Method and System for Analyzing the Proportion of Renal Sclerosis Based on MRI Images

Abstract

The present invention discloses a method and system for analyzing the sclerosis ratio of kidneys based on MRI images, belonging to the technical field of medical image processing, including: selecting an MRI imaging sequence according to the kidney lesion characteristics of a patient; collecting images of the patient's kidneys based on the MRI imaging sequence to obtain different MRI imaging sequence images; fusing the MRI imaging sequence images to obtain a multi-modal image; preprocessing the multi-modal image, including per-pixel signal intensity correction, histogram equalization, adaptive filtering, and non-local means denoising; performing image segmentation on the preprocessed multi-modal image to obtain a binary image including a sclerosis region and a normal region; obtaining the kidney sclerosis ratio based on the binary image. The method of the present invention can improve the recognition accuracy and quantification precision of the kidney sclerosis region, more accurately evaluate the degree of kidney sclerosis, provide strong support for clinical research, and improve the medical level.

Description

Technical Field

[0001] The present invention relates to the technical field of medical image processing, and particularly to a method and system for analyzing the sclerosis ratio of kidneys based on MRI images. Background Art

[0002] In existing medical imaging technologies, MRI (Magnetic Resonance Imaging) is an important means for evaluating the structure and function of kidneys. However, traditional MRI image analysis technologies have certain limitations in identifying kidney sclerosis regions. The current technologies mainly rely on the analysis of images of a single MRI sequence, which often cannot accurately distinguish normal tissues from sclerotic tissues. In addition, the limitations of image processing algorithms also result in low accuracy in segmenting kidney sclerosis regions, and the presence of artifacts and noise further increases the difficulty of analysis. Therefore, there is an urgent need for a method for analyzing the sclerosis ratio of kidneys to improve the image contrast, reduce noise interference, and accurately segment and quantify kidney sclerosis regions. Summary of the Invention

[0003] The purpose of this application is to overcome the defects of the prior art and provide a method and system for analyzing the sclerosis ratio of kidneys based on MRI images.

[0004] In the first aspect, this application provides a method for analyzing the sclerosis ratio of kidneys based on MRI images, including the following steps:

[0005] Select an MRI imaging sequence according to the characteristics of the patient's kidney lesions;

[0006] Collect images of the patient's kidneys based on the selected MRI imaging sequence to obtain different MRI imaging sequence images;

[0007] Fuse the obtained MRI imaging sequence images to obtain a fused multi-modal image, including: performing normalization processing on each of the MRI imaging sequence images to obtain normalized MRI imaging sequence images; performing spatial alignment on the normalized different MRI imaging sequence images; performing weighted fusion on the aligned different MRI imaging sequence images to obtain the multi-modal image;

[0008] Perform preprocessing on the multi-modal image, including: performing per-pixel signal intensity correction on the multi-modal image so that each pixel of the multi-modal image has the same signal intensity level; performing histogram equalization on the corrected multi-modal image to enhance the image contrast; performing adaptive filtering on the histogram-equalized multi-modal image to remove noise and retain edge details; performing non-local means denoising on the adaptively filtered multi-modal image to obtain the preprocessed multi-modal image;

[0009] Perform image segmentation on the preprocessed multimodal image to obtain a binary image including a hardened region and a normal region;

[0010] Obtain the kidney sclerosis ratio based on the binary image.

[0011] Optionally, the selected MRI imaging sequences include: T1-weighted imaging sequence, T2-weighted imaging sequence, diffusion-weighted imaging sequence, and specific kidney imaging sequence; the obtained MRI imaging sequence images include: T1-weighted imaging sequence image, T2-weighted imaging sequence image, diffusion-weighted imaging sequence image, and specific kidney imaging sequence image.

[0012] Optionally, performing per-pixel signal intensity correction on the multimodal image so that each pixel of the multimodal image has the same signal intensity level includes:

[0013] Scanning the same patient or a standard phantom multiple times to obtain the signal intensity relationship between different multimodal images and obtain a correction coefficient;

[0014] Use the correction coefficient to perform per-pixel adjustment on the multimodal image so that each pixel of the multimodal image has the same signal intensity level.

[0015] Optionally, the performing image segmentation on the preprocessed multimodal image to obtain a binary image including a hardened region and a normal region includes:

[0016] Construct a dataset, where the dataset includes a training set, a validation set, and a test set;

[0017] Construct a deep learning model and use the training set to train the deep learning model;

[0018] Optimize the deep learning model using a cross-entropy loss function;

[0019] Use the validation set to evaluate the accuracy of the trained deep learning model and use the test set to test the generalization ability of the model to obtain the final deep learning model;

[0020] Use the deep learning model to perform image segmentation on the preprocessed multimodal image to obtain a binary image including a hardened region and a normal region.

[0021] Optionally, the deep learning model includes a U-Net model.

[0022] Optionally, the hardened region includes a white pixel region; the normal region includes a black pixel region.

[0023] Optionally, obtaining the kidney sclerosis ratio based on the binary image includes:

[0024] Obtaining the number of pixels in the sclerotic region based on the binary image;

[0025] Obtaining the number of pixels of the entire kidney based on the binary image;

[0026] Taking the ratio of the number of pixels in the sclerotic region to the number of pixels of the entire kidney as the kidney sclerosis ratio.

[0027] In a second aspect, the present application also provides a system for analyzing the kidney sclerosis ratio based on MRI images, which is used to execute the method for analyzing the kidney sclerosis ratio based on MRI images as described in any one of the first aspects, including:

[0028] An MRI imaging sequence selection module, configured to select an MRI imaging sequence according to the characteristics of the patient's kidney lesion;

[0029] An image acquisition module, configured to acquire images of the patient's kidney based on the selected MRI imaging sequence to obtain different MRI imaging sequence images;

[0030] A multi-modal fusion module, configured to fuse the obtained MRI imaging sequence images to obtain a fused multi-modal image;

[0031] A preprocessing module, configured to preprocess the multi-modal image;

[0032] An image segmentation module, configured to perform image segmentation on the preprocessed multi-modal image to obtain a binary image including a sclerotic region and a normal region;

[0033] A sclerosis ratio calculation module, configured to obtain the kidney sclerosis ratio based on the binary image.

[0034] The present application provides a method and system for analyzing the kidney sclerosis ratio based on MRI images. By selecting a suitable MRI imaging sequence according to the characteristics of the patient's kidney lesion, more comprehensive and accurate kidney image information can be obtained; by fusing images of different imaging sequences to obtain a multi-modal image, the advantages of each sequence can be integrated; the per-pixel signal intensity correction, histogram equalization, adaptive filtering, and non-local means denoising in the preprocessing process can improve the image quality and provide a good basis for subsequent analysis; using a deep learning model to segment the preprocessed image can accurately distinguish the sclerotic region and the normal region. The method and system of the present application can more accurately evaluate the degree of kidney sclerosis, provide strong support for clinical research, and improve the medical level.

[0035] To make the above features and advantages of the invention more obvious and understandable, specific embodiments are hereinafter given and described in detail in conjunction with the accompanying drawings as follows. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the following will briefly introduce the drawings required for use in the description of the embodiments or the related art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0037] Figure 1 It is a flowchart of a method for analyzing the renal sclerosis ratio based on MRI images provided in an embodiment of the present application.

[0038] Figure 2 It is a flowchart of step S50 in the method for analyzing the renal sclerosis ratio based on MRI images provided in an embodiment of the present application.

[0039] Figure 3 It is a schematic structural diagram of a system for analyzing the renal sclerosis ratio based on MRI images provided in another embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] To make the objectives and technical solutions of the embodiments of the present application clearer, the following will clearly and completely describe the technical solutions of the embodiments of the present application in conjunction with the drawings of the embodiments of the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the described embodiments of the present application without creative efforts fall within the scope of protection of the present application.

[0041] In one embodiment, please refer to Figure 1 , the present application provides a method for analyzing the renal sclerosis ratio based on MRI images. The method for analyzing the renal sclerosis ratio based on MRI images may include the following steps: S10 to S60.

[0042] S10: Select an MRI imaging sequence according to the characteristics of the patient's kidney lesions.

[0043] S20: Collect images of the patient's kidney based on the selected MRI imaging sequence to obtain different MRI imaging sequence images.

[0044] S30: Fuse the obtained MRI imaging sequence images to obtain a fused multimodal image, including: performing normalization processing on each of the MRI imaging sequence images to obtain normalized MRI imaging sequence images; performing spatial alignment on the normalized different MRI imaging sequence images; and performing weighted fusion on the aligned different MRI imaging sequence images to obtain the multimodal image.

[0045] S40: Preprocess the multimodal image, including: performing pixel-by-pixel signal intensity correction on the multimodal image to make the signal intensity levels of the pixels of the multimodal image the same; performing histogram equalization on the corrected multimodal image to enhance the image contrast; performing adaptive filtering on the histogram-equalized multimodal image to remove noise and retain edge details; performing non-local means denoising on the adaptively filtered multimodal image to obtain the preprocessed multimodal image.

[0046] S50: Segment the preprocessed multimodal image to obtain a binary image including a hardened region and a normal region.

[0047] S60: Obtain the kidney sclerosis ratio based on the binary image.

[0048] In the method for analyzing the kidney sclerosis ratio based on MRI images of this application, by selecting MRI imaging sequences according to the characteristics of the patient's kidney lesions, images with strong pertinence can be obtained; fusing images of different imaging sequences into a multimodal image can integrate the advantageous information of each sequence; preprocessing the multimodal image can improve the image quality; segmenting the preprocessed multimodal image can accurately divide the hardened and normal regions; finally, obtaining the sclerosis ratio based on the binary image can accurately evaluate the degree of kidney sclerosis and provide a reliable basis for clinical research.

[0049] In step S10, please refer to Figure 1 step S10 in

[0050] As an example, select the MRI imaging sequence according to the patient's kidney structure, function, and lesion characteristics. Factors such as the patient's age, physical condition, whether there are other underlying diseases, and the type, location, and suspected sclerosis degree of the kidney lesion can be combined to select the most suitable MRI imaging sequence.

[0051] As an example, the selected MRI imaging sequences may include: T1-weighted imaging sequence, T2-weighted imaging sequence, diffusion-weighted imaging sequence. These sequences can provide kidney images with different contrasts and help clearly depict the anatomical structure of the kidney.

[0052] Specifically, the T1-weighted imaging sequence uses the T1 relaxation time difference in tissues to generate contrast. The T1 relaxation time refers to the time required for the hydrogen atomic nuclei in tissues to return from a high-energy state to a low-energy state. In the image generated by the T1-weighted imaging sequence, fat usually appears as a high signal, that is, bright white, while water and liquids appear as low signals, that is, dark. The T1-weighted imaging sequence can provide good soft tissue contrast and help distinguish normal tissue and hardened tissue.

[0053] Furthermore, the T2-weighted imaging sequence uses the T2 relaxation time differences in tissues to generate contrast. The T2 relaxation time refers to the time required for the hydrogen nuclei in tissues to lose phase coherence within the same energy level. In the images generated by the T2-weighted imaging sequence, water and fluids typically appear as high signals, i.e., bright white, while fat and solid tissues appear as low signals, i.e., dark. The T2-weighted imaging sequence is more sensitive to the contrast between fluids and soft tissues, which helps to display the diseased areas of the kidneys.

[0054] Furthermore, the diffusion-weighted imaging sequence is an imaging technique for detecting the Brownian motion of water molecules in tissues. The diffusion-weighted imaging sequence generates images by measuring the degree of diffusion of water molecules after the application of a gradient magnetic field. In the images generated by the diffusion-weighted imaging sequence, areas where the diffusion of water molecules is restricted, such as tumors or infarcted tissues, appear as high signals, i.e., bright white; while normal tissues appear as low signals, i.e., dark. The diffusion-weighted imaging sequence can be used for the early detection of certain tumors. The diffusion-weighted imaging sequence can evaluate the microstructure of tissues by detecting the Brownian motion of water molecules, which is particularly useful for identifying sclerotic areas.

[0055] As an example, the MRI imaging sequence can also include: a specific kidney imaging sequence. By optimizing the specific physiological characteristics of the kidneys through the specific kidney imaging sequence, more accurate information about the sclerotic areas can be obtained, further enhancing the specificity and sensitivity of the images. The specific physiological characteristics of the kidneys can include: blood perfusion characteristics, metabolic function characteristics, tissue composition characteristics, functional reserve characteristics, etc.

[0056] As an example, the specific kidney imaging sequence can include: kidney filtered imaging or kidney diffusion imaging. By selecting different MRI imaging sequences, the sclerotic areas of the kidneys can be better highlighted, providing high-quality raw data for subsequent image analysis and quantification.

[0057] In step S20, please refer to Figure 1 step S20 in, and based on the selected MRI imaging sequence, image acquisition of the patient's kidneys is performed to obtain different MRI imaging sequence images.

[0058] As an example, a doctor can start the MRI device and select one or more of the MRI imaging sequences, including: T1-weighted imaging sequence, T2-weighted imaging sequence, diffusion-weighted imaging sequence, specific kidney imaging sequence, and adjust the relevant parameters of the MRI device according to different MRI imaging sequences for image acquisition to obtain different MRI imaging sequence images.

[0059] As an example, the MRI imaging sequence images may include: T1-weighted imaging sequence images, T2-weighted imaging sequence images, diffusion-weighted imaging sequence images, and specific kidney imaging sequence images.

[0060] As an example, for the T1-weighted imaging sequence, a shorter repetition time TR and echo time TE can be selected. The repetition time TR can be set between 300 - 800 milliseconds, and the echo time TE can be set between 10 - 30 milliseconds to highlight the T1 relaxation characteristics of tissues and display the anatomical structure of the kidney; for the T2-weighted imaging sequence, a longer repetition time TR and echo time TE can be adopted. The repetition time TR can be set between 2000 - 5000 milliseconds, and the echo time TE can be set between 80 - 120 milliseconds to display the T2 relaxation characteristics of tissues and the change in water content of the kidney, which helps to detect kidney lesions; for the diffusion-weighted imaging sequence, different diffusion sensitivity coefficient b values can be set. The b value can be set between 0 - 1000 s / mm² to detect the diffusion motion of water molecules, which helps to observe the microstructural changes of kidney tissues.

[0061] As an example, the collected MRI imaging sequence images can be converted into a digital matrix, and each element in the digital matrix represents the signal intensity of the corresponding pixel in the MRI imaging sequence image. When the MRI device acquires MRI imaging sequence images, the MRI imaging sequence images usually exist in the form of pixels. Each collected MRI imaging sequence image can be converted into a digital matrix, where each pixel in the MRI imaging sequence image corresponds to an element in the digital matrix.

[0062] In one example, a black-and-white image collected by a doctor through an MRI device has a size of 2×2 pixels. This black-and-white image is converted into a 2×2 digital matrix, which is represented as follows:

[0063]

[0064] Among them, P1 represents the signal intensity of the pixel in the upper left corner of the image, P2 represents the signal intensity of the pixel in the upper right corner, P3 represents the signal intensity of the pixel in the lower left corner, and P4 represents the signal intensity of the pixel in the lower right corner. The digital matrix of the black-and-white image can be .

[0065] As an example, for the digital matrix of a black-and-white image, each element represents the signal intensity of the corresponding pixel in the black-and-white image, and the signal intensity is usually related to the brightness and darkness. The digital matrix of a color image is more complex because each pixel of a color image includes three colors: red, green, and blue. Each element of the digital matrix of a color image represents the signal intensities of the red, green, and blue colors of the corresponding pixel.

[0066] In step S30, refer to Figure 1 step S30 in it, and fuse the obtained MRI imaging sequence images to obtain a fused multimodal image, including: performing normalization processing on each of the MRI imaging sequence images to obtain normalized MRI imaging sequence images; performing spatial alignment on the different normalized MRI imaging sequence images; and performing weighted fusion on the aligned different MRI imaging sequence images to obtain the multimodal image.

[0067] Specifically, first perform normalization processing on each individual MRI imaging sequence image to obtain normalized MRI imaging sequence images to ensure the consistency of the MRI imaging sequence images in terms of spatial resolution and contrast.

[0068] As an example, the brightness and contrast of the images can be adjusted to make the images of different sequences visually consistent.

[0069] Furthermore, use an image registration algorithm to perform spatial alignment on the different normalized MRI imaging sequence images for subsequent fusion processing.

[0070] As an example, the image registration algorithm may include: a registration method based on mutual information. Specifically, calculate the transformation parameters between different MRI imaging sequence images to achieve the best alignment.

[0071] Furthermore, perform weighted fusion on the aligned different MRI imaging sequence images to obtain a multimodal image containing more information.

[0072] As an example, linear or non-linear fusion techniques can be used for weighted fusion, combining the information of different MRI imaging sequence images to obtain a multimodal image. The multimodal image retains the complementary information of each MRI imaging sequence image, making the sclerotic area more prominent and easy to identify.

[0073] In another example, a set of T1-weighted imaging sequence images and T2-weighted imaging sequence images collected are fused to obtain a multimodal image. The multimodal image can clearly identify the sclerotic area while showing the kidney structure, providing more accurate information for doctors. Through fusion, the contrast and clarity of the MRI imaging sequence images can be improved, thereby enhancing the recognition accuracy of the kidney sclerotic area. This not only improves the image quality but also lays a solid foundation for subsequent image analysis and sclerotic ratio calculation.

[0074] In step S40, refer to Figure 1In step S40, preprocess the multimodal image, including: performing pixel-by-pixel signal intensity correction on the multimodal image to make the pixels of the multimodal image have the same signal intensity level; performing histogram equalization on the corrected multimodal image to enhance the contrast of the image; performing adaptive filtering on the histogram-equalized multimodal image to remove noise and retain edge details; performing non-local means denoising on the adaptively filtered multimodal image to obtain the preprocessed multimodal image.

[0075] Specifically, first perform pixel-by-pixel signal intensity correction on the multimodal image to eliminate the image intensity deviation caused by inconsistent equipment or scanning parameters, so that the pixels of the multimodal image have the same signal intensity level. Specifically, by scanning the same patient or the same standard phantom multiple times, obtain the signal intensity relationship between different MRI imaging sequence images, and obtain the correction coefficient. Use the correction coefficient to perform pixel-by-pixel adjustment on the multimodal image to make the pixels of the multimodal image have the same signal intensity level.

[0076] Further, perform advanced preprocessing techniques on the corrected multimodal image to further improve the image quality.

[0077] As an example, the advanced preprocessing techniques may include: histogram equalization, adaptive filtering. Specifically, perform histogram equalization on the corrected multimodal image to enhance the contrast of the image; perform adaptive filtering on the histogram-equalized multimodal image to remove noise and retain edge details.

[0078] As an example, the advanced preprocessing techniques may also include: non-local means denoising. Specifically, perform non-local means denoising on the adaptively filtered multimodal image to obtain the preprocessed multimodal image.

[0079] As an example, the histogram equalization changes the histogram distribution of the corrected multimodal image, makes the brightness distribution of the corrected image more uniform, thereby enhancing the contrast of the image and making the difference between the hardened area and the normal tissue more obvious.

[0080] As an example, the adaptive filtering adaptively adjusts the parameters of the filter according to the local characteristics of the image to optimize the denoising effect, and can remove noise while retaining edges and details.

[0081] As an example, the non-local means denoising considers the repeated texture information in the image, performs weighted averaging on similar pixels in the image, effectively reduces noise and protects the structure of the image, can further remove the noise in the image, and lays a solid foundation for subsequent image segmentation and analysis.

[0082] In yet another example, for a set of T1-weighted imaging sequence images after multi-modal fusion, signal intensity correction is used to perform multiple scans to obtain correction coefficients. Based on the correction coefficients, the T1-weighted imaging sequence images are adjusted pixel by pixel to ensure that different sequence images are compared at the same intensity level to achieve a consistent signal intensity level. The histogram of the corrected T1-weighted imaging sequence images is further equalized to make the difference between the hardened area and the normal area more obvious. Adaptive filtering is performed on the histogram-equalized T1-weighted imaging sequence images to adaptively adjust the parameters of the filter according to the local features of the images, so as to remove noise and retain edges and details. Non-local means denoising is performed on the adaptively filtered T1-weighted imaging sequence images, and similar pixels in the T1-weighted imaging sequence images are weighted and averaged to further remove noise in the images, so as to obtain the preprocessed T1-weighted imaging sequence images. After the above preprocessing operations, the contrast and clarity of the T1-weighted imaging sequence images can be significantly improved, and high-quality images with high signal-to-noise ratio can be obtained, enabling subsequent image segmentation to more accurately identify and segment the hardened area of the kidney, thereby calculating the hardening ratio.

[0083] In step S50, refer to Figure 1 step S50 in, and perform image segmentation on the preprocessed multi-modal images to obtain a binary image including the hardened area and the normal area.

[0084] As an example, refer to Figure 2 , step S50 may include the following steps: S501~S505.

[0085] S501: Construct a data set, and the data set includes a training set, a validation set, and a test set.

[0086] S502: Construct a deep learning model, and use the training set to train the deep learning model.

[0087] S503: Optimize the deep learning model using a cross-entropy loss function.

[0088] S504: Use the validation set to evaluate the accuracy of the trained deep learning model, and use the test set to test the generalization ability of the model to obtain the final deep learning model.

[0089] S505: Use the deep learning model to perform image segmentation on the preprocessed multi-modal images to obtain a binary image including the hardened area and the normal area.

[0090] As an example, in step 501, the dataset is constructed based on the historical kidney data obtained in advance. The dataset may include, but is not limited to, 1000 historical kidney images, and each historical kidney image has a corresponding labeled sclerosis area marked in advance. The dataset may include a training set, a validation set, and a test set.

[0091] Further, in step 502, a deep learning model is constructed. The deep learning model may include an architecture based on a convolutional neural network.

[0092] As an example, a deep learning model based on U-Net is constructed. The U-Net deep learning model has a symmetric contraction path and expansion path, which can extract and fuse features at different levels. The skip connections in the model can combine high-resolution feature maps with upsampled feature maps, so as to retain more detailed information during segmentation.

[0093] As an example, the U-Net deep learning model may include: feature extraction, downsampling, upsampling, and skip connections. Specifically, useful features in the image are obtained through feature extraction of the preprocessed multimodal image; features are further extracted through downsampling (i.e., the contraction path) to obtain low-resolution, high-pass reciprocal feature maps. Different-scale low-resolution feature maps are extracted after multiple downsamplings, and the number of channels doubles after each downsampling; through upsampling (i.e., the expansion path), the low-resolution feature maps are restored to high-resolution feature maps. Different-scale low-resolution feature maps are restored after multiple upsamplings, and the number of channels is halved and the feature map size doubles after each upsampling; the skip connections are used to splice the low-resolution feature maps at the corresponding levels in the downsampling with the high-resolution feature maps after upsampling. After splicing, two 3×3 convolutional layers are used for feature fusion and processing, so as to gradually restore the image resolution and utilize the detailed information retained in the skip connections to achieve accurate segmentation of the kidney image.

[0094] Further, the training set is used to train the deep learning model. During the training process, the cross-entropy loss function is used to optimize the deep learning model, and the weights of the deep learning model are continuously adjusted through forward propagation and backward propagation to minimize the loss function.

[0095] As an example, the cross-entropy loss function can measure the difference between the predicted segmentation map and the true label, and can be expressed by the following formula:

[0096]

[0097] where L represents the cross-entropy loss function; N is the total number of pixels in the image; K is the number of categories, that is, two categories: the sclerosis area and the non-sclerosis area; y ik is the true label of the i-th pixel belonging to category k; The probability that the i-th pixel of the predicted segmentation map belongs to class k.

[0098] Further, in step 504, the accuracy of the trained deep learning model is evaluated using a validation set, and the generalization ability of the deep learning model is tested using a test set. The parameters of the deep learning model are fine-tuned according to the test results, and the fine-tuned deep learning model is retrained. Finally, a deep learning model with optimal performance is obtained. The final deep learning model can accurately segment the sclerotic region from the kidney image, achieve image segmentation, and provide a reliable basis for the final calculation of the sclerosis ratio.

[0099] Further, in step 505, the preprocessed multimodal image is segmented using the final deep learning model to obtain a binary image including the sclerotic region and the normal region. The binary image can accurately identify the sclerotic boundary; the sclerotic region may include a white pixel region, and the normal region may include a black pixel region.

[0100] As an example, the obtained binary image can be further used for feature extraction and analysis, so as to provide quantitative information for doctors. The kidney sclerosis ratio can also be obtained by calculating the percentage of the sclerotic region in the entire kidney region, which helps to monitor the disease progression.

[0101] In step S60, refer to Figure 1 step S60 in, and obtain the kidney sclerosis ratio based on the binary image.

[0102] As an example, count all the pixels marked as sclerotic in the binary image to obtain the number of pixels in all sclerotic regions, that is, the number of white pixels in the binary image. Then, based on the binary image, obtain the number of pixels in the entire kidney, where the entire kidney includes the sclerotic region and the normal region, and obtain the total number of all white pixels and black pixels in the binary image. Finally, based on the number of pixels in the sclerotic region and the number of pixels in the entire kidney, obtain the kidney sclerosis ratio. The formula for the kidney sclerosis ratio is as follows:

[0103]

[0104] Among them, the kidney sclerosis ratio is a dimensionless value, which directly reflects the proportion of the sclerotic region in the entire kidney.

[0105] As an example, image processing software or programming languages such as Python combined with the OpenCV library can be used for pixel counting.

[0106] In one example, a binary image is obtained through a deep learning model. The binary image shows that the sclerotic region has 5,000 pixels, while the entire kidney region has 20,000 pixels. Then, the kidney sclerosis ratio can be obtained as , which is 25%.

[0107] It should be understood that although the steps in the flowchart of the accompanying drawings are shown in sequence according to the indication of the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise clearly stated in this article, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, at least a part of the steps in the accompanying drawings may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same moment, but can be executed at different moments. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or sub-steps or stages of other steps.

[0108] In the method for analyzing the kidney sclerosis ratio based on MRI images of the present application, different MRI imaging sequences are selected by combining the characteristics of the patient's kidney lesions, and MRI imaging sequence images are obtained. By using the different imaging principles of each MRI imaging sequence, the kidney structure and lesions are comprehensively presented; the multi-MRI imaging sequence images are fused to obtain a multi-modal image, which can integrate the advantageous information and highlight the sclerotic region; the intensity deviation is eliminated by pixel-by-pixel correction, and then the image quality is further improved through histogram equalization, adaptive filtering, and non-local means denoising; a deep learning model is constructed based on U-Net for image segmentation, and the sclerotic and normal regions are accurately divided by using the contraction, dilation paths, and skip connections; finally, the kidney sclerosis ratio is calculated based on the binary image. The method of the present application can improve the recognition accuracy and quantification precision of the sclerotic region, can accurately evaluate the degree of kidney sclerosis, and provides a more reliable basis and quantitative information for clinical research.

[0109] In another embodiment, please refer to Figure 3, this application also provides a renal sclerosis ratio analysis system based on MRI images, which may include: an MRI imaging sequence selection module 10, an image acquisition module 20, a multimodal fusion module 30, a preprocessing module 40, an image segmentation module 50, and a sclerosis ratio calculation module 60. Among them, the MRI imaging sequence selection module 10 is used to select an MRI imaging sequence according to the renal lesion characteristics of the patient; the image acquisition module 20 is used to acquire images of the patient's kidneys based on the selected MRI imaging sequence to obtain different MRI imaging sequence images; the multimodal fusion module 30 is used to fuse the obtained MRI imaging sequence images to obtain a fused multimodal image; the preprocessing module 40 is used to preprocess the multimodal image; the image segmentation module 50 is used to segment the preprocessed multimodal image to obtain a binary image including a sclerosis region and a normal region; the sclerosis ratio calculation module 60 is used to obtain the renal sclerosis ratio based on the binary image.

[0110] In the above renal sclerosis ratio analysis system based on MRI images, the MRI imaging sequence selection module 10 accurately selects the imaging sequence according to the renal lesion characteristics of the patient, the image acquisition module 20 obtains multi-sequence images, the multimodal fusion module 30 integrates the advantageous information of the images, the preprocessing module 40 improves the further image quality, the image segmentation module 50 accurately divides the sclerosis and normal regions, and the sclerosis ratio calculation module 60 obtains the sclerosis ratio, providing a more reliable basis for clinical research. The system of this application can improve the recognition accuracy and quantification precision of the renal sclerosis region, can provide a quantitative basis for the degree of renal sclerosis, provide intuitive and reliable data for doctors, assist doctors in timely understanding the renal lesion situation of patients, and formulating scientific and reasonable coping strategies.

[0111] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, 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, it should be considered as the scope described in this specification.

[0112] Although this application has been disclosed as above with embodiments, it is not intended to limit this application. Any person with ordinary knowledge in the technical field to which this application pertains can make some modifications and refinements without departing from the spirit and scope of this application. Therefore, the protection scope of this application shall be subject to that defined by the appended patent application scope.

Claims

1. A method for analyzing the sclerosis ratio of kidneys based on MRI images, characterized in that Including the following steps: Select an MRI imaging sequence according to the characteristics of the patient's kidney lesions; Collect images of the patient's kidneys based on the selected MRI imaging sequence to obtain different MRI imaging sequence images; The obtained MRI imaging sequence images include: T1-weighted imaging sequence images, T2-weighted imaging sequence images, diffusion-weighted imaging sequence images, and specific kidney imaging sequence images; Fuse the obtained MRI imaging sequence images to obtain a fused multimodal image, including: performing normalization processing on each of the MRI imaging sequence images to obtain normalized MRI imaging sequence images; performing spatial alignment on the normalized different MRI imaging sequence images; using linear or non-linear fusion technology to perform weighted fusion on the aligned different MRI imaging sequence images to obtain the multimodal image; Preprocess the multimodal image, including: performing per-pixel signal intensity correction on the multimodal image so that each pixel of the multimodal image has the same signal intensity level; performing histogram equalization on the corrected multimodal image to enhance the contrast of the image; performing adaptive filtering on the histogram-equalized multimodal image to remove noise and retain edge details; performing non-local means denoising on the adaptively filtered multimodal image to obtain the preprocessed multimodal image; Perform image segmentation on the preprocessed multimodal image to obtain a binary image including a hardened area and a normal area; Obtain the kidney sclerosis ratio based on the binary image, including: obtaining the number of pixels in the hardened area based on the binary image; obtaining the number of pixels in the entire kidney based on the binary image; taking the ratio of the number of pixels in the hardened area to the number of pixels in the entire kidney as the kidney sclerosis ratio.

2. The method for analyzing the kidney sclerosis ratio based on MRI images according to claim 1, wherein The selected MRI imaging sequence includes: T1-weighted imaging sequence, T2-weighted imaging sequence, diffusion-weighted imaging sequence, and specific kidney imaging sequence.

3. The method for analyzing the kidney sclerosis ratio based on MRI images according to claim 1, characterized in that The performing per-pixel signal intensity correction on the multimodal image so that each pixel of the multimodal image has the same signal intensity level includes: Scanning the same patient or a standard phantom multiple times to obtain the signal intensity relationship between different multimodal images and obtain a correction coefficient; Using the correction coefficient to perform per-pixel adjustment on the multimodal image so that each pixel of the multimodal image has the same signal intensity level.

4. The method for analyzing the renal sclerosis ratio based on MRI images according to claim 1, wherein, The performing image segmentation on the preprocessed multimodal image to obtain a binary image including a hardened area and a normal area includes: Construct a data set, which includes a training set, a validation set, and a test set; Construct a deep learning model and use the training set to train the deep learning model; Optimize the deep learning model using a cross-entropy loss function; Use the validation set to evaluate the accuracy of the trained deep learning model and use the test set to test the generalization ability of the model to obtain the final deep learning model; Use the deep learning model to perform image segmentation on the preprocessed multimodal image to obtain a binary image including a hardened area and a normal area.

5. The method for analyzing the renal sclerosis ratio based on MRI images according to claim 4, wherein The deep learning model includes a U-Net model.

6. The method for analyzing the renal sclerosis ratio based on MRI images according to claim 4, wherein The hardened area includes a white pixel area; the normal area includes a black pixel area.

7. A kidney sclerosis ratio analysis system based on MRI images, characterized in that, For performing the method for analyzing the renal sclerosis ratio based on MRI images according to any one of claims 1 to 6; the system for analyzing the renal sclerosis ratio based on MRI images, comprising: An MRI imaging sequence selection module, configured to select an MRI imaging sequence according to the renal lesion characteristics of the patient; An image acquisition module, configured to acquire images of the patient's kidney based on the selected MRI imaging sequence to obtain different MRI imaging sequence images; A multimodal fusion module, configured to fuse the obtained MRI imaging sequence images to obtain a fused multimodal image; A preprocessing module, configured to preprocess the multimodal image; An image segmentation module, configured to perform image segmentation on the preprocessed multimodal image to obtain a binary image including a hardened area and a normal area; A sclerosis ratio calculation module, configured to obtain the renal sclerosis ratio based on the binary image.

Images