Nuclear magnetic resonance image scanning data reconstruction method and system

Abstract

The present invention discloses a method and system for reconstructing nuclear magnetic resonance (NMR) image scanning data, belonging to the field of medical imaging. The present invention solves the problem of poor quality of reconstructed images in the prior art. During the scanning process of the MRI scanning device, the present invention can correct motion artifacts by using the navigation echo method and the motion compensation method, thereby improving the quality of the reconstructed images. After obtaining the NMR image scanning data, denoising, enhancement, data standardization and normalization are performed, which can improve the reliability and further improve the quality of the reconstructed images. A neural network model is used to reconstruct the NMR image scanning data to analyze the quality changes of the reconstructed images under different MRI scanning devices or different scanning parameters, verify the effects of the reconstructed images, perform image segmentation and image registration processing on the reconstructed images, and evaluate the quality of the reconstructed images by quantitative and qualitative methods, which can more comprehensively and accurately evaluate the quality of the reconstructed images.

Description

Technical Field

[0001] The present invention relates to the technical field of medical imaging, and specifically to a method and system for reconstructing nuclear magnetic image scanning data. Background Art

[0002] Magnetic Resonance Imaging (MRI) is a non-invasive medical imaging technique that uses strong magnetic fields and harmless radio waves to generate detailed images of the human body. However, during the MRI scanning process, data acquisition and image reconstruction are crucial steps that can directly affect image quality and diagnostic accuracy. During data acquisition, due to different devices or different scanning parameters, the quality of the reconstructed images can vary. Additionally, due to the uncertainty of the physiological movements of the research object, motion artifacts are likely to occur during data acquisition, thereby reducing the quality of the reconstructed images. Therefore, the existing requirements are not met, and for this reason, we propose a method and system for reconstructing nuclear magnetic image scanning data. Summary of the Invention

[0003] The purpose of the present invention is to provide a method and system for reconstructing nuclear magnetic image scanning data. During the scanning process of an MRI scanning device, for the artifact problems caused by the physiological activities of the research object, the navigation echo method and the motion compensation method can be used to correct motion artifacts, thereby improving the quality of the reconstructed images. After obtaining the nuclear magnetic image scanning data of the research object, denoising and enhancement, as well as data standardization and normalization, are performed to improve the reliability of the nuclear magnetic image scanning data, thereby further improving the quality of the reconstructed images. Then, a neural network model is used to reconstruct the nuclear magnetic image scanning data, and based on the reconstructed images, the quality changes of the reconstructed images under different MRI scanning devices or different scanning parameters are analyzed to verify the reconstruction image effect. Finally, image segmentation and image registration processing are performed on the reconstructed images, and the quality of the reconstructed images is evaluated through quantitative and qualitative methods, solving the problems raised in the above background art.

[0004] To achieve the above purpose, the present invention provides the following technical solution: A method for reconstructing nuclear magnetic image scanning data, comprising the following steps:

[0005] Set the scanning parameters of the MRI scanning device, including the repetition time, echo time, and matrix size;

[0006] Place the research object on the MRI scanning device, start the MRI scanning device to scan and record the data during the imaging process, and obtain the nuclear magnetic image scanning data of the research object;

[0007] Process the obtained nuclear magnetic image scanning data, including denoising and enhancement, as well as data standardization and normalization;

[0008] Build a neural network model and perform training and testing, and use the trained neural network model to reconstruct the nuclear magnetic image scanning data;

[0009] Perform further processing on the reconstructed image, including segmentation and registration. At the same time, evaluate the quality of the reconstructed image through quantitative and qualitative methods;

[0010] Among them, evaluating the quality of the reconstructed image through quantitative and qualitative methods includes:

[0011] Perform scale transformation on the reconstructed image according to multiple preset sizes to obtain scaled images corresponding to each preset size, extract features from all the scaled images to obtain image features, and unify the image features to obtain standard features;

[0012] Based on the standard features of the reconstructed image and the scaled images, calculate the quality value of the reconstructed image according to the following formula ;

[0013]

[0014] Among them, represents the number of scaled images, represents the number of features, represents the th standard feature value in the reconstructed image, represents the th th standard feature value in the th scaled image, represents the scaling weight value of the

[0015] th scaled image;

[0016] If so, determine that the clarity of the reconstructed image meets the quality requirements;

[0017] Otherwise, determine that the clarity of the reconstructed image does not meet the quality requirements;

[0018] After determining that the clarity of the reconstructed image meets the quality requirements, obtain the initial reconstructed image before registration, and calculate the cross-correlation coefficient between the reconstructed image and the initial reconstructed image according to the following formula;

[0019]

[0020] Among them, represents the cross-correlation coefficient between the reconstructed image and the initial reconstructed image, represents the covariance between the reconstructed image and the initial reconstructed image, represents the variance of the reconstructed image, represents the variance of the initial reconstructed image, represents the expected variance of the reconstructed image based on the initial reconstructed image, represents the expected variance of the initial reconstructed image based on the reconstructed image;

[0021] judge whether the cross - correlation coefficient between the reconstructed image and the initial reconstructed image meets the preset coefficient requirement;

[0022] If so, determine that the artifacts of the reconstructed image meet the quality requirements;

[0023] Otherwise, determine that the artifacts of the reconstructed image do not meet the quality requirements.

[0024] Furthermore, during the scanning process of the MRI scanning device, for the artifact problem caused by the physiological activities of the research object, corresponding correction methods are adopted to correct the motion artifacts. The correction methods include the navigation echo method and the motion compensation method.

[0025] Furthermore, the correction method is specifically:

[0026] The navigation echo method is used for:

[0027] During the scanning process of the MRI scanning device, track and correct the image artifacts caused by respiratory physiological activities, use 2D radio - frequency pulses to track the diaphragm movement, and perform data acquisition when the diaphragm movement is within the set acceptable range;

[0028] The motion compensation method is used for:

[0029] By describing the motion of each small block in adjacent frames of the nuclear magnetic image scanning data, move each small block of the previous frame to the corresponding position in the current frame, thereby reducing the spatial redundancy and motion artifacts in the nuclear magnetic image scanning data sequence.

[0030] Furthermore, the denoising and enhancement are specifically:

[0031] Denoising processing: Use a denoising algorithm to perform denoising processing on the nuclear magnetic image scanning data, remove noise and retain image details. The denoising algorithm includes wavelet transform or non - local mean filtering;

[0032] Enhancement processing: Use an enhancement processing method to enhance the quality, contrast and clarity of the nuclear magnetic image scanning data. The enhancement processing method includes spatial domain enhancement methods and frequency domain enhancement methods.

[0033] Furthermore, the data standardization and normalization are specifically:

[0034] Unifying the data format is used for:

[0035] Unify the nuclear magnetic resonance (NMR) image scan data formats obtained from different MRI scan devices or under different scan parameters;

[0036] Data standardization, for:

[0037] Perform standardization processing on the NMR image scan data to eliminate the differences between different NMR image scan data;

[0038] Data normalization, for:

[0039] Map the NMR image scan data to a specific range, where the specific range is [0, 1] or [-1, 1].

[0040] Furthermore, construct a neural network model and conduct training and testing, specifically including the following processes:

[0041] Collect the NMR image scan data of 10 research subjects to obtain the first data set, and divide the first data set into a training set and a test set;

[0042] By using different MRI scan devices or adjusting scan parameters, collect the NMR image scan data of another 10 research subjects to obtain the second data set, and divide the second data set into a training set and a test set;

[0043] Construct a neural network model, and use the training set and the test set to train and test the neural network model respectively;

[0044] Use the training set and the test set divided from the first data set to train and test the neural network model. The trained neural network model reconstructs the NMR image scan data to obtain the first reconstructed image;

[0045] Use the training set and the test set divided from the second data set to train and test the neural network model. The trained neural network model reconstructs the NMR image scan data to obtain the second reconstructed image;

[0046] Compare the first reconstructed image with the second reconstructed image, and verify the robustness of the neural network model by analyzing the quality changes of the reconstructed images under different MRI scan devices or different scan parameters through comparison;

[0047] After the verification is completed, deploy the neural network model to reconstruct the NMR image scan data.

[0048] Furthermore, perform segmentation and registration processing on the reconstructed images, specifically:

[0049] Image segmentation: Segment the reconstructed image into non-overlapping regions with features through image segmentation;

[0050] Image registration: The reconstructed image is geometrically transformed through image registration so that the reconstructed images are spatially aligned.

[0051] Further, the geometric transformation of the reconstructed image through image registration includes:

[0052] Performing initialization interpolation processing on the reconstructed image according to a preset interpolation method to obtain the pixel positions of the interpolated pixels of the reconstructed image, and determining the pixel gray values of the interpolated pixel positions based on the pixel values of the pixels around the interpolated pixel positions and in combination with the interpolation direction to obtain an interpolated image;

[0053] Based on feature point matching, deforming the reconstructed image into the space where the interpolated image is located to obtain an initial registered image;

[0054] Extracting the image feature information in the initial registered image, and selecting a set of feature points composed of main feature points as the registration basis based on the image feature information, and performing secondary registration on the initial registered image based on the registration basis to obtain a second registered image;

[0055] Extracting the mutual information between the second registered images, and establishing a mutual information matrix based on the image order and information attributes of the second registered images;

[0056] Based on the difference between the horizontal correlation relationship between the mutual information matrices and the preset horizontal relationship, obtaining a horizontal optimization value, and based on the difference between the vertical correlation relationship between the mutual information matrices and the preset vertical relationship, obtaining a vertical optimization value;

[0057] Obtaining the initial optimization parameters of the optimization algorithm, and adjusting the initial optimization parameters based on the horizontal optimization value and the vertical optimization value to obtain target optimization parameters;

[0058] Obtaining the initial conversion relationship from the reconstructed image to the second registered image, and optimizing the initial conversion relationship based on the target optimization parameters to obtain a target conversion relationship;

[0059] Performing geometric transformation on the reconstructed image based on the target conversion relationship to obtain a target registered image.

[0060] A nuclear magnetic resonance (NMR) image scan data reconstruction system for an NMR image scan data reconstruction method, including:

[0061] A data acquisition unit for:

[0062] Setting the scan parameters of the MRI scan device, scanning and recording the data during the imaging process using the MRI scan device, and during the scan, correcting the motion artifacts using the navigation echo method and the motion compensation method, and finally obtaining the NMR image scan data of the research object.

[0063] A data processing unit, configured to:

[0064] Denoise, enhance, standardize and normalize the acquired nuclear magnetic resonance (NMR) image scan data;

[0065] An image reconstruction unit, configured to:

[0066] Construct a neural network model, train and test it, use the trained neural network model to reconstruct the NMR image scan data, and analyze the quality changes of the reconstructed images under different MRI scan devices or different scan parameters according to the reconstructed images;

[0067] A reconstructed image processing unit, configured to:

[0068] Perform image segmentation and image registration on the reconstructed images, and evaluate the quality of the reconstructed images by quantitative and qualitative methods for the processed reconstructed images.

[0069] Compared with the prior art, the beneficial effects of the present invention are:

[0070] During the scanning process of the MRI scan device, the present invention can correct motion artifacts by using the navigation echo method and the motion compensation method, can solve the artifact problem caused by the physiological activities of the research object, thereby effectively reducing the interference of motion artifacts on the image, improving the quality of the reconstructed image. After acquiring the NMR image scan data of the research object and processing it through denoising, enhancement, data standardization and normalization, the reliability of the NMR image scan data can be improved, thereby further improving the quality of the reconstructed image. Then, construct a neural network model, train and test it, use the neural network model to reconstruct the NMR image scan data, and analyze the quality changes of the reconstructed images under different MRI scan devices or different scan parameters according to the reconstructed images to verify the effect of the reconstructed images. Finally, perform image segmentation and image registration on the reconstructed images, which can improve the accuracy and efficiency of the analysis of the reconstructed images, and evaluate the quality of the reconstructed images by quantitative and qualitative methods, which can more comprehensively and accurately evaluate the quality of the reconstructed images. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 is a flowchart of the method for reconstructing NMR image scan data of the present invention;

[0072] Figure 2 is a structural diagram of the system for reconstructing NMR image scan data of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0073] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0074] In order to solve the technical problems that in the prior art during data acquisition, due to different devices or different scanning parameters, the quality of the reconstructed images varies, and in addition, due to the uncertainty of the physiological movements of the research object, motion artifacts are likely to occur during data acquisition, thereby reducing the quality of the reconstructed images, please refer to Figure 1 - Figure 2 , the following technical solutions are provided in this embodiment:

[0075] A method for reconstructing nuclear magnetic resonance (NMR) image scanning data, comprising the following steps:

[0076] Set the scanning parameters of the MRI scanning device, including the repetition time, echo time, and matrix size;

[0077] Place the research object in the MRI scanning device, start the MRI scanning device to scan and record the data during the imaging process, and obtain the NMR image scanning data of the research object;

[0078] Process the obtained NMR image scanning data, including denoising and enhancement, and data standardization and normalization;

[0079] Construct a neural network model and perform training and testing, and use the trained neural network model to reconstruct the NMR image scanning data;

[0080] Further process the reconstructed image, including segmentation and registration. At the same time, evaluate the quality of the reconstructed image through quantitative and qualitative methods;

[0081] Among them, the quantitative evaluation method mainly measures the quality of the image by calculating various indexes, and the qualitative evaluation method mainly evaluates the image quality through human visual perception. The quantitative evaluation method can provide objective data support, but it cannot fully simulate the complex perception process of the human eye. Although the qualitative evaluation method is subjective, it can capture some visual artistry or perceptual advantages and disadvantages that cannot be found by quantitative methods. Therefore, combining the quantitative and qualitative evaluation methods can more comprehensively and accurately evaluate the quality of the reconstructed image.

[0082] In one embodiment, evaluating the quality of the reconstructed image through quantitative and qualitative methods includes:

[0083] The reconstructed image is subjected to scale transformation according to multiple preset sizes to obtain scaled images corresponding to each preset size. Feature extraction is performed on all the scaled images to obtain image features, and the image features are unified to obtain standard features;

[0084] Based on the standard features of the reconstructed image and the scaled images, the quality value of the reconstructed image is calculated according to the following formula ;

[0085]

[0086] where, represents the number of scaled images, represents the number of features, represents the th standard feature value in the reconstructed image, represents the th th standard feature value in the th scaled image, represents the scaling weight value of the th scaled image;

[0087] Judge whether the quality value of the reconstructed image is greater than the preset quality value;

[0088] If so, determine that the clarity of the reconstructed image meets the quality requirements;

[0089] Otherwise, determine that the clarity of the reconstructed image does not meet the quality requirements;

[0090] After determining that the clarity of the reconstructed image meets the quality requirements, obtain the initial reconstructed image before registration, and calculate the cross-correlation coefficient between the reconstructed image and the initial reconstructed image according to the following formula;

[0091]

[0092] where, represents the cross-correlation coefficient between the reconstructed image and the initial reconstructed image, represents the covariance between the reconstructed image and the initial reconstructed image, represents the variance of the reconstructed image, represents the variance of the initial reconstructed image, represents the expected variance of the reconstructed image based on the initial reconstructed image, represents the expected variance of the initial reconstructed image based on the reconstructed image;

[0093] Judge whether the cross-correlation coefficient between the reconstructed image and the initial reconstructed image meets the preset coefficient requirements;

[0094] If so, determine that the artifacts of the reconstructed image meet the quality requirements;

[0095] Otherwise, it is determined that the artifacts of the reconstructed image do not meet the quality requirements.

[0096] In this embodiment, the scaling weight value of the scaled image is determined according to the scaling ratio.

[0097] The beneficial effects of the above design are as follows: By quantitatively evaluating the quality of the image from two aspects of image sharpness and artifacts, a more comprehensive and accurate evaluation of the quality of the reconstructed image is achieved, which is of great significance for improving the quality and diagnostic value of the reconstructed image.

[0098] During the scanning process of the MRI scanning device, for the artifact problem caused by the physiological activities of the research object, corresponding correction methods are used to correct the motion artifacts. The correction methods include the navigation echo method and the motion compensation method.

[0099] Among them, the correction method is specifically:

[0100] The navigation echo method is used for:

[0101] During the scanning process of the MRI scanning device, tracking and correcting the image artifacts caused by respiratory physiological activities, using 2D radiofrequency pulses to track the movement of the diaphragm, and collecting data when the movement of the diaphragm is within the set acceptable range.

[0102] The motion compensation method is used for:

[0103] By describing the movement of each small block in adjacent frames of the nuclear magnetic image scanning data, moving each small block of the previous frame to the corresponding position in the current frame, thereby reducing the spatial redundancy and motion artifacts in the nuclear magnetic image scanning data sequence.

[0104] The technical effects of the above content are as follows: During the scanning process of the MRI scanning device, for the artifact problem caused by the physiological activities of the research object, corresponding correction methods can be used to correct the motion artifacts. The navigation echo method tracks the movement of the diaphragm through 2D radiofrequency pulses and collects data when the movement of the diaphragm is within the set acceptable range. The navigation echo method can effectively reduce the interference of motion artifacts on the image and improve the image quality. The motion compensation method describes the changes in adjacent frames, uses an advanced prediction algorithm to synthesize the lost frames, and moves each small block of the previous frame to the corresponding position in the current frame to reduce motion artifacts. The motion compensation method can reduce or eliminate motion artifacts and significantly improve the image quality. Through the navigation echo method and the motion compensation method, motion artifacts can be reduced or eliminated, which is of great significance for improving the quality and diagnostic value of the reconstructed image.

[0105] Denoising and enhancement are specifically:

[0106] Denoising process: A denoising algorithm is used to denoise the nuclear magnetic resonance (NMR) image scan data, removing noise and retaining image details. The denoising algorithm includes wavelet transform or non-local means filtering;

[0107] Enhancement process: Enhancement processing methods are used to enhance the quality, contrast, and clarity of the NMR image scan data. The enhancement processing methods include spatial domain enhancement methods and frequency domain enhancement methods;

[0108] Spatial domain enhancement methods directly process the pixels of the image. Commonly used spatial domain enhancement methods include:

[0109] Histogram equalization: By changing the histogram distribution of the image, the pixel intensity distribution is made more uniform, thereby enhancing the contrast of the entire image;

[0110] Image sharpening and edge enhancement: By applying a high-pass filter to enhance the edges in the image, making the boundaries clearer;

[0111] Spatial domain filtering: Includes smoothing filtering and sharpening filtering, used to remove image noise or enhance image details;

[0112] Frequency domain enhancement methods are achieved by modifying the spectrum after the Fourier transform of the image. Commonly used frequency domain enhancement methods include:

[0113] Low-pass filtering: Removes the high-frequency part of the image, used to remove noise;

[0114] High-pass filtering: Enhances the high-frequency part of the image, used to sharpen the image;

[0115] Homomorphic filtering: Combines the characteristics of low-pass and high-pass filtering, used to enhance the dynamic range of the image.

[0116] Data standardization and normalization, specifically:

[0117] Unifying data formats, for:

[0118] Unifying the data formats of NMR image scan data obtained under different MRI scanning devices or different scanning parameters;

[0119] Data standardization, for:

[0120] Performing standardization processing on the NMR image scan data to eliminate differences between different NMR image scan data;

[0121] Data normalization, for:

[0122] Mapping the NMR image scan data to a specific range, where the specific range is [0,1] or [-1,1].

[0123] The technical effects of the above content are as follows: After obtaining the nuclear magnetic resonance (NMR) image scan data, denoising, enhancement, data standardization, and normalization are performed. By using wavelet transform or non-local means filtering, the noise in the NMR image scan data can be removed while retaining the image details. By using spatial domain enhancement methods and frequency domain enhancement methods, the quality, contrast, and clarity of the NMR image scan data can be enhanced. Through denoising and enhancement, the quality, contrast, and details of the NMR image scan data can be improved. By unifying the data format, the data formats of NMR image scan data obtained from different MRI scanning devices or under different scanning parameters can be unified. By data standardization, the NMR image scan data can be standardized to eliminate the differences between different NMR image scan data. By data normalization, the NMR image scan data can be mapped to a specific range. By performing data standardization and normalization on the NMR image scan data, the data can have the same dimension and scale, thus better reflecting the true distribution of the data. At the same time, the accuracy and convergence speed of the model can be improved, making the model training more stable and efficient.

[0124] Construct a neural network model and conduct training and testing, which specifically includes the following processes:

[0125] Collect the NMR image scan data of 10 research subjects to obtain the first dataset, and divide the first dataset into a training set and a test set;

[0126] By using different MRI scanning devices or adjusting the scanning parameters, collect the NMR image scan data of another 10 research subjects to obtain the second dataset, and divide the second dataset into a training set and a test set;

[0127] Construct a neural network model, and use the training set and the test set to train and test the neural network model respectively;

[0128] Use the training set and the test set divided from the first dataset to train and test the neural network model. The trained neural network model reconstructs the NMR image scan data to obtain the first reconstructed image;

[0129] Use the training set and the test set divided from the second dataset to train and test the neural network model. The trained neural network model reconstructs the NMR image scan data to obtain the second reconstructed image;

[0130] Compare the first reconstructed image with the second reconstructed image, and verify the robustness of the neural network model by analyzing the quality changes of the reconstructed images under different MRI scanning devices or different scanning parameters through comparison;

[0131] After the verification is completed, deploy the neural network model to reconstruct the NMR image scan data.

[0132] The technical effects of the above content are as follows: A neural network model is constructed and trained and tested. A total of 20 sets of nuclear magnetic resonance (MRI) image scanning data of research subjects are collected. The MRI image scanning data of 10 research subjects are divided into the first data set, and the first data set is divided into a training set and a test set. The MRI image scanning data of the other 10 research subjects are divided into the second data set. The second data set is the MRI image scanning data obtained by using different MRI scanning devices or adjusting scanning parameters. The second data set is the control group of the first data set, and the second data set is also divided into a training set and a test set. First, the neural network model is trained and tested using the training set and test set divided from the first data set. Then, the trained neural network model is used to reconstruct the MRI image scanning data, and a first reconstructed image can be obtained. Then, the neural network model is trained and tested using the training set and test set divided from the second data set. Then, the trained neural network model is used to reconstruct the MRI image scanning data, and a second reconstructed image can be obtained. Finally, the first reconstructed image is compared with the second reconstructed image. By analyzing the quality changes of the reconstructed images under different MRI scanning devices or different scanning parameters, the robustness of the neural network model is verified. Through verification, the MRI scanning device can be reasonably selected and the MRI scanning parameters can be adjusted, so as to effectively improve the image quality. After verification, the neural network model is deployed to reconstruct the MRI image scanning data, so as to realize the reconstruction work of the MRI image scanning data.

[0133] The reconstructed image is subjected to segmentation and registration processing, specifically:

[0134] Image segmentation: The reconstructed image is segmented into non-overlapping regions with features through image segmentation;

[0135] Image registration: The reconstructed image is geometrically transformed through image registration to achieve the purpose of alignment in space.

[0136] In one embodiment, the geometric transformation of the reconstructed image through image registration includes:

[0137] The reconstructed image is initialized and interpolated according to a preset interpolation method to obtain the pixel position of the interpolated pixel points of the reconstructed image. Based on the pixel values of the pixel points around the interpolated pixel point position and in combination with the interpolation direction, the pixel gray value of the interpolated pixel point position is determined to obtain an interpolated image;

[0138] Based on feature point matching, the reconstructed image is deformed into the space where the interpolated image is located to obtain an initial registered image;

[0139] Extracting image feature information from the initial registered image, and selecting a feature point set consisting of main feature points based on the image feature information as a registration basis, and re-registering the initial registered image based on the registration basis to obtain a second registered image;

[0140] Extracting mutual information between the second registered images, and establishing a mutual information matrix based on the image sequence and information attributes of the second registered images;

[0141] Based on the difference between the horizontal correlation relationship between the mutual information matrices and the preset horizontal correlation relationship, a horizontal optimization value is obtained, and based on the difference between the vertical correlation relationship between the mutual information matrices and the preset vertical correlation relationship, a vertical optimization value is obtained;

[0142] Acquire initial optimization parameters of the optimization algorithm, and adjust the initial optimization parameters based on the lateral optimization value and the longitudinal optimization value to obtain target optimization parameters;

[0143] Acquire an initial transformation relationship from the reconstructed image to the second registered image, and optimize the initial transformation relationship based on the target optimization parameter to obtain a target transformation relationship;

[0144] The reconstructed image is geometrically transformed based on the target transformation relationship to obtain a target registration image.

[0145] In this embodiment, the average value of the pixel values ​​of the pixels around the interpolation pixel position corresponding to the interpolation direction is used as the pixel grayscale value of the interpolation pixel position.

[0146] In this embodiment, the reconstructed image is deformed to the space where the interpolated image is located to obtain an initial registration image to achieve a large-scale transformation of the reconstructed image, thereby providing an image basis for subsequent registration.

[0147] In this embodiment, the optimization algorithm is pre-selected or designed according to the optimization target.

[0148] In this embodiment, the second registered image implements feature-based registration between the initial registered images.

[0149] In this embodiment, the initial transformation relationship from the reconstructed image to the second registered image includes the transformation relationship from the reconstructed image to the initial registered image and then to the second registered image.

[0150] In this embodiment, the reconstructed image is geometrically transformed based on the target transformation relationship to obtain a target registered image to achieve accurate registration from two aspects: image features and association between images.

[0151] The beneficial effects of the above design are as follows: By feature point matching, the reconstructed image is deformed into the space where the interpolated image is located to obtain an initial registered image. Based on the registration basis, the initial registered image is registered again to obtain a second registered image. The initial transformation relationship from the reconstructed image to the second registered image is obtained, and the initial transformation relationship is optimized based on the target optimization parameters to obtain a target transformation relationship. Based on the target transformation relationship, geometric transformation is performed on the reconstructed image to obtain a target registered image, achieving accurate registration of the target registered image from two aspects: image features and the association between images, and enabling the reconstructed images to be spatially aligned.

[0152] The technical effects of the above content are as follows: Through image segmentation, the reconstructed image can be segmented into non-overlapping regions with their respective features. The methods of image segmentation mainly include threshold-based segmentation, region-based segmentation, edge-based segmentation, etc. The threshold segmentation method divides the image into different gray-level intervals by selecting one or more thresholds, which is suitable for images with high contrast between the target and the background. The region segmentation method divides pixels with similar attributes into the same region, while edge detection is based on the sudden change properties of the object boundaries in the image for segmentation. Through image registration, geometric transformation is performed on the reconstructed image to make the reconstructed images spatially aligned. The methods of image registration include feature point matching and mutual information-based methods. Feature point matching performs registration by finding common feature points between images, while the mutual information method realizes registration by optimizing the mutual information between images. Performing segmentation and registration processing on the reconstructed image can improve the quality and accuracy of the reconstructed image, and further improve the accuracy and efficiency of the analysis of the reconstructed image.

[0153] A nuclear magnetic resonance (NMR) image scanning data reconstruction system for an NMR image scanning data reconstruction method, comprising:

[0154] A data acquisition unit, for:

[0155] Setting the scanning parameters of the MRI scanning device, using the MRI scanning device to scan and record the data during the imaging process, and during the scanning process, correcting the motion artifacts by using the navigator echo method and the motion compensation method, and finally obtaining the NMR image scanning data of the research object.

[0156] A data processing unit, for:

[0157] Performing denoising, enhancement, data standardization, and normalization on the obtained NMR image scanning data;

[0158] An image reconstruction unit, for:

[0159] Build a neural network model and conduct training and testing. Use the trained neural network model to reconstruct the nuclear magnetic resonance (NMR) image scanning data, and analyze the quality changes of the reconstructed images under different MRI scanning devices or different scanning parameters based on the reconstructed images.

[0160] A reconstructed image processing unit, configured to:

[0161] Perform image segmentation and image registration on the reconstructed image, and evaluate the quality of the reconstructed image by quantitative and qualitative methods for the processed reconstructed image.

[0162] The technical effects of the above content are as follows: The NMR image scanning data reconstruction system consists of a data acquisition unit, a data processing unit, an image reconstruction unit, and a reconstructed image processing unit. And the NMR image scanning data reconstruction system is used to implement the above NMR image scanning data reconstruction method. Through the data acquisition unit, the NMR image scanning data of the research object can be obtained. During the scanning process, motion artifacts can be corrected by the navigation echo method and the motion compensation method, so as to reduce or eliminate motion artifacts, and then the quality of the reconstructed image can be improved. Through the data processing unit, the obtained NMR image scanning data can be denoised, enhanced, data standardized and normalized, which can improve the reliability of the NMR image scanning data, and then the quality of the reconstructed image can be further improved. Then, through the image reconstruction unit, a neural network model is built and trained and tested. The neural network model is used to reconstruct the NMR image scanning data, and the quality changes of the reconstructed images under different MRI scanning devices or different scanning parameters are analyzed based on the reconstructed images, so that the MRI scanning device and the MRI scanning parameters can be reasonably selected and adjusted, thereby effectively improving the image quality. After verification, the neural network model is deployed to reconstruct the NMR image scanning data, so as to realize the reconstruction work of the NMR image scanning data. Finally, through the reconstructed image processing unit, image segmentation and image registration can be performed on the reconstructed image, which can improve the accuracy and efficiency of the analysis of the reconstructed image. And by evaluating the quality of the reconstructed image by quantitative and qualitative methods, the quality of the reconstructed image can be evaluated more comprehensively and accurately.

[0163] Working principle: During the scanning process of the MRI scanning device, the navigation echo method and the motion compensation method are used to reduce or eliminate motion artifacts, thereby improving the quality of the reconstructed image. After obtaining the nuclear magnetic image scanning data of the research object and processing it through denoising, enhancement, data standardization, and normalization, the reliability of the nuclear magnetic image scanning data can be improved, thus further improving the quality of the reconstructed image. Then, a neural network model is constructed and trained and tested. The neural network model is used to reconstruct the nuclear magnetic image scanning data, and the quality changes of the reconstructed images under different MRI scanning devices or different scanning parameters are analyzed based on the reconstructed images to verify the reconstructed image effect, so that the MRI scanning device can be reasonably selected and the MRI scanning parameters can be adjusted, thereby effectively improving the image quality. Finally, image segmentation and image registration processing are performed on the reconstructed image, which can improve the quality and accuracy of the reconstructed image, and further improve the accuracy and efficiency of the analysis of the reconstructed image. By quantitatively and qualitatively evaluating the quality of the reconstructed image, the quality of the reconstructed image can be evaluated more comprehensively and accurately.

[0164] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprising", "including" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device.

[0165] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principle and spirit of the present invention.

Claims

1. A method for reconstructing nuclear magnetic resonance image scanning data, characterized in that: The following steps are involved: Set the scanning parameters of the MRI scanning equipment, including repetition time, echo time and matrix size; Placing the research object on the MRI scanning device, starting the MRI scanning device to scan and record the data during the imaging process, and obtaining the nuclear magnetic resonance image scanning data of the research object; Process the acquired nuclear magnetic resonance image scanning data, including denoising and enhancement and data standardization and normalization; Construct a neural network model and perform training and testing, and use the trained neural network model to reconstruct the MRI scan data; Further processing of the reconstructed image, including segmentation and registration, and evaluation of the quality of the reconstructed image by quantitative and qualitative methods; The quality of the reconstructed image is evaluated by quantitative and qualitative methods, including: The reconstructed image is scaled according to a plurality of preset sizes to obtain a scaled image corresponding to each preset size, feature extraction is performed on all scaled images to obtain image features, and the image features are unified to obtain standard features; Based on the standard features of the reconstructed image and the scaled image, the quality value of the reconstructed image is calculated according to the following formula ; in, Indicates the number of scaled images, represents the number of features, Represents the reconstructed image Standard eigenvalues, Indicates In the zoomed image Standard eigenvalues, Indicates The scaling weight value of the scaled image; Determining whether the quality value of the reconstructed image is greater than a preset quality value; If so, determining that the clarity of the reconstructed image meets the quality requirements; Otherwise, determining that the clarity of the reconstructed image does not meet the quality requirement; After determining that the clarity of the reconstructed image meets the quality requirements, an initial reconstructed image before registration is obtained, and a cross-correlation coefficient between the reconstructed image and the initial reconstructed image is calculated according to the following formula; in, represents the cross-correlation coefficient between the reconstructed image and the initial reconstructed image, represents the covariance between the reconstructed image and the initial reconstructed image, represents the variance of the reconstructed image, represents the variance of the initial reconstructed image, represents the expected variance of the reconstructed image based on the initial reconstructed image, represents the expected variance of the initial reconstructed image based on the reconstructed image; Determining whether the cross-correlation coefficient between the reconstructed image and the initial reconstructed image meets a preset coefficient requirement; If so, determining that the artifacts of the reconstructed image meet the quality requirements; Otherwise, it is determined that the artifacts of the reconstructed image do not meet the quality requirement.

2. The method for reconstructing nuclear magnetic resonance image scanning data according to claim 1, characterized in that: During the scanning process of the MRI scanning device, corresponding correction methods are used to correct motion artifacts. The correction methods include navigation echo methods and motion compensation methods.

3. The method for reconstructing nuclear magnetic resonance image scanning data according to claim 2, characterized in that: The correction method is specifically as follows: Navigation echo method, used for: During the scanning process of the MRI scanning device, 2D radio frequency pulses are used to track the movement of the diaphragm, and data is collected when the diaphragm movement is within the set acceptable range; Motion compensation methods for: By describing the movement of each small block in adjacent frames of nuclear magnetic resonance image scanning data, each small block of the previous frame is moved to the corresponding position of the current frame, thereby reducing spatial redundancy and motion artifacts in the nuclear magnetic resonance image scanning data sequence.

4. The method for reconstructing nuclear magnetic resonance image scanning data according to claim 1, characterized in that: The denoising and enhancement are specifically as follows: Denoising: Denoising algorithms are used to denoise the MRI scan data to remove noise and retain image details. Denoising algorithms include wavelet transform or non-local mean filtering; Enhancement processing: Enhancement processing methods are used to enhance the quality, contrast and clarity of nuclear magnetic resonance image scanning data. Enhancement processing methods include spatial domain enhancement methods and frequency domain enhancement methods.

5. The method for reconstructing nuclear magnetic resonance image scanning data according to claim 1, characterized in that: The data standardization and normalization are specifically as follows: The data format is unified for: Unify the scanning data formats of nuclear magnetic resonance images obtained by different MRI scanning devices or different scanning parameters; Data normalization for: Standardize the MRI scan data to eliminate the differences between different MRI scan data; Data normalization, used to: Map the MRI scan data to a specific range, which is [0,1] or [-1,1].

6. The method for reconstructing nuclear magnetic resonance image scanning data according to claim 1, characterized in that: Build a neural network model and perform training and testing, which includes the following processes: Collecting magnetic resonance imaging scan data of 10 research subjects to obtain a first data set, and dividing the first data set into a training set and a test set; By using different MRI scanning devices or adjusting scanning parameters, magnetic resonance imaging scan data of another 10 research subjects are collected to obtain a second data set, and the second data set is divided into a training set and a test set; Construct a neural network model, and use the training set and test set to train and test the neural network model respectively; The neural network model is trained and tested using the training set and the test set divided by the first data set, and the trained neural network model is used to reconstruct the nuclear magnetic resonance image scanning data to obtain a first reconstructed image; The neural network model is trained and tested using the training set and the test set divided by the second data set, and the trained neural network model is used to reconstruct the nuclear magnetic resonance image scanning data to obtain a second reconstructed image; The first reconstructed image is compared with the second reconstructed image, and the robustness of the neural network model is verified by comparing and analyzing the quality changes of the reconstructed images under different MRI scanning devices or different scanning parameters; After the verification is completed, the MRI scanning equipment and scanning parameters are determined, and the neural network model trained using the MRI scanning equipment and scanning parameters is used to reconstruct the nuclear magnetic resonance image scanning data.

7. The method for reconstructing nuclear magnetic resonance image scanning data according to claim 1, characterized in that: The reconstructed image is segmented and registered, specifically: Image segmentation: Image segmentation is used to divide the reconstructed image into non-overlapping and characteristic regions; Image registration: Image registration is used to geometrically transform the reconstructed image.

8. The method for reconstructing nuclear magnetic resonance image scanning data according to claim 7, characterized in that: The geometric transformation of the reconstructed image through image registration includes: Initializing the interpolation processing of the reconstructed image according to a preset interpolation method to obtain an interpolation pixel position of the reconstructed image, and determining the pixel grayscale value of the interpolation pixel position based on the pixel values ​​of the pixels around the interpolation pixel position and the interpolation direction to obtain an interpolated image; Based on feature point matching, the reconstructed image is deformed to the space where the interpolated image is located to obtain an initial registration image; Extracting image feature information from the initial registered image, and selecting a feature point set consisting of main feature points based on the image feature information as a registration basis, and re-registering the initial registered image based on the registration basis to obtain a second registered image; Extracting mutual information between the second registered images, and establishing a mutual information matrix based on the image sequence and information attributes of the second registered images; Based on the difference between the horizontal correlation relationship between the mutual information matrices and the preset horizontal correlation relationship, a horizontal optimization value is obtained, and based on the difference between the vertical correlation relationship between the mutual information matrices and the preset vertical correlation relationship, a vertical optimization value is obtained; Acquire initial optimization parameters of the optimization algorithm, and adjust the initial optimization parameters based on the lateral optimization value and the longitudinal optimization value to obtain target optimization parameters; Acquire an initial transformation relationship from the reconstructed image to the second registered image, and optimize the initial transformation relationship based on the target optimization parameter to obtain a target transformation relationship; The reconstructed image is geometrically transformed based on the target transformation relationship to obtain a target registration image.

9. A nuclear magnetic image scanning data reconstruction system, used to implement the nuclear magnetic image scanning data reconstruction method according to any one of claims 1 to 7, characterized in that: include: Data acquisition unit for: Setting the scanning parameters of the MRI scanning device, using the MRI scanning device to scan and record the data during the imaging process, and using the navigation echo method and motion compensation method to correct the motion artifacts during the scanning process, and finally obtaining the nuclear magnetic resonance image scanning data of the research object; Data processing unit for: De-noise and enhance the acquired nuclear magnetic resonance image scanning data and perform data standardization and normalization; An image reconstruction unit, for: Construct a neural network model and conduct training and testing. Use the trained neural network model to reconstruct the nuclear magnetic resonance image scanning data, and analyze the quality changes of the reconstructed images under different MRI scanning devices or different scanning parameters based on the reconstructed images. Reconstruction image processing unit for: The reconstructed image is subjected to image segmentation and image registration processing, and the quality of the reconstructed image is evaluated by quantitative and qualitative methods.

Images