Method and system for identifying parathyroid gland lesion based on ultrasonic image

Abstract

The invention provides a parathyroid gland lesion recognition method and system based on an ultrasonic image, and belongs to the technical field of image processing. The parathyroid gland lesion identification method based on the ultrasonic image comprises the following steps: acquiring a current ultrasonic image of a patient and a historical ultrasonic image in a database; determining a feature region in the current ultrasonic image according to the current ultrasonic image and the sample ultrasonic image; obtaining size data, shape data, boundary definition data and echo characteristic quantity data of the characteristic region; calculating basic attention data according to the region perimeter data, the region area data, the shape data, the boundary definition data and the echo characteristic quantity data; calculating trend influence data according to the historical ultrasonic image and the increasing distance data; analyzing comprehensive attention data of the feature region according to the basic attention data and the trend influence data; by adopting the scheme, the feature region can be marked, and a doctor can be helped to observe the ultrasonic image of the patient more visually.

Description

Technical Field

[0001] The present application relates to the technical field of image processing, and in particular to a method and system for identifying parathyroid lesions in ultrasound images. Background Art

[0002] When performing a parathyroid scan on a patient, an ultrasonic probe device is generally used to scan the position of the patient's parathyroid glands, thereby obtaining parathyroid ultrasound images at different angles, and storing these images in a database. These images all carry a timestamp representing the patient's examination time; in this way, for patients who undergo regular examinations, the patient's periodic parathyroid ultrasound images can be retrieved from the database to analyze the patient's parathyroid changes. It can be understood that when comparing and analyzing all the patient's parathyroid ultrasound images, since the scanned parathyroid ultrasound images represent the patient's parathyroid glands at different angles, when comparing and analyzing any two parathyroid ultrasound images at different periods, any two parathyroid ultrasound images representing the same angle are compared and analyzed; and in order to help doctors observe the patient's parathyroid changes more intuitively, it is necessary to specially mark some areas in the parathyroid ultrasound images. Summary of the invention

[0003] The present application provides a method and system for identifying parathyroid lesions in ultrasound images, which can help doctors quickly and intuitively observe patients' ultrasound images.

[0004] In a first aspect, the present application provides a method for identifying parathyroid lesions in ultrasound images. The method comprises: Acquire a current ultrasound image of the patient and a historical ultrasound image in a database; the historical ultrasound image carries a timestamp indicating that the patient underwent a parathyroid scan; and determine a sample ultrasound image based on the historical ultrasound image and the timestamp carried by the historical ultrasound image; the sample ultrasound image is the historical ultrasound image closest to the current moment; Determine a characteristic region in the current ultrasonic image according to the current ultrasonic image and the sample ultrasonic image; the characteristic region is a non-intersecting region in the current ultrasonic image and the sample ultrasonic image; Acquire the size data, shape data, boundary definition data and echo characteristic quantity data of the characteristic area; the size data includes area perimeter data, area area data and growth distance data; and calculate basic attention data according to the area perimeter data, area area data, shape data, boundary definition data and echo characteristic quantity data; calculate trend influence data according to the historical ultrasound image and the growth distance data; The comprehensive attention data of the feature area is analyzed according to the basic attention data and the trend impact data, and the annotation presentation mode of the current ultrasound image corresponding to the feature area is determined based on the comprehensive attention data.

[0005] Further, the acquiring of the size data, shape data, boundary definition data and echo characteristic quantity data of the characteristic area; the size data including area perimeter data, area area data and growth distance data; and calculating the basic attention data according to the area perimeter data, area area data, shape data, boundary definition data and echo characteristic quantity data; and calculating the trend influence data according to the historical ultrasound image and the growth distance data include: Determine shape influence data according to the shape data and a preset shape influence comparison table; Determining clarity impact data according to the boundary clarity data and a preset clarity impact comparison table; The basic attention data is calculated based on the region perimeter data, the region area data, the shape influence data, the clarity influence data and the echo characteristic quantity data.

[0006] Furthermore, the calculation method of the basic attention data includes: ; In the formula, As the basic attention data, is the area perimeter data, is the regional area data, For shape influence data, For clarity, the data is affected. It is the quantitative data of echo characteristics; are respectively a preset first weight, a preset second weight, a preset third weight and a preset fourth weight; and, ; The basic attention data is associated with the value of the area perimeter data, the value of the area area data, the value of the shape data and the boundary clarity data.

[0007] Further, the acquiring of the size data, shape data, boundary definition data and echo characteristic quantity data of the characteristic area; the size data including area perimeter data, area area data and growth distance data; and calculating the basic attention data according to the area perimeter data, area area data, shape data, boundary definition data and echo characteristic quantity data; and calculating the trend influence data according to the historical ultrasound image and the growth distance data include: Analyze the historical difference regions of any two adjacent historical ultrasound images according to the timestamps carried by the historical ultrasound images; determine based on the timestamps that, among any two adjacent historical ultrasound images, the historical ultrasound image close to the current moment is the next historical ultrasound image, and the historical ultrasound image far from the current moment is the previous historical ultrasound image; the historical difference region is the non-intersection region of the next historical ultrasound image with the previous historical ultrasound image; Calculating the historical growth data of each of the historical difference areas; the historical growth data is the length data of the largest parallel line within the historical difference area; Trend impact data is calculated based on the historical growth data and the growth distance data.

[0008] Furthermore, the calculation method of the trend impact data includes: Assume that the total number of the historical growth data plus the growth distance data is n, where the growth distance data is the nth growth data, and the historical growth data farthest from the current moment is the 1st growth data; then the i-th growth data is ; Calculate n-1 growth difference data based on n growth data ;in, ; Determine whether the signs of all growth difference data are greater than 0 or not greater than 0; If so, the trend impact data is calculated as, ; If not, the trend impact data is calculated as, ;in, is the number of growth difference data greater than 0, The number of growth difference data that is not greater than 0.

[0009] In a second aspect, the present application provides a system for identifying parathyroid lesions in ultrasound images, the system comprising: an acquisition module, configured to acquire a current ultrasound image of the patient and a historical ultrasound image in a database; the historical ultrasound image carries a timestamp indicating that the patient underwent a parathyroid scan; and determine a sample ultrasound image based on the historical ultrasound image and the timestamp carried by the historical ultrasound image; the sample ultrasound image is the historical ultrasound image closest to the current moment; A determination module, configured to determine a characteristic region in the current ultrasonic image according to the current ultrasonic image and the sample ultrasonic image; the characteristic region is a non-intersecting region in the current ultrasonic image and the sample ultrasonic image; A calculation module, used to obtain size data, shape data, boundary definition data and echo characteristic quantity data of the characteristic area; the size data includes area perimeter data, area area data and growth distance data; and calculate basic attention data according to the area perimeter data, area area data, shape data, boundary definition data and echo characteristic quantity data; calculate trend influence data according to the historical ultrasound image and the growth distance data; An analysis module is used to analyze the comprehensive attention data of the feature area according to the basic attention data and the trend impact data, and determine the annotation presentation method of the current ultrasound image corresponding to the feature area based on the comprehensive attention data.

[0010] Further, the calculation module is further configured to obtain the size data, shape data, boundary clarity data and echo characteristic quantity data of the feature area; the size data includes area perimeter data, area data and growth distance data; and calculate the basic attention data according to the area perimeter data, area data, shape data, boundary clarity data and echo characteristic quantity data; calculate the trend influence data according to the historical ultrasound image and the growth distance data, including: Determine shape influence data according to the shape data and a preset shape influence comparison table; Determining clarity impact data according to the boundary clarity data and a preset clarity impact comparison table; The basic attention data is calculated based on the region perimeter data, the region area data, the shape influence data, the clarity influence data and the echo characteristic quantity data.

[0011] Furthermore, the calculation module is further configured such that the calculation method of the basic attention data includes: ; In the formula, As the basic attention data, is the area perimeter data, is the regional area data, For shape influence data, For clarity, the data is affected. It is the quantitative data of echo characteristics; are respectively a preset first weight, a preset second weight, a preset third weight and a preset fourth weight; and, ; The basic attention data is associated with the value of the area perimeter data, the value of the area area data, the value of the shape data and the boundary clarity data.

[0012] Further, the calculation module is further configured to obtain the size data, shape data, boundary clarity data and echo characteristic quantity data of the feature area; the size data includes area perimeter data, area data and growth distance data; and calculate the basic attention data according to the area perimeter data, area data, shape data, boundary clarity data and echo characteristic quantity data; calculate the trend influence data according to the historical ultrasound image and the growth distance data, including: Analyze the historical difference regions of any two adjacent historical ultrasound images according to the timestamps carried by the historical ultrasound images; determine based on the timestamps that, among any two adjacent historical ultrasound images, the historical ultrasound image close to the current moment is the next historical ultrasound image, and the historical ultrasound image far from the current moment is the previous historical ultrasound image; the historical difference region is the non-intersection region of the next historical ultrasound image with the previous historical ultrasound image; Calculating the historical growth data of each of the historical difference areas; the historical growth data is the length data of the largest parallel line within the historical difference area; Trend impact data is calculated based on the historical growth data and the growth distance data.

[0013] Furthermore, the calculation module is further configured such that the calculation method of the trend impact data includes: Assume that the total number of the historical growth data plus the growth distance data is n, where the growth distance data is the nth growth data, and the historical growth data farthest from the current moment is the 1st growth data; then the i-th growth data is ; Calculate n-1 growth difference data based on n growth data ;in, ; Determine whether the signs of all growth difference data are greater than 0 or not greater than 0; If so, the trend impact data is calculated as, ; If not, the trend impact data is calculated as, ;in, is the number of growth difference data greater than 0, The number of growth difference data that is not greater than 0.

[0014] It should be understood that the contents described in the Summary of the Invention are not intended to limit the key or important features of the embodiments of the present application, nor are they intended to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The above and other features, advantages and aspects of the embodiments of the present application will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, wherein: Figure 1 A flow chart of a method for identifying parathyroid lesions in an ultrasound image in an embodiment of the present application is shown; Figure 2 A block diagram of a system for identifying parathyroid lesions in ultrasound images in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0017] In addition, the term "and / or" in this article is only a description of the association relationship between the associated objects, indicating that there can 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 character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0018] The present application provides a method and system for identifying parathyroid lesions in ultrasound images, which can quickly mark characteristic areas in the patient's ultrasound image and help doctors observe the patient's ultrasound image more intuitively.

[0019] In a first aspect, the present application provides a method for identifying parathyroid lesions based on ultrasound images. Figure 1 The specific steps included in the method are as follows.

[0020] Step S110: Acquire the patient's current ultrasound image and historical ultrasound images in a database; the historical ultrasound image carries a timestamp indicating when the patient underwent a parathyroid scan; and determine a sample ultrasound image based on the historical ultrasound image and the timestamp carried by the historical ultrasound image; the sample ultrasound image is the historical ultrasound image closest to the current moment.

[0021] It is understandable that, generally speaking, when acquiring ultrasound images of patients, the first step is to collect and preprocess the image data, including standardization, denoising, enhancement, etc.; then, feature extraction is performed on the image, and then a suitable model can be selected and trained accordingly.

[0022] It can be understood that, in the embodiment of the present application, the current ultrasound image obtained corresponds to the historical ultrasound image, including the current ultrasound image at all angles of the historical ultrasound image, that is, when the patient is subjected to parathyroid ultrasound scanning, the same ultrasound images at all angles are obtained, so all historical ultrasound images corresponding to the angles of the current ultrasound image can be found in the database; here, any current ultrasound image is taken as an example to perform analysis and calculation of the comprehensive attention degree, and when analyzing and comparing the current ultrasound image, multiple historical ultrasound images at the same angle are selected from the database for analysis and comparison.

[0023] Step S120: determining a characteristic region in the current ultrasonic image according to the current ultrasonic image and the sample ultrasonic image; the characteristic region is a non-intersecting region of the current ultrasonic image and the sample ultrasonic image.

[0024] It can be understood that the characteristic region here can be understood as the redundant part of the current ultrasound image compared with the sample ultrasound image; that is, the current ultrasound image contains the sample ultrasound image and the additional part, and the additional part is the characteristic region, that is, the region that represents the changes in the parathyroid glands during the patient's examination compared with the last patient examination. It can be understood that the characteristic region here refers to the ultrasound image features in the ultrasound image; generally speaking, the three diseases of the parathyroid gland, parathyroid hyperplasia, parathyroid adenoma and parathyroid carcinoma show different characteristics on ultrasound images; for example, parathyroid hyperplasia usually manifests as uniform enlargement of multiple parathyroid glands, uniform echo and clear boundaries; parathyroid adenomas are mostly single, round or oval, with uniform or uneven echo, clear boundaries, and sometimes visible capsules; the ultrasound image features of parathyroid carcinoma are more complex, usually manifested as irregular morphology, uneven echo, unclear boundaries, and may be accompanied by calcification or infiltration of surrounding tissues.

[0025] Step S130: Obtain the size data, shape data, boundary clarity data and echo characteristic quantity data of the feature area; the size data includes area perimeter data, area area data and growth distance data; and calculate basic attention data based on the area perimeter data, area area data, shape data, boundary clarity data and echo characteristic quantity data; calculate trend impact data based on the historical ultrasound image and the growth distance data.

[0026] In an embodiment of the present application, calculating basic attention data based on the region perimeter data, region area data, shape data, boundary clarity data and echo characteristic quantity data specifically includes: determining shape influence data based on the shape data and a preset shape influence comparison table; determining clarity influence data based on the boundary clarity data and a preset clarity influence comparison table; calculating basic attention data based on the region perimeter data, region area data, shape influence data, clarity influence data and the echo characteristic quantity data.

[0027] It can be understood that in the embodiment of the present application, the preset shape influence comparison table for the feature area includes two cases: regular shape and irregular shape, and the shape influence data corresponding to the two cases are fixed parameters; wherein, the fixed parameters of the irregular shape are not less than the fixed parameters of the regular shape; the boundary clarity data here also includes two cases: clear and blurred, and the boundary clarity data here is determined based on the boundary comparison between the current ultrasound image and the sample ultrasound image. If the boundary clarity of the current ultrasound image is higher than the boundary clarity of the sample ultrasound image, it is clear; otherwise, it is blurred; similarly, the clarity influence data corresponding to the two cases here are fixed parameters, wherein the fixed parameters of the blurred are not less than the fixed parameters of the clear.

[0028] Specifically, in the embodiment of the present application, the shape-affecting data is calculated as follows: a shape descriptor may be used to characterize the shape of a feature region, such as circularity, eccentricity, etc.; taking circularity as an example, ; The circularity value obtained Compared with the preset circularity threshold, if the circularity value is not less than the preset circularity threshold, it means that the shape of the feature area is a regular shape, otherwise it is an irregular shape; the above-mentioned preset shape impact comparison table includes multiple shape descriptors and corresponding comparison thresholds; after determining whether the shape of the feature area is regular, the shape impact data can be determined based on the preset fixed parameters corresponding to regular or irregular.

[0029] The calculation method for clarity influencing data is as follows; specifically, the boundary clarity can be measured by calculating the gradient variance of the boundary of the feature area; the gradient of the boundary pixels of the feature area is calculated, such as using the Sobel operator to obtain a gradient amplitude image, and then the variance of the gradient amplitude image is calculated, and the variance value is used as the boundary clarity data; after the boundary clarity data is calculated, the boundary clarity data is compared with the boundary clarity of the preset sample ultrasound image. If the calculated boundary clarity data of the actual image is not less than the boundary clarity of the preset sample ultrasound image, it means that the boundary of the feature area is clear, otherwise, it means that the boundary of the feature area is blurred; after determining whether the boundary of the feature area is clear, the clarity influencing data can be determined based on the preset fixed parameters corresponding to the clarity or blur of the boundary.

[0030] It should be noted that for echo characteristics, the number may be 0 or 1 or not less than 1. When the number of echo characteristics is 0, it will not affect the basic attention data. Therefore, in this scheme, the default minimum number of echo characteristics is 1; if the number of echo characteristics is 1, it will have a greater impact. If the number of echo characteristics exceeds 1, the fluctuation of its impact will be reduced.

[0031] Calculating the trend influence data based on the historical ultrasonic images and the growth distance data specifically includes: analyzing the historical difference area of ​​any two adjacent historical ultrasonic images based on the timestamps carried by the historical ultrasonic images; determining based on the timestamps that the historical ultrasonic image close to the current moment in any two adjacent historical ultrasonic images is the latter historical ultrasonic image, and the historical ultrasonic image far from the current moment is the former historical ultrasonic image; the historical difference area is the non-intersecting area of ​​the latter historical ultrasonic image with the former historical ultrasonic image; calculating the historical growth data of each of the historical difference areas; the historical growth data is the length data of the largest parallel line within the historical difference area; and calculating the trend influence data based on the historical growth data and the growth distance data.

[0032] The calculation method of basic attention data includes: ; In the formula, As the basic attention data, is the area perimeter data, is the regional area data, For shape influence data, For clarity, the data is affected. It is the quantitative data of echo characteristics; are respectively a preset first weight, a preset second weight, a preset third weight and a preset fourth weight; and, ; The basic attention data is associated with the value of the area perimeter data, the value of the area area data, the value of the shape data and the boundary clarity data.

[0033] It can be understood that in this scheme, the calculation of basic attention data involves five influencing factors, namely, regional perimeter data, regional area data, shape influence data, clarity influence data and echo characteristic quantity data; and in the specific calculation process, for the four data, namely, regional perimeter data, regional area data, shape influence data and clarity influence data, the perimeter and area only use their specific values ​​to participate in the calculation, and do not involve the mathematical units corresponding to the perimeter and area; and the shape influence data and clarity influence data mentioned above participate in the calculation in the form of preset fixed parameters, and only use their specific values ​​for calculation; specifically, the perimeter and area reflect the changes in the characteristic area, which can be understood as the degree of deviation from the standard values ​​of the perimeter and area. The greater the deviation, the more attention the thyroid needs to receive; and the corresponding forms of shape and clarity are only four forms: whether the shape is regular or not and whether the boundary is clear or not. Similarly, irregular shape and blurred boundaries are manifestations of an abnormal state, and the thyroid gland in an abnormal state needs to be paid more attention.

[0034] The calculation methods of trend impact data include: Assume that the total number of the historical growth data plus the growth distance data is n, where the growth distance data is the nth growth data, and the historical growth data farthest from the current moment is the 1st growth data; then the i-th growth data is ; Calculate n-1 growth difference data based on n growth data ;in, ; Determine whether the signs of all growth difference data are greater than 0 or not greater than 0; If so, the trend impact data is calculated as, ; If not, the trend impact data is calculated as, ;in, is the number of growth difference data greater than 0, The number of growth difference data that is not greater than 0.

[0035] It should be noted that in the embodiment of the present application, the size data, shape data, boundary clarity data and echo characteristic quantity data of the characteristic area are used for calculation; in addition to this part of the data, data statistics and data analysis can also be performed using data characteristics such as size data, shape data, echo characteristic data, whether the tumor boundary is clear, blood supply, whether there is calcification, and whether the surrounding tissue is invaded.

[0036] Step S140: analyzing the comprehensive attention data of the feature area according to the basic attention data and the trend impact data, and determining the annotation presentation method of the current ultrasound image corresponding to the feature area based on the comprehensive attention data.

[0037] In an embodiment of the present application, the calculation method of the comprehensive attention data is: comprehensive attention data = basic attention data * trend influence data; after determining the comprehensive attention data of the current ultrasound image, determine whether the current ultrasound image is specially marked and displayed based on a preset attention threshold; if the comprehensive attention data of the current ultrasound image is not less than the preset attention threshold, the characteristic area of ​​the current ultrasound image is specially marked, and the historical ultrasound image closest to the current moment at the same angle as the current ultrasound image is synchronously displayed to reflect the changes in the patient's parathyroid glands at this angle, thereby providing doctors with a more intuitive ultrasound image display.

[0038] Generally speaking, parathyroid diseases mainly include parathyroid hyperplasia, parathyroid cancer and parathyroid adenoma. By specially marking the characteristic areas of the ultrasound image, doctors can better identify whether the current patient's ultrasound image is more consistent with one of the above three diseases.

[0039] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the embodiments of the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required by the present application.

[0040] In a second aspect, the present application provides a system for identifying parathyroid lesions based on ultrasound images. Figure 2As shown, the system includes an acquisition module 210, which is used to acquire a current ultrasound image of a patient and a historical ultrasound image in a database; the historical ultrasound image carries a timestamp marking the time when the patient underwent parathyroid scanning; and a sample ultrasound image is determined based on the historical ultrasound image and the timestamp carried by the historical ultrasound image; the sample ultrasound image is the historical ultrasound image closest to the current moment; a determination module 220, which is used to determine a feature area in the current ultrasound image based on the current ultrasound image and the sample ultrasound image; the feature area is a non-intersecting area in the current ultrasound image with the sample ultrasound image; a calculation module 230, which is used to acquire the feature area The size data, shape data, boundary clarity data and echo characteristic quantity data of the feature area; the size data includes area perimeter data, area area data and growth distance data; and basic attention data is calculated based on the area perimeter data, area area data, shape data, boundary clarity data and echo characteristic quantity data; trend influence data is calculated based on the historical ultrasound image and the growth distance data; an analysis module 240 is used to analyze the comprehensive attention data of the feature area according to the basic attention data and the trend influence data, and determine the annotation presentation method of the current ultrasound image corresponding to the feature area based on the comprehensive attention data.

[0041] Further, the calculation module 230 is further configured to obtain the size data, shape data, boundary clarity data and echo characteristic quantity data of the feature area; the size data includes area perimeter data, area area data and growth distance data; and calculate the basic attention data according to the area perimeter data, area area data, shape data, boundary clarity data and echo characteristic quantity data; calculate the trend influence data according to the historical ultrasound image and the growth distance data including: Determine shape influence data according to the shape data and a preset shape influence comparison table; Determining clarity impact data according to the boundary clarity data and a preset clarity impact comparison table; The basic attention data is calculated based on the region perimeter data, the region area data, the shape influence data, the clarity influence data and the echo characteristic quantity data.

[0042] Furthermore, the calculation module 230 is further configured to calculate the basic attention data in the following manner: ; In the formula, As the basic attention data, is the area perimeter data, is the regional area data, For shape influence data, For clarity, the data is affected. It is the quantitative data of echo characteristics; are respectively a preset first weight, a preset second weight, a preset third weight and a preset fourth weight; and, ; The basic attention data is associated with the value of the area perimeter data, the value of the area area data, the value of the shape data and the boundary clarity data.

[0043] Further, the calculation module 230 is further configured to obtain the size data, shape data, boundary clarity data and echo characteristic quantity data of the feature area; the size data includes area perimeter data, area area data and growth distance data; and calculate the basic attention data according to the area perimeter data, area area data, shape data, boundary clarity data and echo characteristic quantity data; calculate the trend influence data according to the historical ultrasound image and the growth distance data including: Analyze the historical difference regions of any two adjacent historical ultrasound images according to the timestamps carried by the historical ultrasound images; determine based on the timestamps that, among any two adjacent historical ultrasound images, the historical ultrasound image close to the current moment is the next historical ultrasound image, and the historical ultrasound image far from the current moment is the previous historical ultrasound image; the historical difference region is the non-intersection region of the next historical ultrasound image with the previous historical ultrasound image; Calculating the historical growth data of each of the historical difference areas; the historical growth data is the length data of the largest parallel line within the historical difference area; Trend impact data is calculated based on the historical growth data and the growth distance data.

[0044] Furthermore, the calculation module 230 is further configured to calculate the trend impact data in the following manner: Assume that the total number of the historical growth data plus the growth distance data is n, where the growth distance data is the nth growth data, and the historical growth data farthest from the current moment is the 1st growth data; then the i-th growth data is ; Calculate n-1 growth difference data based on n growth data ;in, ; Determine whether the signs of all growth difference data are greater than 0 or not greater than 0; If so, the trend impact data is calculated as, ; If not, the trend impact data is calculated as, ;in, is the number of growth difference data greater than 0, The number of growth difference data that is not greater than 0.

[0045] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the described device can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0046] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the aforementioned disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other to form a technical solution.

Claims

1. A method for identifying parathyroid lesions in ultrasound images, characterized in that: include: Acquire a current ultrasound image of the patient and a historical ultrasound image in a database; the historical ultrasound image carries a timestamp indicating that the patient underwent a parathyroid scan; and determine a sample ultrasound image based on the historical ultrasound image and the timestamp carried by the historical ultrasound image; the sample ultrasound image is the historical ultrasound image closest to the current moment; Determine a characteristic region in the current ultrasonic image according to the current ultrasonic image and the sample ultrasonic image; the characteristic region is a non-intersecting region in the current ultrasonic image and the sample ultrasonic image; Acquire size data, shape data, boundary definition data and echo characteristic quantity data of the feature area; the size data includes area perimeter data, area area data and growth distance data; and calculate basic attention data according to the area perimeter data, area area data, shape data, boundary definition data and echo characteristic quantity data; calculating trend impact data based on the historical ultrasound images and the growth distance data; The comprehensive attention data of the feature area is analyzed according to the basic attention data and the trend impact data, and the annotation presentation mode of the current ultrasound image corresponding to the feature area is determined based on the comprehensive attention data.

2. The method according to claim 1, characterized in that The step of obtaining the size data, shape data, boundary definition data and echo characteristic quantity data of the feature area; the size data includes area perimeter data, area data and growth distance data; and calculating the basic attention data according to the area perimeter data, area data, shape data, boundary definition data and echo characteristic quantity data; Calculating trend impact data according to the historical ultrasound image and the growth distance data includes: Determine shape influence data according to the shape data and a preset shape influence comparison table; Determining clarity impact data according to the boundary clarity data and a preset clarity impact comparison table; The basic attention data is calculated based on the region perimeter data, the region area data, the shape influence data, the clarity influence data and the echo characteristic quantity data.

3. The method according to claim 2, characterized in that The calculation method of the basic attention data includes: ； In the formula, As the basic attention data, is the area perimeter data, is the regional area data, For shape influence data, For clarity, the data is affected. It is the quantitative data of echo characteristics; are respectively a preset first weight, a preset second weight, a preset third weight and a preset fourth weight; and, ; The basic attention data is associated with the value of the area perimeter data, the value of the area area data, the value of the shape data and the boundary clarity data.

4. The method according to claim 3, characterized in that The step of obtaining the size data, shape data, boundary definition data and echo characteristic quantity data of the feature area; the size data includes area perimeter data, area data and growth distance data; and calculating the basic attention data according to the area perimeter data, area data, shape data, boundary definition data and echo characteristic quantity data; Calculating trend impact data according to the historical ultrasound image and the growth distance data includes: Analyze the historical difference regions of any two adjacent historical ultrasound images according to the timestamps carried by the historical ultrasound images; determine based on the timestamps that, among any two adjacent historical ultrasound images, the historical ultrasound image close to the current moment is the next historical ultrasound image, and the historical ultrasound image far from the current moment is the previous historical ultrasound image; the historical difference region is the non-intersection region of the next historical ultrasound image with the previous historical ultrasound image; Calculating the historical growth data of each of the historical difference areas; the historical growth data is the length data of the largest parallel line within the historical difference area; Trend impact data is calculated based on the historical growth data and the growth distance data.

5. The method according to claim 4, characterized in that The calculation method of the trend impact data includes: Assume that the total number of the historical growth data plus the growth distance data is n, where the growth distance data is the nth growth data, and the historical growth data farthest from the current moment is the 1st growth data; then the i-th growth data is ; Calculate n-1 growth difference data based on n growth data ;in, ; Determine whether the signs of all growth difference data are greater than 0 or not greater than 0; If so, the trend impact data is calculated as, ; If not, the trend impact data is calculated as, ;in, is the number of growth difference data greater than 0, The number of growth difference data that is not greater than 0.

6. A parathyroid lesion recognition system based on ultrasound images, characterized in that: include: An acquisition module (210) is used to acquire a current ultrasound image of the patient and a historical ultrasound image in a database; the historical ultrasound image carries a timestamp indicating that the patient underwent a parathyroid scan; and determine a sample ultrasound image based on the historical ultrasound image and the timestamp carried by the historical ultrasound image; the sample ultrasound image is the historical ultrasound image closest to the current moment; A determination module (220), configured to determine a characteristic region in the current ultrasound image according to the current ultrasound image and the sample ultrasound image; the characteristic region is a non-intersecting region in the current ultrasound image and the sample ultrasound image; A calculation module (230) is used to obtain size data, shape data, boundary clarity data and echo characteristic quantity data of the feature area; the size data includes area perimeter data, area area data and growth distance data; and calculate basic attention data based on the area perimeter data, area area data, shape data, boundary clarity data and echo characteristic quantity data; calculating trend impact data based on the historical ultrasound images and the growth distance data; An analysis module (240) is used to analyze the comprehensive attention data of the feature area according to the basic attention data and the trend impact data, and determine the annotation presentation method of the current ultrasound image corresponding to the feature area based on the comprehensive attention data.

7. The system according to claim 6, characterized in that The calculation module (230) is further configured to obtain size data, shape data, boundary definition data and echo characteristic quantity data of the feature area; the size data includes area perimeter data, area data and growth distance data; and calculate basic attention data based on the area perimeter data, area data, shape data, boundary definition data and echo characteristic quantity data; Calculating trend impact data according to the historical ultrasound image and the growth distance data includes: Determine shape influence data according to the shape data and a preset shape influence comparison table; Determining clarity impact data according to the boundary clarity data and a preset clarity impact comparison table; The basic attention data is calculated based on the region perimeter data, the region area data, the shape influence data, the clarity influence data and the echo characteristic quantity data.

8. The system according to claim 7, characterized in that The calculation module (230) is further configured to calculate the basic attention data in the following manner: ； In the formula, As the basic attention data, is the area perimeter data, is the regional area data, For shape influence data, For clarity, the data is affected. It is the quantitative data of echo characteristics; are respectively a preset first weight, a preset second weight, a preset third weight and a preset fourth weight; and, ; The basic attention data is associated with the value of the area perimeter data, the value of the area area data, the value of the shape data and the boundary clarity data.

9. The system according to claim 8, characterized in that The calculation module (230) is further configured to obtain size data, shape data, boundary definition data and echo characteristic quantity data of the feature area; the size data includes area perimeter data, area data and growth distance data; and calculate basic attention data based on the area perimeter data, area data, shape data, boundary definition data and echo characteristic quantity data; Calculating trend impact data according to the historical ultrasound image and the growth distance data includes: Analyze the historical difference regions of any two adjacent historical ultrasound images according to the timestamps carried by the historical ultrasound images; determine based on the timestamps that, among any two adjacent historical ultrasound images, the historical ultrasound image close to the current moment is the next historical ultrasound image, and the historical ultrasound image far from the current moment is the previous historical ultrasound image; the historical difference region is the non-intersection region of the next historical ultrasound image with the previous historical ultrasound image; Calculating the historical growth data of each of the historical difference areas; the historical growth data is the length data of the largest parallel line within the historical difference area; Trend impact data is calculated based on the historical growth data and the growth distance data.

10. The system according to claim 9, characterized in that The calculation module (230) is further configured such that the calculation method of the trend impact data includes: Assume that the total number of the historical growth data plus the growth distance data is n, where the growth distance data is the nth growth data, and the historical growth data farthest from the current moment is the 1st growth data; then the i-th growth data is ; Calculate n-1 growth difference data based on n growth data ;in, ; Determine whether the signs of all growth difference data are greater than 0 or not greater than 0; If so, the trend impact data is calculated as, ; If not, the trend impact data is calculated as, ;in, is the number of growth difference data greater than 0, The number of growth difference data that is not greater than 0.

Images