Information processing method, device and storage medium for ultrasound hyperechoic halo of breast cancer

Abstract

The present disclosure relates to an information processing method, device and storage medium for ultrasound hyperechoic halo of breast cancer, wherein the method comprises determining a lesion area based on a breast ultrasound image; obtaining a grayscale change curve from the lesion area to the peripheral direction of the breast ultrasound image; and determining whether there is a hyperechoic halo around the lesion area based on the grayscale change curve. The embodiments of the present disclosure can realize the detection of hyperechoic halo around the lesion area.

Description

Technical Field

[0001] The present disclosure relates to the technical field of image processing, and in particular to an information processing method, device and storage medium for ultrasound hyperechoic halo of breast cancer. Background Art

[0002] At present, breast cancer is a common malignant tumor. Early and clear diagnosis in clinical practice is crucial for later treatment and prognosis, but misdiagnosis often leads to overtreatment. How to diagnose and identify breast cancer more accurately has always been a difficult problem for imaging doctors.

[0003] Artificial intelligence technology has become an important method for clinical auxiliary diagnosis. However, the stability problem of artificial intelligence network has always existed. Relying solely on artificial intelligence technology to perform breast cancer diagnosis may result in missed diagnosis and misdiagnosis. Therefore, there is an urgent need for a solution that can further analyze and confirm the ultrasound images of detected breast cancer lesions to reduce misdiagnosis. Summary of the invention

[0004] The present disclosure proposes an information processing method, device and storage medium for ultrasound hyperechoic halo of breast cancer. The present disclosure can determine whether the detected lesion area meets the hyperechoic halo condition, thereby helping to reduce the probability of misdiagnosis.

[0005] According to one aspect of the present disclosure, a method for processing information of ultrasound hyperechoic halo of breast cancer is provided, comprising:

[0006] Determine the lesion area based on breast ultrasound images;

[0007] Acquire a grayscale change curve from the lesion area to the periphery of the breast ultrasound image;

[0008] Based on the grayscale change curve, it is determined whether there is a high echo halo around the lesion area.

[0009] In some possible implementations, the grayscale change curve obtained from the lesion area toward the periphery of the breast ultrasound image includes:

[0010] determining the center of the lesion area;

[0011] Taking the center as the starting point, grayscale sampling is performed toward the periphery of the breast ultrasound image to obtain at least one set of grayscale sequences;

[0012] Based on the at least one set of grayscale sequences, the grayscale variation curve is determined.

[0013] In some possible implementations, grayscale sampling is performed in a peripheral direction of the breast ultrasound image starting from the center to obtain at least one set of grayscale sequences, including:

[0014] Expanding a concentric frame from the center toward the periphery of the breast ultrasound image;

[0015] Determine at least one ray starting from the center and heading toward the periphery;

[0016] The intersection point of the ray and the concentric frame is determined as a sampling point, and grayscale sampling is performed to obtain a grayscale sequence corresponding to the ray.

[0017] In some possible implementations, determining the grayscale change curve based on the at least one set of grayscale sequences includes:

[0018] Performing filtering processing on the grayscale sequence;

[0019] Curve fitting is performed on the grayscale sequence after filtering to obtain the grayscale change curve.

[0020] In some possible implementations, determining whether a high echo halo exists around the lesion area based on the grayscale change curve includes:

[0021] In a case where the grayscale change curve has a preset trend, determining that the grayscale change curve has the high echo halo;

[0022] Among them, the preset trend includes rising from the first grayscale interval to the second grayscale interval, and falling from the second grayscale interval to the third grayscale interval, the grayscale value of the second grayscale interval is greater than the grayscale value of the third grayscale interval, and the grayscale value of the third grayscale interval is greater than the grayscale value of the first grayscale interval.

[0023] In some possible implementations, when a set of grayscale change curves are obtained, determining whether there is a high echo halo around the lesion area based on the grayscale change curves further includes:

[0024] Determining the similarity between the grayscale change curve and a standard high echo halo curve;

[0025] When the similarity indicates that the grayscale change curve and the standard hyperechoic halo curve are similar curves, it is determined that a hyperechoic halo exists around the lesion area.

[0026] In some possible implementations, when multiple groups of grayscale change curves are obtained, determining whether there is a high echo halo around the lesion area based on the grayscale change curves further includes at least one of the following methods:

[0027] Performing weighted sum processing on the similarity between the grayscale change curve and the standard hyperechoic halo curve, and determining that a hyperechoic halo exists around the lesion area when the obtained weighted sum indicates that the grayscale change curve and the standard hyperechoic halo curve are similar curves;

[0028] The grayscale change curve and the classification clustering process are performed, and when the distance between two cluster centers obtained by clustering indicates that the grayscale change curve and the standard hyperechoic halo curve are similar curves, it is determined that a hyperechoic halo exists around the lesion area.

[0029] In some possible implementations, the method further includes:

[0030] In the case where the grayscale change curve determines that a high echo halo exists around the lesion area, the lesion area is corrected using the grayscale change curve.

[0031] According to a second aspect of the present disclosure, there is provided an information processing device for ultrasound hyperechoic halo of breast cancer, comprising:

[0032] processor;

[0033] a memory for storing processor-executable instructions;

[0034] The processor is configured to call the instructions stored in the memory to execute the information processing method of the ultrasound hyperechoic halo of breast cancer described in the embodiment of the present disclosure.

[0035] According to a third aspect of the present disclosure, a computer-readable storage medium is provided, on which computer program instructions are stored, characterized in that when the computer program instructions are executed by a processor, the information processing method for ultrasound hyperechoic halo of breast cancer described in the embodiment of the present disclosure is implemented.

[0036] In the embodiment of the present disclosure, the lesion area in the breast ultrasound image can be obtained, and the grayscale change curve of the lesion area toward the periphery of the ultrasound image can be obtained based on the lesion area, and then the existence of a high echo halo in the ultrasound image can be determined according to the grayscale change curve. The existence of the high echo halo can be used as a reference for the accuracy of lesion detection, and the solution of the embodiment of the present disclosure can reduce the misdiagnosis of lesions.

[0037] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.

[0038] Further features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The drawings herein are incorporated into the specification and constitute a part of the specification. These drawings illustrate embodiments consistent with the present disclosure and are used to illustrate the technical solutions of the present disclosure together with the specification.

[0040] Figure 1 A flowchart showing a method for processing information of ultrasound hyperechoic halo of breast cancer according to an embodiment of the present disclosure;

[0041] Figure 2 A schematic diagram of calculating the gray-level co-occurrence matrix of an original image using a sliding window in an embodiment of the present disclosure is shown. In the figure, the matrix on the left is before sliding, and the sliding window of the matrix on the right slides one pixel to the right;

[0042] Figure 3 A flow chart of determining a grayscale change curve according to an embodiment of the present disclosure is shown;

[0043] Figure 4 A schematic diagram showing a concentric frame implemented according to the present disclosure;

[0044] Figure 5 A block diagram showing an information processing device for ultrasound hyperechoic halo of breast cancer according to an embodiment of the present disclosure;

[0045] Figure 6 A block diagram of an electronic device 800 according to an embodiment of the present disclosure is shown;

[0046] Figure 7 A block diagram of another electronic device 1900 according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0047] Various exemplary embodiments, features and aspects of the present disclosure will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise specified.

[0048] The word “exemplary” is used exclusively herein to mean “serving as an example, example, or illustration.” Any embodiment described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.

[0049] The term "and / or" herein is only a description of the association relationship of the associated objects, indicating that there may be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the term "at least one" herein represents any combination of at least two of any one or more of a plurality of. For example, including at least one of A, B, and C can represent including any one or more elements selected from the set consisting of A, B, and C.

[0050] In addition, in order to better illustrate the present disclosure, numerous specific details are given in the following specific embodiments. It should be understood by those skilled in the art that the present disclosure can also be implemented without certain specific details. In some examples, methods, means, components and circuits well known to those skilled in the art are not described in detail in order to highlight the subject matter of the present disclosure.

[0051] The execution subject of the information processing method for ultrasound hyperechoic halo of breast cancer may be an image processing device or an information processing method. For example, the method may be executed by a terminal device or a server or other processing device, wherein the terminal device may be a user equipment (UE), a mobile device, a user terminal, a terminal, a cellular phone, a cordless phone, a personal digital assistant (PDA), a handheld device, a computing device, a vehicle-mounted device, a wearable device, etc. In some possible implementations, the image processing method may be implemented by a processor calling a computer-readable instruction stored in a memory.

[0052] It can be understood that the above-mentioned various method embodiments mentioned in the present disclosure can be combined with each other to form a combined embodiment without violating the principle logic. Due to space limitations, the present disclosure will not go into details.

[0053] In addition, the present disclosure also provides an information processing device, an electronic device, a computer-readable storage medium, and a program for the ultrasound high echo halo of breast cancer, all of which can be used to implement any information processing method for the ultrasound high echo halo of breast cancer provided by the present disclosure. The corresponding technical solutions and descriptions are referred to the corresponding records in the method part and will not be repeated here.

[0054] Figure 1 A flowchart showing a method for processing information of ultrasound hyperechoic halo of breast cancer according to an embodiment of the present disclosure is shown. Figure 1 As shown, the method includes:

[0055] S10: determine the lesion area based on breast ultrasound images;

[0056] In some possible implementations, the lesion area in the breast ultrasound image may be determined by manual annotation by a doctor or by an algorithm-assisted method, and the lesion area may be the area where the breast cancer lesion is located.

[0057] The lesion area in the embodiment of the present disclosure may be circular or square, and the present disclosure does not make any specific limitation on this.

[0058] S20: Acquire a grayscale change curve from the lesion area to the periphery of the breast ultrasound image;

[0059] In some possible implementations, when the lesion area is obtained, the grayscale change of the lesion area and its surroundings may be analyzed to obtain a grayscale change curve.

[0060] S30: Determine whether there is a high echo halo around the lesion area based on the grayscale change curve.

[0061] In some possible implementations, when the grayscale change curve is obtained, it can be further determined whether the grayscale change curve indicates the presence of high echoes (high echo halo) near the lesion area in the ultrasound image. In the case of the presence of the high echo halo, it indicates that the lesion detection is correct and the lesion area corresponds to breast cancer. In the case of the absence of the high echo halo, it indicates that the lesion detection is inaccurate and the lesion area is not breast cancer.

[0062] The embodiments of the present disclosure are described in detail below in conjunction with the accompanying drawings.

[0063] First, a breast ultrasound image may be acquired, which may be connected to an ultrasound instrument to receive the ultrasound image, or the ultrasound image may be read from a database, a server or other electronic device, which is not specifically limited in the present disclosure.

[0064] When an ultrasound image is obtained, lesion detection can be performed on the ultrasound image to detect the lesion area of ​​breast cancer in the ultrasound image. The lesion area can be determined by receiving frame selection information input by the user (doctor). Alternatively, the lesion area can be detected by artificial intelligence.

[0065] Specifically, feature extraction of the ultrasound image may be performed first to obtain image features, and a classifier may be used to perform classification processing to obtain the lesion area and the normal area in the ultrasound image.

[0066] The present invention introduces a semantically interpretable ultrasound image processing method for breast cancer to perform image feature extraction. Through the interpretation of image features and deep learning, it satisfies the interpretation of output results in the clinical practical process of auxiliary identification and enhances the comprehensibility of doctors. The basic idea is to extract the image grayscale co-occurrence matrix, texture features, and morphological features, and combine them with the feature map generated by the interpretable method of deep learning. The low-rank regression model is used to map the abstract features that cannot be interpreted by deep learning to an interpretable semantic feature space. In the interpretable semantic feature space, interpretable ultrasound-assisted identification of breast cancer is performed according to the importance of each feature. The specific content includes:

[0067] The first step is to extract interpretable features from the ultrasound image of the breast, which includes: extracting the gray level co-occurrence matrix and texture features of the ultrasound image, extracting multiple morphological features, the morphological features are the same or similar to the features described by BI-RADS, and the morphological features characterize the shape, orientation, boundary and internal texture features of the tumor;

[0068] Combination Figure 2 , the specific steps of the first step are as follows:

[0069] 1.1. Extracting a grayscale co-occurrence matrix from the ultrasound image, specifically: describing the texture by analyzing the characteristics of repeated alternating changes in grayscale distribution in spatial position, including, for an image represented in matrix form, determining, based on a given direction θ and a step size d, the value of the position (i, j) in the grayscale co-occurrence matrix, the frequency of occurrence of a pixel pair with a pixel value of i and a pixel value of j along the θ direction and a distance of d, so as to calculate the comprehensive information of the ultrasound image in terms of direction, interval, change amplitude and speed, wherein θ is 0°, 45°, 90° or 135°.

[0070] 1.2. A grayscale gradient co-occurrence matrix of pixel grayscale and gradient is extracted based on the ultrasound image. The grayscale gradient co-occurrence matrix of the image is the joint distribution of image grayscale pixels and image gradient magnitudes. The image gradient is obtained through various differential operators and is used to detect the grayscale jump part in the ultrasound image. The fusion of the two can help discover information about image texture arrangement and pixel changes.

[0071] 1.3. Extracting texture features from the ultrasound image of the breast to quantitatively describe the information in the ultrasound image, including extracting five features from the ultrasound image, namely, grayscale standard deviation, energy, grayscale entropy value, grayscale mean, and correlation.

[0072] 1.4. Extract morphological features from breast ultrasound images to describe the shape, orientation, boundary and internal texture features of the tumor, including the roundness, aspect ratio, skin growth angle, lobation, edge roughness, edge puncture, internal calcification number, border blur, etc.

[0073] The second step is to extract abstract features of the ultrasound image, which includes: using a deep neural network and a regression algorithm to learn the extracted features corresponding to each convolution kernel, forming a feature map, using the global mean of the feature map to represent the feature map, and using the feature map mean vector formed by the last convolution layer as the abstract feature vector.

[0074] 2.1. Extracting abstract features of ultrasound images using a deep neural network, including but not limited to extracting abstract features of the ultrasound images using CAM and / or Grad-CAM methods.

[0075] In the third step, for the explainable features and the abstract features, the weight coefficient matrix W of different features in the mapping process is calculated through the semantic low-rank regression algorithm, and the concrete features with weights higher than the preset values ​​are selected to explain the information of interest in the deep learning discriminant model. The information is used to improve the credibility of the feature test model when assisting in the discrimination of breast cancer ultrasound images.

[0076] 3.1. Extract the abstract features through semantic low-rank regression algorithm to find out the features that play an important role in assisting breast cancer ultrasound image discrimination, so as to improve the credibility of the abstract features. Here, the low-rank regression algorithm defines the abstract features The input data of the abstract feature, x t is the deep learning abstract feature corresponding to the tth (1≤t≤N) sample, d is the abstract feature dimension, let is the interpretable feature of the output of the semantic low-rank regression algorithm, c is the dimension of the interpretable feature, y t is the corresponding interpretable feature of the t-th sample, and the mathematical description of the regression model of the semantic low-rank regression algorithm is:

[0077]

[0078] Among them, the first term represents the l2 norm of the mapping loss. The l2 norm represents the difference between the abstract features and the concrete features after the mapping matrix. The greater the difference, the greater the loss, and the worse the performance of the mapping matrix. θ is the regularization penalty coefficient, which is used to regularize the low-rank penalty term RankW. Since the abstract features of deep learning of different samples are generally correlated, according to this prior knowledge, the rank of the W matrix should be low. According to the rank approximation theory, the trace of the matrix needs to be used to approximate the rank of the matrix. TrW is the trace of the matrix, which can be defined as:

[0079]

[0080] σ i is the i-th singular value of W, W T is the transpose of W.

[0081] The weight coefficient matrix W is the weight in the mapping process, specifically the total weight obtained by the comprehensive calculation of the concrete features in the N sample mapping operations, where the influence of the concrete features on the discriminator on the sample set X is:

[0082]

[0083] The calculated W′ selects concrete features with higher weights to explain the highly concerned information in the deep learning discriminant model, which is used to improve the trustworthiness of the feature test model.

[0084] The embodiment of the present application further provides an electronic device, including a memory and a processor, wherein the memory stores an executable program, and the processor executes the executable program to implement the following steps:

[0085] The first step is to extract interpretable features from the ultrasound image of the breast, which includes: extracting the gray level co-occurrence matrix and texture features of the ultrasound image, and extracting multiple morphological features, wherein the morphological features are the same as or similar to the features described by BI-RADS, and the morphological features characterize the shape, orientation, boundary and internal texture features of the tumor;

[0086] The second step is to extract abstract features of the ultrasound image, which includes: using a deep neural network and a regression algorithm to learn the features corresponding to each convolution kernel to form a feature map, using the global mean of the feature map to represent the feature map, and using the mean vector of the feature map formed by the last convolution layer as the abstract feature vector;

[0087] In the third step, for the explainable features and the abstract features, the weight coefficient matrix W of different features in the mapping process is calculated through the semantic low-rank regression algorithm, and the concrete features with weights higher than the preset values ​​are selected to explain the information of interest in the deep learning discriminant model. The information is used to improve the credibility of the feature test model when assisting in the discrimination of breast cancer ultrasound images.

[0088] Through the above configuration, the image features of the ultrasound image can be obtained, and the obtained feature information is input into the classifier (at least one convolution layer) to obtain the location of the lesion area. The present disclosure can utilize the grayscale co-occurrence matrix, texture features, morphological features of the ultrasound image of breast cancer, and use the regression algorithm to extract the features of the deep learning method to form a feature map, and use the global mean of the feature map to represent the feature map. Finally, the importance of different features in image discrimination is calculated using a low-rank regression algorithm to achieve the interpretability of the auxiliary discrimination of breast cancer ultrasound images, and have semantic meanings consistent with clinical features. In view of the problem that the output results in the existing ultrasound breast-assisted identification technology do not have medical interpretability, this method solves the problem that the output results of the deep learning method are unexplainable when performing ultrasound-assisted discrimination of breast cancer, and improves clinical credibility. While using the deep learning method to improve the discrimination effect, the features are selected in a targeted manner, and the results have a semantic interpretation discrimination process, a bridge between doctor knowledge and computer algorithm knowledge is established, and doctors are helped to understand and accept the classification results. The decision-making process and output results of deep learning methods should be consistent with the concept of evidence-based medicine, and the application of deep learning algorithms in breast tumor diagnosis should be promoted.

[0089] When the lesion area is obtained, the grayscale situation around the lesion area can be analyzed to obtain a grayscale change curve.

[0090] Figure 3 A flow chart of determining a grayscale change curve according to an embodiment of the present disclosure is shown, wherein the grayscale change curve obtained from the lesion area to the peripheral direction of the breast ultrasound image includes:

[0091] S21: determining the center of the lesion area;

[0092] S22: starting from the center, performing grayscale sampling in the peripheral direction of the breast ultrasound image to obtain at least one set of grayscale sequences;

[0093] S23: Determine the grayscale change curve based on the at least one set of grayscale sequences.

[0094] In some possible implementations, when the lesion region is a circular frame, the center of the lesion region is determined as the center, and when the lesion region is a square frame, the intersection of the diagonals of the lesion region is determined as the center. When the lesion region is other shapes, the geometric center of the lesion region is determined as the center.

[0095] When the center is determined, the center of the lesion area can be determined as the starting point of the grayscale change, and grayscale sampling can be performed in the peripheral direction of the breast ultrasound image according to the preset direction. At least one set of grayscale sampling values ​​can be obtained in each direction to form a grayscale sequence.

[0096] Specifically, in the disclosed embodiment, grayscale sampling is performed in the peripheral direction of the breast ultrasound image with the center as the starting point to obtain at least one group of grayscale sequences, including: extending a concentric frame in the peripheral direction of the breast ultrasound image according to the center; determining at least one ray with the center as the starting point and in the peripheral direction; determining the intersection of the ray and the concentric frame as a sampling point, performing grayscale sampling, and obtaining a grayscale sequence corresponding to the ray.

[0097] Combination Figure 4 , Figure 4 A schematic diagram of a concentric frame is shown according to the present disclosure, wherein the concentric frame is Figure 4The characterized ultrasound image is described as an example, and the concentric frames of each embodiment in this article are mainly described in detail. Therefore, it can be assumed that the concentric frame centered at point A in the figure (illustrated by concentric circles in the figure) identifies the lesion area and the surrounding area (such as breast cancer lesions, breast tumor areas, etc.) identified by the breast ultrasound image, and the center point A of the lesion area is used as the starting point of the grayscale change curve, and the concentric frame is continuously expanded outward from the periphery of the lesion area until it reaches the boundary of the breast area in the ultrasound image. Among them, the concentric frame can be circular or square, and can be set according to needs. The spacing distance between each concentric frame can be a pre-set distance value, which can be a value less than half the length of the lesion area.

[0098] When the concentric frame is determined, with the center as the starting point, at least one ray is made in the external direction of the ultrasound image. For example, in the embodiment of the present disclosure, the ray may be a horizontal ray pointing to the right, or may be a ray located to the right of the vertical line passing through point A. In the embodiment of the present disclosure, the obtained ray may be one or more. When there is one ray, the camera preferably uses a horizontal ray pointing to the right. When there are multiple rays, the angle between two adjacent rays can be determined according to the quotient between 180 degrees and the number of rays, so that the rays are evenly distributed in the area to the right of the center of the lesion.

[0099] When at least one ray is obtained, the grayscale value corresponding to the intersection point between each ray and the concentric frame can be used to determine the grayscale sequence corresponding to the ray (such as Figure 4 Based on the above, the embodiment of the present disclosure can obtain at least one set of grayscale sequences.

[0100] In some possible implementations, when a grayscale sequence is obtained, a curve fitting may be performed on the grayscale sequence to obtain a grayscale change curve. In the disclosed embodiment, determining the grayscale change curve based on the at least one set of grayscale sequences includes: performing filtering on the grayscale sequence; performing curve fitting on the filtered grayscale sequence to obtain the grayscale change curve.

[0101] The disclosed embodiment can perform filtering processing on the grayscale sequence in a moving mean filtering manner to reduce the influence of noise, and then perform least squares fitting processing on the filtered grayscale sequence to obtain a grayscale change curve.

[0102] Since there may be high echoes around breast cancer lesions, the embodiment of the present disclosure can determine whether there is a high echo based on the grayscale change curve. In the embodiment of the present disclosure, it can be determined that the grayscale change curve has the high echo halo when there is a preset trend in the grayscale change curve; wherein the preset trend includes rising from the first grayscale interval to the second grayscale interval, and falling from the second grayscale interval to the third grayscale interval, the grayscale value of the second grayscale interval is greater than the grayscale value of the third grayscale interval, and the grayscale value of the third grayscale interval is greater than the grayscale value of the first grayscale interval. Wherein, the second grayscale interval is a high echo area relative to the first grayscale interval and the second grayscale interval.

[0103] In addition, in the embodiment of the present disclosure, a standard hyperechoic curve can also be obtained, and according to the similarity between the grayscale change curve and the standard hyperechoic curve, it is determined whether there is a hyperechoic halo, wherein the standard hyperechoic curve satisfies the above-mentioned preset trend. Specifically, the embodiment of the present disclosure can first obtain a determined standard hyperechoic curve, and then perform a curve similarity judgment. Among them, the method for determining the standard hyperechoic halo curve includes: obtaining multiple groups of ultrasound images determined to have breast cancer lesions; for each group of ultrasound images, obtaining a grayscale change curve in the horizontal right direction from the breast cancer lesion area to the periphery of the ultrasound image; performing mean processing on the grayscale change curve to obtain the standard hyperechoic halo curve.

[0104] In the embodiment of the present disclosure, since the grayscale changes of the lesion area and the surrounding area can be fully reflected in the horizontal right direction (horizontal direction toward the areola), the embodiment of the present disclosure can collect a large number of (more than 10,000) grayscale change curves in this direction of breast cancer images, perform curve fitting, and obtain a standard high echo halo curve to ensure the accuracy of the curve change. Alternatively, in the embodiment of the present disclosure, grayscale change curves in multiple directions can also be collected, and fitting can be performed to obtain a standard grayscale curve, which is not specifically limited in the present disclosure.

[0105] In the embodiment of the present disclosure, when a set of grayscale change curves are obtained, determining whether there is a high echo halo around the lesion area based on the grayscale change curve includes: determining the similarity between the grayscale change curve and a standard high echo halo curve; when the similarity indicates that the grayscale change curve and the standard high echo halo curve are similar curves, determining that there is a high echo halo around the lesion area.

[0106] As described in the above embodiment, the obtained ray may be one, and the corresponding grayscale change curve may be a group. At this time, the grayscale change curve may be compared with the standard hyperechoic halo curve corresponding to breast cancer obtained in advance to determine the similarity between the two. The similarity between the grayscale change curve and the standard hyperechoic halo curve may be determined by the distance between the two curves, wherein the distance may be the Euclidean distance or the discrete Fréchet distance, which is not specifically limited in the present disclosure.

[0107] When the similarity between the grayscale change curve and the standard hyperechoic halo curve is obtained, it can be determined whether the grayscale change curve has a hyperechoic halo based on the similarity. Wherein, when the similarity is greater than a similarity threshold, it is determined that the two are similar, that is, there is a hyperechoic halo in the grayscale change curve, otherwise there is no hyperechoic halo. The similarity threshold can be a value greater than or equal to 0.6 and less than 1.

[0108] In some other embodiments of the present disclosure, when multiple groups of grayscale change curves are obtained, determining whether there is a high echo halo around the lesion area based on the grayscale change curves includes at least one of the following methods:

[0109] A) performing weighted sum processing on the similarity between the grayscale change curve and the standard hyperechoic halo curve, and determining that a hyperechoic halo exists around the lesion area when the obtained weighted sum indicates that the grayscale change curve and the standard hyperechoic halo curve are similar curves;

[0110] In the embodiment of the present disclosure, when multiple rays are obtained, multiple groups of grayscale sequences and grayscale change curves are also obtained accordingly. Since the direction of the rays corresponding to each grayscale change curve is different, the grayscale values ​​in the obtained grayscale sequence are also different. In order to improve the accuracy of the grayscale change curve, the embodiment of the present disclosure integrates the grayscale change rules of each ray, wherein a corresponding weight can be assigned to each ray, and the weighted sum of the grayscale change curves of each ray is used to obtain the final grayscale change curve.

[0111] The sum of the weights of the rays is 1, and the weight of the ray pointing horizontally to the right is the largest, the weight of the ray in the vertical direction is the smallest, and the weight of the ray close to the horizontal direction is greater than the weight of the ray close to the vertical direction.

[0112] A weighted sum grayscale curve is obtained by adding the product of each grayscale curve and the corresponding weight, and the similarity between the weighted sum grayscale curve and the standard hyperechoic halo curve is used to determine whether a hyperechoic halo exists.

[0113] B) performing classification and clustering processing on the grayscale change curve and the method, and determining that a hyperechoic halo exists around the lesion area when the distance between two cluster centers obtained by clustering indicates that the grayscale change curve and a standard hyperechoic halo curve are similar curves.

[0114] In addition, in the embodiment of the present disclosure, when each grayscale change curve is obtained, curve feature extraction and clustering processing can be performed on the grayscale change curve and the standard high echo grayscale curve, wherein feature extraction can be performed on each grayscale change curve to obtain curve features, wherein the curve features can include the grayscale sequence on the grayscale change curve and the slope between adjacent sampling points. The curve features obtained for each grayscale change curve and the standard high echo halo curve can be expressed as a feature vector. Then, clustering processing of the feature vector can be performed. The embodiment of the present disclosure can cluster the feature vectors into two categories by using mean clustering, and obtain the cluster centers of the two categories.

[0115] If there is a hyperechoic halo in the grayscale change curve, the distance between the two types of cluster centers will be very close. In the embodiment of the present disclosure, when the distance between the cluster centers is greater than the distance threshold, it can be determined that the grayscale change curve is similar to the standard hyperechoic halo curve, and a hyperechoic halo exists. The distance threshold of the embodiment of the present disclosure is a value less than 0.3 cm, but it is not specifically limited to the present disclosure.

[0116] In the disclosed embodiment, when it is determined that there is a high echo halo around the lesion area, it is determined that the lesion corresponds to breast cancer. At this time, multiple grayscale change curves can be used to correct the lesion area.

[0117] Specifically, in the disclosed embodiment, all pixel points where the grayscale curve changes from the first grayscale interval to the second grayscale interval can be used as boundary points of the lesion area, and the corrected lesion area can be determined using the area connected by the updated boundary points.

[0118] Based on the above configuration, the embodiment of the present disclosure can obtain the lesion area in the breast ultrasound image, and take the lesion area as a reference to obtain the grayscale change curve of the lesion area toward the periphery of the ultrasound image, and then determine whether there is a high echo halo in the ultrasound image according to the grayscale change curve. Among them, the existence of the high echo halo can be used as a reference for the accuracy of lesion detection, and the solution of the embodiment of the present disclosure can reduce the misdiagnosis of lesions.

[0119] Those skilled in the art will appreciate that, in the above method of specific implementation, the order in which the steps are written does not imply a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of the steps should be determined by their functions and possible internal logic.

[0120] Figure 5A block diagram of an information processing device for ultrasound hyperechoic halo of breast cancer according to an embodiment of the present disclosure is shown. Figure 5 As shown, the information processing device for breast cancer ultrasound high echo halo includes:

[0121] A first determination module is used to determine the lesion area based on the breast ultrasound image;

[0122] An acquisition module, used for acquiring a grayscale change curve from the lesion area to the periphery of the breast ultrasound image;

[0123] The second determination module is used to determine whether there is a high echo halo around the lesion area based on the grayscale change curve.

[0124] In some possible implementations, the acquisition module is further used to: determine the center of the lesion area; take the center as a starting point, perform grayscale sampling in the peripheral direction of the breast ultrasound image, and obtain at least one set of grayscale sequences;

[0125] Based on the at least one set of grayscale sequences, the grayscale variation curve is determined.

[0126] In some possible implementations, the acquisition module is also used to expand a concentric frame from the center toward the periphery of the breast ultrasound image; determine at least one ray starting from the center and toward the periphery; determine the intersection of the ray and the concentric frame as a sampling point, perform grayscale sampling, and obtain a grayscale sequence corresponding to the ray.

[0127] In some possible implementations, determining the grayscale change curve based on the at least one set of grayscale sequences includes: performing filtering processing on the grayscale sequence; and performing curve fitting on the filtered grayscale sequence to obtain the grayscale change curve.

[0128] In some possible implementations, the second determining module is further configured to:

[0129] In a case where the grayscale change curve has a preset trend, determining that the grayscale change curve has the high echo halo;

[0130] Among them, the preset trend includes rising from the first grayscale interval to the second grayscale interval, and falling from the second grayscale interval to the third grayscale interval, the grayscale value of the second grayscale interval is greater than the grayscale value of the third grayscale interval, and the grayscale value of the third grayscale interval is greater than the grayscale value of the first grayscale interval.

[0131] In some possible implementations, the second determination module is further configured to, when a set of grayscale change curves are obtained, determine whether there is a high echo halo around the lesion area based on the grayscale change curves, and further include:

[0132] Determining the similarity between the grayscale change curve and a standard high echo halo curve;

[0133] When the similarity indicates that the grayscale change curve and the standard hyperechoic halo curve are similar curves, it is determined that a hyperechoic halo exists around the lesion area.

[0134] In some possible implementations, the second determination module is further configured to, when a plurality of groups of grayscale change curves are obtained, determine whether there is a high echo halo around the lesion area based on the grayscale change curves, and further include at least one of the following methods:

[0135] Performing weighted sum processing on the similarity between the grayscale change curve and the standard hyperechoic halo curve, and determining that a hyperechoic halo exists around the lesion area when the obtained weighted sum indicates that the grayscale change curve and the standard hyperechoic halo curve are similar curves;

[0136] The grayscale change curve and the classification clustering process are performed, and when the distance between two cluster centers obtained by clustering indicates that the grayscale change curve and the standard hyperechoic halo curve are similar curves, it is determined that a hyperechoic halo exists around the lesion area.

[0137] In some possible implementations, the device further includes:

[0138] A correction module is used to correct the lesion area using the grayscale change curve when the grayscale change curve determines that there is a high echo halo around the lesion area.

[0139] In some embodiments, the functions or modules included in the device provided by the embodiments of the present disclosure can be used to execute the method described in the above method embodiments. The specific implementation can refer to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0140] The embodiment of the present disclosure also provides a computer-readable storage medium, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the above method is implemented. The computer-readable storage medium may be a non-volatile computer-readable storage medium.

[0141] The embodiment of the present disclosure further proposes an electronic device, comprising: a processor; and a memory for storing instructions executable by the processor; wherein the processor is configured as the above method.

[0142] The electronic device may be provided as a terminal, a server, or a device in other forms.

[0143] Figure 6A block diagram of an electronic device 800 according to an embodiment of the present disclosure is shown. For example, the electronic device 800 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, or the like.

[0144] Reference Figure 6 , the electronic device 800 may include one or more of the following components: a processing component 802 , a memory 804 , a power component 806 , a multimedia component 808 , an audio component 810 , an input / output (I / O) interface 812 , a sensor component 814 , and a communication component 816 .

[0145] The processing component 802 generally controls the overall operation of the electronic device 800, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 802 may include one or more processors 820 to execute instructions to complete all or part of the steps of the above-mentioned method. In addition, the processing component 802 may include one or more modules to facilitate the interaction between the processing component 802 and other components. For example, the processing component 802 may include a multimedia module to facilitate the interaction between the multimedia component 808 and the processing component 802.

[0146] The memory 804 is configured to store various types of data to support operations on the electronic device 800. Examples of such data include instructions for any application or method operating on the electronic device 800, contact data, phone book data, messages, pictures, videos, etc. The memory 804 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0147] The power supply component 806 provides power to the various components of the electronic device 800. The power supply component 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the electronic device 800.

[0148] The multimedia component 808 includes a screen that provides an output interface between the electronic device 800 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundaries of the touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 808 includes a front camera and / or a rear camera. When the electronic device 800 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera may receive external multimedia data. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.

[0149] The audio component 810 is configured to output and / or input audio signals. For example, the audio component 810 includes a microphone (MIC), and when the electronic device 800 is in an operating mode, such as a call mode, a recording mode, and a voice recognition mode, the microphone is configured to receive an external audio signal. The received audio signal can be further stored in the memory 804 or sent via the communication component 816. In some embodiments, the audio component 810 also includes a speaker for outputting audio signals.

[0150] I / O interface 812 provides an interface between processing component 802 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include but are not limited to: home button, volume button, start button, and lock button.

[0151] The sensor assembly 814 includes one or more sensors for providing various aspects of status assessment for the electronic device 800. For example, the sensor assembly 814 can detect the open / closed state of the electronic device 800, the relative positioning of the components, such as the display and keypad of the electronic device 800, and the sensor assembly 814 can also detect the position change of the electronic device 800 or a component of the electronic device 800, the presence or absence of contact between the user and the electronic device 800, the orientation or acceleration / deceleration of the electronic device 800, and the temperature change of the electronic device 800. The sensor assembly 814 may include a proximity sensor configured to detect the presence of a nearby object without any physical contact. The sensor assembly 814 may also include an optical sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 814 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0152] The communication component 816 is configured to facilitate wired or wireless communication between the electronic device 800 and other devices. The electronic device 800 can access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 816 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 816 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0153] In an exemplary embodiment, the electronic device 800 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above methods.

[0154] In an exemplary embodiment, a non-volatile computer-readable storage medium is also provided, such as a memory 804 including computer program instructions, which can be executed by a processor 820 of an electronic device 800 to perform the above method.

[0155] Figure 7 1 is a block diagram of another electronic device 1900 according to an embodiment of the present disclosure. For example, the electronic device 1900 may be provided as a server. Figure 7 , the electronic device 1900 includes a processing component 1922, which further includes one or more processors, and a memory resource represented by a memory 1932 for storing instructions executable by the processing component 1922, such as an application. The application stored in the memory 1932 may include one or more modules, each corresponding to a set of instructions. In addition, the processing component 1922 is configured to execute instructions to perform the above method.

[0156] The electronic device 1900 may also include a power supply component 1926 configured to perform power management of the electronic device 1900, a wired or wireless network interface 1950 configured to connect the electronic device 1900 to a network, and an input / output (I / O) interface 1958. The electronic device 1900 may operate based on an operating system stored in the memory 1932, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™ or the like.

[0157] In an exemplary embodiment, a non-volatile computer-readable storage medium is also provided, such as a memory 1932 including computer program instructions, which can be executed by the processing component 1922 of the electronic device 1900 to perform the above method.

[0158] The present disclosure may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0159] A computer-readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples of computer-readable storage media (a non-exhaustive list) include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination of the foregoing. As used herein, a computer-readable storage medium is not to be interpreted as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through a wire.

[0160] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0161] The computer program instructions for performing the operation of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. Computer-readable program instructions may be executed completely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be customized by utilizing the state information of the computer-readable program instructions, and the electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.

[0162] Various aspects of the present disclosure are described herein with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer-readable program instructions.

[0163] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0164] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0165] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of the module, program segment or instruction includes one or more executable instructions for realizing the specified logical function. In some alternative implementations, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or action, or can be implemented with a combination of special hardware and computer instructions.

[0166] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A method for processing information of ultrasound hyperechoic halo of breast cancer, characterized in that: include: Determine the lesion area based on breast ultrasound images; Acquire a grayscale change curve from the lesion area to the periphery of the breast ultrasound image; Determining whether there is a high echo halo around the lesion area based on the grayscale change curve; The determining whether there is a high echo halo around the lesion area based on the grayscale change curve includes: In a case where the grayscale change curve has a preset trend, determining that the grayscale change curve has the high echo halo; The preset trend includes increasing from a first grayscale interval to a second grayscale interval and decreasing from the second grayscale interval to a third grayscale interval, the grayscale value of the second grayscale interval is greater than the grayscale value of the third grayscale interval, and the grayscale value of the third grayscale interval is greater than the grayscale value of the first grayscale interval; In the case where multiple groups of grayscale change curves are obtained, determining whether there is a high echo halo around the lesion area based on the grayscale change curves further includes at least one of the following methods: Performing weighted sum processing on the similarity between the grayscale change curve and the standard hyperechoic halo curve, and determining that a hyperechoic halo exists around the lesion area when the obtained weighted sum indicates that the grayscale change curve and the standard hyperechoic halo curve are similar curves; Classification and clustering processing is performed on the grayscale change curve, and when the distance between two cluster centers obtained by clustering indicates that the grayscale change curve and the standard hyperechoic halo curve are similar curves, it is determined that a hyperechoic halo exists around the lesion area.

2. The method according to claim 1, characterized in that The grayscale change curve obtained from the lesion area toward the peripheral direction of the breast ultrasound image includes: determining the center of the lesion area; Taking the center as the starting point, grayscale sampling is performed toward the periphery of the breast ultrasound image to obtain at least one set of grayscale sequences; Based on the at least one set of grayscale sequences, the grayscale variation curve is determined.

3. The method according to claim 2, characterized in that The grayscale sampling is performed in the peripheral direction of the breast ultrasound image with the center as the starting point to obtain at least one set of grayscale sequences, including: Expanding a concentric frame from the center toward the periphery of the breast ultrasound image; Determine at least one ray starting from the center and heading toward the periphery; The intersection point of the ray and the concentric frame is determined as a sampling point, and grayscale sampling is performed to obtain a grayscale sequence corresponding to the ray.

4. The method according to claim 2 or 3, characterized in that: The step of determining the grayscale change curve based on the at least one set of grayscale sequences includes: Performing filtering processing on the grayscale sequence; Curve fitting is performed on the grayscale sequence after filtering to obtain the grayscale change curve.

5. The method according to claim 1, characterized in that In the case where a set of grayscale change curves are obtained, determining whether there is a high echo halo around the lesion area based on the grayscale change curves further includes: Determining the similarity between the grayscale change curve and a standard high echo halo curve; When the similarity indicates that the grayscale change curve and the standard hyperechoic halo curve are similar curves, it is determined that a hyperechoic halo exists around the lesion area.

6. The method according to claim 1, characterized in that Also includes: In the case where the grayscale change curve determines that a high echo halo exists around the lesion area, the lesion area is corrected using the grayscale change curve.

7. An information processing device for ultrasound hyperechoic halo of breast cancer, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the method described in any one of claims 1-6.

8. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images