Medical image data processing method and device, computer device and storage medium

Abstract

This application relates to a medical image data processing method, apparatus, computer equipment, storage medium, and computer program product. The method includes: acquiring a fetal ultrasound cross-sectional image; performing target detection on the fetal ultrasound cross-sectional image based on a target detection model to obtain target structural features; the target structural features include the target structural image, target structural category, and target structural location information; segmenting the fetal ultrasound cross-sectional image based on a segmentation model to obtain a fetal contour image; and extracting features based on the target structural features and the fetal contour image to obtain crown-rump length data. This entire approach extracts both the fetal target structural features and the fetal contour image, and the feature extraction based on these two dimensions results in more accurate crown-rump length data.

Description

Technical Field

[0001] This application relates to the field of medical image technology, and in particular to a medical image data processing method, apparatus, computer equipment, storage medium, and computer program product. Background Technology

[0002] Crown-rump length (CRL) measurement refers to the measurement of the maximum distance from the fetal head to the buttocks when the fetus is in a natural position. It is an important part of early pregnancy ultrasound examination. The latest guidelines issued by the International Society of Ultrasound in Obstetrics and Gynecology (ISUOG) state that, except for assisted reproductive technologies such as in-vitro fertilization, CRL measurements should be used to determine gestational age in all cases.

[0003] However, fetal movement is greater in early pregnancy, and when the fetus is in an unnatural position, it will lead to significant errors in the measurement of the fetal crown-rump length. Summary of the Invention

[0004] Therefore, it is necessary to provide a medical image data processing method, apparatus, computer equipment, computer-readable storage medium, and computer program product that can accurately measure the head-rump length, addressing the aforementioned technical problems.

[0005] Firstly, this application provides a method for processing medical image data. The method includes:

[0006] Obtain fetal ultrasound cross-sectional images;

[0007] Based on the target detection model, target detection is performed on fetal ultrasound cross-sectional images to obtain target structural features; the target structural features include target structural image, target structural category, and target structural location information.

[0008] Based on the segmentation model, the fetal ultrasound cross-sectional image is segmented to obtain the fetal contour image;

[0009] Feature extraction was performed based on the target structural features and fetal contour images to obtain crown-rump length data.

[0010] In one embodiment, the target structural features include head structural features and buttock structural features; feature extraction based on the target structural features and the fetal contour image to obtain crown-rump length data includes: extracting head structural features and buttock structural features from the target structural features; determining the fetal head orientation based on the head structural features and buttock structural features; determining the position of the head vertex and the position of the buttock vertex in the fetal contour image based on the fetal head orientation; and calculating the crown-rump length data based on the positions of the head vertex and the buttock vertex.

[0011] In one embodiment, the method further includes: extracting features based on the target structure, its corresponding category and location information, and the fetal outline image to obtain fetal posture data; and comparing the fetal posture data with preset natural posture data to obtain fetal posture state information.

[0012] In one embodiment, the target structural features further include mandibular structural features; feature extraction based on the target structure, its corresponding category and location information, and the fetal contour image to obtain fetal posture data includes: extracting mandibular structural features from the target structural features; extracting back target points based on the mandibular structural features and the fetal contour image; and calculating the angle between the head vertex, the buttocks vertex, and the back target points, using the back target points as vertices, to obtain fetal posture data.

[0013] In one embodiment, comparing the fetal posture data with preset natural posture data to obtain fetal posture state information includes: when the fetal posture data is within the range of preset natural posture data, the fetal posture state information is natural; when the fetal posture data is less than the minimum value of preset natural posture data, the fetal posture state information is overflexed, and the crown-rump length data needs to be corrected; when the fetal posture data is greater than the maximum value of preset natural posture data, the fetal posture state information is overextended, and the crown-rump length data needs to be corrected.

[0014] In one embodiment, the head structure features include head structure location information; the buttock structure features include buttock structure location information; determining the fetal head orientation based on the head structure features and the buttock structure features includes: determining the fetal head orientation according to the coordinate relationship between the head structure location information and the buttock structure location information.

[0015] Secondly, this application also provides a medical image data processing apparatus. The apparatus includes:

[0016] The acquisition module is used to acquire ultrasound cross-sectional images of the fetus;

[0017] The detection module is used to perform target detection on fetal ultrasound cross-sectional images based on a target detection model to obtain target structural features; the target structural features include target structural image, target structural category, and target structural location information.

[0018] The segmentation module is used to segment fetal ultrasound cross-sectional images based on a segmentation model to obtain fetal contour images;

[0019] The determination module is used to extract features based on the target structural features and the fetal contour image to obtain crown-rump length data.

[0020] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the following steps:

[0021] Obtain fetal ultrasound cross-sectional images;

[0022] Based on the target detection model, target detection is performed on fetal ultrasound cross-sectional images to obtain target structural features; the target structural features include target structural image, target structural category, and target structural location information.

[0023] Based on the segmentation model, the fetal ultrasound cross-sectional image is segmented to obtain the fetal contour image;

[0024] Feature extraction was performed based on the target structural features and fetal contour images to obtain crown-rump length data.

[0025] Fourthly, this application also provides a computer-readable storage medium. This computer-readable storage medium stores a computer program thereon, which, when executed by a processor, performs the following steps:

[0026] Obtain fetal ultrasound cross-sectional images;

[0027] Based on the target detection model, target detection is performed on fetal ultrasound cross-sectional images to obtain target structural features; the target structural features include target structural image, target structural category, and target structural location information.

[0028] Based on the segmentation model, the fetal ultrasound cross-sectional image is segmented to obtain the fetal contour image;

[0029] Feature extraction was performed based on the target structural features and fetal contour images to obtain crown-rump length data.

[0030] Fifthly, this application also provides a computer program product. This computer program product includes a computer program that, when executed by a processor, performs the following steps:

[0031] Obtain fetal ultrasound cross-sectional images;

[0032] Based on the target detection model, target detection is performed on fetal ultrasound cross-sectional images to obtain target structural features; the target structural features include target structural image, target structural category, and target structural location information.

[0033] Based on the segmentation model, the fetal ultrasound cross-sectional image is segmented to obtain the fetal contour image;

[0034] Feature extraction was performed based on the target structural features and fetal contour images to obtain crown-rump length data.

[0035] The aforementioned medical image data processing method, apparatus, computer equipment, storage medium, and computer program product acquire fetal ultrasound cross-sectional images; based on a target detection model, target detection is performed on the fetal ultrasound cross-sectional images to obtain target structural features; the target structural features include the target structural image, target structural category, and target structural location information; based on a segmentation model, the fetal ultrasound cross-sectional images are segmented to obtain fetal contour images; and feature extraction is performed based on the target structural features and the fetal contour images to obtain crown-rump length data. The entire scheme extracts fetal target structural features on one hand, and fetal contour images on the other, performing feature extraction based on data from both dimensions, resulting in more accurate crown-rump length data. Attached Figure Description

[0036] Figure 1 This is an application environment diagram of a medical image data processing method in one embodiment;

[0037] Figure 2 This is a flowchart illustrating a medical image data processing method in one embodiment;

[0038] Figure 3 This is a schematic diagram of a fetal outline image in one embodiment;

[0039] Figure 4 This is a schematic diagram of the fetal sitting posture in one embodiment;

[0040] Figure 5 This is a flowchart illustrating a medical image data processing method in another embodiment;

[0041] Figure 6 This is a structural block diagram of a medical image data processing device in one embodiment;

[0042] Figure 7 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0044] Crown-rump length (CRL) measurement refers to the measurement of the maximum distance from the fetal head to the buttocks when the fetus is in a natural position. It is an important part of early pregnancy ultrasound examination. The latest guidelines issued by the International Society of Ultrasound in Obstetrics and Gynecology (ISUOG) state that, except for assisted reproductive technologies such as in-vitro fertilization (IVF), CRL measurements should be used to determine gestational age in all cases. However, fetal activity is high in early pregnancy, and the fetus is often in an unnatural position. Studies have shown that when the fetus is in a hyperextended or hyperflexed position, the average CRL measurement can increase by 5.7 mm or decrease by 4.7 mm. Therefore, determining whether the fetus is in a natural position is crucial for accurate measurement of CRL in early pregnancy.

[0045] The medical image data processing method provided in this application embodiment can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104 or placed on a cloud or other network server. Terminal 102 acquires fetal ultrasound cross-sectional images; based on the target detection model trained on server 104, it performs target detection on the fetal ultrasound cross-sectional images to obtain target structural features; the target structural features include the target structural image, target structural category, and target structural location information; based on the segmentation model trained on server 104, it segments the fetal ultrasound cross-sectional images to obtain a fetal contour image; based on the target structural features and the fetal contour image, it performs feature extraction to obtain crown-rump length data. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can be smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, etc. Portable wearable devices can be smartwatches, smart bracelets, head-mounted devices, etc. Server 104 can be implemented using a standalone server or a server cluster composed of multiple servers.

[0046] In one embodiment, such as Figure 2 As shown, a medical image data processing method is provided, which can be applied to... Figure 1 Taking the terminal in the example, the explanation includes the following steps:

[0047] Step 202: Obtain fetal ultrasound cross-sectional images.

[0048] Ultrasound cross-sectional images refer to images containing multiple anatomical sections of internal organs obtained by scanning a living organism with an ultrasound beam and receiving and processing the reflected signals. In this embodiment, fetal ultrasound cross-sectional images refer to images containing multiple anatomical sections of the fetus obtained by scanning the pregnant woman's abdomen with an ultrasound beam and receiving and processing the reflected signals. The fetal ultrasound cross-sectional images in this embodiment are pre-processed images, meaning the original fetal ultrasound cross-sectional images undergo quality checks to ensure they contain images of all parts of the fetus and can be used for measuring fetal crown-rump length (CRL). Fetal ultrasound cross-sectional images include fetal crown-rump length (CRL) cross-sectional images.

[0049] Specifically, the terminal sends a command to the medical ultrasound scanning device to acquire a fetal ultrasound cross-sectional image. When the medical ultrasound scanning device receives the command, it performs an ultrasound beam scan on the fetus to obtain a crown-rump cross-sectional image.

[0050] Step 204: Based on the target detection model, target detection is performed on the fetal ultrasound cross-sectional image to obtain target structural features.

[0051] The target structural features include the target structural image, target structural category, and target structural location information. The target detection model is used to detect the structural information of various parts of the fetus. Specifically, the structural information can be the bounding rectangle of each fetal part and the confidence score of each rectangle. Further, the structural information can include the length, width, and coordinate information of the bounding rectangle of each fetal part. The target detection model is trained based on the YOLOv5 network and a fetal ultrasound cross-sectional image dataset. The fetal ultrasound cross-sectional image dataset includes a fetal crown-rump length cross-sectional image dataset and annotation information for each fetal part. The fetal parts include the head, chest, buttocks, abdomen, maxilla, mandible, genitals, rhomboid brain, and diencephalon. The annotation information includes the bounding rectangle of each fetal part in the fetal crown-rump length cross-sectional image and its category information. The category information is used to distinguish which type of fetal part belongs to among the head, chest, buttocks, abdomen, maxilla, mandible, genitals, rhomboid brain, and diencephalon. The target detection model can also be built based on other target detection neural network models; this embodiment does not limit its application.

[0052] Specifically, the terminal inputs the fetal crown-rump length cross-sectional image into the target detection model for target detection and obtains the output results of the target detection model, which includes the bounding rectangles of each part of the fetus, the fetal structure category, and the position information of each part of the fetus.

[0053] Step 206: Based on the segmentation model, segment the fetal ultrasound cross-sectional image to obtain the fetal contour image.

[0054] The segmentation model is used to segment the fetal contour. It is trained on a Unet network and a fetal ultrasound cross-sectional image dataset. The fetal ultrasound cross-sectional image dataset includes fetal crown-rump length cross-sectional images and polygonal regions representing the entire fetal contour, such as... Figure 3 As shown.

[0055] Specifically, the terminal inputs the fetal ultrasound cross-sectional image into the segmentation model, obtains the output of the segmentation model, which is a polygonal region containing the entire fetal outline, and obtains the fetal outline image.

[0056] Step 208: Feature extraction is performed based on the target structural features and the fetal contour image to obtain crown-rump length data.

[0057] Specifically, fetal head and buttock features are extracted from the target structural features. Based on the fetal contour image, the positions of the head apex and buttock apex are determined from the head features. Then, the straight line containing the positions of the head apex and buttock apex is determined, and the length from the head apex to the buttock apex in the straight line is calculated to obtain the fetal crown-rump length data.

[0058] The aforementioned medical image data processing method involves acquiring fetal ultrasound cross-sectional images; performing target detection on the fetal ultrasound cross-sectional images based on a target detection model to obtain target structural features; these features include the target structural image, target structural category, and target structural location information; segmenting the fetal ultrasound cross-sectional images based on a segmentation model to obtain fetal contour images; and extracting features based on the target structural features and fetal contour images to obtain crown-rump length data. This entire approach extracts fetal target structural features and fetal contour images simultaneously, resulting in more accurate crown-rump length data.

[0059] In an optional embodiment, feature extraction based on target structural features and fetal contour image to obtain crown-rump length data includes: extracting head structure features and buttock structure features from the target structural features; determining the fetal head orientation based on the head structure features and buttock structure features; determining the position of the head vertex and the buttock vertex in the fetal contour image based on the fetal head orientation; and calculating the crown-rump length data based on the position of the head vertex and the position of the buttock vertex.

[0060] The target structural features include head and buttock features. In this embodiment, the fetus is facing upwards in the fetal ultrasound image, meaning the longitudinal coordinates of the fetal chest and abdomen are greater than the longitudinal coordinates of the buttocks. The direction of the fetal head is used to determine whether the fetal head is on the left or right side of the fetal ultrasound image. The fetal chest and abdominal structural features can also be used to determine whether the fetus is facing upwards; if the chest and abdominal structural features are facing upwards, then the fetus is facing upwards.

[0061] Specifically, after the terminal obtains the target structural features from the output of the target detection model, it extracts head and buttock structural features based on the target structural category to obtain head and buttock position information. Based on this information, the fetal head orientation is determined. Using the fetal contour image, the two farthest vertices are identified as the head and buttock vertices. Then, based on the fetal head orientation, the head and buttock vertices are distinguished. Using the coordinates of the head and buttock vertices, the crown-rump length line containing these coordinates is determined. The length of the line segment between the head and buttock vertices on this line is calculated to obtain the fetal crown-rump length data.

[0062] In this embodiment, based on the fetal contour image, the two points furthest apart in the fetal contour image are determined as the head vertex and the buttocks vertex, and the accurate head vertex and buttocks vertex positions are obtained. Then, based on the fetal head structure features and buttock structure features, the direction of the fetal head is determined, thereby accurately distinguishing the head vertex and the buttocks vertex. Finally, based on the head vertex and the buttocks vertex, accurate head-rump length data is obtained.

[0063] In an optional embodiment, determining the fetal head orientation based on head structure features and buttock structure features includes: determining the fetal head orientation according to the coordinate relationship between head structure position information and buttock structure position information.

[0064] The head structure features include head structure location information; the buttock structure features include buttock structure location information.

[0065] Specifically, in determining the direction of the fetal head, head structure features and buttock structure features are extracted from the target structural features to obtain head structure position information and buttock structure position information. The horizontal coordinates of the head structure and the buttock structure are compared. When the horizontal coordinate of the head structure is less than the horizontal coordinate of the buttock structure, the direction of the fetal head is determined to be on the left. When the horizontal coordinate of the head structure is greater than the horizontal coordinate of the buttock structure, the direction of the fetal head is determined to be on the right.

[0066] In this embodiment, the direction of the fetal head is accurately determined based on the relationship between the horizontal coordinates of the fetal head structure and the buttock structure, thereby accurately distinguishing the head apex and the buttock apex. Based on the accurate head apex and buttock apex, accurate head-rump length data is obtained.

[0067] In an optional embodiment, the above method further includes: extracting features based on the target structure and its corresponding category and location information and the fetal contour image to obtain fetal posture data; and comparing the fetal posture data with preset natural posture data to obtain fetal posture state information.

[0068] Fetal posture data refers to the angle between the fetus's head, back, and buttocks, specifically the angle between the head-back line and the buttock-back line. The head-back line is the straight line connecting the apex of the head to the target point on the back, and the buttock-back line is the straight line connecting the target point on the back to the apex of the buttocks. Fetal posture status information includes the natural state, hyperflexed state, and hyperextended state. Studies have found that the posture data in the natural state ranges from 118° to 141°, with the preset natural posture data set at [118°, 141°].

[0069] Specifically, the terminal extracts the head apex, back target point, and buttocks apex based on the fetal head structure features, mandibular structure features, and fetal contour image. It then determines the posterior-posterior line by connecting the back target point and the buttocks apex, and calculates the angle between the head-posterior line and the posterior-posterior line to obtain fetal posture data. The back target point is the intersection of the perpendicular line from the center point of the mandible and the fetal back contour in the fetal contour image. Finally, the fetal posture data is compared with preset natural posture data to obtain fetal posture state information.

[0070] In an optional embodiment, the target structural features further include mandibular structural features; feature extraction based on the target structure, its corresponding category and location information, and the fetal contour image to obtain fetal posture data includes: extracting mandibular structural features from the target structural features; extracting back target points based on the mandibular structural features and the fetal contour image; and calculating the angle between the head vertex, the buttocks vertex, and the back target points, using the back target points as vertices, to obtain fetal posture data.

[0071] Specifically, such as Figure 4As shown, the terminal extracts the head vertex, mandibular center point, and buttocks vertex based on the fetal head structure features, mandibular structure features, and fetal contour image. A perpendicular line is drawn from the mandibular center point along the vertical direction. The intersection of this perpendicular line and the fetal contour image yields the back target point. The head-back line is determined by connecting the back target point to the head vertex, and the buttocks-back line is determined by connecting the back target point to the buttocks vertex. The angle between the head-back line and the buttocks-back line is calculated to obtain the fetal posture data. The mandibular center point refers to the center point of the circumscribed rectangle of the mandibular structure features, specifically the intersection of the two diagonals of the center point of the circumscribed rectangle.

[0072] In this embodiment, by determining the fetal posture data, it is determined whether the fetus is in a natural state. If the fetus is in an unnatural state, that is, in a hyperflexed or hyperextended state, fetal posture correction is required to obtain accurate head vertex coordinates. Then, based on the accurate head vertex position coordinates and buttock vertex position coordinates, accurate crown-rump length data can be obtained, reducing the inaccuracy of crown-rump length data measurement caused by the unnatural state of the fetus.

[0073] In an optional embodiment, comparing fetal posture data with preset natural posture data to obtain fetal posture state information includes: when the fetal posture data is within the range of preset natural posture data, the fetal posture state information is natural; when the fetal posture data is less than the minimum value of preset natural posture data, the fetal posture state information is overflexed, and the crown-rump length data needs to be corrected; when the fetal posture data is greater than the maximum value of preset natural posture data, the fetal posture state information is overextended, and the crown-rump length data needs to be corrected.

[0074] The preset natural posture data is [118°, 141°], with the minimum value of the preset natural posture data being 118° and the maximum value being 141°.

[0075] Specifically, since the fetal posture data measured in a natural position is between 118° and 141°, an angle less than 118° is considered hyperflexion, and an angle greater than 141° is considered hyperextension. Both hyperflexion and hyperextension require correction of the measured crown-rump length data to obtain accurate crown-rump length data and more accurately calculate gestational age.

[0076] To facilitate understanding of the technical solutions provided in the embodiments of this application, such as Figure 5 As shown, the medical image data processing method provided in this application embodiment will be briefly described using a complete medical image data processing process:

[0077] (1) Obtain fetal ultrasound cross-sectional images.

[0078] (2) Based on the target detection model, target detection is performed on the fetal ultrasound cross-sectional images to obtain target structural features; the target structural features include the target structural image, the target structural category, and the target structural location information. The target structural features include head structural features and buttock structural features.

[0079] (3) Based on the segmentation model, the fetal ultrasound cross-sectional image is segmented to obtain the fetal contour image.

[0080] (4) Extract head structure features and buttock structure features from the target structure features; determine the direction of the fetal head based on the coordinate relationship between the head structure position information and the buttock structure position information; determine the position of the head vertex and the position of the buttock vertex in the fetal contour image based on the direction of the fetal head; calculate the head-rump length data based on the position of the head vertex and the position of the buttock vertex.

[0081] (5) Extract mandibular structural features from target structural features; extract back target points based on mandibular structural features and fetal contour images; calculate the angle between head vertex, buttock vertex and back target points using back target points as vertices to obtain fetal posture data.

[0082] (6) When the fetal posture data is within the range of the preset natural posture data, the fetal posture status information is natural.

[0083] (7) When the fetal posture data is less than the minimum value of the preset natural posture data, the fetal posture status information is overflexed, and the crown-rump length data needs to be corrected.

[0084] (8) When the fetal posture data is greater than the maximum value of the preset natural posture data, the fetal posture status information is hyperextension, and the crown-rump length data needs to be corrected.

[0085] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0086] Based on the same inventive concept, this application also provides a medical image data processing apparatus for implementing the medical image data processing method described above. The solution provided by this apparatus is similar to the implementation scheme described in the above method; therefore, the specific limitations in one or more embodiments of the medical image data processing apparatus provided below can be found in the limitations of the medical image data processing method described above, and will not be repeated here.

[0087] In one embodiment, such as Figure 6 As shown, a medical image data processing device is provided, including: an acquisition module 602, a detection module 604, a segmentation module 606, and a determination module 608, wherein:

[0088] The acquisition module 602 is used to acquire fetal ultrasound cross-sectional images.

[0089] The detection module 604 is used to perform target detection on the fetal ultrasound cross-sectional image based on the target detection model to obtain target structural features; the target structural features include the target structural image, the target structural category, and the target structural location information.

[0090] The segmentation module 606 is used to segment the fetal ultrasound cross-sectional image based on the segmentation model to obtain the fetal contour image.

[0091] The determination module 608 is used to extract features based on the target structural features and the fetal contour image to obtain crown-rump length data.

[0092] In one embodiment, the target structural features include head structural features and buttock structural features; the determining module 608 is further configured to extract head structural features and buttock structural features from the target structural features; determine the fetal head orientation based on the head structural features and buttock structural features; determine the position of the head vertex and the position of the buttock vertex in the fetal contour image based on the fetal head orientation; and calculate the head-rump length data based on the position of the head vertex and the position of the buttock vertex.

[0093] In one embodiment, the medical image data processing device further includes a fetal posture determination module, which is used to extract features based on the target structure and corresponding category and location information and the fetal contour image to obtain fetal posture data; and compare the fetal posture data with preset natural posture data to obtain fetal posture state information.

[0094] In one embodiment, the target structural features also include mandibular structural features; the fetal posture determination module is further used to extract mandibular structural features from the target structural features; extract back target points based on mandibular structural features and fetal contour images; and calculate the angle between the head vertex, the buttocks vertex and the back target points using the back target points as vertices to obtain fetal posture data.

[0095] In one embodiment, the fetal posture determination module is further configured to: when the fetal posture data is within the range of preset natural posture data, the fetal posture status information is natural; when the fetal posture data is less than the minimum value of the preset natural posture data, the fetal posture status information is overflexed, and the crown-rump length data needs to be corrected; when the fetal posture data is greater than the maximum value of the preset natural posture data, the fetal posture status information is overextended, and the crown-rump length data needs to be corrected.

[0096] In one embodiment, the head structure features include head structure position information; the buttock structure features include buttock structure position information; the determining module 608 is further configured to determine the direction of the fetal head based on the coordinate relationship between the head structure position information and the buttock structure position information.

[0097] Each module in the aforementioned medical image data processing device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0098] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 7 As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a medical image data processing method. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0099] Those skilled in the art will understand that Figure 7 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0100] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0101] Obtain fetal ultrasound cross-sectional images;

[0102] Based on the target detection model, target detection is performed on fetal ultrasound cross-sectional images to obtain target structural features; the target structural features include target structural image, target structural category, and target structural location information.

[0103] Based on the segmentation model, the fetal ultrasound cross-sectional image is segmented to obtain the fetal contour image;

[0104] Feature extraction was performed based on the target structural features and fetal contour images to obtain crown-rump length data.

[0105] In one embodiment, when the processor executes the computer program, it further performs the following steps: the target structural features include head structural features and buttock structural features; feature extraction based on the target structural features and the fetal contour image to obtain crown-rump length data includes: extracting head structural features and buttock structural features from the target structural features; determining the fetal head orientation based on the head structural features and buttock structural features; determining the position of the head vertex and the position of the buttock vertex in the fetal contour image based on the fetal head orientation; and calculating the crown-rump length data based on the positions of the head vertex and the buttock vertex.

[0106] In one embodiment, when the processor executes the computer program, it further performs the following steps: extracting features based on the target structure and its corresponding category and location information and the fetal outline image to obtain fetal posture data; comparing the fetal posture data with preset natural posture data to obtain fetal posture state information.

[0107] In one embodiment, when the processor executes the computer program, it further performs the following steps: the target structural features also include mandibular structural features; feature extraction based on the target structure and its corresponding category and location information and the fetal contour image to obtain fetal posture data includes: extracting mandibular structural features from the target structural features; extracting back target points based on the mandibular structural features and the fetal contour image; and calculating the angle between the head vertex, the buttocks vertex and the back target points, using the back target points as vertices, to obtain fetal posture data.

[0108] In one embodiment, when the processor executes the computer program, it further performs the following steps: comparing fetal posture data with preset natural posture data to obtain fetal posture state information, including: when the posture data is within the range of preset natural posture data, the fetal posture state information is natural; when the fetal posture data is less than the minimum value of preset natural posture data, the fetal posture state information is overflexed, and the crown-rump length data needs to be corrected; when the fetal posture data is greater than the maximum value of preset natural posture data, the fetal posture state information is overextended, and the crown-rump length data needs to be corrected.

[0109] In one embodiment, when the processor executes the computer program, it further performs the following steps: the head structure features include head structure position information; the buttock structure features include buttock structure position information; determining the fetal head orientation based on the head structure features and the buttock structure features includes: determining the fetal head orientation according to the coordinate relationship between the head structure position information and the buttock structure position information.

[0110] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0111] Obtain fetal ultrasound cross-sectional images;

[0112] Based on the target detection model, target detection is performed on fetal ultrasound cross-sectional images to obtain target structural features; the target structural features include target structural image, target structural category, and target structural location information.

[0113] Based on the segmentation model, the fetal ultrasound cross-sectional image is segmented to obtain the fetal contour image;

[0114] Feature extraction was performed based on the target structural features and fetal contour images to obtain crown-rump length data.

[0115] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: the target structural features include head structural features and buttock structural features; feature extraction based on the target structural features and the fetal contour image to obtain crown-rump length data includes: extracting head structural features and buttock structural features from the target structural features; determining the fetal head orientation based on the head structural features and buttock structural features; determining the position of the head vertex and the position of the buttock vertex in the fetal contour image based on the fetal head orientation; and calculating the crown-rump length data based on the positions of the head vertex and the buttock vertex.

[0116] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: extracting features based on the target structure and corresponding category and location information and the fetal outline image to obtain fetal posture data; comparing the fetal posture data with preset natural posture data to obtain fetal posture state information.

[0117] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: the target structural features also include mandibular structural features; feature extraction based on the target structure and its corresponding category and location information and the fetal contour image to obtain fetal posture data includes: extracting mandibular structural features from the target structural features; extracting back target points based on the mandibular structural features and the fetal contour image; and calculating the angle between the head vertex, the buttocks vertex and the back target point, using the back target point as the vertex, to obtain fetal posture data.

[0118] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: comparing fetal posture data with preset natural posture data to obtain fetal posture state information, including: when the fetal posture data is within the range of the preset natural posture data, the fetal posture state information is natural; when the fetal posture data is less than the minimum value of the preset natural posture data, the fetal posture state information is overflexed, and the crown-rump length data needs to be corrected; when the fetal posture data is greater than the maximum value of the preset natural posture data, the fetal posture state information is overextended, and the crown-rump length data needs to be corrected.

[0119] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: the head structure features include head structure position information; the buttock structure features include buttock structure position information; determining the fetal head orientation based on the head structure features and the buttock structure features includes: determining the fetal head orientation according to the coordinate relationship between the head structure position information and the buttock structure position information.

[0120] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0121] Obtain fetal ultrasound cross-sectional images;

[0122] Based on the target detection model, target detection is performed on fetal ultrasound cross-sectional images to obtain target structural features; the target structural features include target structural image, target structural category, and target structural location information.

[0123] Based on the segmentation model, the fetal ultrasound cross-sectional image is segmented to obtain the fetal contour image;

[0124] Feature extraction was performed based on the target structural features and fetal contour images to obtain crown-rump length data.

[0125] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: the target structural features include head structural features and buttock structural features; feature extraction based on the target structural features and the fetal contour image to obtain crown-rump length data includes: extracting head structural features and buttock structural features from the target structural features; determining the fetal head orientation based on the head structural features and buttock structural features; determining the position of the head vertex and the position of the buttock vertex in the fetal contour image based on the fetal head orientation; and calculating the crown-rump length data based on the positions of the head vertex and the buttock vertex.

[0126] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: extracting features based on the target structure and corresponding category and location information and the fetal outline image to obtain fetal posture data; comparing the fetal posture data with preset natural fetal posture data to obtain fetal posture state information.

[0127] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: the target structural features also include mandibular structural features; feature extraction based on the target structure and its corresponding category and location information and the fetal contour image to obtain fetal posture data includes: extracting mandibular structural features from the target structural features; extracting back target points based on the mandibular structural features and the fetal contour image; and calculating the angle between the head vertex, the buttocks vertex and the back target point, using the back target point as the vertex, to obtain fetal posture data.

[0128] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: comparing fetal posture data with preset natural posture data to obtain fetal posture state information, including: when the fetal posture data is within the range of the preset natural posture data, the fetal posture information is natural; when the fetal posture data is less than the minimum value of the preset natural posture data, the fetal posture state information is overflexed, and the crown-rump length data needs to be corrected; when the fetal posture data is greater than the maximum value of the preset natural posture data, the fetal posture state information is overextended, and the crown-rump length data needs to be corrected.

[0129] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: the head structure features include head structure position information; the buttock structure features include buttock structure position information; determining the fetal head orientation based on the head structure features and the buttock structure features includes: determining the fetal head orientation according to the coordinate relationship between the head structure position information and the buttock structure position information.

[0130] 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, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0131] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. 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 embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this 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), magnetic 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. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0132] The technical features of the above embodiments can be combined in any way. For the sake of brevity, 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.

[0133] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for processing medical image data, characterized in that, The method includes: Obtain fetal ultrasound cross-sectional images; Based on the target detection model, target detection is performed on the fetal ultrasound cross-sectional image to obtain target structural features; the target structural features include target structural image, target structural category, and target structural location information. Based on the segmentation model, the fetal ultrasound cross-sectional image is segmented to obtain a fetal contour image; Feature extraction is performed based on the target structural features and the fetal contour image to obtain crown-rump length data; Extract mandibular structural features from the target structural features; Based on the mandibular structure features and the fetal contour image, the head apex, mandibular center point, and buttock apex are extracted. A perpendicular line is drawn from the position coordinates of the mandibular center point along the vertical direction. The back target point is obtained based on the intersection of the perpendicular line and the fetal contour image. The head-back line is determined by connecting the back target point and the head apex, and the buttock-back line is determined by connecting the back target point and the buttock apex. The angle between the head-back line and the buttock-back line is calculated to obtain fetal posture data. When the fetal posture data is less than the minimum value of the preset natural posture data, or greater than the maximum value of the preset natural posture data, the crown-rump length data is corrected.

2. The method according to claim 1, characterized in that, The target structural features include head structural features and buttock structural features; The step of extracting features based on the target structural features and the fetal contour image to obtain crown-rump length data includes: Extract head and buttock structural features from the target structural features; Based on the aforementioned head and buttock structural features, the orientation of the fetal head is determined; Based on the direction of the fetal head, determine the position of the head apex and the position of the buttocks apex in the fetal contour image; The head-to-hip length data is obtained by calculating based on the positions of the head apex and the hip apex.

3. The method according to claim 1, characterized in that, After obtaining the fetal position data, the method further includes: When the fetal posture data is within the range of preset natural posture data, the fetal posture status information is natural.

4. The method according to claim 1, characterized in that, When the fetal posture data is less than the minimum value of the preset natural posture data, or greater than the maximum value of the preset natural posture data, the crown-rump length data is corrected, including: When the fetal posture data is less than the minimum value of the preset natural posture data, the fetal posture status information is overflexed, and the crown-rump length data needs to be corrected. When the fetal posture data is greater than the maximum value of the preset natural posture data, the fetal posture status information is hyperextension, and the crown-rump length data needs to be corrected.

5. The method according to claim 2, characterized in that, The head structure features include head structure location information; the buttock structure features include buttock structure location information; The determination of the fetal head orientation based on the head structure features and buttock structure features includes: The orientation of the fetal head is determined based on the coordinate relationship between the head structure position information and the buttock structure position information.

6. A medical image data processing device, characterized in that, The device includes: The acquisition module is used to acquire ultrasound cross-sectional images of the fetus; The detection module is used to perform target detection on the fetal ultrasound cross-sectional image based on the target detection model to obtain target structural features; the target structural features include target structural image, target structural category, and target structural location information; The segmentation module is used to segment the fetal ultrasound cross-sectional image based on a segmentation model to obtain a fetal contour image; The determination module is used to extract features based on the target structural features and the fetal contour image to obtain crown-rump length data; The fetal posture determination module is used to extract mandibular structural features from the target structural features; based on the mandibular structural features and the fetal contour image, it extracts the head vertex, mandibular center point, and buttock vertex; it draws a perpendicular line from the position coordinates of the mandibular center point along the vertical direction; and obtains the back target point based on the intersection of the perpendicular line and the fetal contour image; it determines the head-back line based on the line connecting the back target point and the head vertex, and determines the buttock-back line based on the line connecting the back target point and the buttock vertex; it calculates the angle between the head-back line and the buttock-back line to obtain fetal posture data; when the fetal posture data is less than the minimum value of the preset natural posture data, or greater than the maximum value of the preset natural posture data, it corrects the head-rump length data.

7. The apparatus according to claim 6, characterized in that, The target structural features include head structural features and buttock structural features; the determining module is further configured to: Extract head and buttock structural features from the target structural features; Based on the aforementioned head and buttock structural features, the orientation of the fetal head is determined; Based on the direction of the fetal head, determine the position of the head apex and the position of the buttocks apex in the fetal contour image; The head-to-hip length data is obtained by calculating based on the positions of the head apex and the hip apex.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

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

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

Images