operation ends.

[0123] It can be understood that, for each floating frame, the initial deformation field s1 is 0. During the next iterative process, the initial deformation field s1 is equal to the updated deformation field s2.

[0124] For example, if the floating frame is an image including a heart region, and Figure 7 The reference frame includes the heart region. The reference frame is an image of the diaphragm region below the heart without motion offset. Figure 7 The white curve in the middle represents the boundary line of the diaphragm area. Figure 7 Before the motion correction process is performed on the diaphragm region in the floating frame, the diaphragm region below the heart region has motion offset, as shown in FIG. Figure 8 As shown ( Figure 8 The white curve in the middle represents the offset line of the diaphragm region due to the movement offset, which causes the boundary line of the diaphragm region to shift). Correspondingly, Figure 9 Shown is a motion-corrected image obtained after motion correction processing is performed on the floating frame; Figure 9 and Figure 7 Roughly the same, that is, Figure 8 After the motion correction processing of the diaphragm region in the image, the boundary of the diaphragm region is basically consistent with the boundary of the diaphragm region in the reference frame. Figure 10 Shown Figure 8 The corresponding deformed image, Figure 10 The squares in the figure represent the areas without motion deformation, the deformed grids represent the areas with motion deformation, and the white area represents the heart area. In this embodiment, the heart area in the floating frame is protected and the diaphragm area is motion corrected.

[0125] The above-mentioned medical image processing method can obtain the optimal updated floating frame through iterative loop processing, and determine the optimal updated floating frame as the motion correction image of the floating frame. This method can not only correct the motion influence of the background area in the floating frame, but also protect the motion of the target area, thereby improving the universal applicability of the respiratory motion correction method; in addition, this method can directly process the real medical image to determine the motion correction image corresponding to the target organ in the floating frame, without the need to use other information to achieve motion correction, so that the motion state of the target organ in the motion correction image can be consistent with the actual motion state of the target organ of the imaging object, thereby improving the accuracy of the acquired respiratory motion correction image.

[0126] To facilitate understanding by those skilled in the art, the medical image processing method provided in this application is described by taking a computer device as an example. Specifically, the method includes:

[0127] (1) Acquire a medical image, where the medical image includes a reference frame and at least one floating frame.

[0128] (2) The reference frame is segmented to obtain a mask of the target area in the reference frame. The floating frame includes the target area and the background area.

[0129] For each floating frame, perform the following steps:

[0130] (3) According to the pixel values ​​of the pixels in the reference frame and the pixel values ​​of the pixels in the floating frame, the variance of the reference frame and the variance of the floating frame are calculated, and the covariance between the reference frame and the floating frame is calculated.

[0131] (4) Determine the correlation parameters between each pixel in the reference frame and the pixel in the floating frame through the covariance between the reference frame and the floating frame, the variance of the reference frame, and the variance of the floating frame.

[0132] (5) Perform negative gradient operation on the floating frame through the correlation coefficient to obtain the initial driving force.

[0133] (6) Perform subtraction operation using the preset standard matrix and the mask to obtain the subtraction result.

[0134] (7) The initial driving force and the subtraction result are multiplied to obtain the target driving force.

[0135] (8) Regularize the target driving force and determine the deformation field update amount of the floating frame.

[0136] (9) The initial deformation field of the floating frame is updated by the deformation field update amount to obtain an updated deformation field.

[0137] (10) Based on the updated deformation field, the pixel values ​​in the floating frame are resampled to obtain an updated floating frame.

[0138] (11) Add the updated deformation field and the pixel values ​​in the floating frame to obtain the updated floating frame.

[0139] (12) The updated floating frame is determined as the current floating frame. For the current floating frame, the above steps (3) to (11) are continued to be iteratively executed. The convergence condition of the iterative operation is that the number of iterations reaches a preset number threshold or the accuracy of the relevant parameters reaches a preset accuracy.

[0140] The execution process of the above (1) to (13) can be specifically referred to the description of the above embodiment. The implementation principles and technical effects are similar and will not be repeated here.

[0141] It should be understood that although Figure 2-6 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. In addition, Figure 2-6 At least part of the steps may include multiple steps or multiple stages. These steps or stages are not necessarily performed at the same time, but can be performed at different times. The order of execution of these steps or stages is not necessarily one by one, but can be performed in turn or alternately with other steps or at least part of the steps or stages in other steps.

[0142] In one embodiment, Figure 11 As shown, a medical image processing device is provided, comprising: an image acquisition module 11, a segmentation module 12, a registration processing module 13 and a motion correction module 14, wherein:

[0143] An image acquisition module 11 is configured to acquire a medical image, wherein the medical image includes a reference frame and at least one floating frame;

[0144] The segmentation module 12 is used to perform segmentation processing on the reference frame to obtain a mask of the target area in the reference frame;

[0145] A registration processing module 13 is configured to perform registration processing on the background area in the floating frame with reference to the frame and the mask to obtain an updated floating frame, where the floating frame includes the target area and the background area;

[0146] The motion correction module 14 is configured to perform motion correction processing on the background area in the updated floating frame to obtain a motion corrected image corresponding to the floating frame, and perform motion analysis on the target area image in the motion corrected image.

[0147] The medical image processing device provided in this embodiment can execute the above method embodiment. Its implementation principles and technical effects are similar and will not be repeated here.

[0148] In one embodiment, the registration processing module 13 includes: an initial driving force determination unit, a target driving force determination unit and a processing unit, wherein:

[0149] an initial driving force determining unit, configured to determine an initial driving force according to a reference frame and a floating frame;

[0150] a target driving force determining unit, configured to determine a target driving force based on the initial driving force and the mask;

[0151] The processing unit is used to process the target driving force and the floating frame to determine an updated floating frame.

[0152] The medical image processing device provided in this embodiment can execute the above method embodiment. Its implementation principles and technical effects are similar and will not be repeated here.

[0153] In one embodiment, the initial driving force determination unit includes: a processing subunit and a negative gradient calculation subunit, wherein:

[0154] A processing subunit, configured to obtain a correlation coefficient between a pixel point in a reference frame and a pixel point in a floating frame;

[0155] The negative gradient operation subunit is used to perform negative gradient operation on the floating frame through the correlation coefficient to obtain the initial driving force.

[0156] The medical image processing device provided in this embodiment can execute the above method embodiment. Its implementation principles and technical effects are similar and will not be repeated here.

[0157] In one embodiment, the processing subunit is specifically used to calculate the variance of the reference frame and the variance of the floating frame based on the pixel values ​​in the reference frame and the pixel values ​​in the floating frame, and calculate the covariance between the reference frame and the floating frame, and determine the relevant parameters between each pixel point in the reference frame and the pixel point in the floating frame through the covariance between the reference frame and the floating frame, the variance of the reference frame, and the variance of the floating frame.

[0158] The medical image processing device provided in this embodiment can execute the above method embodiment. Its implementation principles and technical effects are similar and will not be repeated here.

[0159] In one embodiment, the target driving force determination unit includes: a subtraction operator unit and a multiplication operator unit, wherein:

[0160] A subtraction operator unit, configured to perform a subtraction operation on a preset standard matrix and a mask to obtain a subtraction result;

[0161] The multiplication operation subunit is used to perform a multiplication operation on the initial driving force and the subtraction operation result to obtain the target driving force.

[0162] The medical image processing device provided in this embodiment can execute the above method embodiment. Its implementation principles and technical effects are similar and will not be repeated here.

[0163] In one embodiment, the processing unit includes: a regularization processing subunit, a deformation field updating subunit and a resampling subunit, wherein:

[0164] A regularization processing subunit, used to perform regularization processing on the target driving force and determine the deformation field update amount of the floating frame;

[0165] The deformation field updating subunit is used to update the initial deformation field of the floating frame by using the deformation field update amount to obtain an updated deformation field;

[0166] The resampling subunit is used to resample the pixel values ​​in the floating frame according to the updated deformation field to obtain an updated floating frame.

[0167] The medical image processing device provided in this embodiment can execute the above method embodiment. Its implementation principles and technical effects are similar and will not be repeated here.

[0168] In one embodiment, the motion correction module 14 includes an iterative processing unit, wherein:

[0169] The iterative processing unit is used to determine the updated floating frame as the current floating frame, perform registration processing on the background area in the current floating frame based on the reference frame and the mask to obtain a motion-corrected image of the current floating frame, and update the current floating frame based on the motion-corrected image of the current floating frame; the convergence condition of the iterative operation is that the number of iterations reaches a preset number threshold or the accuracy of the relevant parameters reaches a preset accuracy.

[0170] The medical image processing device provided in this embodiment can execute the above method embodiment. Its implementation principles and technical effects are similar and will not be repeated here.

[0171] The specific definition of the medical image processing apparatus can be found in the definition of the medical image processing method above and will not be repeated here. Each module in the aforementioned medical image processing apparatus may be implemented in whole or in part through software, hardware, or a combination thereof. Each of these modules may be embedded in or independent of a processor in a computer device in hardware form, or may be stored in a computer device memory in software form, so that the processor can call and execute the corresponding operations of each module.

[0172] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 12As shown. The computer device includes a processor, a memory and a network interface connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store medical images, image frames and image frames to be corrected. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a medical image processing method is implemented.

[0173] Those skilled in the art will understand that Figure 12 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0174] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0175] Acquire a medical image, the medical image including a reference frame and at least one floating frame, wherein the reference frame and the floating frame respectively correspond to different physiological phases of a scan of the imaging subject;

[0176] Acquire a mask of a target region, where the target region corresponds to a region of interest in the reference frame or the floating frame;

[0177] performing registration processing on at least a portion of the floating frame based on the reference frame and the mask to obtain an updated floating frame;

[0178] A motion correction process is performed on at least a portion of the updated floating frame to obtain a motion-corrected image corresponding to the floating frame; the corresponding area of ​​the region of interest in the motion-corrected image remains consistent with the corresponding area in the floating frame.

[0179] In one embodiment, a readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0180] Acquiring a medical image, the medical image comprising a reference frame and at least one floating frame;

[0181] Perform segmentation processing on the reference frame to obtain a mask of the target area in the reference frame;

[0182] Based on the reference frame and the mask, the background area in the floating frame is registered to obtain an updated floating frame, where the floating frame includes the target area and the background area;

[0183] A motion correction process is performed on the background area in the updated floating frame to obtain a motion corrected image corresponding to the floating frame, and a motion analysis is performed on the target area image in the motion corrected image.

[0184] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0185] Acquiring a medical image, the medical image comprising a reference frame and at least one floating frame;

[0186] Perform segmentation processing on the reference frame to obtain a mask of the target area in the reference frame;

[0187] Based on the reference frame and the mask, the background area in the floating frame is registered to obtain an updated floating frame, where the floating frame includes the target area and the background area;

[0188] A motion correction process is performed on the background area in the updated floating frame to obtain a motion corrected image corresponding to the floating frame, and a motion analysis is performed on the target area image in the motion corrected image.

[0189] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0190] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0191] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A medical image processing method, characterized in that: The method comprises: Acquiring a medical image, wherein the medical image includes a reference frame and at least one floating frame; performing segmentation processing on the reference frame to obtain a mask of a target area in the reference frame; For each floating frame, determining an initial driving force based on the reference frame and the floating frame; the initial driving force represents the deformation amount of the target organ and other organs in the reference frame and the floating frame; determining a target driving force based on the initial driving force and the mask; the target driving force represents a change in the deformation field of the target organ during movement; Processing the target driving force and the floating frame to determine an updated floating frame; the floating frame includes the target area and the background area; A motion correction process is performed on the background area in the updated floating frame to obtain a motion-corrected image corresponding to the floating frame, and a motion analysis is performed on the target area image in the motion-corrected image.

2. The method according to claim 1, characterized in that The target area is a region of the organ where local motion occurs.

3. The method according to claim 1, characterized in that The determining of the initial driving force based on the reference frame and the floating frame comprises: Obtaining a correlation coefficient between a pixel point in the reference frame and a pixel point in the floating frame; A negative gradient operation is performed on the floating frame using the correlation coefficient to obtain the initial driving force.

4. The method according to claim 3, characterized in that The obtaining of the correlation coefficient between the pixel points in the reference frame and the pixel points in the floating frame includes: Calculating the variance of the reference frame and the variance of the floating frame according to the pixel values ​​of the pixel points in the reference frame and the pixel values ​​of the pixel points in the floating frame, and calculating the covariance between the reference frame and the floating frame; Correlation parameters between each pixel point in the reference frame and the pixel point in the floating frame are determined by the covariance between the reference frame and the floating frame, the variance of the reference frame, and the variance of the floating frame.

5. The method according to claim 1, wherein The determining of the target driving force based on the initial driving force and the mask includes: Performing a subtraction operation on the mask using a preset standard matrix to obtain a subtraction operation result; The initial driving force and the subtraction result are multiplied to obtain the target driving force.

6. The method according to any one of claims 1 to 5, characterized in that The processing of the target driving force and the floating frame to determine the updated floating frame includes: performing regularization processing on the target driving force to determine an update amount of the deformation field of the floating frame; updating the initial deformation field of the floating frame by using the deformation field update amount to obtain an updated deformation field; Based on the updated deformation field, pixel values ​​in the floating frame are resampled to obtain the updated floating frame.

7. The method according to claim 1, characterized in that The performing motion correction processing on the background area in the updated floating frame includes: one or more iterative operations; At least one iteration operation includes: determining the updated floating frame as the current floating frame; performing registration processing on the background area in the current floating frame based on the reference frame and the mask to obtain a motion-corrected image of the current floating frame; The current floating frame is updated according to the motion corrected image of the current floating frame; and the convergence condition of the iterative operation is that the number of iterations reaches a preset iteration threshold or the accuracy of related parameters reaches a preset accuracy.

8. A medical image processing device, characterized in that: The device comprises: An image acquisition module, configured to acquire a medical image, wherein the medical image includes a reference frame and at least one floating frame; a segmentation module, configured to perform segmentation processing on the reference frame to obtain a mask of a target area in the reference frame; a registration processing module configured to determine, for each floating frame, an initial driving force based on the reference frame and the floating frame; the initial driving force representing the deformation amount of the target organ and other organs in the reference frame and the floating frame; determine a target driving force based on the initial driving force and the mask; the target driving force representing the amount of change in the deformation field of the target organ under motion; process the target driving force and the floating frame to determine an updated floating frame; the floating frame including the target area and the background area; The motion correction module is used to perform motion correction processing on the background area in the updated floating frame to obtain a motion corrected image corresponding to the floating frame, and perform motion analysis on the target area image in the motion corrected image.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the following steps are implemented: Acquire a medical image, wherein the medical image includes a reference frame and at least one floating frame, wherein the reference frame and the floating frame respectively correspond to different physiological phases of a scan of an imaging subject; Acquire a mask of a target region, where the target region corresponds to a region of interest in the reference frame or the floating frame; For each floating frame, determining an initial driving force based on the reference frame and the floating frame; the initial driving force represents the deformation amount of the target organ and other organs in the reference frame and the floating frame; determining a target driving force based on the initial driving force and the mask; the target driving force represents a change in the deformation field of the target organ during movement; Processing the target driving force and the floating frame to determine an updated floating frame; the floating frame includes the target area and the background area; A motion correction process is performed on at least a portion of the updated floating frame to obtain a motion-corrected image corresponding to the floating frame; and a corresponding area of ​​the region of interest in the motion-corrected image remains consistent with a corresponding area in the floating frame.

10. A 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 according to any one of claims 1 to 7 are implemented.

Images