Method, device and computer equipment for determining standard cross-section images of femur and humerus

Abstract

The present application relates to a method, device and computer equipment for determining a standard section image of a femur and humerus. The method comprises: obtaining an ultrasonic section image of a detection object; extracting section structure features of the ultrasonic section image; if the ultrasonic section image is preliminarily determined to be a standard section image based on the section structure features, extracting the bone structure contour of the ultrasonic section image to obtain the position information of the bone structure contour; if the bone structure contour is determined to be complete based on the position information of the bone structure contour, the ultrasonic section image is used as the target standard section image. The entire scheme obtains the section structure features of the ultrasonic section image, performs a preliminary standard section image judgment based on the section structure features, extracts the bone structure contour when the ultrasonic section image is a preliminary standard section image, and then performs a bone contour integrity judgment based on the bone structure contour, obtains the standard section image through two judgments, and improves the accuracy of determining the standard section image.

Description

Technical Field

[0001] The present application relates to the technical field of image processing, and in particular to a method, device and computer equipment for determining a standard section image of a femur and humerus. Background Art

[0002] In recent years, with the development of computer technology, deep learning has provided tremendous technical support for medical image segmentation and played an important role in improving the level of medical technology.

[0003] Ultrasound imaging technology has the advantages of simple operation, safety, non-destructiveness, and low cost. It has important value in reducing the rate of congenital malformations, maternal mortality, and fetal mortality. It is currently recognized as the preferred imaging method for prenatal diagnosis and birth defect screening. The detection and measurement of the femur and humerus is a routine item in obstetric examinations and is one of the important indicators for evaluating the development of the fetal skeletal system. For fetal ultrasound images, doctors mainly obtain the detection information of the target object through standard section images.

[0004] Since the anatomical structures to be segmented have huge differences in shape and size, the current image segmentation technology cannot obtain accurate femoral and humeral cross-sectional images when the target size varies greatly. Summary of the invention

[0005] Based on this, it is necessary to provide a method, device, computer equipment, computer-readable storage medium and computer program product for determining standard femoral and humeral cross-section images that can obtain accurate femoral and humeral cross-section images in real time from a video stream to address the above technical problems.

[0006] In a first aspect, the present application provides a method for determining a standard section image of a femur and humerus. The method comprises:

[0007] Acquire an ultrasonic cross-sectional image of the detected object;

[0008] Extracting the section structure features of the ultrasound section image;

[0009] If the ultrasonic section image is preliminarily determined to be a standard section image based on the section structure characteristics, the bone structure contour of the ultrasonic section image and the position information of the bone structure contour are extracted;

[0010] If the bone structure contour is determined to be complete based on the position information of the bone structure contour, the ultrasonic section image is used as the target standard section image.

[0011] In one of the embodiments, it also includes: determining the section type data of the ultrasound section image based on the section structure characteristics; obtaining the section structure feature weight corresponding to the section type data; obtaining the confidence of the section structure feature, and calculating according to the confidence of the section structure feature and the section structure feature weight to obtain the ultrasound section image score; and preliminarily determining whether the ultrasound section image is a standard section image based on the relationship between the ultrasound section image score and a preset score.

[0012] In one embodiment, the ultrasound section image includes at least one section image; if the bone structure contour is determined to be complete based on the position information of the bone structure contour, then using the ultrasound section image as the target standard section image includes: if the bone structure contour is determined to be complete based on the position information of the bone structure contour, then scoring the bone structure contour to obtain a bone contour score; calculating based on the ultrasound section image score and the bone contour score to obtain a section image score; determining a maximum section image score based on the section image score; and using the ultrasound section image corresponding to the maximum section image score as the target standard section image.

[0013] In one of the embodiments, extracting the bone structure contour of the ultrasonic section image and obtaining the position information of the bone structure contour includes: intercepting the ultrasonic section image according to the section structure characteristics to obtain the section intercepted image; extracting the bone structure contour of the section intercepted image, determining the minimum circumscribed rectangle of the bone structure contour, and obtaining the position information of the bone structure contour based on the minimum circumscribed rectangle of the bone structure contour.

[0014] In one of the embodiments, extracting the bone structure contour of the cross-section image, determining the minimum circumscribed rectangle of the bone structure contour, and obtaining the position information of the bone structure contour based on the minimum circumscribed rectangle of the bone structure contour includes: determining a binarization threshold based on an average pixel value of the cross-section image; binarizing the cross-section image based on the binarization threshold to obtain a binary cross-section image; extracting the bone structure contour of the binary cross-section image, determining the minimum circumscribed rectangle of the bone structure contour, and obtaining the position information of the bone structure contour based on the minimum circumscribed rectangle of the bone structure contour.

[0015] In one embodiment, the method further includes: extracting the length data, width data, first boundary data, and second boundary data of the bone structure contour according to the position information of the bone structure contour; judging whether the bone structure contour is complete according to the length data, width data, first boundary data, second boundary data of the bone structure contour and a preset contour integrity judgment rule; if the bone structure contour is determined to be complete according to the position information of the bone structure contour, after taking the ultrasonic section image as the target standard section image, the method further includes:

[0016] The length data of the bone structure contour is used as the measured length of the bone in the target standard section image.

[0017] In a second aspect, the present application also provides a device for determining a standard section image of a femur and humerus. The device comprises:

[0018] An acquisition module, used for acquiring an ultrasonic cross-sectional image of the detected object;

[0019] A feature extraction module, used to extract the section structure features of the ultrasound section image;

[0020] A preliminary judgment module is used to extract the bone structure contour of the ultrasonic section image and obtain the position information of the bone structure contour if the ultrasonic section image is preliminarily judged to be a standard section image according to the section structure characteristics;

[0021] The contour judgment module is used to use the ultrasonic section image as the target standard section image if it is judged that the bone structure contour is complete according to the position information of the bone structure contour.

[0022] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:

[0023] Acquire an ultrasonic cross-sectional image of the detected object;

[0024] Extracting the section structure features of the ultrasound section image;

[0025] If the ultrasonic section image is preliminarily determined to be a standard section image based on the section structure characteristics, the bone structure contour of the ultrasonic section image is extracted to obtain the position information of the bone structure contour;

[0026] If the bone structure contour is determined to be complete based on the position information of the bone structure contour, the ultrasonic section image is used as the target standard section image.

[0027] In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0028] Acquire an ultrasonic cross-sectional image of the detected object;

[0029] Extracting the section structure features of the ultrasound section image;

[0030] If the ultrasonic section image is preliminarily determined to be a standard section image based on the section structure characteristics, the bone structure contour of the ultrasonic section image is extracted to obtain the position information of the bone structure contour;

[0031] If the bone structure contour is determined to be complete based on the position information of the bone structure contour, the ultrasonic section image is used as the target standard section image.

[0032] In a fifth aspect, the present application further provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0033] Acquire an ultrasonic cross-sectional image of the detected object;

[0034] Extracting the section structure features of the ultrasound section image;

[0035] If the ultrasonic section image is preliminarily determined to be a standard section image based on the section structure characteristics, the bone structure contour of the ultrasonic section image is extracted to obtain the position information of the bone structure contour;

[0036] If the bone structure contour is determined to be complete based on the position information of the bone structure contour, the ultrasonic section image is used as the target standard section image.

[0037] The above-mentioned method, device, computer equipment, storage medium and computer program product for determining the standard section image of the femur and humerus obtain an ultrasonic section image of the detection object; extract the section structure features of the ultrasonic section image; if the ultrasonic section image is preliminarily determined to be a standard section image based on the section structure features, the bone structure contour of the ultrasonic section image is extracted to obtain the position information of the bone structure contour; if the bone structure contour is determined to be complete based on the position information of the bone structure contour, the ultrasonic section image is used as the target standard section image. The entire scheme obtains the section structure features of the ultrasonic section image, makes a preliminary standard section image judgment based on the section structure features, extracts the bone structure contour when the ultrasonic section image is a preliminary standard section image, and then makes a bone contour integrity judgment based on the bone structure contour, obtains the standard section image through two judgments, and improves the accuracy of determining the standard section image. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 A schematic diagram of a process of determining a standard section image of a femur and humerus in one embodiment;

[0039] Figure 2 It is a schematic diagram of a process of determining whether an ultrasound section image is a standard section image in one embodiment;

[0040] Figure 3 is a schematic diagram of a cross-sectional structural feature in another embodiment;

[0041] Figure 4 A schematic diagram of a binary section intercepting an image in one embodiment;

[0042] Figure 5is a schematic diagram of position information of a bone contour in one embodiment;

[0043] Figure 6 It is a structural block diagram of a device for determining a standard section image of a femur and humerus in one embodiment;

[0044] Figure 7 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0046] At present, during prenatal ultrasound examinations, in addition to obtaining standard long-axis cross-sectional images of the fetus' femur and humerus, doctors also need to manually measure the grooves of the femur and humerus. The above operations require many steps, time-consuming and laborious, and doctors need to perform about 10 steps and spend 3-5 minutes. In addition, there may be obvious individual differences in scanning techniques, cross-sectional judgments, and measurement standards among doctors. There is a lack of unified objective standards, and the measurement results of different doctors are poorly reproducible, resulting in frequent missed and misdiagnoses in clinical practice. Traditional methods cannot obtain accurate standard cross-sectional images.

[0047] In order to solve the problems of existing doctors' standard scanning techniques, judgment criteria and measurement of femur and humerus long axis sections, lack of unified objective standards, many manual measurement steps, poor reproducibility between doctors, etc., this application provides a method for determining femur and humerus standard section images, and bone measurement can also be performed based on the obtained standard section images. This application can help ultrasound physicians automatically obtain femur and humerus standard section images and automatically measure bone length, while meeting clinical real-time needs.

[0048] In one embodiment, Figure 1 As shown, a method for determining a standard cross-section image of a femur and humerus is provided. This embodiment is illustrated by applying the method to a terminal. It is understandable that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0049] Step 102: Acquire an ultrasonic cross-sectional image of the detection object.

[0050] Specifically, using a three-dimensional ultrasound device, the doctor moves the ultrasound probe to scan the object to be tested, obtains an ultrasound video stream, and the terminal obtains a frame of image from the ultrasound video stream in real time to obtain an ultrasound section image of the object to be tested. The object to be tested may be the femur or humerus of a fetus, and the ultrasound section image is an ultrasound section image of the femur and humerus.

[0051] Step 104: extracting the section structure features of the ultrasound section image.

[0052] The section structure characteristics are the section structure characteristics of the femur long axis or the section structure characteristics of the humerus long axis. The section structure characteristics of the femur long axis include the femur, the distal femoral epiphysis, the proximal femoral epiphysis, and the femoral contour; the section structure characteristics of the humerus long axis include the humerus, the distal humeral epiphysis, the proximal humeral epiphysis, and the humeral contour.

[0053] Specifically, the terminal inputs the ultrasonic section image into the section detection model for detection, obtains the section structure characteristics, and determines the section type according to the section structure characteristics. The section type is the femoral long axis section or the humeral long axis section.

[0054] The number of section types determined by the section structure features detected by our section detection model. For example, a frame of ultrasound section image is input into the section detection model for detection, and the detection results are: 'femur', 'proximal femoral epiphysis', 'femoral contour' and 'distal humeral epiphysis'. In the detection results, there are 3 small structures belonging to the long axis section of the femur, and only 1 small structure belonging to the long axis section of the humerus. Because 3>1, we judge this section as the long axis section of the femur. If the number of small structures belonging to the two sections is equal, this frame of the image is discarded and no subsequent processing is performed on it.

[0055] The section detection model is trained based on a deep learning model and a standard section image dataset. The deep learning model can be a yolov5 model or other target detection models. The training dataset is obtained from the fetal ultrasound volume obtained by the three-dimensional ultrasound device and the section structure features of the two sections (femur long axis section, humerus long axis section) manually annotated by ultrasound experts for each three-dimensional ultrasound volume, as well as the size data of the section structure features (a total of 8 small structures in the femur long axis section and the humerus long axis section, and the annotation information is the location information and category information of the small structures in the image). After the ultrasound doctor annotates the structural features of the standard section images, 80% of the standard section images in the dataset are used as the training set, 10% of the standard section images are used as the validation set, and 10% as the test set. The trained model is used to detect the small structures of each section, and the section type recognition is realized according to the existence of small structure detection (the number of small structures belonging to each section), and the more standard section images are preliminarily screened out.

[0056] Step 106: If the ultrasonic section image is preliminarily determined to be a standard section image based on the section structure characteristics, the bone structure contour of the ultrasonic section image is extracted to obtain the position information of the bone structure contour.

[0057] Specifically, the terminal determines the section type according to the number of section structures belonging to each section type, calculates the score of the ultrasound section image under the corresponding section type, and preliminarily determines whether the ultrasound section image is a standard section image according to the ultrasound section image score. If the ultrasound section image is preliminarily determined to be a standard section image according to the ultrasound section image score, the image of the area within the bone detection frame is further extracted, and the image of the area is binarized, and the binarization threshold is 1.65 times the average pixel value of the area, and then all contours are extracted from the binarized image area according to the findContours method in the openCV library, and the largest contour among these contours is used as the bone contour, and the minAreaRect method in the openCV library is used to obtain the minimum circumscribed rectangle of the bone contour to obtain the position information of the bone contour.

[0058] Step 108: If the bone structure contour is determined to be complete based on the position information of the bone structure contour, the ultrasonic section image is used as the target standard section image.

[0059] Specifically, the terminal determines whether the bone contour is complete based on the position information of the bone contour. If the bone contour is complete, the ultrasonic section image is used as the target standard section image. After that, the center point connection line of the two short sides of the minimum circumscribed rectangle of the bone structure contour is used as the measured length of the bone in the section, thereby realizing automatic bone measurement.

[0060] In the above-mentioned method for determining the standard section image of the femur and humerus, an ultrasonic section image of the detection object is obtained; the section structure features of the ultrasonic section image are extracted; if the ultrasonic section image is preliminarily determined to be a standard section image based on the section structure features, the bone structure contour of the ultrasonic section image is extracted, and the minimum circumscribed rectangle of the bone structure contour is determined, and the position information of the bone structure contour is obtained based on the minimum circumscribed rectangle; if the bone structure contour is determined to be complete based on the position information of the bone structure contour, the ultrasonic section image is used as the target standard section image. The entire scheme obtains the section structure features of the ultrasonic section image, performs a preliminary standard section image judgment based on the section structure features, extracts the bone structure contour when the ultrasonic section image is a preliminary standard section image, and then performs a bone contour integrity judgment based on the bone structure contour, obtains the standard section image through two judgments, and improves the accuracy of determining the standard section image.

[0061] In an optional embodiment, if Figure 2 As shown, it also includes:

[0062] Step 202: Determine the slice type data of the ultrasound slice image according to the slice structure characteristics.

[0063] Among them, the section types include femoral long axis section and humeral long axis section. The small structures of femoral long axis section include femur, distal femoral epiphysis, proximal femoral epiphysis, and femoral contour; the small structures of humeral long axis section include humerus, distal humeral epiphysis, proximal humeral epiphysis, and humeral contour.

[0064] Specifically, the terminal determines the section type according to the number of section structures belonging to each section type. If the number of small structures belonging to the femoral section is greater than the number of small structures of the humeral section, the section type of the ultrasound section image is the femoral section. If the number of small structures belonging to the humeral section is greater than the number of small structures of the femoral section, the section type of the ultrasound section image is the humeral section. If the number of small structures belonging to the humeral section is equal to the number of small structures of the femoral section, the type of the ultrasound section image cannot be determined and the ultrasound section image is discarded.

[0065] Step 204: Obtain the section structure feature weight corresponding to the section type data.

[0066] Specifically, the terminal obtains the locally stored section structure feature weight corresponding to the section type data.

[0067] Step 206, obtaining the confidence of the section structure feature, and performing calculation according to the confidence of the section structure feature and the section structure feature weight to obtain the ultrasound section image score.

[0068] The section structure feature weight includes the weight of each small structure corresponding to the section type.

[0069] Specifically, the confidence of each small structure corresponding to the section type in the ultrasound section image is obtained, and the confidence of each small structure corresponding to the section type in the ultrasound section image is multiplied by the corresponding small structure weight to obtain the ultrasound section image score.

[0070] For each ultrasound section image, score is given according to the existence of small structures detected by the section detection model and the confidence of each small structure (the model will give a corresponding confidence for each small structure detected, which is a decimal between 0 and 1). The score is 100, and scores greater than or equal to 80 are considered standard. The higher the score, the more standard it is. A weight is assigned to each small structure in advance. For example, in the femoral section type, the weight of the femur is 50, the weight of the proximal femoral epiphysis is 30, the weight of the distal femoral epiphysis is 30, and the weight of the femoral contour is 20; in the humeral section type, the weight of the humerus is 50, the weight of the proximal humeral epiphysis is 30, the weight of the distal humeral epiphysis is 30, and the weight of the humeral contour is 20. This weight can be determined according to the actual situation of your own model.

[0071] For example, if the confidence level detected by our model is: femur: 0.8, proximal femoral epiphysis: 0.7, distal femoral epiphysis: 0.5, femoral contour: 0.7, then the section is judged as the femoral long axis section, and the score is: score = min (100, 50*0.8 + 30*0.7 + 30*0.5 + 20*0.7) = 90. The formula for calculating the score is also determined according to the actual situation.

[0072] If a frame of image contains both small structures of the femoral long axis section and small structures of the humeral long axis section, for example, the model detects: femur, proximal femoral epiphysis, femoral contour, distal humeral epiphysis, then the frame of image is judged as the femoral long axis section, and the small structure of the distal humeral epiphysis is not taken into account when calculating the score.

[0073] Step 208: Preliminarily determine whether the ultrasound section image is a standard section image based on the relationship between the ultrasound section image score and a preset score threshold.

[0074] The preset score threshold is the corresponding minimum score of the standard section image.

[0075] Specifically, if the ultrasound section image score is greater than or equal to a preset score threshold, it is preliminarily determined whether the ultrasound section image is a standard section image.

[0076] In an optional embodiment, the ultrasound section image includes at least one section image; if the bone structure contour is determined to be complete based on the position information of the bone structure contour, then using the ultrasound section image as the target standard section image includes: if the bone structure contour is determined to be complete based on the position information of the bone structure contour, then scoring the bone structure contour to obtain a bone contour score; calculating based on the ultrasound section image score and the bone contour score to obtain a section image score; determining a maximum section image score based on the section image score; and using the ultrasound section image corresponding to the maximum section image score as the target standard section image.

[0077] Specifically, the terminal obtains multiple frames of section images from the ultrasound video stream, and first extracts the section structure features of the ultrasound section image for each frame of the section image. If the ultrasound section image is preliminarily determined to be a standard section image based on the section structure features, the bone structure contour of the ultrasound section image is extracted to obtain the position information of the bone structure contour. Whether the bone structure contour is complete is determined based on the position information of the bone structure contour. If the bone contour is determined to be complete based on the position information of the bone structure contour, the bone structure contour score is obtained by multiplying the average pixel value in the bone detection frame in the ultrasound image by a preset value. The ultrasound section image score is added to the bone structure contour score to obtain the section image score. The highest score is selected from the section image scores as the maximum section image score, and the ultrasound section image corresponding to the maximum section image score is used as the target standard section image.

[0078] Since the higher the pixel value of the bone, the more obvious the bone contour is, it is considered to be a more standard section. The contour score calculation formula is θ / 30 (θ is the average pixel value of the bone detection frame). After obtaining the target standard section image, the length of the bone to be measured can be calculated based on the target standard section image. The length of the bone in the bone detection frame is multiplied by the scale corresponding to the section to obtain the actual bone length.

[0079] In an optional embodiment, extracting the bone structure contour of the ultrasonic section image and obtaining the position information of the bone structure contour includes: intercepting the ultrasonic section image according to the section structure characteristics to obtain a section intercept image; extracting the bone structure contour of the section intercept image to obtain the position information of the bone structure contour.

[0080] Specifically, when the section detection model detects and outputs the section structure features of the ultrasound section image, it also outputs the location information of the small structure of the section. The terminal extracts the image of the small structure corresponding to the section type, that is, extracts the image area of ​​the bone detection frame to obtain the section interception image, extracts the bone contour of the section interception image according to the findContours method in the openCV library, and uses the minAreaRect method in the openCV library to obtain the minimum circumscribed rectangle of the bone contour to obtain the location information of the bone contour. Figure 3 As shown, the images where the three small structures are located are cut out to obtain the cross-sectional images.

[0081] In an optional embodiment, extracting the bone structure contour of the cross-section image, determining the minimum circumscribed rectangle of the bone structure contour, and obtaining the position information of the bone structure contour based on the minimum circumscribed rectangle of the bone structure contour includes: determining a binarization threshold based on an average pixel value of the cross-section image; binarizing the cross-section image based on the binarization threshold to obtain a binary cross-section image; extracting the bone structure contour of the binary cross-section image, determining the minimum circumscribed rectangle of the bone structure contour, and obtaining the position information of the bone structure contour based on the minimum circumscribed rectangle of the bone structure contour.

[0082] Specifically, since different cross-sectional images have obvious differences in brightness, a binarization threshold is dynamically selected according to different images to binarize the cross-sectional images. The terminal calculates the average pixel value of the cross-sectional image, multiplies the average pixel value by a preset binarization coefficient, and obtains a binarization threshold. According to the binarization threshold, the cross-sectional image is binarized to obtain a binary cross-sectional image. Then, the bone contour and the minimum circumscribed rectangle of the bone contour of the binary cross-sectional image are extracted from the binary cross-sectional image according to the findContours method and minAreaRect method in the openCV library, respectively, to obtain the bone contour and the position information of the bone contour. The preset binarization coefficient can be used as 1.65, which is data obtained through a large number of experiments.

[0083] For a binary slice image, there may be multiple contours, such as Figure 4 As shown, there are contours 1, 2, 3, 4... and so on. For each contour, calculate its length in the horizontal direction, and then find the contour with the largest length value, which is the bone contour (contour 2 is the bone contour).

[0084] In this embodiment, the binarization threshold is dynamically determined according to the average pixel value of the cut-plane intercepted image, so that more accurate bone contours and position information of the bone contours can be extracted, and the applicability of the method is enhanced.

[0085] In an optional embodiment, the method further includes: extracting the length data, width data, first boundary data, and second boundary data of the bone structure contour according to the position information of the bone structure contour; and judging whether the bone structure contour is complete according to the length data, width data, first boundary data, second boundary data of the bone structure contour and a preset contour integrity judgment rule.

[0086] The bone structure contour is the bone contour, the position information of the bone structure contour is the position information of the bone contour, and the position information of the bone contour includes the length data, width data, first boundary data, and second boundary data of the bone contour. The first boundary data is the distance from the leftmost side of the bone contour to the left boundary of the ultrasonic section image, and the second boundary data is the distance from the rightmost side of the bone contour to the right boundary of the ultrasonic section image.

[0087] The preset contour integrity judgment rules include that the first boundary data is less than the contour transverse data multiplied by the preset length ratio, the second boundary data is less than the contour length data multiplied by the preset length ratio, and the ratio of the length data to the width data of the bone contour is within the preset ratio range. The contour transverse data is the sum of the first boundary data, the second boundary data and the length data of the bone contour. The preset length ratio can be 0.25, and the preset ratio range can be (3,9). Figure 5 As shown, if the first boundary data is w1, the length data of the bone contour is w2, the second boundary data is w3, and the width data is w4, then the preset contour integrity judgment rule is:

[0088] (1)w1<(w1+w2+w3)*0.25;

[0089] (2)w3<(w1+w2+w3)*0.25;

[0090] (3)3 <w2 / w4<9。

[0091] Specifically, the position information of the bone structure contour is extracted to obtain the length data, width data, first boundary data, and second boundary data of the bone structure contour. The contour transverse data is calculated based on the length data, first boundary data, and second boundary data of the bone structure contour, and it is determined whether the first boundary data is less than the contour transverse data multiplied by a preset length ratio, whether the second boundary data is less than the contour length data multiplied by a preset length ratio, and whether the ratio of the length data to the width data of the bone contour is within a preset ratio range, and the bone structure contour is determined to be complete. If the first boundary data is less than the contour transverse data multiplied by a preset length ratio, the second boundary data is less than the contour length data multiplied by a preset length ratio, and the ratio of the length data to the width data of the bone contour is within a preset ratio range, the bone structure contour is determined to be complete, and the ultrasonic section image is used as the target standard section image. The preset contour integrity judgment rule is obtained through a large number of clinical practice summaries.

[0092] In order to facilitate understanding of the technical solution provided in the embodiment of the present application, the method for determining the standard section image of the femur and humerus provided in the embodiment of the present application is briefly described with the complete process of determining the standard section image of the femur and humerus:

[0093] (1) Acquire an ultrasonic cross-sectional image of the detection object, wherein the ultrasonic cross-sectional image includes at least one cross-sectional image.

[0094] (2) Extract the cross-sectional structural features of the ultrasound cross-sectional image.

[0095] (3) Determine the section type data of the ultrasound section image based on the section structure characteristics; obtain the section structure feature weight corresponding to the section type data; obtain the confidence of the section structure feature, and calculate the ultrasound section image score based on the confidence of the section structure feature and the section structure feature weight; and preliminarily determine whether the ultrasound section image is a standard section image based on the relationship between the ultrasound section image score and a preset score threshold.

[0096] (4) If the ultrasound section image is preliminarily determined to be a standard section image based on the section structure characteristics, the ultrasound section image is intercepted according to the section structure characteristics to obtain a section intercepted image; a binarization threshold is determined based on the average pixel value of the section intercepted image; the section intercepted image is binarized based on the binarization threshold to obtain a binary section intercepted image; the bone structure contour of the binary section intercepted image is extracted to determine the minimum circumscribed rectangle of the bone structure contour, and the position information of the bone structure contour is obtained based on the minimum circumscribed rectangle of the bone structure contour.

[0097] (5) Extracting the position information of the bone structure contour to obtain the length data, width data, first boundary data, and second boundary data of the bone structure contour; judging whether the bone structure contour is complete based on the length data, width data, first boundary data, and second boundary data of the bone structure contour and preset contour integrity judgment rules.

[0098] (6) If the bone structure contour is determined to be complete based on the length data, width data, first boundary data, second boundary data and preset contour integrity judgment rules, a score is calculated for the bone structure contour to obtain a bone contour score; a calculation is performed based on the ultrasound section image score and the bone contour score to obtain a section image score; a maximum section image score is determined based on the section image score; and the ultrasound section image corresponding to the maximum section image score is used as the target standard section image.

[0099] (7) Obtain bone measurement data of the object to be detected based on the length data of the bone structure contour in the target standard section image.

[0100] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0101] Based on the same inventive concept, the embodiment of the present application also provides a femur-humerus standard section image determination device for implementing the femur-humerus standard section image determination method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more femur-humerus standard section image determination device embodiments provided below can refer to the limitations of the femur-humerus standard section image determination method above, and will not be repeated here.

[0102] In one embodiment, Figure 6 As shown, a device for determining a standard section image of a femur and humerus is provided, comprising: an acquisition module 602, a feature extraction module 604, a preliminary judgment module 606 and a contour judgment module 608, wherein:

[0103] The acquisition module 602 is used to acquire an ultrasonic cross-sectional image of the detection object.

[0104] The feature extraction module 604 is used to extract the section structure features of the ultrasound section image.

[0105] The preliminary determination module 606 is used to extract the bone structure contour of the ultrasonic section image and obtain the position information of the bone structure contour if the ultrasonic section image is preliminarily determined to be a standard section image according to the section structure characteristics.

[0106] The contour judgment module 608 is used to use the ultrasonic section image as the target standard section image if it is determined that the bone structure contour is complete according to the position information of the bone structure contour.

[0107] In an optional embodiment, the preliminary judgment module 606 is also used to determine the section type data of the ultrasound section image based on the section structure characteristics; obtain the section structure feature weight corresponding to the section type data; obtain the confidence of the section structure feature, and calculate according to the confidence of the section structure feature and the section structure feature weight to obtain the ultrasound section image score; and preliminarily judge whether the ultrasound section image is a standard section image based on the relationship between the ultrasound section image score and a preset score threshold.

[0108] In an optional embodiment, the ultrasound section image includes at least one section image; the contour judgment module 608 is also used to calculate the score of the bone structure contour to obtain a bone contour score if the bone structure contour is determined to be complete based on the position information of the bone structure contour; calculate based on the ultrasound section image score and the bone contour score to obtain a section image score; determine the maximum section image score based on the section image score; and use the ultrasound section image corresponding to the maximum section image score as the target standard section image.

[0109] In an optional embodiment, the preliminary judgment module 606 is also used to intercept the ultrasonic section image according to the section structure characteristics to obtain a section intercepted image; extract the bone structure contour of the section intercepted image, determine the minimum circumscribed rectangle of the bone structure contour, and obtain the position information of the bone structure contour based on the minimum circumscribed rectangle of the bone structure contour.

[0110] In an optional embodiment, the preliminary judgment module 606 is also used to determine a binarization threshold based on the average pixel value of the cross-section image; binarize the cross-section image based on the binarization threshold to obtain a binary cross-section image; extract the bone structure contour of the binary cross-section image, determine the minimum circumscribed rectangle of the bone structure contour, and obtain the position information of the bone structure contour based on the minimum circumscribed rectangle of the bone structure contour.

[0111] In an optional embodiment, the contour judgment module 608 is also used to extract the position information of the bone structure contour to obtain the length data, width data, first boundary data, and second boundary data of the bone structure contour; and judge whether the bone structure contour is complete based on the length data, width data, first boundary data, second boundary data of the bone structure contour and preset contour integrity judgment rules.

[0112] Each module in the above-mentioned standard section image determination device can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0113] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 7 As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer 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 the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a standard section image determination method is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a button, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0114] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0115] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0116] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0117] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0118] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0119] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0120] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0121] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for determining a standard section image of a femur and humerus, characterized in that: The method comprises: Acquire an ultrasonic cross-sectional image of the detection object; the ultrasonic cross-sectional image is an ultrasonic cross-sectional image of the femur and humerus; Extracting the section structure feature of the ultrasonic section image; the section structure feature is the long axis section structure feature of the femur or the long axis section structure feature of the humerus; According to the section structure characteristics, determine the section type data of the ultrasonic section image, the section types include the femoral long axis section and the humeral long axis section, the small structures of the femoral long axis section include the femur, the distal femoral epiphysis, the proximal femoral epiphysis, and the femoral contour, and the small structures of the humeral long axis section include: the humerus, the distal humeral epiphysis, the proximal humeral epiphysis, and the humeral contour; obtain the section structure feature weight corresponding to the section type data, the section structure feature weight includes the weight of each of the small structures corresponding to the section type data; obtain the confidence of each of the small structures corresponding to the section type data in the ultrasonic section image, calculate according to the confidence of each of the small structures corresponding to the section type data and the weight of each of the small structures, and obtain the ultrasonic section image score; according to the relationship between the ultrasonic section image score and the preset score, preliminarily judge whether the ultrasonic section image is a standard section image; If the ultrasonic section image is preliminarily determined to be a standard section image based on the section structure characteristics, the bone structure contour of the ultrasonic section image is extracted to obtain position information of the bone structure contour; If it is determined that the bone structure contour is complete according to the position information of the bone structure contour, the ultrasonic section image is used as the target standard section image.

2. The method according to claim 1, characterized in that The ultrasonic section image includes at least one section image; If it is determined according to the position information of the bone structure contour that the bone structure contour is complete, then using the ultrasonic section image as the target standard section image comprises: If it is determined according to the position information of the bone structure contour that the bone structure contour is complete, a score is calculated for the bone structure contour to obtain a bone contour score; Calculating according to the ultrasound section image score and the bone contour score to obtain a section image score; Determining a maximum section image score according to the section image score; The ultrasound section image corresponding to the maximum section image score is used as the target standard section image.

3. The method according to claim 1, characterized in that Extracting the bone structure contour of the ultrasonic section image to obtain the position information of the bone structure contour includes: intercepting the ultrasonic section image according to the section structure characteristics to obtain a section intercept image; The bone structure contour of the cut surface intercepted image is extracted, the minimum circumscribed rectangle of the bone structure contour is determined, and the position information of the bone structure contour is obtained according to the minimum circumscribed rectangle of the bone structure contour.

4. The method according to claim 3, characterized in that The step of extracting the bone structure contour of the cut surface intercepted image, determining the minimum circumscribed rectangle of the bone structure contour, and obtaining the position information of the bone structure contour according to the minimum circumscribed rectangle of the bone structure contour comprises: Determining a binarization threshold value according to an average pixel value of the image intercepted by the section; Binarizing the section intercepted image according to the binarization threshold to obtain a binary section intercepted image; The bone structure contour of the binary section intercepted image is extracted, the minimum circumscribed rectangle of the bone structure contour is determined, and the position information of the bone structure contour is obtained according to the minimum circumscribed rectangle of the bone structure contour.

5. The method according to claim 1, characterized in that Also includes: Extracting according to the position information of the bone structure contour to obtain length data, width data, first boundary data, and second boundary data of the bone structure contour; Determining whether the bone structure contour is complete according to the length data, width data, first boundary data, second boundary data and preset contour integrity judgment rules of the bone structure contour; If the bone structure contour is determined to be complete according to the position information of the bone structure contour, after taking the ultrasonic section image as the target standard section image, the method further includes: The length data of the bone structure contour is used as the measured length of the bone in the target standard section image.

6. A device for determining a standard section image of a femur and humerus, characterized in that: The device comprises: An acquisition module is used to acquire an ultrasonic cross-sectional image of the detection object; the ultrasonic cross-sectional image is an ultrasonic cross-sectional image of the femur and humerus; A feature extraction module, used to extract the section structure feature of the ultrasonic section image; the section structure feature is the femoral long axis section structure feature or the humeral long axis section structure feature; A preliminary judgment module is used to determine the section type data of the ultrasonic section image according to the section structure characteristics, the section types include the femoral long axis section and the humeral long axis section, the small structures of the femoral long axis section include the femur, the distal femoral epiphysis, the proximal femoral epiphysis, and the femoral contour, and the small structures of the humeral long axis section include: the humerus, the distal humeral epiphysis, the proximal humeral epiphysis, and the humeral contour; obtain the section structure feature weight corresponding to the section type data, the section structure feature weight includes the weight of each of the small structures corresponding to the section type data; obtain the confidence of each of the small structures corresponding to the section type data in the ultrasonic section image, calculate according to the confidence of each of the small structures corresponding to the section type data and the weight of each of the small structures, and obtain the score of the ultrasonic section image; according to the relationship between the ultrasonic section image score and the preset score, preliminarily judge whether the ultrasonic section image is a standard section image; The preliminary judgment module is further used to extract the bone structure contour of the ultrasonic section image and obtain the position information of the bone structure contour if the ultrasonic section image is preliminarily judged to be a standard section image according to the section structure characteristics; The contour judgment module is used to use the ultrasonic section image as a target standard section image if it is determined that the bone structure contour is complete according to the position information of the bone structure contour.

7. The device according to claim 6, characterized in that The ultrasonic section image includes at least one section image; The contour judgment module is further used to calculate the score of the bone structure contour to obtain a bone contour score if it is determined that the bone structure contour is complete based on the position information of the bone structure contour; calculate based on the ultrasound section image score and the bone contour score to obtain a section image score; determine the maximum section image score based on the section image score; and use the ultrasound section image corresponding to the maximum section image score as the target standard section image.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images