Analysis Method of DR Image and Electronic Device

Abstract

A method for analyzing DR images and an electronic device, the method comprising: obtaining a digital radiography (DR) image, wherein the DR image includes at least one of a DR original image and an image obtained after processing the DR original image; extracting image features from the DR image, the image features including at least one of a gray entropy feature, a texture feature, a noise feature, a gradient feature, and a divergence feature; inputting at least one of the gray entropy feature, the texture feature, the noise feature, the gradient feature, and the divergence feature into a target model, and outputting a DR image index reflecting the basic feature information content of the DR image. The present application obtains a DR image index that can objectively reflect the basic feature information content of the DR image from the image features extracted from the DR image, thereby providing an objective basis for image judgment for the operator.

Description

Technical Field

[0001] The present application relates to the field of medical image technology, and more specifically to a DR image analysis method and electronic device. Background Art

[0002] Digital Radiography (DR) images are a common type of medical digital images and are widely used in physical examinations and routine medical imaging diagnosis. With the development of digital X-ray detectors and digital image processing systems, DR images can often present diagnostic image effects within a wide range of exposure doses. However, during the image acquisition process, the final image effect is often determined by the experience of the operator, and the accuracy of the judgment is easily affected by various factors such as experience differences, subjective differences, and post-processing.

[0003] In order to reflect the effect of DR images, one way is to use the exposure index (EI) to indicate the size of a single exposure. However, the exposure index usually has energy dependence on a specific line quality, and usually the exposure index only uses the grayscale information of the image. Due to the complexity and diversity of clinical images and human body structures, the exposure index cannot well reflect the image information actually used for diagnosis in DR images. Summary of the invention

[0004] A series of simplified concepts are introduced in the Summary of the Invention section, which will be further described in detail in the Detailed Description of the Invention section. The Summary of the Invention section of this application does not mean to attempt to define the key features and essential technical features of the claimed technical solution, nor does it mean to attempt to determine the scope of protection of the claimed technical solution.

[0005] On one hand, the present application provides a DR image analysis method, the method comprising:

[0006] Emitting X-rays to a target tissue site, and controlling the X-rays to pass through the target tissue site;

[0007] receiving X-rays after passing through the target tissue site;

[0008] Processing the X-rays after passing through the target tissue site to obtain a digital X-ray photography DR image, wherein the DR image includes at least one of a DR original image and an image obtained by processing the DR original image;

[0009] Extracting image features from the DR image, the image features comprising at least one of grayscale entropy features, texture features, noise features, gradient features, and divergence features;

[0010] Input at least one of the image features including the grayscale entropy feature, texture feature, noise feature, gradient feature, and divergence feature into the target model, and output a DR image index reflecting the basic feature information content of the DR image through the target model.

[0011] On the other hand, the present application provides an analysis method for DR images, and the method includes:

[0012] Obtain a digital radiography (DR) image, where the DR image includes at least one of a DR original image and an image obtained by processing the DR original image;

[0013] Extract image features from at least one of the DR original image and the image obtained by processing the DR original image, where the image features include at least one of a grayscale entropy feature, texture feature, noise feature, gradient feature, and divergence feature;

[0014] Determine a DR image index reflecting the basic feature information content of the DR image according to at least one of the grayscale entropy feature, texture feature, noise feature, gradient feature, and divergence feature;

[0015] Output a DR image index reflecting the basic feature information content of the DR image.

[0016] On the other hand, the present application provides an analysis method for DR images, and the method includes:

[0017] Obtain a digital radiography (DR) image, where the DR image includes at least one of a DR original image and an image obtained by processing the DR original image;

[0018] Determine a DR image index reflecting the basic feature information content of the DR image according to at least one of the DR original image and the image obtained by processing the DR original image;

[0019] Output a DR image index reflecting the basic feature information content of the DR image.

[0020] On the other hand, the present application provides an analysis method for DR images, and the method includes:

[0021] Obtain a digital radiography (DR) image and the image features corresponding to the DR image, where the DR image includes at least one of a DR original image and an image obtained by processing the DR original image, and the image features include at least one of a grayscale entropy feature, texture feature, noise feature, gradient feature, and divergence feature;

[0022] Determine a DR image index reflecting the basic feature information content of the DR image according to the DR image and the image features corresponding to the DR image;

[0023] Output a DR image index reflecting the basic feature information content of the DR image.

[0024] On the other hand, the present application provides a DR imaging device, which includes:

[0025] An X-ray generator, a detector, a processor and a display;

[0026] The X-ray generator is used to generate X-rays, emit X-rays to the target tissue site, and control the X-rays to pass through the target tissue site;

[0027] The detector is used to receive the X-rays after passing through the target tissue site, and process the X-rays after passing through the target tissue site to obtain a digital radiography (DR) image, where the DR image includes at least one of a DR original image and an image obtained by processing the DR original image;

[0028] The processor is used to input at least one of the image features of gray entropy feature, texture feature, noise feature, gradient feature and divergence feature into the target model, and output a DR image index reflecting the basic feature information content of the DR image through the target model;

[0029] The display is used to display the DR image index reflecting the basic feature information content of the DR image.

[0030] On the other hand, the present application provides an electronic device, which includes a memory and a processor. A computer program is stored on the memory and run by the processor. When the computer program is run by the processor, it executes the steps of the analysis method of the DR image provided by the present application.

[0031] In the present application, a DR image index that can objectively reflect the basic feature information content of the DR image is obtained according to the image features extracted from the DR image, so as to provide an objective basis for image judgment for the operator. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other objects, features and advantages of the present application will become more obvious. The accompanying drawings are used to provide a further understanding of the embodiments of the present application, and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation to the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0033] Figure 1 A schematic flowchart showing the analysis method of the DR image according to an embodiment of the present invention;

[0034] Figure 2 Schematic diagram showing model training according to an embodiment of the present invention and DR image analysis using the trained network model

[0035] Figure 3 Schematic flowchart showing an analysis method for DR images according to another embodiment of the present invention

[0036] Figure 4 Block diagram showing the structure of an electronic device according to an embodiment of the present invention

[0037] Figure 5 Schematic flowchart showing an analysis method for DR images according to another embodiment of the present invention

[0038] Figure 6 Schematic flowchart showing an analysis method for DR images according to another embodiment of the present invention

[0039] Figure 7 Block diagram showing the structure of a DR imaging device according to an embodiment of the present invention Detailed implementation manners

[0040] In order to make the objectives, technical solutions and advantages of the present application more apparent, exemplary embodiments according to the present application will be described in detail below with reference to the accompanying drawings. Apparently, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited by the exemplary embodiments described herein. Based on the embodiments of the present application described herein, all other embodiments obtained by those skilled in the art without creative efforts shall fall within the protection scope of the present application.

[0041] In the following description, numerous specific details are given to provide a more thorough understanding of the present application. However, it is obvious to those skilled in the art that the present application can be implemented without one or more of these details. In other instances, in order to avoid confusion with the present application, some well-known technical features are not described.

[0042] It should be understood that the present application can be implemented in different forms and should not be construed as limited to the embodiments presented herein. On the contrary, providing these embodiments will make the disclosure thorough and complete, and will fully convey the scope of the present application to those skilled in the art.

[0043] The purpose of the terms used herein is only to describe specific embodiments and is not a limitation of the present application. As used herein, the singular forms "a", "an" and "the" are also intended to include the plural forms unless the context clearly dictates otherwise. It should also be understood that the terms "comprising" and / or "including", when used in this specification, specify the presence of the stated features, integers, steps, operations, elements and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups. As used herein, the term "and / or" includes any and all combinations of the associated listed items.

[0044] To fully understand the present application, detailed structures will be presented in the following description to illustrate the technical solutions proposed by the present application. The optional embodiments of the present application are described in detail below. However, in addition to these detailed descriptions, the present application may also have other embodiments.

[0045] Next, first refer to Figure 1 Describe an analysis method 100 of a DR image according to an embodiment of the present application. This analysis method can be applied to a DR imaging device or other electronic devices, such as a computer, a smart terminal, etc. As Figure 1 shown, the analysis method 100 of the DR image may include the following steps:

[0046] In step S110, obtain a digital radiography (DR) image;

[0047] In the embodiment of the present application, the device obtaining the DR image can be obtained locally or from other external devices, and no specific limitation is made here. Among them, obtaining from local can be that the device obtains the DR image in real time or obtains it non-real-time and stores it locally.

[0048] In a possible implementation manner, the specific process for the device to obtain the DR image locally is:

[0049] Emit X-rays to the target tissue site and control the X-rays to pass through the target tissue site; receive the X-rays after passing through the target tissue site; process the X-rays after passing through the target tissue site to obtain a digital radiography (DR) image. Among them, the target tissue site can be a tissue site to be examined of a human body or other animal body. For example, the head, abdomen, etc.

[0050] In this application, the DR image includes at least one of a DR original image and an image obtained after processing the DR original image. Among them, the DR original image is a grayscale image that has not undergone post-processing such as contrast adjustment and brightness adjustment. For example, the DR original image can be digital image information obtained by converting X-rays into visible light and then converting the visible light into an electrical signal. The image obtained after processing the DR original image includes an image that has undergone a certain transformation relationship, or post-processing such as image contrast and image brightness adjustment. That is, the image obtained after processing the DR original image can be regarded as an image obtained after arbitrary post-processing of the DR original image, and no specific limitation is made here.

[0051] In step S120, image features are extracted from the DR image, and the image features include at least one of a gray entropy feature, a texture feature, a noise feature, a gradient feature, and a divergence feature.

[0052] In step S130, at least one of the gray entropy feature, the texture feature, the noise feature, the gradient feature, and the divergence feature is input into the target model, and a DR image index reflecting the basic feature information content of the DR image is output.

[0053] In the analysis method 100 of the DR image according to the embodiment of this application, at least one of a gray entropy feature, a texture feature, a noise feature, a gradient feature, and a divergence feature is extracted from the DR image, and a DR image index that can objectively reflect the basic feature information content of the DR image, or a so-called Image Feature Index (IFI), is obtained based on the at least one extracted image feature. This DR image index can provide an objective judgment basis for the operator and can guide directions such as in-hospital quality control and dose reduction imaging.

[0054] Specifically, the DR image can be an original image collected by a DR imaging system. In this application, the original image can be an image displayed on a display, or original image data or original data, etc., and no specific limitation is made here. The DR system mainly consists of parts such as an X-ray generator, a detector (such as a flat panel detector), a workstation, and a mechanical device. Its working process mainly includes: the X-rays generated by the X-ray generator pass through the target part of the object to be measured and are attenuated, and the attenuated X-rays are projected onto the detector. The flat panel detector converts the X-rays into visible light and then converts the visible light into an electrical signal, thereby obtaining digital image information and synchronously transmitting the digital image information to the workstation. In addition, the DR image can also be an image obtained after processing the DR original image. For example, the workstation performs post-processing on the image, and finally obtains an image obtained after processing the DR original image.

[0055] In step S120, image features are extracted from the DR image. The image features include at least one of gray entropy features, texture features, noise features, gradient features, and divergence features. That is to say, the image features extracted in step S120 can be gray entropy features, texture features, noise features, gradient features, and divergence features, or any one or any two of these image features. Among them, gray entropy features, texture features, noise features, gradient features, and divergence features respectively reflect the gray scale, texture, noise information, gradient information, and divergence information of the DR image. The multi-dimensional image features are beneficial to obtaining more accurate DR image metrics.

[0056] In one embodiment, the gray entropy feature is the information content obtained by statistically screening out redundant gray level information sources in the DR image. According to the transfer characteristics of X-ray quanta being converted into digital signals by the detector of the DR system, each independent gray level in the DR image can be regarded as an information source, and the DR imaging process can be regarded as a process of transmitting information through the gray level information source. Among all the gray levels, there are some gray levels without gray values, so they are regarded as redundant gray level information sources. Since the extraction process of the gray entropy feature in this application screens out redundant gray level information sources, more effective image information can be extracted.

[0057] Exemplarily, when the gray entropy feature is the information content obtained by statistically screening out redundant gray level information sources in the DR image, the following method can be used to extract the gray entropy feature from the DR image:

[0058] First, obtain the first probability statistical distribution pH(i) of each gray level information source in the DR image, where i is the image gray value of the DR image; screen the first probability statistical distribution of the redundant gray level information sources among the gray level information sources to obtain the first probability statistical distribution pNH(i) of the non-redundant gray level information sources, that is:

[0059]

[0060] According to the ratio of the first probability statistical distribution pNH(i) of each non-redundant gray level information source to the sum ∑pNH(i) of the first probability statistical distributions of all non-redundant gray level information sources, obtain the second probability statistical distribution pH’(i) of each non-redundant gray level information source:

[0061] pH'(i) = pNH(i) / ∑pNH(i) Formula (2)

[0062] Finally, perform entropy calculation according to the second probability statistical distribution pH’(i) of the non-redundant gray level information sources to obtain the gray entropy feature H(Image). The method of entropy calculation includes but is not limited to taking the natural logarithm ln with e as the base or taking the logarithm with other numerical values as the base. It is expressed by the formula:

[0063]

[0064] Thus, the image gray entropy feature after screening out the redundant gray level source can be obtained.

[0065] The texture feature represents the change relationship of different gray values on the DR image in space, and it can reflect the abstract features of the change law of human tissues. Exemplarily, statistical methods can be used to extract the texture features of the DR image, and this method obtains the statistical characteristics of the texture region based on the gray attributes of the pixel and its neighborhood. The following mainly describes the specific details of extracting texture features by statistical methods, but the methods for extracting texture features can also include geometric methods, model methods, or other suitable texture feature extraction methods.

[0066] Exemplarily, when using statistical methods to extract texture features from a DR image, first, a texture feature description matrix is obtained according to the gray values of each pixel in the DR image; then, at least one two-dimensional component of the texture feature description matrix is extracted as at least one texture feature.

[0067] In one embodiment, the texture feature description matrix describes the change of gray values in different distances and different directions on the DR image. Let the DR image Image = f(x, y), then its texture feature description matrix P(i, j) is:

[0068] P(i, j) = #{(x1, y1), (x2, y2) ∈ M * N|f(x1, y1) = i, f(x2, y2) = j} Formula (4)

[0069] Where #(x) represents the number of elements in set x. Let the distance between two points (x1, y1) and (x2, y2) in the image be k, then the texture feature description matrix in different directions l can be extended to P(i, j, k, l). Statistically analyzing the change of gray values in different distances and different directions to obtain the texture feature description matrix can expand the texture feature description matrix to obtain information in more dimensions. Exemplarily, in order to calculate the actual eigenvalue in different angles, super-resolution interpolation can be performed on the DR image in each direction to obtain the sub-pixel gray value corresponding to the new target position, and the texture feature description matrix in different directions can be obtained according to the sub-pixel gray value.

[0070] After obtaining the texture feature description matrix, at least one two-dimensional component of the texture feature description matrix is extracted to obtain the texture feature, and the texture feature reflects the key characteristics of the texture feature description matrix. Exemplarily, the texture feature includes at least one of the following:

[0071] The first texture feature P1 is used to statistically analyze the value distribution of the texture feature description matrix and the overall distribution of texture changes in the DR image; the magnitude of the P1 value can reflect the clarity of the DR image and the depth of the texture grooves. The deeper the texture grooves, the larger the P1 value; conversely, the shallower the grooves, the smaller the P1 value.

[0072] The second texture feature P2 is used to statistically analyze the value distribution in the texture feature description matrix and the similarity degree in the parallel and normal directions in the DR image. The larger the P2 value, the greater the similarity degree of the gray levels of the image in different directions.

[0073] The third texture feature P3 is used to statistically analyze the value distribution in the texture feature description matrix and the uniformity of the gray level change distribution in the DR image. The magnitude of the P3 value can reflect the uniformity of the image gray level distribution and the coarseness of the texture. The larger the P3 value, the more stable the texture change of the DR image.

[0074] The fourth texture feature P4 is used to statistically analyze the value distribution in the texture feature description matrix and the measurement of the local texture change amount of the DR image. The larger the P4 value, the stronger the regularity of the texture.

[0075] The texture features extracted from the texture feature description matrix are not limited to the above four. In other embodiments, other two-dimensional components of the texture feature description matrix can also be extracted as texture features.

[0076] Image noise mainly includes X-ray quantum noise, and the distribution of X-ray quanta follows a Poisson distribution. Its variance is proportional to the average quantum detection number. According to this characteristic of X-ray quantum noise, the fluctuation degree of X-ray quanta can be statistically analyzed from the DR image as the image noise feature, that is, the image noise feature reflects the fluctuation degree of X-ray quanta in the DR image.

[0077] In one embodiment, extracting the image noise feature from the DR image includes the following steps: obtaining a high-frequency image from the DR image, and the high-frequency image obtained from the DR image mainly contains noise information; extracting the effective information in the high-frequency image to obtain a noise distribution image; statistically analyzing the noise value distribution in the noise distribution image to obtain the image noise feature.

[0078] As a way of implementation, obtaining the high-frequency image from the DR image includes: performing low-frequency filtering on the DR image I to filter out its low-frequency components to obtain a low-frequency image I1; obtaining the high-frequency image I2 according to the difference between the low-frequency image and the DR image. In other ways of implementation, the DR image I can also be directly subjected to high-frequency filtering to obtain the high-frequency image I2.

[0079] Exemplarily, the DR image I can be subjected to Gaussian low-pass filtering to obtain a low-frequency image I1. Among them, since the image is a two-dimensional signal, a two-dimensional Gaussian function is used for Gaussian low-pass filtering, and the two-dimensional Gaussian filter kernel is:

[0080]

[0081] By calculating the difference between the low-frequency image I1 and the DR image I, a high-frequency image I2 can be obtained, that is, I2 = I1 - I. By extracting the effective information in the high-frequency image I2, a noise distribution image I3 can be obtained. In one embodiment, the noise distribution image I3 is an image composed of the local root mean square of each pixel point in the high-frequency image, and the method for calculating the local root mean square includes using the L1 norm or other approximate or equivalent measures, such as:

[0082]

[0083] Among them, I3(i, j) is the value of the noise distribution image I3 at the pixel point (i, j), and I2(l, k) is the value of the high-frequency I2 at the pixel point (l, k).

[0084] After obtaining the noise distribution image I3, the noise value distribution thereof is statistically analyzed to obtain the image noise characteristics. Exemplarily, the noise value interval with the highest noise value distribution probability in the noise distribution image can be determined, and the noise value in the noise value interval with the highest noise value distribution probability can be used as the image noise characteristic.

[0085] Specifically, first, the pixel values of the noise distribution image I3 are divided into M intervals, the pixel value value interval of each interval is d, the histogram vector is initialized as h, the length is M, and each component h(i) of the initialized histogram represents the number of pixel value values in the i-th interval. For the pixel point (i, j) in the noise distribution image I3, the interval corresponding to the pixel value of this point is R[I3(i, j)]. Traverse each pixel point of the noise distribution image I3, statistically analyze h(R(I3(i, j))), after obtaining h, the maximum value of its main peak is max(h), the corresponding interval R0 = arg max(h), then calculate the noise distribution probability R0 * d as the noise characteristic.

[0086] In one embodiment, the gradient feature is used to characterize the clarity of the boundaries of different tissues in the DR image. Extracting the gradient feature from the DR image includes: determining the regional distribution of different tissues in the DR image; obtaining the clarity of the boundaries of the regional distribution of different tissues as the gradient feature.

[0087] Exemplarily, according to the different absorption coefficients of X-rays passing through different tissues of the human body, regional distributions of different tissues are formed in the DR image. There are boundaries of varying degrees between different tissue regions, and the clarity of these boundaries can be quantified by extracting gradient features. Specifically, it can be calculated according to the following formula:

[0088]

[0089] Among them, Grad(x, y) is used to represent the gradient magnitude at the coordinate (x, y), and Image(x, y) is used to represent the pixel value at the coordinate (x, y).

[0090] In one embodiment, the divergence feature is used to represent the transition strength and / or trend consistency of the boundaries between different tissues in the DR image. Extracting the divergence feature from the DR image includes: determining the regional distribution of different tissues in the DR image; obtaining the transition strength and / or trend consistency of the boundaries of the regional distribution of different tissues as the divergence feature.

[0091] Exemplarily, in the DR image, the boundaries between different tissues are not strict boundaries but have a certain degree of transition. The strength and / or trend consistency of this transition can be quantified by extracting the divergence feature. Specifically, it can be calculated according to the following formula:

[0092]

[0093] Among them, Diver(x, y) is used to represent the divergence magnitude at the coordinate (x, y), Image(x, y) is used to represent the pixel value at the coordinate (x, y), and actan represents the arctangent function.

[0094] In some embodiments, in addition to at least one of the above several image features, other image features can also be extracted from the DR image for obtaining DR image metrics, such as gray gradient, gray mean, variance, pixel value information, signal-to-noise ratio, contrast-to-noise ratio, etc. Specifically, it can be determined according to the actual situation and is not specifically limited here.

[0095] In step S130, as Figure 2As shown, at least one of the image features such as the grayscale entropy feature, texture feature, noise feature, gradient feature, and divergence feature is input into the target model, and a DR image index reflecting the basic feature information content of the DR image is output. Among them, the target model can be a trained model or an untrained model. Through this target model, a DR image index reflecting the basic feature information content of the DR image can be output. In a possible implementation manner, the target model includes a pre-trained network model. Among them, the network model can be trained in an offline manner, and its training method mainly includes:

[0096] First, obtain a set of DR images, where the set of DR images includes multiple DR images. An image feature set can be obtained according to the set of DR images, and the image feature set includes multiple image features extracted from the multiple DR images. Specifically, for each DR image in the set of DR images, at least one of the grayscale entropy feature, texture feature, noise feature, gradient feature, and divergence feature is extracted respectively. The multiple image features extracted from the multiple DR images in the set of DR images together constitute the image feature set, and each DR image in the set of DR images corresponds to at least one image feature in the image feature set.

[0097] Obtain a set of DR diagnostic images according to the set of DR images. The set of DR diagnostic images includes multiple DR diagnostic images obtained by performing image processing on the multiple DR images. Specifically, a DR image processing system can perform image processing on the multiple DR images in the set of DR images respectively to obtain multiple DR diagnostic images, that is, the visual images used by doctors for diagnosis.

[0098] Obtain a score set according to the set of DR diagnostic images. The score set includes the scores obtained by evaluating the multiple DR diagnostic images. In some embodiments, clinical experts can evaluate the DR diagnostic images in the set of DR diagnostic images according to the quality of the DR diagnostic images and give expert scores. The higher the score of the DR diagnostic image, the more basic feature information content is included in the corresponding DR image; conversely, the lower the score of the DR diagnostic image, the less basic feature information content is included in the corresponding DR image. The expert scores of the DR diagnostic images constitute the expert score set.

[0099] After that, the obtained image feature set and score set are used as the training sample set to train the network model, so as to obtain a trained network model. Among them, the network model can be a traditional machine learning model or a deep learning model, including but not limited to neural networks, support vector machines, linear discriminant analysis, etc.; the training methods include but not limited to model training methods such as linear regression and gradient descent. Exemplarily, an optimal mapping function from image features to scores can be learned to minimize the error between the DR image metrics mapped from the image features and the actually calibrated expert scores. Executing this optimal mapping function on the image features obtained in step S120 can obtain the prediction result of the DR image metrics closest to the expert scores.

[0100] In one embodiment, when performing classification and regression training using the DR image feature set and the expert score set, the classification and regression training formula adopted is:

[0101]

[0102] where x is the feature vector, {x i} i＝1,...,m is the support vector, α i is the weighting coefficient, b is the bias, k is the kernel function, and each value of the feature vector x needs to be obtained through a linear transfer function:

[0103]

[0104] where is the input feature, w and s are the scaling and translation parameter vectors respectively, and the kernel function k adopts or

[0105] It should be noted that this application places no restrictions on the linear transfer function, kernel function or parameters adopted in the regression training method.

[0106] After obtaining the trained network model in the model training stage, in the actual application process, input at least one of the image features such as the gray entropy feature, texture feature, noise feature, gradient feature and divergence feature extracted in step S120 into the trained network model, and the DR image metrics reflecting the basic feature information content of the DR image can be obtained, and the DR image metrics can be displayed through a display device or output in other ways.

[0107] Based on the above description, the method for analyzing a DR image according to an embodiment of the present application extracts at least one of gray entropy features, texture features, and noise features from the DR image, and obtains a DR image index that can objectively reflect the basic feature information content of the DR image according to the at least one image feature extracted. This DR image index can provide an objective basis for judgment for the operator and can guide directions such as in-hospital quality control and dose-reduced imaging.

[0108] Next, refer to Figure 3 to describe a method for analyzing a DR image according to another embodiment of the present application. Figure 3 is a schematic flowchart of an analysis method 300 for a DR image according to an embodiment of the present application.

[0109] As Figure 3 shown, the analysis method 300 for the DR image includes the following steps:

[0110] In step S310, a digital radiography (DR) image is obtained, where the DR image includes at least one of a DR original image and an image obtained after processing the DR original image;

[0111] In step S320, image features are extracted from the DR image, and the image features include at least one of gray entropy features, texture features, noise features, gradient features, and divergence features;

[0112] In step S330, a DR image index reflecting the basic feature information content of the DR image is determined according to at least one of the gray entropy features, texture features, noise features, gradient features, and divergence features;

[0113] It should be noted that in step S330, there are many ways to determine a DR image index reflecting the basic feature information content of the DR image according to at least one of the gray entropy features, texture features, noise features, gradient features, and divergence features. It can be Figure 1 and Figure 2 shown to be determined by inputting into a target model, or it can be processed by other processors on the device, or obtained after being processed by other processors outside the device. No specific limitation is made here.

[0114] In step S340, a DR image index reflecting the basic feature information content of the DR image is output.

[0115] The analysis method 300 of DR images is similar to the analysis method 100 of DR images in the above text. However, the analysis method 300 of DR images does not limit the specific manner of determining the DR image index according to the image features. After extracting at least one of the image features such as gray entropy feature, texture feature, noise feature, gradient feature, and divergence feature, it can be directly processed and calculated to obtain the DR image index, or the DR image index can be obtained through the trained network model as described above, or any other suitable manner can be used to obtain the DR image index. In addition, in step S340, the manner of outputting the DR image index can be to display the DR image index through a display device, or to output the DR image index through a speaker, a printer, or other output devices, which is not specifically limited here.

[0116] There are still many identical or similar contents in each step of the analysis method 300 of DR images and the analysis method 100 of DR images. For specific details, please refer to the relevant descriptions above, which will not be elaborated here.

[0117] Based on the above description, the analysis method 300 of DR images according to the embodiments of the present application extracts at least one of the gray entropy feature, texture feature, noise feature, gradient feature, and divergence feature from the DR image, and obtains a DR image index that can objectively reflect the basic feature information content of the DR image according to the at least one extracted image feature. The DR image index can provide an objective judgment basis for the operator and can guide directions such as in-hospital quality control and dose reduction imaging.

[0118] As Figure 5 shown, the analysis method 500 of DR images includes the following steps:

[0119] In step S510, a digital radiography (DR) image is obtained, where the DR image includes at least one of a DR original image and an image obtained after processing the DR original image;

[0120] In step S520, a DR image index reflecting the basic feature information content of the DR image is determined according to at least one of the DR original image and the image obtained after processing the DR original image;

[0121] It should be noted that in the analysis method 500 of DR images, it is not necessary to extract the image features in the DR image, but the DR image index reflecting the basic feature information content of the DR image can be directly determined according to the image. This processing method is efficient and the operation process is simple.

[0122] In some possible implementation manners, the DR image index reflecting the basic feature information content of the DR image is determined based on at least one of the DR original image and the image after the DR original image is processed, and includes: inputting at least one of the DR original image and the image after the DR original image is processed into a target model, and outputting, through the target model, the DR image index reflecting the basic feature information content of the DR image. Of course, after determining the DR image index by inputting it into the target model, it may also be processed by other processors on the device, or may be obtained after being processed by other processors outside the device, and specific limitations are not made here.

[0123] In step S530, output the DR image index reflecting the basic feature information content of the DR image.

[0124] As Figure 6 shown, the analysis method 600 of the DR image includes the following steps:

[0125] In step S610, obtain a digital radiography (DR) image and the image features corresponding to the DR image, where the DR image includes at least one of a DR original image and an image after the DR original image is processed, and the image features include at least one of a gray entropy feature, a texture feature, a noise feature, a gradient feature, and a divergence feature;

[0126] In step S620, determine the DR image index reflecting the basic feature information content of the DR image according to the DR image and the image features corresponding to the DR image;

[0127] It should be noted that in the analysis method 600 of the DR image, the DR image index needs to be determined based on these two contents, namely the DR image and the image features corresponding to the DR image. The DR image index obtained by this processing method is relatively accurate. Instead of determining the DR image index only through the DR image or the image features corresponding to the DR image.

[0128] In a possible implementation manner, determining the DR image index reflecting the basic feature information content of the DR image according to the DR image and the image features corresponding to the DR image includes:

[0129] Input the DR image and the image features corresponding to the DR image into a target model, and output the DR image index reflecting the basic feature information content of the DR image. Of course, after determining the DR image index by inputting it into the target model, it may also be processed by other processors on the device, or may be obtained after being processed by other processors outside the device, and specific limitations are not made here.

[0130] In step S630, a DR image index reflecting the basic feature information content of the DR image is output.

[0131] It should be noted that there are still many identical or similar contents in the steps of the DR image analysis methods 500 and 600 and the DR image analysis method 100. For specific details, please refer to the relevant descriptions above and will not be elaborated here.

[0132] Refer to Figure 4 In addition, an electronic device 400 is provided in an embodiment of the present application. The electronic device 400 can be used to implement the DR image analysis method 100, the DR image analysis method 300, the DR image analysis method 500, or the DR image analysis method 600 described above. Among them, the electronic device 400 can be a DR device for DR imaging, such as a flat panel detector, an automatic exposure controller, a mobile DR device, a fixed DR device, etc., or other computers or terminal devices, such as mobile phones, computers, palmtop computers, etc., which are not specifically limited here. The electronic device 400 includes a memory 410 and a processor 420. A computer program run by the processor 420 is stored on the memory 410. When the computer program is run by the processor, it executes the steps of the DR image analysis method 100 or the DR image analysis method 300. When used to implement the DR image analysis method 100, the computer program stored on the memory 410 executes the following steps when run by the processor 420: obtaining a digital radiography (DR) image; extracting image features from the DR image, where the image features include at least one of gray entropy features, texture features, noise features, gradient features, and divergence features; inputting at least one of the gray entropy features, texture features, noise features, gradient features, and divergence features into a pre-trained network model, and outputting a DR image index reflecting the basic feature information content of the DR image. Of course, in practical applications, the DR image index can also be displayed through a display device. When used to implement the DR image analysis method 300, the computer program stored on the memory 410 executes the following steps when run by the processor 420: obtaining a digital radiography DR image; extracting image features from the DR image, where the image features include at least one of gray entropy features, texture features, noise features, gradient features, and divergence features; determining a DR image index reflecting the basic feature information content of the DR image according to at least one of the gray entropy features, texture features, noise features, gradient features, and divergence features; outputting a DR image index reflecting the basic feature information content of the DR image. Other specific details of the DR image analysis method 100 and the DR image analysis method 300 can be found above and will not be elaborated here.

[0133] Among them, the processor 420 can be implemented by software, hardware, firmware, or any combination thereof. Circuits, single or multiple application-specific integrated circuits, single or multiple general-purpose integrated circuits, single or multiple microprocessors, single or multiple programmable logic devices, or any combination of the foregoing circuits and / or devices, or other suitable circuits or devices can be used. And the processor 420 can control other components in the electronic device 400 to perform desired functions.

[0134] The memory 410 may include one or more computer program products, and the computer program products may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory and / or cache memory, etc. The non-volatile memory may include, for example, read-only memory, hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 420 may run the program instructions to implement the DR image analysis method and / or other various desired functions in this application. Various application programs and various data may also be stored in the computer-readable storage medium, such as various data used and / or generated by the application programs, etc.

[0135] In addition, according to an embodiment of the present application, a computer storage medium is further provided. Program instructions are stored on the computer storage medium and are used to execute the corresponding steps of the DR image analysis method according to any embodiment of the present application when the program instructions are run by a computer or a processor. In some embodiments, the computer storage medium is a non-volatile computer-readable storage medium, which may include a program storage area and a data storage area. Among them, the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created during the process of executing the program instructions, etc. In addition, the non-volatile computer-readable storage medium may include high-speed random access memory and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid-state storage devices. In some embodiments, the non-volatile computer-readable storage medium may optionally include a memory remotely provided with respect to the processor.

[0136] Exemplarily, the computer storage medium may include, for example, the hard disk of a personal computer, the storage component of a tablet computer, read-only memory (ROM), erasable programmable read-only memory (EPROM), portable compact disc read-only memory (CD-ROM), USB memory, the memory card of a smart phone, or any combination of the above storage media. The computer-readable storage medium may be any combination of one or more computer-readable storage media.

[0137] In a possible implementation manner, refer toFigure 7 The electronic device may be a DR imaging device 700, which includes an X-ray generator 710, a detector 720, a processor 730, and a display 740. Among them, the processor 730 is communicatively connected to the X-ray generator 710, the detector 720, and the display 740; the X-ray generator 710 is configured to generate X-rays, emit X-rays to a target tissue site, and control the X-rays to pass through the target tissue site;

[0138] The detector 720 is configured to receive the X-rays after passing through the target tissue site, and process the X-rays after passing through the target tissue site to obtain a digital radiography (DR) image, where the DR image includes at least one of a DR original image and an image obtained by processing the DR original image;

[0139] The processor 730 is configured to input at least one of the gray entropy feature, texture feature, noise feature, gradient feature, and divergence feature into a target model, and output a DR image index reflecting the basic feature information content of the DR image;

[0140] The display 740 is configured to display the DR image index reflecting the basic feature information content of the DR image.

[0141] In addition, according to an embodiment of the present application, a computer program is also provided. The computer program can be stored on a cloud or local storage medium. When the computer program is run by a computer or a processor, it is used to execute the corresponding steps of the DR image analysis method of the embodiment of the present application.

[0142] In summary, according to the DR image analysis method and device of the present application, a DR image index that can objectively reflect the basic feature information content of the DR image is obtained based on the image features extracted from the DR image, thereby providing an objective judgment basis for the operator.

[0143] Although example embodiments have been described herein with reference to the accompanying drawings, it should be understood that the above example embodiments are merely exemplary and are not intended to limit the scope of the present application thereto. Those of ordinary skill in the art can make various changes and modifications therein without departing from the scope and spirit of the present application. All such changes and modifications are intended to be included within the scope of the present application as claimed in the appended claims.

[0144] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled artisans may use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of this application.

[0145] In several embodiments provided by this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed.

[0146] In the specification provided herein, a large number of specific details are set forth. However, it can be understood that the embodiments of this application can be practiced without these specific details. In some instances, well-known methods, structures, and technologies have not been shown in detail so as not to obscure the understanding of this specification.

[0147] Similarly, it should be understood that, in order to streamline this application and assist in understanding one or more of the various inventive aspects, in the description of the exemplary embodiments of this application, the various features of this application are sometimes grouped together into a single embodiment, figure, or description thereof. However, the methods of this application should not be construed as reflecting the intention that the claimed application requires more features than are expressly recited in each claim. Rather, as reflected by the corresponding claims, the inventive point lies in being able to solve the corresponding technical problems with fewer features than all the features of a single disclosed embodiment. Therefore, the claims following the detailed description are hereby expressly incorporated into the detailed description, where each claim itself serves as a separate embodiment of this application.

[0148] Those skilled in the art can understand that, except for features that are mutually exclusive, any combination can be used for all the features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all the processes or units of any method or device so disclosed. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) can be replaced by an alternative feature that provides the same, equivalent, or similar purpose.

[0149] In addition, those skilled in the art can understand that although some embodiments described herein include certain features included in other embodiments rather than other features, the combination of features of different embodiments means that it is within the scope of this application and forms different embodiments. For example, in the claims, any one of the claimed embodiments can be used in any combination.

[0150] Each component embodiment of this application can be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art should understand that a microprocessor or a digital signal processor (DSP) can be used in practice to implement some or all of the functions of some modules in the article analysis device according to the embodiments of this application. This application can also be implemented as a device program (for example, a computer program and a computer program product) for executing part or all of the methods described herein. Such a program for implementing this application can be stored on a computer-readable medium, or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0151] It should be noted that the above embodiments illustrate rather than limit this application, and those skilled in the art can design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in the claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. This application can be implemented by means of hardware including several different elements and by means of a suitably programmed computer. In the unit claims listing several devices, several of these devices can be embodied by the same item of hardware. The use of the words first, second, and third, etc. does not denote any order. These words can be interpreted as names.

[0152] As described above, it is only the specific implementation manner of this application or the description of the specific implementation manner. The protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by this application, and all of them should be covered by the protection scope of this application. The protection scope of this application shall be subject to the protection scope of the claims.

Claims

1. A method for analyzing DR images, characterized in that, The method includes: Emitting X-rays to a target tissue site and controlling the X-rays to pass through the target tissue site; Receiving the X-rays after passing through the target tissue site; Processing the X-rays after passing through the target tissue site to obtain a digital radiography (DR) image, where the DR image includes at least one of a DR original image and an image obtained after processing the DR original image; Extracting image features from the DR image, where the image features include at least two of a gray entropy feature, a texture feature, a noise feature, a gradient feature, and a divergence feature; where the gray entropy feature is the information content obtained by statistically screening redundant gray level information sources in the DR image; the noise feature includes X-ray quantum noise, and the noise feature reflects the fluctuation degree of X-ray quanta in the DR image; the texture feature represents the variation relationship of different gray values in space on the DR image; the gradient feature is used to characterize the sharpness of different tissue boundaries in the DR image; the divergence feature is used to characterize the transition strength degree and / or trend consistency of different tissue boundaries in the DR image; Inputting at least two of the gray entropy feature, the texture feature, the noise feature, the gradient feature, and the divergence feature into a target model and outputting a DR image index reflecting the basic feature information content of the DR image.

2. The method according to claim 1, wherein Extracting the gray entropy feature from the DR image includes: Obtaining a first probability statistical distribution of each gray level information source in the DR image; Screening the first probability statistical distribution of redundant gray level information sources in the gray level information sources to obtain the first probability statistical distribution of non-redundant gray level information sources; Obtaining a second probability statistical distribution of each non-redundant gray level information source according to the ratio of the first probability statistical distribution of each non-redundant gray level information source to the sum of the first probability statistical distributions of all non-redundant gray level information sources; Calculating entropy according to the second probability statistical distribution of the non-redundant gray level information sources to obtain the gray entropy feature.

3. The method according to claim 1, wherein Extracting the texture feature from the DR image includes: Obtaining a texture feature description matrix according to the gray values of each pixel in the DR image; Extracting at least one two-dimensional component of the texture feature description matrix as at least one of the texture features.

4. The method according to claim 3, wherein The texture feature includes at least one of the following: A first texture feature for statistically analyzing the value distribution of the texture feature description matrix and the overall distribution of texture changes in the DR image; A second texture feature for statistically analyzing the similarity degree of the value distribution in the texture feature description matrix in the parallel and normal directions in the DR image; A third texture feature for statistically analyzing the uniformity of the value distribution in the texture feature description matrix and the gray change distribution in the DR image; A fourth texture feature for statistically analyzing the measure of the local texture change amount of the value distribution in the texture feature description matrix and the DR image.

5. The method according to claim 3, characterized in that The texture feature description matrix describes the variation of gray values at different distances and in different directions in the DR image.

6. The method according to claim 5, wherein It further includes: Perform super-resolution interpolation on the DR image in different directions to obtain sub-pixel gray values; Obtain the texture feature description matrix in different directions according to the sub-pixel gray values.

7. The method according to claim 1, wherein Extract the image noise feature from the DR image, including: Obtain a high-frequency image from the DR image; Extract the effective information in the high-frequency image to obtain a noise distribution image, where the noise distribution image is an image composed of the local root mean square of each pixel point in the high-frequency image; Statistically analyze the noise value distribution in the noise distribution image to obtain the image noise feature.

8. The method according to claim 7, wherein The statistically analyzing the noise value distribution in the noise value distribution image to obtain the image noise feature includes: Determine the noise value interval with the highest noise value distribution probability in the noise distribution image; Use the noise value in the noise value interval with the highest noise value distribution probability as the image noise feature.

9. The method according to claim 7, wherein The obtaining a high-frequency image from the DR image includes: Perform low-frequency filtering on the DR image to obtain a low-frequency image, and obtain the high-frequency image according to the difference between the low-frequency image and the DR image; or, Perform high-frequency filtering on the DR image to obtain the high-frequency image.

10. The method according to claim 1, characterized in that, The target model includes a pre-trained network model, and the training process of the network model includes: Obtain a set of DR images, where the set of DR images includes multiple DR images; Obtain a set of image features according to the set of DR images, where the set of image features includes multiple image features extracted from multiple DR images; Obtain a set of DR diagnostic images according to the set of DR images, where the set of DR diagnostic images includes multiple DR diagnostic images obtained by performing image processing on multiple DR images; Obtain a set of scores according to the set of DR diagnostic images, where the set of scores includes the scores obtained by evaluating multiple DR diagnostic images; Use the set of image features and the set of scores as a training sample set to train the network model to obtain a trained network model.

11. The method according to claim 1, characterized in that, The extracting the gradient feature from the DR image includes: Determine the regional distribution of different tissues in the DR image; Obtain the sharpness of the boundary of the regional distribution of different tissues as the gradient feature.

12. The method according to claim 1, characterized in that, The extracting the divergence feature from the DR image includes: Determine the regional distribution of different tissues in the DR image; Obtain the transition strength and / or trend consistency of the boundary of the regional distribution of different tissues as the divergence feature.

13. A method for analyzing DR images, characterized in that, The method includes: Obtain a digital radiography (DR) image, where the DR image includes at least one of a DR original image and an image obtained by processing the DR original image; Extract image features from at least one of the DR original image and the image after the DR original image is processed. The image features include at least two of gray entropy features, texture features, noise features, gradient features, and divergence features. Among them, the gray entropy feature is the information content obtained by statistically screening out redundant gray-level information sources in the DR image. The noise feature includes X-ray quantum noise, and the noise feature reflects the fluctuation degree of X-ray quanta in the DR image. The texture feature represents the spatial variation relationship of different gray values on the DR image. The gradient feature is used to characterize the sharpness of different tissue boundaries in the DR image. The divergence feature is used to characterize the transition strength degree and / or trend consistency of different tissue boundaries in the DR image. Determine a DR image index that reflects the basic feature information content of the DR image according to at least one of the gray entropy feature, texture feature, noise feature, gradient feature, and divergence feature. Output a DR image index that reflects the basic feature information content of the DR image.

14. A method for analyzing DR images, characterized in that, The method includes: Obtain a digital radiography (DR) image and the image features corresponding to the DR image. Among them, the DR image includes at least one of the DR original image and the image after the DR original image is processed. The image features include at least two of gray entropy features, texture features, noise features, gradient features, and divergence features. Among them, the gray entropy feature is the information content obtained by statistically screening out redundant gray-level information sources in the DR image. The noise feature includes X-ray quantum noise, and the noise feature reflects the fluctuation degree of X-ray quanta in the DR image. The texture feature represents the spatial variation relationship of different gray values on the DR image. The gradient feature is used to characterize the sharpness of different tissue boundaries in the DR image. The divergence feature is used to characterize the transition strength degree and / or trend consistency of different tissue boundaries in the DR image. Determine a DR image index that reflects the basic feature information content of the DR image according to the DR image and the image features corresponding to the DR image. Output a DR image index that reflects the basic feature information content of the DR image.

15. The method according to claim 14, wherein The determining a DR image index that reflects the basic feature information content of the DR image according to the DR image and the image features corresponding to the DR image includes: Input the DR image and the image features corresponding to the DR image into a target model, and output a DR image index that reflects the basic feature information content of the DR image.

16. A DR imaging device, characterized in that, The DR imaging device includes: an X-ray generator, a detector, a processor, and a display; The X-ray generator is used to generate X-rays, emit X-rays to a target tissue site, and control the X-rays to pass through the target tissue site. The detector is used to receive the X-rays after passing through the target tissue site, and process the X-rays after passing through the target tissue site to obtain a digital radiography (DR) image, where the DR image includes at least one of a DR original image and an image obtained by processing the DR original image; The processor is used to extract image features from the DR image, and the image features include at least two of gray entropy features, texture features, noise features, gradient features, and divergence features; the processor is further used to input at least one of the gray entropy features, texture features, noise features, gradient features, and divergence features into a target model, and output a DR image index reflecting the basic feature information content of the DR image; where the gray entropy feature is the information content obtained by statistically screening out redundant gray level information sources in the DR image; the noise feature includes X-ray quantum noise, and the noise feature reflects the fluctuation degree of X-ray quanta in the DR image; the texture feature represents the change relationship of different gray values on the DR image in space; the gradient feature is used to characterize the sharpness of different tissue boundaries in the DR image; the divergence feature is used to characterize the transition strength degree and / or trend consistency of different tissue boundaries in the DR image; The display is used to display the DR image index reflecting the basic feature information content of the DR image.

17. An electronic device, characterized in that, The electronic device includes a memory and a processor, and a computer program run by the processor is stored on the memory, and the computer program executes the steps of the analysis method of the DR image according to any one of claims 1-16 when being run by the processor.

Images