Method and device for measuring standard section of fetal four-chamber heart, and ultrasonic equipment and medium

Abstract

The application discloses a kind of fetal four-chamber heart standard section automatic measurement method, device and a kind of ultrasonic equipment and computer readable storage medium, the method comprises: fetal four-chamber heart standard section is input integrated detection model;Integrated detection model outputs the profile of chest and heart in fetal four-chamber heart standard section, the position data of key point;Position data is post-processed, and the profile trace line, axis line, transverse diameter line of chest and heart, target measurement item result is obtained, and output to display.The application realizes the automatic measurement of the profile trace line, axis line, transverse diameter line of chest and heart, target measurement item, improves the measurement efficiency of fetal four-chamber heart standard section.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ultrasound, more particularly, to an automatic measurement method and device for a fetal four-chamber heart standard section, an ultrasound device and a computer readable storage medium. BACKGROUND

[0002] In fetal ultrasound screening, the fetal four-chamber heart section is a very basic and important section for diagnosing whether the fetal heart is malformed. The best opportunity for fetal heart ultrasound screening is 20-24 weeks of pregnancy. The main observation contents include: heart position, atrium and ventricle position, size, shape and data, atrioventricular connection relationship, two and three mitral valve shape position, size and opening and closing movement.

[0003] At present, doctors mainly measure the fetal four-chamber heart standard section manually, including heart axis measurement, cardiothoracic area ratio, cardiothoracic transverse diameter ratio, etc., which is difficult and time-consuming for doctors. It can be seen that the measurement efficiency of the fetal four-chamber heart standard section is low in the related art.

[0004] Therefore, how to improve the measurement efficiency of the fetal four-chamber heart standard section is a technical problem to be solved by those skilled in the art. SUMMARY

[0005] The present application aims to provide an automatic measurement method and device for a fetal four-chamber heart standard section, an ultrasound device and a computer readable storage medium, which improves the measurement efficiency of the fetal four-chamber heart standard section.

[0006] To achieve the above-mentioned purpose, the present application provides an automatic measurement method for a fetal four-chamber heart standard section, comprising:

[0007] inputting the fetal four-chamber heart standard section into an integrated detection model;

[0008] the integrated detection model outputs the contour of the thoracic cavity and the heart, and the position data of the key points in the fetal four-chamber heart standard section;

[0009] post-processing the position data to obtain the contour trace line, axis line, transverse diameter line and target measurement item result of the thoracic cavity and the heart, and outputting to a display.

[0010] The semantic output channels of the integrated detection model include an output channel corresponding to the thoracic cavity contour, an output channel corresponding to the heart contour, and an output channel corresponding to each key point.

[0011] The integrated detection model comprises a plurality of branches connected in parallel, each of the branches comprises a plurality of feature maps with the same resolution connected in series, different branches contain feature maps with different resolutions, and the feature maps in different branches interact through up-sampling layers or down-sampling layers.

[0012] The method further comprises:

[0013] Obtaining training fetal four-chamber standard sections and corresponding measurement annotations, the measurement annotations comprising a chest contour, a heart contour, a chest axis, a heart axis, a chest transverse diameter and a heart transverse diameter;

[0014] Training an integrated detection model by using the training fetal four-chamber standard sections and the corresponding measurement annotations.

[0015] Before the fetal four-chamber standard section is input into the integrated detection model, the method further comprises:

[0016] Obtaining an ultrasound video, and inputting the ultrasound video into a standard section recognition model to obtain the fetal four-chamber standard section.

[0017] The method further comprises:

[0018] Obtaining training ultrasound videos and corresponding fetal four-chamber standard section annotations;

[0019] Training a fetal four-chamber standard section recognition model by using the training ultrasound videos and the corresponding fetal four-chamber standard section annotations.

[0020] The key points comprise an apex point and a base point of the heart axis, and a first end point and a second end point of the chest axis, the first end point being a spine end point.

[0021] The position data is post-processed to obtain axis and transverse diameter results of the chest and the heart, comprising:

[0022] Determining a chest axis based on the first end point and the second end point of the chest axis, and determining a heart axis based on the apex point and the base point of the heart axis;

[0023] Determining a first intersection line segment of a perpendicular line of the chest axis and the chest contour, and determining a chest transverse diameter line as a first intersection line segment with the maximum length;

[0024] Determining a second intersection line segment of a perpendicular line of the heart axis and the heart contour, and determining a heart transverse diameter line as a second intersection line segment with the maximum length.

[0025] The target measurement item comprises any one or a combination of a plurality of items selected from the group consisting of a heart axis angle, a heart-chest area ratio and a heart-chest transverse diameter ratio, and the position data is post-processed to obtain a target measurement item result, comprising:

[0026] calculating an angle between the heart axis and the thoracic axis as a heart-axis angle;

[0027] and / or, calculating a ratio between a heart area and a thoracic area as a heart-thoracic area ratio according to the heart profile and the thoracic profile;

[0028] and / or, calculating a ratio between a length of the heart transverse diameter and a length of the thoracic transverse diameter as a heart-thoracic transverse diameter ratio.

[0029] wherein further comprising:

[0030] if the heart-axis angle exceeds a preset angle range, and / or the heart-thoracic area ratio exceeds a first threshold range, and / or the heart-thoracic transverse diameter ratio exceeds a second threshold range, outputting an abnormal prompt.

[0031] To achieve the above object, the present application provides a device for automatically measuring a standard section of a four-chamber heart of a fetus, comprising:

[0032] an input module for inputting a standard section of a four-chamber heart of a fetus into an integrated detection model;

[0033] an output module for outputting, by the integrated detection model, profile data of a thoracic cavity and a heart and position data of key points in the standard section of the four-chamber heart of the fetus;

[0034] a processing module for post-processing the position data to obtain profile traces, axis lines, transverse diameter lines, and target measurement item results of the thoracic cavity and the heart, and outputting to a display.

[0035] To achieve the above object, the present application provides an ultrasound device, comprising:

[0036] a memory for storing a computer program;

[0037] a processor for executing the computer program to implement the steps of the automatic measurement method of a standard section of a four-chamber heart of a fetus as described above.

[0038] To achieve the above object, the present application provides a computer readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the steps of the automatic measurement method of a standard section of a four-chamber heart of a fetus as described above.

[0039] It can be known from the above scheme that the automatic measurement method of the fetal four-chamber heart standard section provided in the application comprises: inputting the fetal four-chamber heart standard section into an integrated detection model; the integrated detection model outputs the contour of the chest cavity and the heart and the position data of the key points in the fetal four-chamber heart standard section; the position data is post-processed to obtain the contour trace line, the axis line, the transverse diameter line and the target measurement item result of the chest cavity and the heart, and is output to a display.

[0040] The automatic measurement method of the fetal four-chamber heart standard section provided in the application automatically identifies the contour of the chest cavity and the heart and the position data of the key points in the fetal four-chamber heart standard section through the integrated detection model, and further post-processes the identified position data of the key points to automatically generate the contour trace line, the axis line, the transverse diameter line and the target measurement item result of the chest cavity and the heart. It can be seen that the automatic measurement method of the fetal four-chamber heart standard section provided in the application realizes the automatic measurement of the contour trace line, the axis line, the transverse diameter line and the target measurement item of the chest cavity and the heart, and improves the measurement efficiency of the fetal four-chamber heart standard section. The application also discloses an automatic measurement device for the fetal four-chamber heart standard section, an ultrasonic equipment and a computer readable storage medium, which can also achieve the above technical effects.

[0041] It should be understood that the above general description and the following detailed description are only exemplary and cannot limit the application. BRIEF DESCRIPTION OF DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description only some embodiments of the application, and for those skilled in the art, without creative labor, can also obtain other drawings according to these drawings. The drawings are used to provide further understanding of the present disclosure and constitute a part of the specification, and together with the following specific embodiments, they are used to explain the present disclosure, but do not constitute a limitation on the present disclosure. In the drawings:

[0043] Figure 1 The flow chart of a fetal four-chamber heart standard section measurement method according to an exemplary embodiment is shown;

[0044] Figure 2 The structure diagram of a key point detection model according to an exemplary embodiment is shown;

[0045] Figure 3 The flow chart of another fetal four-chamber heart standard section measurement method according to an exemplary embodiment is shown;

[0046] Figure 4A structural diagram of a fetal four-chamber heart standard section measurement system according to an exemplary embodiment is shown.

[0047] Figure 5 A structural diagram of a fetal four-chamber heart standard section measurement device according to an exemplary embodiment is shown.

[0048] Figure 6 A structural diagram of an ultrasound device according to an exemplary embodiment is shown. DETAILED DESCRIPTION

[0049] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without any creative work fall within the scope of protection of the present application. In addition, in the embodiments of the present application, “first”, “second”, and the like are used to distinguish similar objects, rather than necessarily used to describe a specific order or sequence.

[0050] The embodiments of the present application disclose an automatic measurement method of a fetal four-chamber heart standard section, which improves the measurement efficiency of the fetal four-chamber heart standard section.

[0051] Referring to Figure 1 , a flowchart of a fetal four-chamber heart standard section measurement method according to an exemplary embodiment is shown in FIG. 1, which includes the following steps. Figure 1

[0052] S101: inputting a fetal four-chamber heart standard section into an integrated detection model;

[0053] S102: outputting, by the integrated detection model, contour and key point position data of a thoracic cavity and a heart in the fetal four-chamber heart standard section;

[0054] The execution subject of the present embodiment is an ultrasound device, and the purpose is to automatically trace and measure a fetal four-chamber heart standard section. Of course, the execution subject of the present embodiment can also be other electronic devices with storage and operation functions, which can obtain a fetal four-chamber heart standard section by communicating with an ultrasound device or connecting with a data cloud, and then perform independent reasoning and operation.

[0055] ​In practice, the standard cross-section of the fetal four-chamber heart is input into the integrated detection model to obtain the contours of the thoracic cavity and heart, as well as the location data of key points. This integrated detection model can be improved upon the key point detection model by adding two semantic segmentation channels to the output channels of the key point detection model for outputting the thoracic cavity contour and the heart contour. That is, the semantic output channels of the integrated detection model include the output channel corresponding to the thoracic cavity contour, the output channel corresponding to the heart contour, and the output channel corresponding to each key point. The key points here can include the apex and basal endpoints of the cardiac axis, and the first and second endpoints of the thoracic axis, with the first endpoint being the spinal endpoint. In other words, the integrated detection model contains six semantic output channels, respectively outputting the thoracic cavity contour, the heart contour, the spinal endpoint and the other endpoint of the thoracic axis, and the apex and basal endpoints of the cardiac axis.

[0056] The training process of the keypoint detection model includes: acquiring a standard cross-section of the four chambers of the fetus and corresponding measurement annotations, wherein the measurement annotations include the chest cavity contour, heart contour, chest cavity axis, heart axis, chest cavity transverse diameter, and heart transverse diameter; and training the keypoint detection model using the standard cross-section of the four chambers of the fetus and the corresponding measurement annotations. In specific implementation, the standard cross-section of the four chambers of the fetus is acquired, and the standard cross-section of the four chambers of the fetus is measured and annotated using annotation software, including the chest cavity contour, heart contour, chest cavity axis, heart axis, chest cavity transverse diameter, and heart transverse diameter. The measured and annotated standard cross-section of the four chambers of the fetus is then input into the keypoint detection model for training.

[0057] In a preferred embodiment, the keypoint detection model includes multiple branches in parallel, each branch comprising multiple feature maps of the same resolution connected in series. Different branches contain feature maps with different resolutions, and the feature maps in different branches interact through upsampling or downsampling layers. In a specific implementation, the keypoint detection model connects feature maps of different resolutions in parallel, and then adds interactions between these feature maps based on the parallel connection. For example, ... Figure 2 As shown, feature maps are feature maps, convolutional units are convolutional units, downsampling is a downsampling layer, and upsampling is an upsampling layer. Figure 2 The first parallel branch consists of feature maps with the same input resolution, the second parallel branch consists of feature maps with half the input resolution, and the third parallel branch consists of feature maps with 1 / 4 the input resolution. Throughout the process, information in the parallel multi-resolution sub-networks is repeatedly exchanged to perform repeated multi-scale fusion.

[0058] S103: Post-process the location data to obtain the contour lines, axis lines, transverse diameter lines, and target measurement results of the chest cavity and heart, and output them to the display.

[0059] In this step, the chest cavity contour and the heart contour are automatically traced in the fetal four-chamber standard section on the ultrasound device interface to obtain the contour tracing lines of the chest cavity and the heart. Further, the key points are marked on the ultrasound device interface based on the position information of the output key points, and the position information of the key points is post-processed to obtain the results of the chest cavity axis, the heart axis, the chest cavity transverse diameter line, the heart transverse diameter line and other target measurement items, and marked on the ultrasound device interface.

[0060] Specifically, the position data is post-processed to obtain the axis and transverse diameter line results of the chest cavity and the heart, including: determining the chest cavity axis based on the first end point and the second end point of the chest cavity axis, and determining the heart axis based on the apical end point and the basal end point of the heart axis; determining the first intersection line segment of the perpendicular line of the chest cavity axis and the chest cavity contour, and determining the longest first intersection line segment as the chest cavity transverse diameter line; determining the second intersection line segment of the perpendicular line of the heart axis and the heart contour, and determining the longest second intersection line segment as the heart transverse diameter line.

[0061] In a specific implementation, the spine end point and the other end point based on the chest cavity axis can trace the chest cavity axis in the fetal four-chamber standard section, and the apical end point and the basal end point based on the heart axis can trace the heart axis in the fetal four-chamber standard section.

[0062] It can be understood that the chest cavity transverse diameter line is the longest line segment in the chest cavity contour perpendicular to the chest cavity axis, so after obtaining the chest cavity contour and the chest cavity axis, the points on the chest cavity axis are traversed, the first intersection line segment of the perpendicular line of the chest cavity axis and the chest cavity contour is calculated, and the length of the first intersection line segment is calculated, and the longest first intersection line segment is taken as the final chest cavity transverse diameter line.

[0063] Similarly, the heart transverse diameter line is the longest line segment in the heart contour perpendicular to the heart axis, so after obtaining the heart contour and the heart axis, the points on the heart axis are traversed, the second intersection line segment of the perpendicular line of the heart axis and the heart contour is calculated, and the length of the second intersection line segment is calculated, and the longest second intersection line segment is taken as the final heart transverse diameter line.

[0064] As a feasible implementation, the target measurement items in the embodiment can include any one or a combination of the heart axis angle, the heart-chest area ratio and the heart-chest transverse diameter ratio, and the position data is post-processed to obtain the target measurement item results, including: calculating the included angle between the heart axis and the chest cavity axis as the heart axis angle; and / or, calculating the ratio between the heart area and the chest cavity area as the heart-chest area ratio according to the chest cavity contour and the heart contour; and / or, calculating the ratio between the length of the heart transverse diameter and the length of the chest cavity transverse diameter as the heart-chest transverse diameter ratio.

[0065] In specific implementations, other measurement parameters can be calculated according to the thoracic cavity contour, the heart contour, the thoracic cavity axis, the heart axis, the thoracic cavity transverse diameter and the heart transverse diameter, including the heart axis angle, the heart-thoracic area ratio and the heart-thoracic transverse diameter ratio. The heart axis angle is the included angle between the heart axis and the thoracic cavity axis; the heart-thoracic area ratio is the ratio between the heart area and the thoracic cavity area, that is, the thoracic cavity area is calculated according to the thoracic cavity contour, the heart area is calculated according to the heart contour, and the ratio between the heart area and the thoracic cavity area is taken as the heart-thoracic area ratio; the heart-thoracic transverse diameter ratio is the ratio between the length of the heart transverse diameter and the length of the thoracic cavity transverse diameter.

[0066] Further, as a preferred embodiment, the embodiment further includes: if the heart axis angle exceeds a preset angle range, and / or the heart-thoracic area ratio exceeds a first threshold range, and / or the heart-thoracic transverse diameter ratio exceeds a second threshold range, an abnormal prompt is output. In specific implementations, a corresponding preset angle range is set for the heart axis angle, a corresponding first threshold range is set for the heart-thoracic area ratio, and a corresponding second threshold range is set for the heart-thoracic transverse diameter ratio, for example, the preset angle range is 45°±20°, the first threshold range is 0.25-0.33, and the second threshold range is 0.38-0.53, and the ultrasound device outputs an abnormal prompt when any of them exceeds the corresponding range.

[0067] The automatic measurement method of the fetal four-chamber heart standard section provided in the embodiment of the application automatically identifies the contour of the thoracic cavity and the heart, the position data of the key points in the fetal four-chamber heart standard section through an integrated detection model, and further processes the position data of the identified key points to automatically generate the contour trace line, the axis, the transverse diameter line and the target measurement item result of the thoracic cavity and the heart. As can be seen, the automatic measurement method of the fetal four-chamber heart standard section provided in the embodiment of the application realizes the automatic measurement of the contour trace line, the axis, the transverse diameter line and the target measurement item of the thoracic cavity and the heart, and improves the measurement efficiency of the fetal four-chamber heart standard section.

[0068] The embodiment of the application discloses an automatic measurement method of a fetal four-chamber heart standard section. Compared with the previous embodiment, the technical solution is further described and optimized. Specifically,

[0069] Referring to Figure 3 , the flow chart of another automatic measurement method of a fetal four-chamber heart standard section according to an example embodiment is shown in Figure 3 , which includes

[0070] S201: An ultrasound video is acquired, and the ultrasound video is input into a fetal four-chamber heart standard section identification model to obtain a fetal four-chamber heart standard section;

[0071] In this step, the ultrasound device collects an ultrasound video of the fetus, inputs the ultrasound video into a fetal four-chamber heart standard section recognition model, and obtains a fetal four-chamber heart standard section. The fetal four-chamber heart standard section needs to meet the following conditions: clearly showing the four chambers of the heart, being able to show the left and right pulmonary veins, being able to simultaneously show the opening and closing movements of the tricuspid and mitral valves, and being able to simultaneously show complete rib bone images and a spine as three strong echo points.

[0072] The process of training the fetal four-chamber heart standard section recognition model is as follows: obtaining training ultrasound videos and corresponding fetal four-chamber heart standard section annotations; and training the fetal four-chamber heart standard section recognition model by using the training ultrasound videos and the corresponding fetal four-chamber heart standard section annotations. In a specific implementation, training ultrasound videos are collected, fetal four-chamber heart standard section annotations of the training ultrasound videos are obtained on annotation software, the training ultrasound videos and the corresponding fetal four-chamber heart standard section annotations are input into the fetal four-chamber heart standard section recognition model for training, and iterative optimization is performed, so that a fetal four-chamber heart standard section recognition model with the highest recognition accuracy is finally trained. The fetal four-chamber heart standard section recognition model can use a target detection algorithm such as an RCNN series, a YOLO series, or RetinaNet, and is not specifically limited here.

[0073] S202: inputting the fetal four-chamber heart standard section into an integrated detection model;

[0074] S203: the integrated detection model outputs position data of outlines of a thoracic cavity and a heart and key points in the fetal four-chamber heart standard section; the key points include an apical end point and a basal end point of a heart axis and a first end point and a second end point of a thoracic cavity axis, and the first end point is a spine end point;

[0075] In this embodiment, a fetal four-chamber heart standard section measurement system is as shown in FIG. 8. Figure 4 As shown in FIG. 8, video annotation and model training, i.e., training of the fetal four-chamber heart standard section recognition model, image annotation and model training, i.e., training of the integrated detection model, inputting obstetric ultrasound videos into the fetal four-chamber heart standard section recognition model to obtain a fetal four-chamber heart standard section, and inputting the fetal four-chamber heart standard section into the integrated detection model to obtain and output position data of a thoracic cavity outline, a heart outline, and key points, including position data of a spine end point and another end point of a thoracic cavity axis and an apical end point and a basal end point of a heart axis.

[0076] S204: determining a thoracic cavity axis based on the first end point and the second end point of the thoracic cavity axis and determining a heart axis based on the apical end point and the basal end point of the heart axis;

[0077] S205: determining a first intersection line segment of a perpendicular line of the thoracic cavity axis and the thoracic cavity outline, and determining a thoracic cavity transverse diameter line as a first intersection line segment with the maximum length;

[0078] S206: determine the second intersection line segments of the perpendicular line of the heart axis and the heart contour, and determine the second intersection line segment with the maximum length as the heart transverse diameter line;

[0079] S207: calculate the included angle between the heart axis and the chest axis as the heart axis angle, calculate the ratio between the heart area and the chest area as the heart-chest area ratio according to the chest contour and the heart contour, and calculate the ratio between the length of the heart transverse diameter and the length of the chest transverse diameter as the heart-chest transverse diameter ratio.

[0080] The automatic measurement method of the fetal four-chamber heart standard section provided in the embodiments of the present application automatically identifies the fetal four-chamber heart standard section through a fetal four-chamber heart standard section identification model, automatically identifies the contour of the chest and the heart and the position data of the key points in the fetal four-chamber heart standard section through an integrated detection model, and further post-processes the position data of the identified key points to automatically generate the contour trace line of the chest and the heart, the axis, the transverse diameter line, and the target measurement item result. As can be seen, the measurement method of the fetal four-chamber heart standard section provided in the embodiments of the present application can automatically obtain the contour trace line of the chest and the heart, the axis, the transverse diameter line, and the target measurement item result at the same time after inputting the ultrasound video of the fetus, thereby improving the efficiency of obtaining the fetal four-chamber heart standard section and the measurement efficiency of the fetal four-chamber heart standard section, reducing the workload of the doctor, and thereby improving the examination efficiency and accuracy of the doctor.

[0081] A fetal four-chamber heart standard section measurement device provided in the embodiments of the present application is described below. The fetal four-chamber heart standard section measurement device described below can be mutually referred to with the fetal four-chamber heart standard section measurement method described above.

[0082] Referring to Figure 5 , a structure diagram of a fetal four-chamber heart standard section measurement device according to an exemplary embodiment is shown, as Figure 5 indicated, comprising:

[0083] The input module 501 is configured to input the fetal four-chamber heart standard section into an integrated detection model.

[0084] The output module 502 is configured to output the contour of the chest and the heart and the position data of the key points in the fetal four-chamber heart standard section by the integrated detection model.

[0085] The processing module 503 is configured to post-process the position data to obtain the contour trace line of the chest and the heart, the axis, the transverse diameter line, and the target measurement item result, and output to the display.

[0086] The measuring device for the standard section of the four-chamber heart of a fetus provided in the embodiments of the present application automatically identifies the position data of the contours of the thoracic cavity and the heart and the key points in the standard section of the four-chamber heart of a fetus through an integrated detection model, and further post-processes the position data of the identified key points to automatically generate the contour trace lines, axis lines, transverse diameter lines and target measurement item results of the thoracic cavity and the heart. It can be seen that the automatic measurement method for the standard section of the four-chamber heart of a fetus provided in the embodiments of the present application realizes the automatic measurement of the contour trace lines, axis lines, transverse diameter lines and target measurement items of the thoracic cavity and the heart, and improves the measurement efficiency of the standard section of the four-chamber heart of a fetus.

[0087] On the basis of the above-mentioned embodiments, as a preferred implementation manner, the semantic output channels of the integrated detection model include an output channel corresponding to the thoracic cavity contour, an output channel corresponding to the heart contour and an output channel corresponding to each key point.

[0088] On the basis of the above-mentioned embodiments, as a preferred implementation manner, the integrated detection model includes a plurality of branches in parallel, each of the branches includes a plurality of feature maps with the same resolution in series, the feature maps included in different branches are different in resolution, and the feature maps in different branches interact through up-sampling layers or down-sampling layers.

[0089] On the basis of the above-mentioned embodiments, as a preferred implementation manner, the method further comprises:

[0090] The first obtaining module is configured to obtain a training standard section of a four-chamber heart of a fetus and corresponding measurement annotations, the measurement annotations including a thoracic cavity contour, a heart contour, a thoracic cavity axis, a heart axis, a thoracic cavity transverse diameter and a heart transverse diameter.

[0091] The first training module is configured to train an integrated detection model by using the training standard section of the four-chamber heart of a fetus and the corresponding measurement annotations.

[0092] On the basis of the above-mentioned embodiments, as a preferred implementation manner, the method further comprises:

[0093] The standard section of the four-chamber heart of a fetus determining module is configured to obtain an ultrasound video and input the ultrasound video into a standard section recognition model to obtain the standard section of the four-chamber heart of a fetus.

[0094] On the basis of the above-mentioned embodiments, as a preferred implementation manner, the method further comprises:

[0095] The second obtaining module is configured to obtain a training ultrasound video and corresponding standard section annotations of a four-chamber heart of a fetus.

[0096] The second training module is configured to train a standard section recognition model of a four-chamber heart of a fetus by using the training ultrasound video and the corresponding standard section annotations of a four-chamber heart of a fetus.

[0097] In the above embodiment, as a preferred implementation, the key points include an apex end point and a base end point of a heart axis, and a first end point and a second end point of a thoracic axis, the first end point being a spine end point.

[0098] In the above embodiment, as a preferred implementation, the processing module 503 includes:

[0099] a first determining unit, configured to determine a thoracic axis based on the first end point and the second end point of the thoracic axis, and determine a heart axis based on the apex end point and the base end point of the heart axis;

[0100] a second determining unit, configured to determine a first intersection line segment of a perpendicular line of the thoracic axis and the thoracic contour, and determine a thoracic transverse diameter line as a first intersection line segment with the maximum length;

[0101] a third determining unit, configured to determine a second intersection line segment of a perpendicular line of the heart axis and the heart contour, and determine a heart transverse diameter line as a second intersection line segment with the maximum length.

[0102] In the above embodiment, as a preferred implementation, the target measurement items include any one or a combination of the following items: a heart-axis angle, a cardiothoracic area ratio, and a cardiothoracic transverse diameter ratio, and the processing module 503 includes:

[0103] a first calculating unit, configured to calculate an included angle between the heart axis and the thoracic axis as a heart-axis angle;

[0104] and / or a second calculating unit, configured to calculate a ratio between a heart area and a thoracic area as a cardiothoracic area ratio according to the thoracic contour and the heart contour;

[0105] and / or a third calculating unit, configured to calculate a ratio between a length of the heart transverse diameter and a length of the thoracic transverse diameter as a cardiothoracic transverse diameter ratio.

[0106] In the above embodiment, as a preferred implementation, the device further includes:

[0107] an abnormality prompting module, configured to output an abnormality prompt when the heart-axis angle exceeds a preset angle range, and / or the cardiothoracic area ratio exceeds a first threshold range, and / or the cardiothoracic transverse diameter ratio exceeds a second threshold range.

[0108] As to the device in the above embodiment, the specific manners in which the various modules perform operations have been described in detail in the embodiments of the method, and thus will not be described in detail here.

[0109] Based on the hardware implementation of the above program modules, and in order to implement the method of the embodiment of the application, the embodiment of the application further provides an ultrasonic device, Figure 6 A structural diagram of an ultrasonic device according to an exemplary embodiment is shown in FIG. 1. As shown in FIG. 1, the ultrasonic device includes: Figure 6

[0110] A communication interface 1 capable of information interaction with other devices such as network devices and the like;

[0111] A processor 2 connected with the communication interface 1 to realize information interaction with other devices, for running a computer program, and executing the measurement method of the fetal four-chamber heart standard section provided in one or more of the above technical solutions. The computer program is stored on a memory 3.

[0112] Of course, in actual application, various components in the ultrasonic device are coupled together through a bus system 4. It can be understood that the bus system 4 is used to realize the connection and communication between the components. The bus system 4 includes not only a data bus, but also a power supply bus, a control bus and a state signal bus. However, in order to clearly illustrate, all kinds of buses are marked as the bus system 4 in the Figure 6

[0113] The memory 3 in the embodiment of the application is used to store various types of data to support the operation of the ultrasonic device. Examples of these data include: any computer program used to operate on the ultrasonic device.

[0114] ​​It can be understood that the memory 3 can be a volatile memory or a non-volatile memory, and can also include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a ferromagnetic random access memory (FRAM), a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM). The magnetic surface memory can be a disk memory or a tape memory. The volatile memory can be a random access memory (RAM) used as an external cache. By way of example but not limitation, many forms of RAM can be used, such as static random access memory (SRAM), synchronous static random access memory (SSRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), sync link dynamic random access memory (SLDRAM), and direct rambus random access memory (DRRAM).The memory 3 described in the embodiments of the present application is intended to include, but not limited to, these and any other suitable types of memory.

[0115] The method disclosed in the embodiments of the present application can be applied to the processor 2 or implemented by the processor 2. The processor 2 can be an integrated circuit chip with processing capability. In the implementation process, each step of the above method can be completed by the integrated logic circuit of hardware in the processor 2 or the instruction in the form of software. The processor 2 described above can be a general processor, a DSP, or other programmable logic device, discrete gate or transistor logic device, discrete hardware component, etc. The processor 2 can implement or execute the disclosed methods, steps and logic block diagrams in the embodiments of the present application. The general processor can be a microprocessor or any conventional processor, etc. In combination with the steps of the method disclosed in the embodiments of the present application, the execution can be directly embodied as hardware decoding processor or executed by the combination of hardware and software modules in the decoding processor. The software module can be located in the storage medium, which is located in the memory 3. The processor 2 reads the program in the memory 3 and combines the hardware to complete the steps of the above method.

[0116] The processor 2 implements the corresponding flow in each method of the embodiments of the present application when executing the program. For brevity, it will not be repeated here.

[0117] In the exemplary embodiments, the embodiments of the present application also provide a storage medium, i.e. a computer storage medium, specifically a computer readable storage medium, such as a memory 3 storing a computer program, which can be executed by the processor 2 to complete the steps of the above method. The computer readable storage medium can be FRAM, ROM, PROM, EPROM, EEPROM, Flash Memory, magnetic surface memory, optical disc, or CD-ROM, etc.

[0118] Those skilled in the art can understand that all or part of the steps of the above method embodiments can be completed by program instruction related hardware, and the above program can be stored in a computer readable storage medium. When the program is executed, the steps of the above method embodiments are executed; and the above storage medium includes mobile storage device, ROM, RAM, magnetic disc or optical disc, etc. various storage medium which can store program code.

[0119] Alternatively, the above-mentioned integrated units of the present application, if realized in the form of software function modules and sold or used as independent products, can also be stored in a computer-readable storage medium. Based on such understanding, the technical solutions of the embodiments of the present application can be embodied in the form of a software product, which is stored in a storage medium and includes a plurality of instructions for causing an ultrasonic device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: mobile storage devices, ROM, RAM, magnetic disks or optical disks, and various media that can store program codes.

[0120] The above is merely specific embodiments of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. An automatic measurement method for a standard section of a fetal four-chamber heart, characterized in that, include: Input the standard cross-section of the fetal four-chamber heart into the integrated detection model; The integrated detection model outputs the outline of the thoracic cavity and heart, and the location data of key points in the standard cross-section of the fetal four-chamber heart; wherein, the key points include the apex and base of the heart axis, the first and second endpoints of the thoracic axis, and the first endpoint being the spinal endpoint; The location data is post-processed to obtain the contour lines, axis lines, transverse diameter lines, and target measurement results of the chest cavity and heart, and then output to the display. The location data is post-processed to obtain the axis and transverse diameter results of the thoracic cavity and heart, including: The thoracic axis is determined based on the first and second endpoints of the thoracic axis, and the cardiac axis is determined based on the apex and base of the cardiac axis. Determine the first intersection line segment between the perpendicular line of the thoracic cavity axis and the thoracic cavity contour, and determine the first intersection line segment with the longest length as the transverse diameter line of the thoracic cavity; Determine the second intersection segment between the perpendicular line of the heart axis and the heart contour, and determine the second intersection segment with the longest length as the transverse diameter line of the heart; The target measurement items include any one or a combination of any of the following: cardioaxial angle, cardiothoracic area ratio, and cardiothoracic transverse diameter ratio.

2. The automatic measurement method according to claim 1, characterized in that, The semantic output channels of the integrated detection model include the output channel corresponding to the chest cavity contour, the output channel corresponding to the heart contour, and the output channel corresponding to each key point.

3. The automatic measurement method according to claim 1, characterized in that, The integrated detection model includes multiple branches in parallel, each branch including multiple feature maps of the same resolution in series, different branches contain feature maps with different resolutions, and feature maps in different branches interact through upsampling layers or downsampling layers.

4. The automatic measurement method according to claim 1, characterized in that, Also includes: Obtain the standard cross-section of the four-chamber heart of the training fetus and the corresponding measurement markings, the measurement markings including the chest cavity contour, heart contour, chest cavity axis, heart axis, chest cavity transverse diameter, and heart transverse diameter; The integrated detection model was trained using the standard cross-section of the four chambers of the fetus and the corresponding measurement annotations.

5. The automatic measurement method according to claim 1, characterized in that, Before inputting the standard four-chamber view of the fetus into the integrated detection model, the following steps are also included: Acquire ultrasound video and input the ultrasound video into a standard section recognition model to obtain the standard section of the fetal four-chamber heart.

6. The automatic measurement method according to claim 5, characterized in that, Also includes: Acquire training ultrasound videos and corresponding standard cross-sectional annotations of the fetal four-chamber heart; The training ultrasound video and the corresponding standard section of the fetal four-chamber heart were used to train the standard section recognition model.

7. The automatic measurement method according to claim 1, characterized in that, The location data is post-processed to obtain the target measurement results, including: The angle between the cardiac axis and the thoracic axis is calculated as the cardiac axis angle. And / or, the ratio between the heart area and the chest area is calculated based on the chest cavity contour and the heart contour as the cardiothoracic area ratio; And / or, calculate the ratio between the length of the transverse diameter of the heart and the length of the transverse diameter of the thoracic cavity as the cardiothoracic diameter ratio.

8. The automatic measurement method according to claim 1, characterized in that, Also includes: If the axial angle exceeds a preset angle range, and / or the cardiothoracic area ratio exceeds a first threshold range, and / or the cardiothoracic transverse diameter ratio exceeds a second threshold range, an abnormal prompt will be output.

9. An automatic measuring device for a standard section of a fetal four-chamber heart, characterized in that, include: The input module is used to input the standard cross-section of the fetal four-chamber heart into the integrated detection model; The output module is used by the integrated detection model to output the contours of the thoracic cavity and heart and the location data of key points in the standard cross-section of the fetal four-chamber heart; wherein, the key points include the apex and base of the heart axis, the first and second endpoints of the thoracic axis, and the first endpoint being the spinal endpoint; The processing module is used to post-process the location data to obtain the contour lines, axis lines, transverse diameter lines, and target measurement results of the chest cavity and heart, and output them to the display. The processing module includes: The first determining unit is used to determine the thoracic axis based on the first and second endpoints of the thoracic axis, and to determine the cardiac axis based on the apex and base of the cardiac axis. The second determining unit is used to determine the first intersection line segment between the vertical line of the thoracic cavity axis and the thoracic cavity contour, and to determine the first intersection line segment with the longest length as the transverse diameter line of the thoracic cavity. The third determining unit is used to determine the second intersection line segment between the perpendicular line of the heart axis and the heart contour, and to determine the second intersection line segment with the longest length as the transverse diameter line of the heart. The target measurement items include any one or a combination of any of the following: cardioaxial angle, cardiothoracic area ratio, and cardiothoracic transverse diameter ratio.

10. An ultrasonic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the steps of the automatic measurement method for the standard section of the fetal four-chamber heart as described in any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the automatic measurement method for the standard section of the fetal four-chamber heart as described in any one of claims 1 to 8.

Images