Medical image enhancement method, device, electronic device and readable storage medium

Abstract

The medical image enhancement method, device, electronic device and readable storage medium provided by the present invention include: obtaining a first background estimation image and a second background estimation image corresponding to each of multiple channel images of a medical image to be processed; determining a detail information image corresponding to each channel image based on the first background estimation image corresponding to each channel image, and determining a background compensation factor corresponding to each channel image based on all detail information images and all second background estimation images; performing background compensation on each channel image based on the background compensation factor corresponding to each channel image, and performing image enhancement on each channel image after background compensation; fusing each channel image after image enhancement to obtain an image-enhanced medical image to be processed. The present invention achieves the purpose of image enhancement through background estimation images, detail image information and background compensation factors, thereby improving the medical image enhancement effect.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to a medical image enhancement method, device, electronic device and readable storage medium. Background Art

[0002] Medical image enhancement is an essential part of medical acquisition and imaging equipment. The quality of the generated medical images may affect the experimenter or doctor's in-depth understanding of the medical problems behind the medical images.

[0003] Medical image enhancement requires strong robustness. While maintaining high image quality, it must ensure that the essential content of the enhanced image is not altered by the model. In some use cases, medical image enhancement algorithms also require extremely high computational efficiency, allowing users to quickly obtain high-quality images and more time to analyze the underlying pathologies.

[0004] Traditional methods often use body sensors to determine the usage environment when studying medical image enhancement. Deep learning-based methods typically construct training data specifically based on usage scenarios and goals. However, medical image labeling is an extremely demanding process, and relying on data labeling to train models is time-consuming and laborious. Furthermore, methods that rely on transfer learning and generative adversarial networks produce uncertain results. Summary of the Invention

[0005] One of the purposes of the present invention is to provide a medical image enhancement method, device, electronic device and readable storage medium to unify the medical image enhancement problem into a framework, while not relying on deep learning methods to establish an efficient enhancement method based on traditional image processing.

[0006] In a first aspect, the present invention provides a medical image enhancement method, which includes: obtaining multiple channel images of a medical image to be processed, and respectively determining a first background estimation image and a second background estimation image corresponding to each of the channel images; determining a detail information image corresponding to each of the channel images based on the first background estimation image corresponding to each of the channel images, and determining a background compensation factor corresponding to each of the channel images based on all of the detail information images and all of the second background estimation images; wherein the detail information image contains the image content of interest in the medical image to be processed; the first background estimation image and the second background estimation image do not contain the image content of interest; based on the background compensation factor, performing background compensation on the channel image corresponding to the background compensation factor, and performing image enhancement on each of the channel images after background compensation; fusing each of the channel images after image enhancement to obtain the medical image to be processed after image enhancement.

[0007] In a second aspect, the present invention provides a medical image enhancement device, comprising: a determination module, configured to obtain multiple channel images of a medical image to be processed, and respectively determine a first background estimation image and a second background estimation image corresponding to each channel image; the determination module is further configured to determine a detail information image corresponding to each channel image based on the first background estimation image corresponding to each channel image, and determine a background compensation factor corresponding to each channel image based on all the detail information images and all the second background estimation images; wherein the detail information image contains the image content of interest in the medical image to be processed; the first background estimation image and the second background estimation image do not contain the image content of interest; an enhancement module, configured to perform background compensation on the channel image corresponding to the background compensation factor based on the background compensation factor, and perform image enhancement on each channel image after background compensation; the enhancement module is further configured to fuse each channel image after image enhancement to obtain the medical image after image enhancement.

[0008] In a third aspect, the present invention provides an electronic device comprising a processor and a memory, wherein the memory stores a computer program executable by the processor, and the processor can execute the computer program to implement the medical image enhancement method provided in the first aspect.

[0009] In a fourth aspect, the present invention provides a readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the medical image enhancement method provided in the first aspect.

[0010] The medical image enhancement method, device, electronic device and readable storage medium provided by the present invention include: first, performing background estimation image calculation on multiple channel images to obtain a first background estimation image and a second background estimation image corresponding to each of the multiple channel images; then, subtracting the first background estimation image from the channel image to obtain a detail information image, combining the detail information image and the second background estimation image to obtain a background compensation factor; combining the background compensation factor to perform background compensation on each channel image; and finally, normalizing each channel image after background compensation to obtain the final enhancement result corresponding to each channel image. The enhanced channel images are then re-fused to obtain an enhanced medical image to be processed. The method achieves the purpose of image enhancement by performing operations such as simulated background estimation image extraction, detail image information extraction, and background compensation factor calculation on the image. This method does not require special optimization for different tissue sites or imaging environments. The entire process is fast and efficient, improving the enhancement effect of medical images. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.

[0012] Figure 1 A structural block diagram of an electronic device provided by an embodiment of the present invention;

[0013] Figure 2 A schematic flow chart of a medical image enhancement method provided in an embodiment of the present invention;

[0014] Figure 3 A schematic flowchart of step S201 provided in an embodiment of the present invention;

[0015] Figure 4 A schematic flowchart of step S202 provided in an embodiment of the present invention;

[0016] Figure 5 A schematic flowchart of step S203 provided in an embodiment of the present invention;

[0017] Figure 6 A schematic diagram showing the effect of the medical image enhancement method provided by an embodiment of the present invention;

[0018] Figure 7 This is a functional module diagram of the medical image enhancement device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations.

[0020] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention as claimed, but rather merely represents selected embodiments of the present invention. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without creative effort shall fall within the scope of protection of the present invention.

[0021] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings.

[0022] In the description of the present invention, it should be noted that if the terms "upper", "lower", "inside", "outside", etc. appear, the orientation or position relationship indicated is based on the orientation or position relationship shown in the accompanying drawings, or is the orientation or position relationship in which the product of the invention is usually placed when in use. It is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be understood as a limitation on the present invention.

[0023] In addition, the terms "first", "second", etc., if used, are merely used to distinguish and describe, and should not be understood as indicating or implying relative importance.

[0024] It should be noted that, in the absence of conflict, the features in the embodiments of the present invention may be combined with each other.

[0025] First, the relevant terms involved in the embodiments of the present invention are explained.

[0026] RGB color space: RGB color space is based on the three primary colors R (Red), G (Green), and B (Blue). By superimposing these colors to varying degrees, it produces a rich and wide range of colors. The RGB color space is the most commonly used model in everyday life, and most televisions, CRT monitors, and other computer displays use this model. Any color in nature can be created by mixing red, green, and blue light. Most colors we see in real life are mixed colors.

[0027] Deep Learning: Deep Learning (DL) in English uses deep neural networks to abstract the characteristics of data in order to more accurately represent the distribution and characteristics of the data.

[0028] Transfer learning (TL) is a machine learning method that uses the model developed for task A as the starting point and reuses it in the process of developing a model for task B.

[0029] Convolutional Neural Network (CNN) is a feedforward neural network that includes convolution operations. It has outstanding performance in large-scale image processing and is one of the representative algorithms of deep learning.

[0030] Generative Adversarial Networks (GANs) are a type of deep learning model that learns through the interaction of (at least) two modules in the framework: a generative model and a discriminative model.

[0031] Gaussian filter: Gaussian filter is a low-pass filter that is applied to the field of digital signal processing by simulating Gaussian function.

[0032] Medical images are commonly used by medical researchers and physicians to interpret and infer pathological information presented in these images. The imaging process is extremely complex, and their use cases vary widely, ranging from lung CT scans to head MRIs and even cellular protein imaging.

[0033] Medical image enhancement is an essential component of medical acquisition and imaging equipment. The quality of the generated medical images may affect the experimenter or doctor's understanding of the medical issues underlying the medical images, resulting in a lack of in-depth understanding. Medical image enhancement must be highly robust. While ensuring high quality of the enhanced image, it must also ensure that the essential content of the enhanced image is not altered by the model. In some use cases, medical image enhancement algorithms also require extremely high computational efficiency so that users can quickly obtain high-quality images and have more time to analyze the pathological causes behind the images.

[0034] Currently, relevant technologies provide a variety of medical image enhancement methods, but these methods all have limitations.

[0035] The first method receives a selection of an observed body part, determines the target pixel width by determining the physical width of the target tissue, the image reconstruction field of view, and the image matrix size. Based on the characteristics of the observed part, a sharpening filter operator is selected from a plurality of different filter operators. The selected filter operator is then used to calculate a sharpened image. This achieves the desired enhancement effect for the medical image under reconstruction conditions.

[0036] The above method requires calculating the physical width of the tissue being observed to determine the matrix size and corresponding sharpening filter for the enhanced image. Using different sharpening filters for different tissues results in image enhancement requiring additional expert information about the tissue. Furthermore, this information can be inaccurate due to differences between individuals, and even within individuals with different physical conditions. This can lead to unstable image enhancement results.

[0037] The second method: After preliminary processing and preprocessing of the medical image to obtain medical image I, it is input into the trained deep learning model to obtain medical image III. After post-processing of medical image III to obtain medical image IV, medical image I and medical image IV are fused and reconstructed to obtain a high-definition medical image V; the deep learning model consists of a VGG16 network (medical image II is obtained from medical image I) and a DenseUnet network (medical image III is obtained from medical image II); the training process uses medical image I as the input of the deep learning model and the theoretical output medical image III as the output of the deep learning model, and continuously adjusts all parameters of the deep learning model to achieve the purpose of enhancing the input image to the output image.

[0038] The above method relies on deep learning methods for model training. This requires a large amount of medical image data verified by medical experts, which consumes a great deal of manpower and resources. It also relies on the basic VGG neural network and DenseUnet neural network, which are designed and trained based on routine, everyday data. Directly using these neural networks for medical image enhancement using transfer learning methods may result in unstable output due to the significant differences in morphology between everyday and medical data.

[0039] The third method uses a combined input of a clear medical image and random noise, and processes the combined input features using deep learning model 1 to generate a blood-stained image. Deep learning model 2 processes the generated blood-stained image and the original pure blood-stained image to determine whether the two images are similar. If not, the neural network gradient of deep learning model 2 is updated. If similar, the next step is performed; the blood-stained image is input into a cascade neural network model, and the final clear image result is output. This patent uses a generative adversarial network in deep learning to simulate and generate blood-stained images, and uses an improved variational autoencoder network to remove blood stains from the blood-stained image, thereby improving the clarity of medical images.

[0040] The above method simulates a low-quality image to be enhanced by using a combined input of a clear medical image and random noise. The amount and type of random noise will determine the degree of low image quality. The data simulated by this simple method is difficult to generalize to the data generated in real-world environments due to complex factors such as physical environment, lighting, and imaging materials. At the same time, the model relies on a generative adversarial network to reconstruct a high-quality medical image from a low-quality medical image. The generative adversarial network itself uses data randomness for learning, and the network learns the data distribution of the entire training data. For pixel-level calculations such as image enhancement or restoration, its output makes it difficult to achieve stable image enhancement and restoration effects.

[0041] In view of this, in order to solve the problem of enhancing medical images in different scenarios, the medical image enhancement problem is unified into a framework; at the same time, it does not rely on deep learning methods, and establishes an efficient method based on traditional image processing. The embodiment of the present invention provides a simple, fast and efficient medical image enhancement method, which achieves the purpose of image enhancement by performing operations such as simulated background estimation image extraction, detailed image information extraction, and background compensation factor calculation on the image.

[0042] First see Figure 1 , Figure 1 This is a structural block diagram of an electronic device provided in an embodiment of the present invention. The electronic device can be used to execute the medical image enhancement method provided in an embodiment of the present invention.

[0043] like Figure 1 As shown, electronic device 100 includes memory 101, processor 102, and communication interface 103. Memory 101, processor 102, and communication interface 103 are electrically connected to each other directly or indirectly to enable data transmission or interaction. For example, these components can be electrically connected to each other via one or more communication buses or signal lines.

[0044] The memory 101 can be used to store software programs and modules, such as the instructions / modules of the medical image enhancement device 400 provided in an embodiment of the present invention. These can be stored in the memory 101 in the form of software or firmware or embedded in the operating system (OS) of the electronic device 100. The processor 102 executes the software programs and modules stored in the memory 101 to perform various functional applications and data processing. The communication interface 103 can be used for signaling or data communication with other node devices.

[0045] Among them, the memory 101 can be, but is not limited to, random access memory (RAM), read only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), etc.

[0046] The processor 102 may be an integrated circuit chip with signal processing capabilities. The processor 102 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0047] I understand. Figure 1 The structure shown is for illustration only. The electronic device 100 may also include Figure 1 More or fewer components than shown, or with Figure 1 Different configurations shown. Figure 1 The components shown may be implemented in hardware, software, or a combination thereof.

[0048] Below Figure 1 The electronic device shown is the execution subject. The medical image enhancement method provided by the embodiment of the present invention is described in detail. Figure 2 , Figure 2 A schematic flow chart of a medical image enhancement method provided in an embodiment of the present invention, the method comprising:

[0049] S201, obtaining multiple channel images of a medical image to be processed, and respectively determining a first background estimation image and a second background estimation image corresponding to each channel image;

[0050] S202, determining a detail information image corresponding to each channel image based on the first background estimation image corresponding to each channel image, and determining a background compensation factor corresponding to each channel image based on all detail information images and all second background estimation images;

[0051] The detail information image contains the image content of interest in the medical image to be processed; the first background estimation image and the second background estimation image do not contain the image content of interest;

[0052] S203, performing background compensation on the channel image corresponding to the background compensation factor based on the background compensation factor, and performing image enhancement on each channel image after background compensation;

[0053] S204: Fusing each channel image after image enhancement to obtain an image-enhanced medical image to be processed.

[0054] In the above-mentioned medical image enhancement method, background estimation images are first calculated for multiple channel images to obtain first and second background estimation images corresponding to each of the multiple channel images. The first background estimation image is then subtracted from the channel image to obtain a detail information image. The detail information image and the second background estimation image are combined to obtain a background compensation factor. Based on the background compensation factor, background compensation is performed on each channel image. Finally, each channel image after background compensation is normalized to obtain the final enhancement result corresponding to each channel image. The enhanced channel images are then re-fused to obtain the enhanced medical image to be processed. The entire process is fast and efficient, improving the enhancement effect of medical images.

[0055] The above steps S201 to S204 are described in detail below.

[0056] In step S201, a plurality of channel images of a medical image to be processed are obtained, and a first background estimation image and a second background estimation image corresponding to each channel image are respectively determined.

[0057] In this embodiment, the medical image to be processed may be a medical image captured in real time or a medical image pre-stored locally. The plurality of channel images are respectively an R channel image, a G channel image and a B channel image.

[0058] As an optional implementation, the implementation of step S201 can refer to Figure 3 , Figure 3 A schematic flow chart of step S201 provided in an embodiment of the present application:

[0059] S201-1, performing RGB channel separation on the medical image to be processed to obtain multiple channel images.

[0060] S201 - 2 , inputting the first background estimation parameter into the Gaussian function, and performing convolution calculation on the Gaussian function and the channel image to obtain a first background estimation image.

[0061] S201-3, inputting the second background estimation parameter into the Gaussian function, and performing convolution calculation on the Gaussian function and the channel image to obtain a second background estimation image; the first background estimation parameter and the second background estimation parameter are different.

[0062] For ease of understanding, assume that the medical image to be processed is represented as f(x,y), and RGB channels are separated to obtain multiple channel images, that is, f(x,y) = [R(x,y), G(x,y), B(x,y)], because the calculations of the three components R(x,y), G(x,y), and B(x,y) are exactly the same. i(x,y) is used to represent R(x,y), G(x,y), and B(x,y), where (x,y) represents the spatial coordinates of the two-dimensional image in the x and y directions.

[0063] The Gaussian function in the embodiment of the present application is as follows:

[0064]

[0065] Wherein, k∫∫G(x,y)dxdy=1; σ is used here to represent the scale of the Gaussian function; different background estimation parameters can be obtained by setting different values for σ. In the embodiment of the present application, σ is set to i and k i is the first background estimation parameter, σ c and k c is the second background estimation parameter, σ c ≠σ i σ i , σ c It can be set according to actual needs and is not limited here.

[0066] Based on the above Gaussian function, the first background estimation parameter σ i and k i , and the second background estimation parameter σ c and k c , for c i (x,y) performs background estimation calculation.

[0067] Among them, the first background estimation image is IG i (x,y):

[0068] IG i (x,y)=conv(c i (x,y),G i (x,y))

[0069] Among them, conv represents image convolution calculation.

[0070] The second background estimation image is like IGC i (x,y), where

[0071] IGC i (x,y)=conv(c i (x,y),G c (x, y)) Through the above implementation, the first background estimation image and the second background estimation image corresponding to R(x, y), G(x, y), and B(x, y) can be obtained respectively, and then step S202 can be executed.

[0072] In step S202, based on the first background estimation image corresponding to each channel image, the detail information image corresponding to each channel image is determined, and based on all the detail information images and all the second background estimation images, the background compensation factor corresponding to each channel image is determined.

[0073] In this embodiment, the detail information image includes the image content of interest in the medical image to be processed, and the first background estimation image and the second background estimation image do not include the image content of interest; the image content of interest can be different tissue parts, such as bones, brain structure, chest structure, etc., which is not limited here.

[0074] The first background estimation image is used to determine the detail information image, and the second background image is used to determine the background compensation factor corresponding to each channel image and perform background compensation on the channel image.

[0075] This embodiment also provides an implementation method for determining the detail information image and the background compensation factor, see Figure 4 , Figure 4 A schematic flow chart of step S202 provided in an embodiment of the present invention:

[0076] S202-1: For each channel image, subtract the channel image from a first background estimation image corresponding to the channel image to obtain a detail information image corresponding to the channel image.

[0077] S202-2: Taking the ratio of the sum of all pixel values of the second background estimation image corresponding to the channel image to the sum of all pixel values of the detail information image as the background compensation factor corresponding to the channel image.

[0078] Continuing with the above example, the calculation method of the detail information image is as follows:

[0079] f i (x,y)=k c [c i (x,y)-IG i (x,y)]

[0080] where k c is a natural constant.

[0081] In order to make the final output more prominent in the details of the image and enhance the contrast of the image, after obtaining the detail information image, each detail information image can be optimized based on the preset optimization function to obtain a generalized detail information image, which is expressed as SDf i (x,y):

[0082] Df i (x,y)=Sig(Df i (x,y))

[0083] The optimization function is set as:

[0084]

[0085] Among them, A,k g ,z0 are natural parameters, and their values will not have a big impact on the final output.

[0086] After obtaining the detail information image, set the background compensation factor k b , the background compensation factor k can be obtained according to the following relationship b :

[0087]

[0088] Here, ΣΣ represents a summation symbol.

[0089] It is understood that the images in the embodiments of the present application are all represented in the form of a matrix, where the rows of the matrix correspond to the height of the image (in pixels), the columns of the matrix correspond to the width of the image (in pixels), the elements of the matrix correspond to the pixels of the image, and the values of the matrix elements are the pixel values of the pixels. For example, the matrix corresponding to the R channel image is: R = [[1,1], [2,2]…], then the above summation means: sum(sum(R)) = (1+1) + (2+2)…

[0090] Get the background compensation factor k b Afterwards, in order to improve the image compensation effect, the background compensation factor can be corrected first. The correction method can be: the minimum value of the background compensation factor and the reciprocal of the background compensation factor is used as the corrected background compensation factor, that is:

[0091]

[0092] After the background compensation factor is obtained, step S203 may be executed.

[0093] In step S203, background compensation is performed on each channel image based on the background compensation factor corresponding to each channel image, and image enhancement is performed on each channel image after background compensation.

[0094] As an optional implementation, the above step S203 can refer to Figure 5 , Figure 5 A schematic flow chart of step S203 provided in an embodiment of the present application:

[0095] S203-1, for each channel image, multiply the background compensation factor by the second background estimation image corresponding to the channel image, and add the second background estimation image obtained after the multiplication to the detail information image corresponding to the channel image to obtain the channel image after background compensation.

[0096] S203-2, normalizing each channel image after background compensation to obtain each channel image after image enhancement.

[0097] As an optional implementation, the normalization method may be: dividing each pixel value of the background-compensated channel image by the maximum pixel value in the channel image to obtain the image-enhanced channel image.

[0098] In the embodiment of the present application, the detail information image SDf i (x,y) and the second background estimation image IGC i (x,y), we can get the channel image Ncrf(x,y) after background compensation:

[0099] Ncrf(x,y)=SDf i (x,y)+k n *IGC i (x,y)

[0100] Get Ncrf(x,y) corresponding to R(x,y), G(x,y), and B(x,y) respectively, and then perform image enhancement on the channel image after background compensation, that is, normalize Ncrf(x,y) to get:

[0101]

[0102] Here, max represents the maximum value. For example, if Ncrf(x,y) = [[2,3],[4,100]], then max(Ncrf(x,y)) is 100.

[0103] After obtaining the background-compensated channel images corresponding to R(x,y), G(x,y), and B(x,y), respectively, step S204 may be executed.

[0104] In step S204, each channel image after image enhancement is fused to obtain an image-enhanced medical image to be processed.

[0105] As an optional implementation, each channel image after background compensation is divided by a maximum value channel image among all channel images after background compensation to obtain each channel image after image enhancement.

[0106] After obtaining each channel image of the image enhancement, the channel images can be fused using existing conventional image fusion methods, for example, pixel-level fusion, feature-level fusion, etc., and the resulting image is the medical image after image enhancement implemented in the embodiment of the present application.

[0107] The medical image enhancement method in the above embodiment can enhance the medical image to be processed quickly and efficiently. This method does not require special optimization for different tissue parts or imaging environments. It is a unified medical image enhancement calculation method. The enhancement effect of the embodiment of the present invention can be seen in Figure 6 , Figure 6 This is a schematic diagram of the effect of the medical image enhancement method provided in an embodiment of the present invention, where the image on the left is the medical image to be processed, and the image on the right is the effect of the image after the medical image enhancement processing provided by an embodiment of the present invention. It can be clearly seen that the image effect after enhancement by this method is significantly better than the original image and can provide more detailed information.

[0108] The medical image enhancement method provided in the embodiment of the present application can be executed in a hardware device or implemented in the form of a software module. When the medical image enhancement method is implemented in the form of a software module, the embodiment of the present application also provides a medical image enhancement method device, see Figure 7 , Figure 7 This is a functional module diagram of a medical image enhancement device provided in an embodiment of the present application. The medical image enhancement device 400 may include:

[0109] The determination module 410 is configured to obtain a plurality of channel images of the medical image to be processed, and respectively determine a first background estimation image and a second background estimation image corresponding to each channel image;

[0110] The determination module 410 is further configured to: determine a detail information image corresponding to each channel image based on the first background estimation image corresponding to each channel image, and determine a background compensation factor corresponding to each channel image based on all detail information images and all second background estimation images;

[0111] The detail information image contains the image content of interest in the medical image to be processed; the first background estimation image and the second background estimation image do not contain the image content of interest;

[0112] The enhancement module 420 is configured to perform background compensation on each channel image based on a background compensation factor corresponding to each channel image, and perform image enhancement on each channel image after background compensation.

[0113] The enhancement module 420 is further configured to fuse each channel image after image enhancement to obtain an enhanced medical image.

[0114] It is understood that the determination module 410 and the enhancement module 420 can be executed in a coordinated manner. Figure 2 Each step in the process is performed to achieve the corresponding technical effects.

[0115] In an optional embodiment, the determination module 410 is specifically used to: for each channel image, subtract the channel image from the first background estimation image corresponding to the channel image to obtain a detail information image corresponding to the channel image; and use the ratio of the sum of all pixel values of the second background estimation image corresponding to the channel image to the sum of all pixel values of the detail information image as the background compensation factor corresponding to the channel image.

[0116] In an optional embodiment, the medical image enhancement device 400 may include an optimization module for optimizing each detail information image based on a preset optimization function.

[0117] In an optional embodiment, the enhancement module 420 is specifically used to: for each channel image, multiply the background compensation factor by the second background estimation image corresponding to the channel image, and add the second background estimation image obtained after the multiplication to the detail information image corresponding to the channel image to obtain the channel image after background compensation; normalize each channel image after background compensation to obtain each channel image after image enhancement.

[0118] In an optional embodiment, the medical image enhancement apparatus 400 may include a correction module, configured to use the minimum value between the background compensation factor and the reciprocal of the background compensation factor as the corrected background compensation factor.

[0119] In an optional embodiment, the enhancement module 420 is specifically configured to: divide each pixel value of the channel image after background compensation by the maximum pixel value in the channel image to obtain the channel image after image enhancement.

[0120] In an optional embodiment, the determination module 410 is specifically used to: perform RGB channel separation on the medical image to be processed to obtain multiple channel images; input the first background estimation parameter into the Gaussian function, and perform convolution calculation on the Gaussian function and the channel image to obtain a first background estimation image; input the second background estimation parameter into the Gaussian function, and perform convolution calculation on the Gaussian function and the channel image to obtain a second background estimation image; the first background estimation parameter and the second background estimation parameter are different.

[0121] The present application also provides a computer-readable storage medium having a computer program stored thereon. When executed by a processor, the computer program implements the medical image enhancement method described in any of the aforementioned embodiments. The computer-readable storage medium may be, but is not limited to, a USB flash drive, a mobile hard drive, ROM, RAM, PROM, EPROM, EEPROM, a magnetic disk, or an optical disk, among other media capable of storing program code.

[0122] It should be understood that the apparatus and method disclosed in the present invention may also be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of the apparatus, methods, and computer program products according to various embodiments of the present invention. In this regard, each box in the flowchart or block diagram may represent a module, a program segment, or a portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the boxes may also occur in an order different from that marked in the accompanying drawings. For example, two consecutive boxes may actually be executed substantially in parallel, or they may sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or action, or may be implemented using a combination of dedicated hardware and computer instructions.

[0123] In addition, the functional modules in the various embodiments of the present invention may be integrated together to form an independent part, or each module may exist independently, or two or more modules may be integrated to form an independent part.

[0124] If the function is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes various media that can store program code, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk. 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 actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not preclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0125] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, improvements, and the like made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention. It should be noted that similar reference numerals and letters represent similar items in the following figures. Therefore, once an item is defined in one figure, it need not be further defined or explained in subsequent figures.

Claims

1. A medical image enhancement method, characterized in that: The method comprises: Obtaining a plurality of channel images of a medical image to be processed, and respectively determining a first background estimation image and a second background estimation image corresponding to each of the channel images; Determining a detail information image corresponding to each channel image based on the first background estimation image corresponding to each channel image, and determining a background compensation factor corresponding to each channel image based on all the detail information images and all the second background estimation images, including: for each channel image, subtracting the channel image from the first background estimation image corresponding to the channel image to obtain the detail information image corresponding to the channel image; and using a ratio of a sum of all pixel values of the second background estimation image corresponding to the channel image to a sum of all pixel values of the detail information image as the background compensation factor corresponding to the channel image; The detail information image includes the image content of interest in the medical image to be processed; the first background estimation image and the second background estimation image do not include the image content of interest; Based on the background compensation factor, background compensation is performed on the channel image corresponding to the background compensation factor, and image enhancement is performed on each of the channel images after background compensation, including: for each channel image, multiplying the background compensation factor by the second background estimation image corresponding to the channel image, and adding the second background estimation image obtained after the multiplication to the detail information image corresponding to the channel image to obtain the channel image after background compensation; normalizing each of the channel images after background compensation to obtain each of the channel images after image enhancement; Each channel image after image enhancement is fused to obtain the medical image to be processed after image enhancement.

2. The medical image enhancement method according to claim 1, characterized in that: Before taking the ratio of the sum of all pixel values of the second background estimation image corresponding to the channel image to the sum of all pixel values of the detail information image as the background compensation factor corresponding to the channel image, the method further includes: Based on a preset optimization function, each of the detail information images is optimized.

3. The medical image enhancement method according to claim 1, wherein: Before multiplying the background compensation factor by the second background estimation image corresponding to each channel image, and adding the multiplied second background estimation image to the detail information image corresponding to the channel image to obtain the background-compensated channel image, the method further includes: The minimum value between the background compensation factor and the reciprocal of the background compensation factor is used as the corrected background compensation factor.

4. The medical image enhancement method according to claim 1, wherein: Normalizing each of the channel images after background compensation to obtain each of the channel images after image enhancement, comprising: Each pixel value of the channel image after background compensation is divided by the maximum pixel value in the channel image to obtain the channel image after image enhancement.

5. The medical image enhancement method according to claim 1, wherein: Obtaining multiple channel images of a medical image to be processed, and determining a first background estimation image and a second background estimation image corresponding to each channel image, respectively, including: Performing RGB channel separation on the medical image to be processed to obtain a plurality of channel images; Inputting a first background estimation parameter into a Gaussian function, and performing a convolution calculation on the Gaussian function and the channel image to obtain the first background estimation image; A second background estimation parameter is input into the Gaussian function, and a convolution calculation is performed on the Gaussian function and the channel image to obtain the second background estimation image; the first background estimation parameter and the second background estimation parameter are different.

6. A medical image enhancement device, characterized in that: include: A determination module is used to obtain multiple channel images of the medical image to be processed and respectively determine a first background estimation image and a second background estimation image corresponding to each channel image; The determination module is further configured to: determine a detail information image corresponding to each channel image based on the first background estimation image corresponding to each channel image, and determine a background compensation factor corresponding to each channel image based on all the detail information images and all the second background estimation images, including: for each channel image, subtracting the channel image from the first background estimation image corresponding to the channel image to obtain the detail information image corresponding to the channel image; and using a ratio of a sum of all pixel values of the second background estimation image corresponding to the channel image to a sum of all pixel values of the detail information image as the background compensation factor corresponding to the channel image; The detail information image includes the image content of interest in the medical image to be processed; the first background estimation image and the second background estimation image do not include the image content of interest; an enhancement module, configured to: perform background compensation on the channel image corresponding to the background compensation factor based on the background compensation factor, and perform image enhancement on each of the channel images after background compensation, including: multiplying the background compensation factor by the second background estimation image corresponding to the channel image for each channel image, and adding the second background estimation image obtained after the multiplication to the detail information image corresponding to the channel image to obtain the channel image after background compensation; and normalizing each channel image after background compensation to obtain each channel image after image enhancement; The enhancement module is further used to fuse each channel image after image enhancement to obtain the medical image after image enhancement.

7. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores a computer program that can be executed by the processor, and the processor can execute the computer program to implement the method according to any one of claims 1 to 5.

8. A readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images