Ultrasonic image quality evaluation method, device, equipment and medium

Abstract

The invention relates to the technical field of image quality evaluation, in particular to an ultrasonic image quality evaluation method and device, equipment and a medium. The method comprises the following steps: firstly, extracting a key structure of an individual from an ultrasonic image, evaluating the key structure to obtain a first evaluation score, dividing a display area for displaying the individual from the ultrasonic image, and then evaluating the display area to obtain a second evaluation score. And finally, according to the first evaluation score and the second evaluation score, obtaining an overall evaluation score, and using the overall evaluation score as a quality evaluation result of the ultrasonic image. The display area and the key structure on the ultrasonic image are main consideration factors of diagnosis, so that the key structure and the display area directly influence the quality of the ultrasonic image, and the accuracy of an evaluation result can be improved by evaluating the quality of the ultrasonic image according to the display area and the key structure.

Description

Technical Field

[0001] The present invention relates to the technical field of image quality assessment, and in particular to an ultrasonic image quality assessment method, device, equipment and medium. Background Art

[0002] Ultrasound imaging has become an important diagnostic tool for disease screening due to its immediate diagnostic results, low cost, and non-invasive nature. However, ultrasound image quality impacts diagnostic results, necessitating an assessment of ultrasound image quality to select high-quality images for subsequent diagnosis. Existing techniques assess ultrasound image quality based on image clarity. However, ultrasound image quality is not only dependent on clarity but also on key structures depicted. Therefore, assessing image quality solely based on clarity reduces the accuracy of the assessment.

[0003] In summary, the existing technology reduces the accuracy of ultrasound image quality assessment.

[0004] Therefore, the existing technology needs to be improved and enhanced. Summary of the Invention

[0005] To solve the above technical problems, the present invention provides an ultrasound image quality assessment method, device, equipment and medium, which solves the problem that the existing technology reduces the accuracy of ultrasound image quality assessment.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] In a first aspect, the present invention provides a method for evaluating ultrasound image quality, comprising:

[0008] Acquiring an ultrasound image of an individual, obtaining each key structure of the individual based on the ultrasound image, and evaluating the quality of each key structure to obtain a first evaluation score;

[0009] determining a display area on the ultrasound image for displaying the individual, and evaluating a quality of the display area based on a size of the display area to obtain a second evaluation score;

[0010] The quality of the ultrasound image is evaluated based on the first evaluation score and the second evaluation score to obtain an overall evaluation score of the ultrasound image.

[0011] In one implementation, obtaining various key structures of the individual based on the ultrasound image includes:

[0012] Extracting each original structure from the ultrasound image, and determining the category and confidence of each original structure;

[0013] The original structures are filtered according to the categories and the confidence levels to obtain the key structures of the individuals.

[0014] In one implementation, evaluating the quality of each of the key structures to obtain a first evaluation score includes:

[0015] A multi-branch regression model is applied to each of the key structures to obtain the clarity of each of the key structures, and the clarity of each of the key structures is used as the first evaluation score of each of the key structures.

[0016] In one implementation, determining the number of branches of the multi-branch regression model includes:

[0017] Determining the section type corresponding to the display area;

[0018] Determine the number of branches corresponding to the section type.

[0019] In one implementation, evaluating the quality of the display area based on the size of the display area to obtain a second evaluation score includes:

[0020] The size of the ultrasonic image is acquired, and a ratio of the size of the display area to the size of the ultrasonic image is calculated, and the ratio is used as a second evaluation score.

[0021] In one implementation, evaluating the quality of the ultrasound image according to the first evaluation score and the second evaluation score to obtain an overall evaluation score of the ultrasound image includes:

[0022] Determining the section type corresponding to the display area;

[0023] Judging whether there are mutually exclusive structures in each of the key structures according to the section type, and obtaining a mutual exclusion judgment result;

[0024] An overall evaluation score of the ultrasound image is obtained according to the first evaluation score, the second evaluation score, and the mutually exclusive determination result.

[0025] In one implementation, obtaining an overall evaluation score of the ultrasound image according to the first evaluation score, the second evaluation score, and the mutually exclusive determination result includes:

[0026] When the mutual exclusion determination result is that there are mutually exclusive structures, unqualified is used as the overall evaluation score of the ultrasound image;

[0027] When the mutual exclusion determination result is that there is no mutually exclusive structure, a weighted calculation is performed on the first evaluation score, the second evaluation score, and the mutual exclusion determination result to obtain an overall evaluation score of the ultrasound image.

[0028] In a second aspect, an embodiment of the present invention further provides an ultrasound image quality assessment device, wherein the device includes the following components:

[0029] a first evaluation module, configured to obtain an ultrasound image of an individual, obtain each key structure of the individual based on the ultrasound image, and evaluate the quality of each key structure to obtain a first evaluation score;

[0030] a second evaluation module, configured to determine a display area on the ultrasound image for displaying the individual, and evaluate a quality of the display area based on a size of the display area to obtain a second evaluation score;

[0031] An overall evaluation module is used to evaluate the quality of the ultrasound image according to the first evaluation score and the second evaluation score to obtain an overall evaluation score of the ultrasound image.

[0032] In a third aspect, an embodiment of the present invention further provides a terminal device, wherein the terminal device includes a memory, a processor, and an ultrasound image quality assessment program stored in the memory and executable on the processor, and when the processor executes the ultrasound image quality assessment program, the steps of the ultrasound image quality assessment method described above are implemented.

[0033] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which an ultrasound image quality assessment program is stored. When the ultrasound image quality assessment program is executed by a processor, the steps of the ultrasound image quality assessment method described above are implemented.

[0034] Beneficial Effects: The present invention first extracts the key structures of an individual from an ultrasound image and evaluates the key structures to obtain a first evaluation score. It also divides the ultrasound image into a display area for displaying the individual, then evaluates the display area to obtain a second evaluation score. Finally, an overall evaluation score is obtained based on the first and second evaluation scores, and the overall evaluation score is used as the quality assessment result of the ultrasound image. Because the display area and key structures on an ultrasound image are primary considerations for diagnosis, the key structures and display area directly affect the quality of the ultrasound image. Therefore, the present invention evaluates the quality of the ultrasound image based on the display area and key structures, thereby improving the accuracy of the assessment results. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is the overall flow chart of the present invention;

[0036] Figure 2 Schematic diagram of data interaction among the image structure detection module, image quality assessment regression module, and quality judgment feedback module in an embodiment of the present invention;

[0037] Figure 3 is a flowchart of an image structure detection module in an embodiment of the present invention;

[0038] Figure 4 is a schematic diagram of a display area in an embodiment of the present invention;

[0039] Figure 5 This is a flowchart of an image quality assessment regression module in an embodiment of the present invention;

[0040] Figure 6 This is a flow chart of a quality determination feedback module in an embodiment of the present invention;

[0041] Figure 7 This is a structural diagram of the ultrasonic image quality assessment device provided by the present invention;

[0042] Figure 8 This is a block diagram of the internal structure of a terminal device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The following is a clear and complete description of the technical solutions of the present invention in conjunction with the embodiments and the accompanying drawings. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0044] Research has found that ultrasound imaging has become an important diagnostic tool for disease screening due to its immediate diagnostic results, low cost, and non-invasive nature. However, ultrasound image quality impacts diagnostic results, necessitating an assessment of ultrasound image quality to select high-quality images for subsequent diagnosis. Existing techniques assess ultrasound image quality based on image clarity. However, ultrasound image quality is not only dependent on clarity but also on key structures depicted. Therefore, assessing image quality solely based on clarity reduces the accuracy of the assessment.

[0045] To solve the above technical problems, the present invention provides an ultrasound image quality assessment method, device, equipment and medium, which solves the problem that the existing technology reduces the accuracy of ultrasound image quality assessment.

[0046] The ultrasound image quality assessment method of this embodiment can be applied to a terminal device, which can be a terminal product with image processing function, such as a computer. Figure 1 As shown in , the ultrasound image quality assessment method specifically includes the following steps:

[0047] S100, acquiring an ultrasound image of an individual, obtaining key structures of the individual based on the ultrasound image, and evaluating the quality of each key structure to obtain a first evaluation score;

[0048] S200, determining a display area on the ultrasound image for displaying the individual, and evaluating a quality of the display area based on a size of the display area to obtain a second evaluation score;

[0049] S300 : Evaluate the quality of the ultrasound image according to the first evaluation score and the second evaluation score to obtain an overall evaluation score of the ultrasound image.

[0050] In this embodiment, Figure 2 As shown, the ultrasound image is input into the image structure detection module, which outputs the fetal display structure and display area, including key structures. The key structures and display area are then input into the image quality assessment regression module, which outputs an overall assessment score. Finally, the overall assessment score is input into the quality judgment feedback module, which determines whether the ultrasound image is acceptable and, in other words, whether it can be used for disease diagnosis.

[0051] The ultrasound image of this embodiment is not limited to a prenatal ultrasound image, but may also be other medical ultrasound images.

[0052] In this embodiment, extracting each key structure in step S100 includes the following specific steps: extracting each original structure from the ultrasound image and determining the category and confidence of each original structure; filtering each original structure based on the category and the confidence to obtain each key structure of the individual.

[0053] The ultrasound image in this embodiment is an ultrasound image after preprocessing. When the ultrasound image is an RGB image, the preprocessing includes converting the ultrasound image into a data element in the form of (H, W, C), where H and W represent the height and width of the ultrasound image, respectively, and C represents the number of channels of the ultrasound image.

[0054] When the ultrasound image is a grayscale image, the preprocessing includes converting the grayscale image into an RGB image, and then converting the RGB image into a data element in the form of (H, W, C).

[0055] The input size of the image structure detection module (ie, YOLO) is adjusted through the data element so that the module can accept the input of ultrasound images.

[0056] like Figure 3As shown in the figure, the ultrasound image after preprocessing is subjected to structure recognition and ROI (ROI is the display region of interest) recognition to obtain the original structure (the original structure is all the human body structures appearing on the ultrasound image) represented by the form of (x, y, h, w, c, p), where x and y are the center coordinates of the structure area, h and w are the height and width of the structure, and c and p represent the category and confidence of the structure. When the confidence is too low, it means that the currently detected structure is unreliable. The confidence threshold can be set to filter out unnecessary structures, so as to screen out key structures with a confidence greater than the threshold from the original structure.

[0057] Crop the display area including key structures from the ultrasound image, such as Figure 4 The fetal ultrasound image shown, Figure 4 The display area includes not only key structures such as the fetal spine, but also other structures such as the fetal umbilical cord and fetal abdomen. In other words, the display area presents the image of the entire fetus.

[0058] In this embodiment, step S100 evaluates the quality of the key structures, including the following specific steps: applying a multi-branch regression model to each of the key structures to obtain the clarity of each of the key structures, and using the clarity of each of the key structures as the first evaluation score of each of the key structures.

[0059] The number of key structures is equal to the number of branch networks contained in the multi-branch regression model, that is, one branch network processes one key structure to evaluate the clarity of the key structure to obtain a first evaluation score of the key structure.

[0060] The extracted key structures are input into the pre-trained image encoder, which extracts the feature information of the key structures, such as Figure 5 As shown in Figure 1, the features of the key structure are input into the branch network, and the branch network outputs a score that can represent the quality of the key structure (the value range of the score is: 0-1). Among them, the image encoder includes a convolutional network and a Transformer.

[0061] The method for determining the number of branch networks includes: determining the section type corresponding to the display area; and determining the number of branches corresponding to the section type.

[0062] That is, according to the section type of the display area, the number of key structures contained in the display area can be determined, and the number of branches is set to the number of key structures to obtain a branch network matching the key structures.

[0063] The section types in this embodiment include the fetal midsagittal section, the upper abdominal section, and the renal coronal section. The pre-stored "section: number of branches" file records the correspondence between the section type and the number of branches, so the number of branches can be known based on this file.

[0064] In this embodiment, each branch network is trained in the following manner:

[0065] Based on the pixel density and sharpness of key structures in the sample image, the sharpness labels of key structures in the sample image are annotated. The sample image is input into the branch network, and the predicted sharpness output by the branch network is obtained. A loss function is constructed based on the sharpness labels and the predicted sharpness, and the branch network is trained based on the loss function.

[0066] The branch network obtained by the above training can fully consider pixel density and sharpness when calculating clarity, where sharpness is the edge clarity of the key structure, and the edge clarity of the key structure is related to the image quality density. Therefore, the trained branch network can accurately calculate the clarity of the key structure.

[0067] In this embodiment, calculating the second evaluation score in step S200 includes: acquiring the size of the ultrasound image, calculating the ratio of the size of the display area to the size of the ultrasound image, and using the ratio as the second evaluation score.

[0068] This embodiment uses the pixel area s of the display area as the size of the display area, and the total pixel area S of the ultrasound image as the size of the ultrasound image. Where s = h*w, h is the height of the display area, w is the width of the display area; S = H*W, H is the height of the ultrasound image, W is the width of the ultrasound image. The second evaluation score is the ratio of s to S (that is, Figure 5 The second evaluation score represents the magnification ratio of the display area. A larger magnification ratio indicates a larger proportion of the display area in the ultrasound image, resulting in a clearer display area and better identification of the fetus in the ultrasound image. Therefore, this embodiment considers the magnification ratio of the display area when evaluating image quality, thereby improving the accuracy of image quality assessment.

[0069] In this embodiment, step S300 includes the following specific steps: determining the section type corresponding to the display area; judging whether there are mutually exclusive structures in each of the key structures based on the section type, and obtaining a mutual exclusion judgment result; when the mutual exclusion judgment result is that there are mutually exclusive structures, unqualified is used as the overall evaluation score of the ultrasound image; when the mutual exclusion judgment result is that there are no mutually exclusive structures, the first evaluation score, the second evaluation score and the mutual exclusion judgment result are weightedly calculated to obtain the overall evaluation score of the ultrasound image.

[0070] The types of sections to which the display area belongs include the fetal midsagittal section, the upper abdominal section, and the renal coronary section. In theory, each section has a corresponding fetal structure. For example, only fetal structures such as arms and chest should appear on the upper abdominal section, and the fetal structure such as legs should not appear. If two mutually exclusive fetal structures, arms and legs, appear on the upper abdominal section at the same time, the ultrasound image in the display area will be directly judged as unqualified, that is, the ultrasound image is not suitable for disease diagnosis.

[0071] When no mutually exclusive structures appear in the display area, scores and weights are assigned to the mutual exclusion determination results to participate in the calculation of the final overall evaluation score.

[0072] Finally, a weighted calculation is performed on the mutual exclusion judgment result in which no mutually exclusive structure exists, the first evaluation score of each key structure, and the second evaluation score corresponding to the display area to obtain an overall evaluation score.

[0073] This embodiment inputs the overall evaluation score output by the image quality evaluation regression module into the quality judgment feedback module, such as Figure 6 As shown in the figure, the quality judgment feedback module classifies the quality of ultrasound images according to the overall evaluation score. When the overall evaluation score is greater than or equal to 90, the image quality is judged to be excellent; when the overall evaluation score is greater than or equal to 80 and less than 90, the image quality is judged to be good; when the overall evaluation score is greater than or equal to 60 and less than 80, the image quality is judged to be qualified; when the overall evaluation score is less than 60, the image quality is judged to be unqualified.

[0074] The quality assessment feedback module also determines the structures that should appear in the section based on the section classification. It also uses key structural information to determine whether there are organs that should not be displayed simultaneously in the section. If mutually exclusive structures coexist in the image section (for example, organs that should not be displayed together), the image will be forcibly judged as unqualified to prevent incorrect images from passing the review. Mutually exclusive structures (such as the jaw and spine) are defined according to the prenatal standard section guidelines (ISUOG standards).

[0075] This embodiment also evaluates whether there is significant occlusion in the image and whether the boundary grayscale of key structures is appropriate to obtain the final evaluation result of the ultrasound image. The boundary grayscale refers to the division of different key structures in the image and is used to describe the brightness and darkness of the pixels in the image. Therefore, this embodiment uses boundary grayscale to evaluate ultrasound image quality. The resulting evaluation result can accurately reflect whether there are clear boundaries between key structures.

[0076] This embodiment also provides an ultrasound image quality assessment device, such as Figure 7 As shown, the device includes the following components:

[0077] A first evaluation module 01 is configured to obtain an ultrasound image of an individual, obtain key structures of the individual based on the ultrasound image, and evaluate the quality of each key structure to obtain a first evaluation score;

[0078] A second evaluation module 02 is configured to determine a display area on the ultrasound image for displaying the individual, and to evaluate the quality of the display area based on a size of the display area to obtain a second evaluation score;

[0079] The overall evaluation module 03 is configured to evaluate the quality of the ultrasound image according to the first evaluation score and the second evaluation score to obtain an overall evaluation score of the ultrasound image.

[0080] Based on the above embodiment, the present invention further provides a terminal device, whose principle block diagram can be shown as follows: Figure 8 As shown. The terminal device includes a processor, a memory, a network interface, and a display screen connected via a system bus. The processor of the terminal device is used to provide computing and control capabilities. The memory of the terminal device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The network interface of the terminal device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, an ultrasound image quality assessment method is implemented. The display screen of the terminal device can be a liquid crystal display or an electronic ink display.

[0081] Those skilled in the art will understand that Figure 8 The principle block diagram shown in the figure is only a block diagram of a partial structure related to the solution of the present invention, and does not constitute a limitation on the terminal device to which the solution of the present invention is applied. The specific terminal device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0082] In one embodiment, a terminal device is provided. The terminal device includes a memory, a processor, and an ultrasound image quality assessment program stored in the memory and executable on the processor. When the processor executes the ultrasound image quality assessment program, the following operating instructions are implemented:

[0083] Acquiring an ultrasound image of an individual, obtaining each key structure of the individual based on the ultrasound image, and evaluating the quality of each key structure to obtain a first evaluation score;

[0084] determining a display area on the ultrasound image for displaying the individual, and evaluating a quality of the display area based on a size of the display area to obtain a second evaluation score;

[0085] The quality of the ultrasound image is evaluated based on the first evaluation score and the second evaluation score to obtain an overall evaluation score of the ultrasound image.

[0086] Those skilled in the art will appreciate that all or part of the processes in the above-described embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the above-described embodiments. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAM bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM).

[0087] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for evaluating ultrasound image quality, characterized in that: include: Acquiring an ultrasound image of an individual, obtaining each key structure of the individual based on the ultrasound image, and evaluating the quality of each key structure to obtain a first evaluation score; determining a display area on the ultrasound image for displaying the individual, and evaluating a quality of the display area based on a size of the display area to obtain a second evaluation score; The quality of the ultrasound image is evaluated based on the first evaluation score and the second evaluation score to obtain an overall evaluation score of the ultrasound image.

2. The ultrasound image quality assessment method according to claim 1, wherein: Based on the ultrasound image, key structures of the individual are obtained, including: Extracting each original structure from the ultrasound image, and determining the category and confidence of each original structure; The original structures are filtered according to the categories and the confidence levels to obtain the key structures of the individuals.

3. The ultrasound image quality assessment method according to claim 1, wherein: Evaluate the quality of each of the key structures to obtain a first evaluation score, including: A multi-branch regression model is applied to each of the key structures to obtain the clarity of each of the key structures, and the clarity of each of the key structures is used as the first evaluation score of each of the key structures.

4. The ultrasound image quality assessment method according to claim 3, wherein: Determining the number of branches of the multi-branch regression model includes: Determining the section type corresponding to the display area; Determine the number of branches corresponding to the section type.

5. The ultrasound image quality assessment method according to claim 1, wherein: Evaluating the quality of the display area based on the size of the display area to obtain a second evaluation score includes: The size of the ultrasonic image is acquired, and a ratio of the size of the display area to the size of the ultrasonic image is calculated, and the ratio is used as a second evaluation score.

6. The ultrasound image quality assessment method according to claim 1, wherein: Evaluating the quality of the ultrasound image according to the first evaluation score and the second evaluation score to obtain an overall evaluation score of the ultrasound image includes: Determining the section type corresponding to the display area; Judging whether there are mutually exclusive structures in each of the key structures according to the section type, and obtaining a mutual exclusion judgment result; An overall evaluation score of the ultrasound image is obtained according to the first evaluation score, the second evaluation score, and the mutually exclusive determination result.

7. The ultrasound image quality assessment method according to claim 6, wherein: Obtaining an overall evaluation score of the ultrasound image according to the first evaluation score, the second evaluation score, and the mutually exclusive determination result, including: When the mutual exclusion determination result is that there are mutually exclusive structures, unqualified is used as the overall evaluation score of the ultrasound image; When the mutual exclusion determination result is that there is no mutually exclusive structure, a weighted calculation is performed on the first evaluation score, the second evaluation score, and the mutual exclusion determination result to obtain an overall evaluation score of the ultrasound image.

8. An ultrasound image quality assessment device, characterized in that: The device comprises the following components: a first evaluation module, configured to obtain an ultrasound image of an individual, obtain each key structure of the individual based on the ultrasound image, and evaluate the quality of each key structure to obtain a first evaluation score; a second evaluation module, configured to determine a display area on the ultrasound image for displaying the individual, and evaluate a quality of the display area based on a size of the display area to obtain a second evaluation score; An overall evaluation module is used to evaluate the quality of the ultrasound image according to the first evaluation score and the second evaluation score to obtain an overall evaluation score of the ultrasound image.

9. A terminal device, characterized in that: The terminal device includes a memory, a processor, and an ultrasound image quality assessment program stored in the memory and executable on the processor. When the processor executes the ultrasound image quality assessment program, the steps of the ultrasound image quality assessment method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores an ultrasound image quality assessment program, and when the ultrasound image quality assessment program is executed by the processor, the steps of the ultrasound image quality assessment method according to any one of claims 1 to 7 are implemented.

Images