Method, device and equipment for ultrasonically detecting kidney and storage medium

Abstract

The invention relates to the technical field of ultrasonic kidney detection, in particular to a kidney ultrasonic detection method, device and equipment and a storage medium, and the method comprises the steps: 1, collecting kidney ultrasonic images, and carrying out the data annotation of a plurality of ultrasonic images, and obtaining a data set; step 2, performing data processing on the data in the data set to generate ultrasonic image data with uniform data; performing feature extraction on the ultrasonic image data to obtain multi-scale features; 3, presetting a kidney model ultrasonic image, and performing comparative analysis on the multi-scale features and the preset kidney model ultrasonic image to obtain a kidney comparison result on a single-frame image; and 4, obtaining a kidney detection result. The detection method provided by the invention has the advantages that the operation is convenient, the detection result is intelligently analyzed, the accuracy of the detection result is high, and the lesion position and the lesion degree of the kidney can be timely found.

Description

Technical Field

[0001] The present invention relates to the technical field of ultrasonic kidney detection, and in particular to a method, device, equipment and storage medium for ultrasonic kidney detection. Background Art

[0002] Ultrasound testing refers to the use of weak ultrasonic waves to irradiate the body, image the reflected waves of the tissue, and measure the morphology and data of physiological tissues to provide a basis for disease discovery and diagnosis. In related technologies, ultrasound can be used to collect kidney images to obtain structural data such as its size, morphology, and renal pelvis. These structural data can then be used to determine whether there is a disease in the kidney. A renal ultrasound examination device is a medical device used to examine the structure, morphology, and function of the human kidney using ultrasound. This device can produce ultrasound images to help doctors diagnose various kidney diseases such as stones, cysts, and tumors. As a non-invasive, real-time, and cost-effective diagnostic tool, ultrasound examination is widely used in clinical practice.

[0003] However, traditional manual ultrasound scanning relies on the operator's skill and experience, and its results can be influenced by subjective factors. In recent years, with the development of robotics and artificial intelligence, ultrasound scanning robots have become a research hotspot, aiming to improve the accuracy, reliability, and efficiency of ultrasound examinations while reducing the operator's workload. Despite this, current ultrasound robotic systems still face many challenges, especially in terms of autonomy, intelligent detection, and test result analysis, and are still some distance away from full clinical application. Summary of the Invention

[0004] One of the purposes of the present invention is to provide a method for ultrasonic kidney detection, which has the advantages of easy operation, intelligent analysis of detection results, high accuracy of detection results, and the ability to promptly detect the location and degree of kidney lesions.

[0005] To achieve the above-mentioned purpose, the present invention provides the following technical solution: A method for ultrasonic detection of kidneys, comprising: step 1: collecting kidney ultrasound images, and data-labeling multiple ultrasound images to obtain a data set; step 2: performing data processing on the data in the data set to generate unified ultrasound image data; performing feature extraction on the ultrasound image data to obtain multi-scale features; step 3: presetting a kidney model ultrasound image, and comparing and analyzing the multi-scale features with the preset kidney model ultrasound image based on an attention mechanism to obtain a kidney comparison result on a single-frame image; step 4: collecting the single-frame images with different comparison results, and forming an ultrasound image comparison data set, and obtaining a kidney detection result based on the analysis of the comparison data set.

[0006] This solution obtains a data set by collecting kidney ultrasound images and performing data annotation. Ultrasound image data is generated by processing the data in the data set. The generated image is compared with the preset ultrasound image for analysis to obtain the kidney detection results. This method is easy to operate, has high detection efficiency, can see the test results in a timely manner, and has high accuracy. Through comparative analysis, the location of kidney lesions and the specific state of the lesions can be clearly found.

[0007] In step three, the preset kidney model ultrasound image includes: the data range of the overall kidney image and the specific location information coordinates of the kidney. Step three also includes screening, through comparative analysis, target features that differ between the ultrasound image and the preset kidney model ultrasound image, and performing analysis and image magnification and annotation based on the target features. This solution can more accurately compare the differences and specific location information between the collected kidney ultrasound image and the preset kidney model ultrasound image. By using image magnification and annotation, the differences between the kidney ultrasound images can be clearly compared, and the location and extent of kidney lesions can be promptly discovered.

[0008] The comparative analysis is performed based on the coordinates of the position information of the preset kidney model ultrasound image. The data processing of the data in the dataset to generate unified ultrasound image data includes: scaling the images in the dataset frame by frame and cropping them based on the image center to obtain cropped image data; enhancing the cropped image data using a preset data enhancement method to obtain enhanced image data; and normalizing the size and grayscale of the enhanced image data to generate the unified ultrasound image data.

[0009] The multi-scale features are obtained by extracting features from the ultrasound image data, including: reducing the size of the ultrasound image layer by layer and applying Gaussian blur, obtaining high-frequency details through the difference of Gaussian pyramids, so as to respectively extract features of different modalities in the ultrasound image data to obtain the multi-scale features.

[0010] One of the objects of the present invention is to provide a kidney ultrasound detection device, comprising: a data set acquisition unit, for acquiring kidney ultrasound images, and performing data annotation on multiple ultrasound images to obtain a data set; a data processing unit, for performing data processing on the data in the data set to generate unified ultrasound image data; a feature extraction unit, for performing feature extraction on the ultrasound image data to obtain multi-scale features; a comparative analysis unit, for presetting a kidney model ultrasound image, and performing comparative analysis on the multi-scale features and the preset kidney model ultrasound image based on an attention mechanism to obtain a kidney comparison result on a single-frame image; a detection result generation unit, for collecting the single-frame images with different comparison results, and forming an ultrasound image comparison data set, and obtaining a kidney detection result based on the analysis of the comparison data set.

[0011] One of the objects of the present invention is to provide a computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it implements any of the above-mentioned methods for ultrasonic detection of kidneys.

[0012] One of the objects of the present invention is to provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any one of the above-mentioned methods for ultrasonically detecting kidneys.

[0013] Compared with the existing technology, the beneficial effects of the present invention are: the present invention obtains a data set by collecting kidney ultrasound images and performing data annotation, generates ultrasound image data by processing the data in the data set, and compares and analyzes the generated image with the preset ultrasound image to obtain the kidney detection result. The method is easy to operate, has high detection efficiency, can see the test results in time, and has high accuracy. Through comparative analysis, the location of kidney lesions and the specific state of the lesions can be clearly found. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 A flow chart of the method for ultrasonically detecting kidneys provided by the present invention; Figure 2 A structural block diagram of a device for ultrasonically detecting kidneys provided in an embodiment of the present application; Figure 3 The structure of an electronic device for ultrasonically detecting kidneys provided in an embodiment of the present application is shown in FIG. intention. DETAILED DESCRIPTION

[0015] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0016] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0017] Next, the present invention is described in detail with reference to schematic diagrams. For ease of illustration, cross-sectional views illustrating device structures may be partially enlarged and not to scale when describing the embodiments of the present invention. Furthermore, the schematic diagrams are merely illustrative and should not limit the scope of protection of the present invention. Furthermore, in actual production, three-dimensional dimensions, including length, width, and depth, should be included.

[0018] To make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0019] Example 1 This embodiment provides a method for ultrasonically detecting kidneys. Figure 1 Shown, including: Step 1: Acquire renal ultrasound images and annotate multiple of the ultrasound images to obtain a data set. Specifically, in some application scenarios, the operator can place the ultrasound probe of the ultrasound instrument at the patient's kidney position and move the ultrasound probe to acquire renal ultrasound images. In some scenarios, multiple frames of ultrasound images can also form an ultrasound video, which also belongs to the collection of renal ultrasound images in this embodiment.

[0020] Step 2: Process the data in the dataset to generate unified ultrasound image data; extract features from the ultrasound image data to obtain multi-scale features; specifically, the embodiment of the present application uses a deep learning model to classify kidney ultrasound images. In the actual data collection process, due to differences in the inspection equipment and operating standards used, the resolution, aspect ratio, etc. of the collected images cannot be completely unified, while the input size of the deep learning model is fixed. Therefore, in order to reduce the impact of data inconsistency on model training and evaluation results, the embodiment of the present application performs data preprocessing. The preprocessing is: unify the aspect ratio, size, etc. of all kidney ultrasound images to be processed to generate unified ultrasound image data.

[0021] Step 3: Preset a kidney model ultrasound image, and compare and analyze the multi-scale features with the preset kidney model ultrasound image based on the attention mechanism to obtain a kidney comparison result on a single frame image; The above-mentioned preset kidney model ultrasound image can be regarded as a complete kidney image, and each frame of the preset ultrasound image can reflect the structural data of a certain part of the normal kidney.

[0022] After the ultrasound device acquires the renal ultrasound image to be processed, it can search for a target ultrasound image that matches each frame of the preset renal model ultrasound image within the renal ultrasound image. For example, if the renal ultrasound image to be processed consists of 100 frames of preset ultrasound images, the ultrasound device can search for target renal ultrasound images that match each of the 100 frames of renal ultrasound images to be processed from the preset renal model ultrasound images. In some application scenarios, for example, ultrasound images with identical image information can be considered target ultrasound images that match the preset ultrasound images, i.e., normal renal images, while renal ultrasound images to be processed that have a low degree of match with the preset ultrasound images are considered ultrasound renal images with different results. The above-mentioned image information may include, for example, texture, color, and other information.

[0023] Step 4: Collect the single-frame images with different comparison results and form an ultrasound image comparison data set, and obtain the kidney detection result based on the analysis of the comparison data set.

[0024] This solution obtains a data set by collecting kidney ultrasound images and performing data annotation. Ultrasound image data is generated by processing the data in the data set. The generated image is compared with the preset ultrasound image for analysis to obtain the kidney detection results. This method is easy to operate, has high detection efficiency, can see the test results in a timely manner, and has high accuracy. Through comparative analysis, the location of kidney lesions and the specific state of the lesions can be clearly found.

[0025] Example 2 This embodiment provides a method for ultrasonically detecting kidneys, comprising: Step 1: Acquire kidney ultrasound images and annotate multiple ultrasound images to obtain a data set; Step 2: Processing the data in the dataset to generate unified ultrasound image data, specifically including: scaling the images in the dataset frame by frame, and cropping them based on the image center to obtain cropped image data; enhancing the cropped image data using a preset data enhancement method to obtain enhanced image data; and normalizing the size and grayscale of the enhanced image data to generate unified ultrasound image data.

[0026] Multi-scale features are obtained by performing feature extraction on the ultrasound image data, specifically including: reducing the size of the ultrasound image layer by layer and applying Gaussian blur, obtaining high-frequency details through the difference of Gaussian pyramids, so as to respectively extract features of different modalities in the ultrasound image data to obtain the multi-scale features.

[0027] Step 3: A preset kidney model ultrasound image is generated. The preset kidney model ultrasound image includes the data range of the entire kidney image and the specific location information coordinates of the kidney. Based on the attention mechanism, the multi-scale features are compared and analyzed with the preset kidney model ultrasound image. The location information coordinates of the preset kidney model ultrasound image are used as the standard for the comparative analysis. Target features that differ between the ultrasound image and the preset kidney model ultrasound image are screened out. These target features are analyzed and image magnification and annotation are performed based on the target features. A kidney comparison result is obtained on a single frame.

[0028] Step 4: Collect the single-frame images with different comparison results and form an ultrasound image comparison data set, and obtain the kidney detection result based on the analysis of the comparison data set.

[0029] The advantage of the solution provided by this embodiment is that it can more accurately compare the differences between the collected kidney ultrasound image and the preset kidney model ultrasound image and specific location information, and further clearly reflect the differences in the kidney ultrasound image through image magnification and annotation, so as to timely discover the location and degree of kidney lesions, reduce the errors that may be caused by manual comparison, and improve the efficiency and overall accuracy of kidney detection.

[0030] Example 3 Reference Figure 2 As shown, this embodiment provides a kidney ultrasound detection device, including: a data set acquisition unit 201, used to acquire kidney ultrasound images and perform data annotation on multiple ultrasound images to obtain a data set; a data processing unit 202, used to generate unified ultrasound image data by performing data processing on the data in the data set; a feature extraction unit 203, used to obtain multi-scale features by extracting features from the ultrasound image data; a comparison and analysis unit 204, which presets a kidney model ultrasound image, compares and analyzes the multi-scale features with the preset kidney model ultrasound image based on an attention mechanism, and obtains a kidney comparison result on a single frame image; a detection result generation unit 205, used to collect the single frame images with different comparison results, form an ultrasound image comparison data set, and obtain a kidney detection result based on the analysis of the comparison data set.

[0031] Example 4 This embodiment provides a computer device, referring to Figure 3 As shown, it includes a memory 301, a processor 302 and a communication interface 303. The memory 301 stores a computer program, and when the processor 302 executes the computer program, the method for ultrasonic detection of kidneys in the above embodiment is implemented.

[0032] The communication interface 303 in the embodiment of the present application is used to communicate signaling or data with other node devices. The memory 301 can be a read-only memory (ROM): a semiconductor memory whose stored content is fixed and can only be read but not written. It can also be a random read-write memory (RAM): a semiconductor memory that can be both read and written. Or a non-permanent memory memory: a memory in which information disappears after power is cut off. Or a permanent memory memory: a memory that can still save information after power is cut off. The memory 301 can optionally be a storage device located away from the processor 302. The memory 301 stores computer-readable instructions. When the computer-readable instructions are executed by the processor 302, the computer device can execute the above-mentioned instructions. Figure 1 The method process shown.

[0033] I understand. Figure 3 The structure shown is only for illustration, and the computer device may also include Figure 3 More or fewer components than shown, or with Figure 3 Different configurations shown. Figure 3 Each component shown in the figure can be implemented by hardware, software or a combination thereof.

[0034] Example 5 This embodiment provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for ultrasonically detecting kidneys in the above method embodiment is implemented.

[0035] The computer program includes program instructions. When the program instructions are executed by the computer, the computer can execute the methods provided by the above-mentioned method embodiments. For example, the method may include: a data set acquisition unit 201, used to acquire kidney ultrasound images and perform data annotation on multiple ultrasound images to obtain a data set; a data processing unit 202, used to generate unified ultrasound image data by performing data processing on the data in the data set; a feature extraction unit 203, used to obtain multi-scale features by extracting features from the ultrasound image data; a comparative analysis unit 204, which presets a kidney model ultrasound image, and compares and analyzes the multi-scale features with the preset kidney model ultrasound image based on an attention mechanism to obtain a kidney comparison result on a single frame image; a detection result generation unit 205, used to collect the single frame images with different comparison results, form an ultrasound image comparison data set, and obtain a kidney detection result based on the analysis of the comparison data set.

[0036] In the embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interface, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0037] In addition, the units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0038] Furthermore, the functional modules in each embodiment of the present application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0039] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A method for ultrasonically detecting kidneys, characterized in that: include: Step 1: Acquire kidney ultrasound images and annotate multiple ultrasound images to obtain a data set; Step 2: Processing the data in the dataset to generate unified ultrasound image data; extracting features from the ultrasound image data to obtain multi-scale features; Step 3: Preset a kidney model ultrasound image, and compare and analyze the multi-scale features with the preset kidney model ultrasound image based on the attention mechanism to obtain a kidney comparison result on a single frame image; Step 4: Collect the single-frame images with different comparison results and form an ultrasound image comparison data set, and obtain the kidney detection result based on the analysis of the comparison data set.

2. The method for ultrasonically detecting kidneys according to claim 1, wherein: In the step 3, the preset kidney model ultrasound image includes: the data range of the whole kidney image and the specific location information coordinates of the kidney.

3. The method for ultrasonic detection of kidneys according to claim 1 or 2, characterized in that: The step three also includes screening out target features that are different between the ultrasound image and the preset kidney model ultrasound image results through comparative analysis, and performing analysis and image enlargement and annotation based on the target features.

4. The method for ultrasonically detecting kidneys according to claim 3, wherein: The comparative analysis is performed based on the position information coordinates of the preset kidney model ultrasound image.

5. The method for ultrasonically detecting kidneys according to claim 1, wherein: The step of processing the data in the data set to generate unified ultrasound image data includes: Scaling the images in the data set frame by frame, and cropping them based on the center of the images to obtain cropped image data; Performing image enhancement on the cropped image data using a preset data enhancement method to obtain enhanced image data; The size and grayscale of the enhanced image data are normalized to generate the ultrasound image data with unified data.

6. The method for ultrasonic detection of kidneys according to claim 1, characterized in that: The multi-scale features are obtained by extracting features from the ultrasound image data, including: reducing the size of the ultrasound image layer by layer and applying Gaussian blur, obtaining high-frequency details through the difference of Gaussian pyramids, so as to respectively extract features of different modalities in the ultrasound image data to obtain the multi-scale features.

7. A device for ultrasonically detecting kidneys, characterized in that: include: a data set acquisition unit, configured to acquire kidney ultrasound images and perform data annotation on a plurality of the ultrasound images to obtain a data set; a data processing unit, configured to process the data in the data set to generate unified ultrasound image data; a feature extraction unit, configured to extract features from the ultrasound image data to obtain multi-scale features; A comparison and analysis unit is configured to preset a kidney model ultrasound image, and to compare and analyze the multi-scale features with the preset kidney model ultrasound image based on an attention mechanism to obtain a kidney comparison result on a single frame image; The detection result generating unit is used to collect the single frame images with different comparison results and form an ultrasound image comparison data set, and obtain the kidney detection result based on the analysis of the comparison data set.

8. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the method for ultrasonically detecting the kidney according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for ultrasonically detecting kidneys according to any one of claims 1 to 6 is implemented.

Images