Method and device for determining thickness of nuchal translucency

Abstract

The present invention discloses a method and device for determining the thickness of a nuchal translucency layer, the method comprising: acquiring a target ultrasound image, determining a target nuchal translucency layer image from the target ultrasound image; determining the image edge of the target nuchal translucency layer image, and determining the midpoint connection line of the target nuchal translucency layer image according to the image edge; and determining the thickness of the nuchal translucency layer corresponding to the target nuchal translucency layer image according to the distance between all pixel points on the midpoint connection line and the image edge in the corresponding normal direction. It can be seen that the present invention can determine the thickness of the nuchal translucency layer based on the distance between the pixel points on the midpoint connection line of the nuchal translucency layer image and the image edge in the corresponding normal direction, so that when targeting a curved nuchal translucency layer image, accurate thickness information can be calculated, thereby providing an accurate data basis for subsequent fetal clinical diagnosis.

Description

Technical Field

[0001] The present invention relates to the field of medical imaging technology, and in particular to a method and device for determining the thickness of a nuchal translucency layer. Background Art

[0002] Nuchal Translucency (NT) ultrasound examination is the preferred method for early pregnancy prenatal screening of fetal abnormalities, especially in the screening of fetal chromosomal abnormalities. The thickening of the fetal nuchal translucency has a high sensitivity to fetal abnormalities. The measurement of the thickness of the fetal nuchal translucency in early pregnancy has an important clinical diagnostic value in detecting fetal abnormalities, which is helpful to guide the prognosis of the fetus and improve eugenics.

[0003] Traditional medical technology generally relies on manual measurement when measuring the thickness of the nuchal translucency layer of fetal ultrasound images. Since manual measurement of NT thickness depends on the doctor's experience and operating skills, manual errors often occur in the measurement results. In addition, doctors need to repeat the operation, which is monotonous and boring and easy to get tired. Therefore, in the prior art, realizing automatic measurement of the thickness of the fetal nuchal translucency layer is the development direction, but the methods for measuring the nuchal translucency layer of the fetus disclosed in the prior art usually cannot solve the problem of measuring the curved nuchal translucency layer, that is, when the nuchal translucency layer is curved, the error of the thickness measurement result is large. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a method and device for determining the thickness of the nuchal translucency layer, which can calculate accurate thickness information for curved nuchal translucency images, thereby providing an accurate data basis for subsequent fetal clinical diagnosis.

[0005] In order to solve the above technical problems, the first aspect of the present invention discloses a method for determining the thickness of the nuchal transparent layer, the method comprising:

[0006] Acquire a target ultrasound image, and determine a target nuchal translucency image from the target ultrasound image;

[0007] Determine the image edge of the target nuchal transparent layer image, and determine the midpoint connection line of the target nuchal transparent layer image according to the image edge;

[0008] The thickness of the neck transparent layer corresponding to the target neck transparent layer image is determined according to the distance between all the pixel points on the midpoint line and the image edge in the corresponding normal direction.

[0009] As an optional implementation, in the first aspect of the present invention, determining the midpoint line of the target nuchal translucency image according to the image edge comprises:

[0010] Determining the image vertical direction of the target nuchal transparent layer image; the image vertical direction is substantially perpendicular to the extension direction of the target nuchal transparent layer image;

[0011] A midpoint line of the target nuchal transparent layer image is determined according to a plurality of pixel point pairs of the image edge of the target nuchal transparent layer image in a vertical direction of the image.

[0012] As an optional implementation, in the first aspect of the present invention, determining the vertical direction of the target nuchal translucency image comprises:

[0013] Calculate the minimum circumscribed rectangle corresponding to the target neck transparent layer image;

[0014] The extension direction of the wide side of the minimum circumscribed rectangle is determined as the image vertical direction of the target nuchal transparent layer image.

[0015] As an optional implementation, in the first aspect of the present invention, determining the midpoint connection line of the target nuchal translucency layer image according to a plurality of pixel point pairs of the image edge of the target nuchal translucency layer image in the vertical direction of the image comprises:

[0016] Determine a plurality of pixel pairs of the image edge of the target nuchal translucency image in the vertical direction of the image; each of the pixel pairs includes two pixel points where the image edge intersects with a virtual line extending in a direction parallel to the vertical direction of the image;

[0017] For any of the pixel point pairs, calculate the midpoint of a virtual line connecting two pixel points included in the pixel point pair;

[0018] A midpoint line of the target nuchal transparent layer image is determined according to the midpoints of all the pixel pairs.

[0019] As an optional implementation, in the first aspect of the present invention, determining the thickness of the nuchal transparent layer corresponding to the target nuchal transparent layer image according to the distance between all pixel points on the midpoint line and the image edge in the corresponding normal direction comprises:

[0020] Generate normal lines corresponding to all pixel points on the midpoint line;

[0021] Calculating the intersection distance between two intersection points of each normal line and the edge of the image;

[0022] The maximum value among the intersection distances of all the normal lines is determined as the thickness of the neck translucency layer corresponding to the target neck translucency layer image.

[0023] As an optional embodiment, in the first aspect of the present invention, after determining the target nuchal translucency image from the target ultrasound image, the method further includes:

[0024] Determining the curvature of the target nuchal translucency image, and judging whether the curvature is less than a preset curvature threshold;

[0025] When the judgment result is yes, proceed to the next step.

[0026] As an optional implementation, in the first aspect of the present invention, the target ultrasound image includes a plurality of ultrasound image frames; and determining a target nuchal translucency image from the target ultrasound image includes:

[0027] The target ultrasound image is input into a feature detection model; the feature detection model is used to select a target ultrasound image frame that meets a preset midsagittal plane feature condition from the multiple ultrasound image frames, and determine a target nuchal translucency image from the target ultrasound image frame;

[0028] The target nuchal translucency image determined by the feature detection model is acquired.

[0029] A second aspect of an embodiment of the present invention discloses a device for determining the thickness of a nuchal transparent layer, the device comprising:

[0030] An image determination module, used to obtain a target ultrasound image and determine a target nuchal translucency image from the target ultrasound image;

[0031] A midline determination module, used to determine the image edge of the target nuchal transparent layer image, and determine the midpoint connection line of the target nuchal transparent layer image according to the image edge;

[0032] The thickness determination module is used to determine the thickness of the neck transparent layer corresponding to the target neck transparent layer image according to the distance between all pixel points on the midpoint line and the image edge in the corresponding normal direction.

[0033] As an optional implementation, in the second aspect of the present invention, the specific manner in which the midline determination module determines the midpoint connection line of the target nuchal translucency image according to the image edge includes:

[0034] Determining the image vertical direction of the target nuchal transparent layer image; the image vertical direction is substantially perpendicular to the extension direction of the target nuchal transparent layer image;

[0035] A midpoint line of the target nuchal transparent layer image is determined according to a plurality of pixel point pairs of the image edge of the target nuchal transparent layer image in a vertical direction of the image.

[0036] As an optional implementation, in the second aspect of the present invention, the specific manner in which the midline determination module determines the vertical direction of the target nuchal translucency image comprises:

[0037] Calculate the minimum circumscribed rectangle corresponding to the target neck transparent layer image;

[0038] The extension direction of the wide side of the minimum circumscribed rectangle is determined as the image vertical direction of the target nuchal transparent layer image.

[0039] As an optional implementation, in the second aspect of the present invention, the midline determination module determines the specific method of connecting the midpoints of the target nuchal translucency image according to a plurality of pixel point pairs of the image edge of the target nuchal translucency image in the vertical direction of the image, including:

[0040] Determine a plurality of pixel pairs of the image edge of the target nuchal translucency image in the vertical direction of the image; each of the pixel pairs includes two pixel points where the image edge intersects with a virtual line extending in a direction parallel to the vertical direction of the image;

[0041] For any of the pixel point pairs, calculate the midpoint of a virtual line connecting two pixel points included in the pixel point pair;

[0042] A midpoint line of the target nuchal transparent layer image is determined according to the midpoints of all the pixel pairs.

[0043] As an optional implementation, in the second aspect of the present invention, the thickness determination module determines the specific manner in which the thickness of the neck transparent layer corresponding to the target neck transparent layer image is determined according to the distance between all pixel points on the midpoint line and the image edge in the corresponding normal direction, including:

[0044] Generate normal lines corresponding to all pixel points on the midpoint line;

[0045] Calculating the intersection distance between two intersection points of each normal line and the edge of the image;

[0046] The maximum value among the intersection distances of all the normal lines is determined as the thickness of the neck translucency layer corresponding to the target neck translucency layer image.

[0047] As an optional embodiment, in the second aspect of the present invention, the device also includes a curvature judgment module, which is used to determine the curvature of the target nuchal translucency image and determine whether the curvature is less than a preset curvature threshold, and when the judgment result is yes, enable the centerline determination module and / or the thickness determination module to perform their functions.

[0048] As an optional implementation, in the second aspect of the present invention, the target ultrasound image includes a plurality of ultrasound image frames; the specific manner in which the image determination module determines the target nuchal translucency image from the target ultrasound image includes:

[0049] The target ultrasound image is input into a feature detection model; the feature detection model is used to select a target ultrasound image frame that meets a preset midsagittal plane feature condition from the multiple ultrasound image frames, and determine a target nuchal translucency image from the target ultrasound image frame;

[0050] The target nuchal translucency image determined by the feature detection model is acquired.

[0051] The third aspect of the present invention discloses another device for determining the thickness of the nuchal translucency layer, the device comprising:

[0052] A memory storing executable program code;

[0053] a processor coupled to the memory;

[0054] The processor calls the executable program code stored in the memory to execute part or all of the steps in the method for determining the thickness of the nuchal transparent layer disclosed in the first aspect of the present invention.

[0055] The fourth aspect of the present invention discloses a computer storage medium, which stores computer instructions. When the computer instructions are called, they are used to execute some or all of the steps in the method for determining the thickness of the neck transparent layer disclosed in the first aspect of the present invention.

[0056] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0057] In an embodiment of the present invention, a method and device for determining the thickness of the nuchal translucency layer is disclosed, the method comprising: obtaining a target ultrasound image, determining a target nuchal translucency layer image from the target ultrasound image; determining the image edge of the target nuchal translucency layer image, and determining the midpoint line of the target nuchal translucency layer image according to the image edge; and determining the thickness of the nuchal translucency layer corresponding to the target nuchal translucency layer image according to the distance between all pixel points on the midpoint line and the image edge in the corresponding normal direction. It can be seen that the embodiment of the present invention can determine the thickness of the nuchal translucency layer based on the distance between the pixel points on the midpoint line of the nuchal translucency layer image and the image edge in the corresponding normal direction, so that when targeting a curved nuchal translucency layer image, accurate thickness information can be calculated, thereby providing an accurate data basis for subsequent fetal clinical diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0059] Figure 1 It is a flow chart of a method for determining the thickness of a neck transparent layer disclosed in an embodiment of the present invention.

[0060] Figure 2 It is a flow chart of another method for determining the thickness of the nuchal transparent layer disclosed in an embodiment of the present invention.

[0061] Figure 3 It is a structural schematic diagram of a device for determining the thickness of a neck transparent layer disclosed in an embodiment of the present invention.

[0062] Figure 4 It is a schematic structural diagram of another device for determining the thickness of the neck transparent layer disclosed in an embodiment of the present invention.

[0063] Figure 5 It is a structural schematic diagram of another device for determining the thickness of the neck transparent layer disclosed in an embodiment of the present invention.

[0064] Figure 6 It is a schematic diagram for determining the thickness of the neck transparent layer disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0065] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0066] The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, device, product or equipment that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or equipment.

[0067] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0068] The present invention discloses a method and device for determining the thickness of the nuchal translucency layer, which can determine the thickness of the nuchal translucency layer based on the distance between the pixel points on the midpoint line of the nuchal translucency layer image and the image edge in the corresponding normal direction, so that when targeting a curved nuchal translucency layer image, accurate thickness information can be calculated, thereby providing an accurate data basis for subsequent fetal clinical diagnosis. The following are detailed descriptions.

[0069] Embodiment 1

[0070] See also Figure 1 , Figure 1 : is a flow chart of a method for determining the thickness of the nuchal translucency layer disclosed in an embodiment of the present invention. Figure 1 The method for determining the thickness of the nuchal translucency layer described is applied to an image processing system / image processing device / image processing server (wherein the image processing server includes a local image processing server or a cloud image processing server). Figure 1 As shown, the method for determining the thickness of the nuchal transparent layer may include the following operations:

[0071] 101. Obtain a target ultrasound image, and determine a target nuchal translucency image from the target ultrasound image.

[0072] In an embodiment of the present invention, the target ultrasound image is a fetal ultrasound image at least including the nuchal translucency portion, which can be obtained from one of a variety of media such as an ultrasound system, DICOM, and a USB flash drive. Specifically, the target ultrasound image is mainly collected by a transducer of an ultrasound imaging device. The ultrasound imaging device at least includes a transducer, an ultrasound host, an input unit, a control unit, and a memory. The ultrasound imaging device may include a display screen, and the display screen of the ultrasound imaging device may be a display of a recognition system. The transducer is used to transmit and receive ultrasound waves. The transducer is stimulated by a transmission pulse to transmit ultrasound waves to a target tissue (such as an organ, tissue, blood vessel, etc. in the fetus), and after a certain delay, receives an ultrasound echo with information of the target tissue reflected from the target area, and reconverts the ultrasound echo into an electrical signal to obtain an ultrasound image or video. The transducer can be connected to the ultrasound host in a wired or wireless manner. Among them, the input unit is used to input control instructions of the operator. The input unit may be at least one of a keyboard, a tracking ball, a mouse, a touch panel, a handle, a dial, a joystick, and a foot switch. The input unit may also input non-contact signals, such as sound, gestures, sight, or brain wave signals. Specifically, the control unit can at least control the focus information, the driving frequency information, the driving voltage information and the scanning information such as the imaging mode. The control unit processes the signal differently according to the different imaging modes required by the user, obtains ultrasonic image data of different modes, and then forms ultrasonic images of different modes through logarithmic compression, dynamic range adjustment, digital scan conversion and other processing, such as B image, C image, D image, Doppler blood flow image, elastic image containing tissue elastic characteristics, etc., or other types of two-dimensional ultrasonic images or three-dimensional ultrasonic images. It should be understood that the target ultrasonic image can also be an ultrasonic image stored in a storage medium, such as a cloud server, a U disk or a hard disk. Specifically, first, after receiving an external signal through the ultrasonic diagnostic system, the signal is processed to generate an ultrasonic image containing the neck transparent layer to be tested; then, the neck transparent layer is placed in the middle of the image, and the approximate range containing the neck transparent layer is determined by a trackball, a mouse or a button, and the ultrasonic image is cropped to obtain a rough target neck transparent layer image, which not only reduces interference information, but also enhances the real-time performance of the algorithm.

[0073] 102. Determine the image edge of the target nuchal transparent layer image, and determine the midpoint connection line of the target nuchal transparent layer image according to the image edge.

[0074] In the embodiment of the present invention, the image edge of the target neck transparent layer image can be obtained by edge detection algorithm or neural network model. For example, canny, sobel and other gradient-based edge detection methods can be used to determine the image edge of the target neck transparent layer image. For example, a phase asymmetric edge detection method that is more robust to noise and grayscale non-uniformity can also be selected to determine the image edge of the target neck transparent layer image. Specifically, the phase asymmetric edge detection method can include:

[0075] Construct a multi-scale, multi-directional filter, convolve the target neck clear layer image with the multi-scale, multi-directional filter group, obtain a group of even filter responses and odd filter responses for each target neck clear layer image pixel, combine the filter responses of each target neck clear layer image pixel using phase asymmetry measurement, and thus generate the image edge of the target neck clear layer image. Optionally, the filter can be selected as Gabor or LogGabor.

[0076] 103. Determine the thickness of the neck transparent layer corresponding to the target neck transparent layer image according to the distance between all pixel points on the midpoint line and the image edge in the corresponding normal direction.

[0077] By such setting, since what is calculated is the distance between the pixel point and the edge of the image in the normal direction, compared with the prior art method of estimating the thickness of the entire neck transparent layer in the width direction, this method can obviously achieve a more accurate thickness measurement effect when targeting a curved neck transparent layer image.

[0078] It can be seen that the above-mentioned embodiment of the invention can determine the thickness of the nuchal translucency layer based on the distance between the pixel points on the line connecting the midpoints of the nuchal translucency layer image and the image edge in the corresponding normal direction, so that when the nuchal translucency layer image is curved, accurate thickness information can be calculated, thereby providing an accurate data basis for subsequent fetal clinical diagnosis.

[0079] As an optional implementation, in step 102, determining the midpoint line of the target nuchal translucency image according to the image edge includes:

[0080] Determine the image vertical direction of the target nuchal translucency image;

[0081] According to a plurality of pixel point pairs of the image edge of the target nuchal transparent layer image in the vertical direction of the image, a midpoint connecting line of the target nuchal transparent layer image is determined.

[0082] In the embodiment of the present invention, the vertical direction of the image is approximately perpendicular to the extension direction of the target neck transparent layer image. The extension direction of the target neck transparent layer image can be estimated by a direction detection algorithm, and the vertical direction of the image can be determined by drawing a line segment perpendicular to the extension direction.

[0083] It can be seen that through this optional implementation, the midpoint line of the target neck transparent layer image can be determined according to the multiple pixel point pairs of the image edge of the target neck transparent layer image in the vertical direction of the image, so that the determined midpoint line can be better used to characterize the center position of the transparent layer image, and then the thickness of the neck transparent layer finally determined can achieve a more accurate thickness measurement effect when targeting a curved neck transparent layer image.

[0084] As an optional implementation, in the above step, determining the vertical direction of the target nuchal translucency image includes:

[0085] Calculate the minimum enclosing rectangle corresponding to the target neck transparent layer image;

[0086] The extension direction of the wide side of the minimum circumscribed rectangle is determined as the image vertical direction of the target nuchal transparent layer image.

[0087] Optionally, the minimum circumscribed rectangle of the image edge of the target neck transparent layer image can also be calculated to obtain a more accurate effect. Optionally, the wide side of the minimum circumscribed rectangle should be the shorter pair of sides between the two pairs of sides of the rectangle. Through the above settings, the long side of the minimum circumscribed rectangle, that is, the longer pair of sides of the two pairs of sides, can be used to characterize the approximate extension direction of the target neck transparent layer image, so that the wide side of the minimum circumscribed rectangle perpendicular to it can be used to characterize the vertical direction of the image of the target neck transparent layer image.

[0088] It can be seen that through this optional implementation, the extension direction of the wide side of the minimum circumscribed rectangle can be determined as the image vertical direction of the target neck transparent layer image, so that the determined image vertical direction can be better used to characterize the vertical direction of the transparent layer image, and then the subsequently determined midpoint connection line can be better used to characterize the center position of the transparent layer image.

[0089] As an optional implementation manner, in the above step, determining the midpoint connection line of the target nuchal translucency layer image according to a plurality of pixel point pairs of the image edge of the target nuchal translucency layer image in the vertical direction of the image includes:

[0090] Determine a plurality of pixel point pairs at the image edge of the target nuchal transparent layer image in the vertical direction of the image;

[0091] For any pixel pair, calculate the midpoint of the virtual line connecting the two pixels included in the pixel pair;

[0092] According to the midpoints of all pixel pairs, the midpoint connecting line of the target nuchal transparent layer image is determined.

[0093] In the embodiment of the present invention, each pixel point pair includes two pixel points where the image edge intersects with a virtual line extending in a direction parallel to the image vertical direction. Specifically, multiple virtual lines extending in a direction parallel to the image vertical direction may be drawn along the image edge to obtain multiple pixel point pairs of the image edge in the image vertical direction.

[0094] In the embodiment of the present invention, the midpoint line of the target nuchal transparent layer image is determined based on the midpoints of all pixel pairs. The midpoint line can be obtained by directly connecting the midpoints of all pixel pairs, or by fitting the midpoints of all pixel pairs using a fitting algorithm to obtain a fitting curve of all midpoints to obtain the midpoint line. The former is more suitable for the case where there are a large number of midpoints, and the latter is suitable for the case where there are a small number of midpoints. The technician can make a choice according to the actual situation, and the present invention does not limit it.

[0095] It can be seen that through this optional implementation, the midpoint line of the target neck transparent layer image can be determined according to the midpoints of all pixel pairs, so that the determined midpoint line can be better used to characterize the center position of the transparent layer image, and then the thickness of the neck transparent layer finally determined can achieve a more accurate thickness measurement effect when targeting a curved neck transparent layer image.

[0096] As an optional implementation, in the above step 103, determining the thickness of the neck transparent layer corresponding to the target neck transparent layer image according to the distance between all pixel points on the midpoint line and the image edge in the corresponding normal direction includes:

[0097] Generate normal lines corresponding to all pixels on the midpoint line;

[0098] Calculate the intersection distance between the two intersection points of each normal line and the image edge;

[0099] The maximum value among the intersection distances of all normal lines is determined as the thickness of the nuchal translucency layer corresponding to the target nuchal translucency layer image.

[0100] Specifically, the schematic diagram for determining the thickness of the nuchal translucency layer can refer to Figure 6 .

[0101] It can be seen that through this optional implementation, the maximum value of the intersection distances of all normal lines can be determined as the thickness of the nuchal translucency layer corresponding to the target nuchal translucency layer image, so that when targeting a curved nuchal translucency layer image, accurate thickness information can be calculated, thereby providing an accurate data basis for subsequent fetal clinical diagnosis.

[0102] As an optional implementation, the target ultrasound image includes a plurality of ultrasound image frames, which may be an ultrasound video acquired by an ultrasound diagnostic system. Specifically, in the above step 101, determining the target nuchal translucency image from the target ultrasound image includes:

[0103] Inputting the target ultrasound image into the feature detection model;

[0104] Acquire the target nuchal translucency image determined by the feature detection model.

[0105] In an embodiment of the present invention, a feature detection model is used to filter out target ultrasound image frames that meet preset median sagittal plane feature conditions from multiple ultrasound image frames, and determine a target nuchal translucency image from the target ultrasound image frames. Optionally, the feature detection model can be a trained neural network model, which can adopt a convolutional network model and be trained using a pre-set training image set. Preferably, the feature detection model can include an image frame selection network and an image segmentation network, and these two networks can be trained together until convergence, or trained separately until convergence and then connected for use, which is not limited by the present invention. Among them, the image frame selection network is used to filter out target ultrasound image frames that meet preset median sagittal plane feature conditions from multiple ultrasound image frames, and the image segmentation network is used to determine the target nuchal translucency image from the target ultrasound image frame.

[0106] Optionally, the preset median sagittal plane feature condition may be that the ultrasound image of the fetus displayed by the target ultrasound image frame is the median sagittal section of the fetus, and preferably, the median sagittal section is the median sagittal section of the cranial brain. Specifically, the image frame selection network may determine whether the current ultrasound image frame is the target ultrasound image frame by determining whether the current ultrasound image frame includes one or more of the fetal nose tip image, nasal bone image, maxillary bone image, diencephalon image, and nuchal translucency image.

[0107] It can be seen that through this optional implementation, the target ultrasound image can be input into the feature detection model to obtain the target nuchal translucency image determined by the feature detection model, thereby obtaining an accurate target nuchal translucency image, so that accurate thickness information can be calculated subsequently, thereby providing an accurate data basis for subsequent fetal clinical diagnosis.

[0108] Embodiment 2

[0109] See also Figure 2 , Figure 2 FIG. 1 is a flow chart of another method for determining the thickness of the nuchal translucency layer disclosed in an embodiment of the present invention. Figure 2The method for determining the thickness of the nuchal translucency layer described is applied to an image processing system / image processing device / image processing server (wherein the image processing server includes a local image processing server or a cloud image processing server). Figure 2 As shown, the method for determining the thickness of the nuchal transparent layer may include the following operations:

[0110] 201. Obtain a target ultrasound image, and determine a target nuchal translucency image from the target ultrasound image.

[0111] 202. Determine the curvature of the target nuchal translucency image, and determine whether the curvature is less than a preset curvature threshold.

[0112] 203. When the judgment result is yes, execute step 204.

[0113] 204. Determine the image edge of the target neck transparent layer image, and determine the midpoint connection line of the target neck transparent layer image according to the image edge.

[0114] 205. Determine the thickness of the neck transparent layer corresponding to the target neck transparent layer image according to the distance between all pixel points on the midpoint line and the image edge in the corresponding normal direction.

[0115] In the embodiment of the present invention, for the relevant description of steps 201, 204-205, please refer to the detailed description of steps 101-103 in the first embodiment, and the embodiment of the present invention will not be repeated.

[0116] Optionally, when the judgment result in step 202 is no, the thickness determination method can be directly terminated, that is, the thickness measurement process can be directly terminated to select other images with smaller curvature for thickness measurement.

[0117] In the embodiment of the present invention, the curvature of the target nuchal translucency image is used to characterize the curvature of the image, and can also be further used to characterize the curvature of the fetal body. Through the above steps, the fetal irradiation state corresponding to the fetal ultrasound image can be judged in advance, and the subsequent steps are performed only when it is judged that the curvature is small, thereby avoiding the inaccurate thickness information measured by the image with a large curvature. This is because the thickness of the nuchal translucency will become thinner due to the curvature when the curvature is large, so this situation needs to be avoided. Optionally, the execution order of steps 202-203 is not necessarily after step 201, and it can also be executed after step 204, which can also play a certain avoidance effect, and the present invention does not limit it.

[0118] It can be seen that the above-mentioned embodiment of the invention can perform the subsequent thickness measurement step only when it is determined that the curvature of the target nuchal translucency image is less than the preset curvature threshold, thereby screening out the nuchal translucency image that is too curved to improve the accuracy of the thickness information obtained by subsequent calculation, thereby providing an accurate data basis for subsequent fetal clinical diagnosis.

[0119] In an optional embodiment, in step 202, determining the curvature of the target nuchal translucency image and judging whether the curvature is less than a preset curvature threshold value includes:

[0120] Calculate the curvature of the image edge of the target nuchal translucency image;

[0121] Determine whether the arc is less than a preset arc threshold.

[0122] Optionally, the method for determining the image edge can refer to the method for determining the image edge of the target neck transparent layer image in the first embodiment, which will not be described in detail here. Optionally, the curvature of the image edge can be the curvature of any edge portion of the image edge or the average value of the curvatures of multiple edge portions of the image edge. For example, the image edge of the target neck transparent layer can include two upper and lower edges, and its curvature can be determined by calculating the curvature of the upper edge or the lower edge. Specifically, the determination of the curvature of the upper edge or the lower edge can be determined based on a curvature solution algorithm.

[0123] Optionally, the image edge of the target neck transparent layer is divided into the upper and lower edges by scanning the image edge of the target neck transparent layer column by column, and selecting the boundary positions where the pixel values ​​change as the upper and lower boundaries. Optionally, the upper and lower boundaries can be corrected respectively using a Gaussian smoothing window.

[0124] Optionally, the method of dividing the image edge of the target neck transparent layer into two upper and lower edges can be: calculating the minimum circumscribed rectangle of the image edge, calculating a dividing line that passes through the wide side and divides the minimum circumscribed rectangle into two equal rectangles, and using the dividing line to divide the image edge into two upper and lower edges.

[0125] In an optional embodiment, in step 202, determining the curvature of the target nuchal translucency image and judging whether the curvature is less than a preset curvature threshold value includes:

[0126] Determine the midpoint line of the target nuchal transparent layer image and calculate the arc of the midpoint line;

[0127] Determine whether the arc of the midpoint connection line is less than the preset arc threshold.

[0128] Optionally, the method for determining the midpoint line can refer to the method for determining the midpoint line of the target nuchal transparent layer image in Example 1, which will not be repeated here. Optionally, the radian of the midpoint line can be the radian of any line segment of the midpoint line or the average radian of multiple line segment portions of the midpoint line. Specifically, the radian of the midpoint line can be determined based on a radian solution algorithm.

[0129] Embodiment 3

[0130] See also Figure 3 , Figure 3 : is a schematic diagram of a structure of a device for determining the thickness of a nuchal transparent layer disclosed in an embodiment of the present invention. Figure 3 The apparatus for determining the thickness of the nuchal translucency layer described is applied to an image processing system / image processing device / image processing server (wherein the image processing server includes a local image processing server or a cloud image processing server). Figure 3 As shown, the device for determining the thickness of the nuchal translucency layer may include:

[0131] The image determination module 301 is used to obtain a target ultrasound image and determine a target nuchal translucency image from the target ultrasound image.

[0132] In an embodiment of the present invention, the target ultrasound image is a fetal ultrasound image at least including the nuchal translucency portion, which can be obtained from one of a variety of media such as an ultrasound system, DICOM, and a USB flash drive. Specifically, the target ultrasound image is mainly collected by a transducer of an ultrasound imaging device. The ultrasound imaging device at least includes a transducer, an ultrasound host, an input unit, a control unit, and a memory. The ultrasound imaging device may include a display screen, and the display screen of the ultrasound imaging device may be a display of a recognition system. The transducer is used to transmit and receive ultrasound waves. The transducer is stimulated by a transmission pulse to transmit ultrasound waves to a target tissue (such as an organ, tissue, blood vessel, etc. in the fetus), and after a certain delay, receives an ultrasound echo with information of the target tissue reflected from the target area, and reconverts the ultrasound echo into an electrical signal to obtain an ultrasound image or video. The transducer can be connected to the ultrasound host in a wired or wireless manner. Among them, the input unit is used to input control instructions of the operator. The input unit may be at least one of a keyboard, a tracking ball, a mouse, a touch panel, a handle, a dial, a joystick, and a foot switch. The input unit may also input non-contact signals, such as sound, gestures, sight, or brain wave signals. Specifically, the control unit can at least control the focus information, the driving frequency information, the driving voltage information and the scanning information such as the imaging mode. The control unit processes the signal differently according to the different imaging modes required by the user, obtains ultrasonic image data of different modes, and then forms ultrasonic images of different modes through logarithmic compression, dynamic range adjustment, digital scan conversion and other processing, such as B image, C image, D image, Doppler blood flow image, elastic image containing tissue elastic characteristics, etc., or other types of two-dimensional ultrasonic images or three-dimensional ultrasonic images. It should be understood that the target ultrasonic image can also be an ultrasonic image stored in a storage medium, such as a cloud server, a U disk or a hard disk. Specifically, first, after receiving an external signal through the ultrasonic diagnostic system, the signal is processed to generate an ultrasonic image containing the neck transparent layer to be tested; then, the neck transparent layer is placed in the middle of the image, and the approximate range containing the neck transparent layer is determined by a trackball, a mouse or a button, and the ultrasonic image is cropped to obtain a rough target neck transparent layer image, which not only reduces interference information, but also enhances the real-time performance of the algorithm.

[0133] The midline determination module 302 is used to determine the image edge of the target nuchal transparent layer image, and determine the midpoint connection line of the target nuchal transparent layer image according to the image edge.

[0134] In the embodiment of the present invention, the image edge of the target neck transparent layer image can be obtained by edge detection algorithm or neural network model. For example, canny, sobel and other gradient-based edge detection methods can be used to determine the image edge of the target neck transparent layer image. For example, a phase asymmetric edge detection method that is more robust to noise and grayscale non-uniformity can also be selected to determine the image edge of the target neck transparent layer image. Specifically, the phase asymmetric edge detection method can include:

[0135] Construct a multi-scale, multi-directional filter, convolve the target neck clear layer image with the multi-scale, multi-directional filter group, obtain a group of even filter responses and odd filter responses for each target neck clear layer image pixel, combine the filter responses of each target neck clear layer image pixel using phase asymmetry measurement, and thus generate the image edge of the target neck clear layer image. Optionally, the filter can be selected as Gabor or LogGabor.

[0136] The thickness determination module 303 is used to determine the thickness of the neck transparent layer corresponding to the target neck transparent layer image according to the distance between all the pixels on the midpoint line and the image edge in the corresponding normal direction.

[0137] By such setting, since what is calculated is the distance between the pixel point and the edge of the image in the normal direction, compared with the prior art method of estimating the thickness of the entire neck transparent layer in the width direction, this method can obviously achieve a more accurate thickness measurement effect when targeting a curved neck transparent layer image.

[0138] It can be seen that the above-mentioned embodiment of the invention can determine the thickness of the nuchal translucency layer based on the distance between the pixel points on the line connecting the midpoints of the nuchal translucency layer image and the image edge in the corresponding normal direction, so that when the nuchal translucency layer image is curved, accurate thickness information can be calculated, thereby providing an accurate data basis for subsequent fetal clinical diagnosis.

[0139] As an optional implementation, the specific method in which the midline determination module 302 determines the midpoint connection line of the target nuchal translucency image according to the image edge includes:

[0140] Determine the image vertical direction of the target nuchal translucency image;

[0141] According to a plurality of pixel point pairs of the image edge of the target nuchal transparent layer image in the vertical direction of the image, a midpoint connecting line of the target nuchal transparent layer image is determined.

[0142] In the embodiment of the present invention, the vertical direction of the image is approximately perpendicular to the extension direction of the target neck transparent layer image. The extension direction of the target neck transparent layer image can be estimated by a direction detection algorithm, and the vertical direction of the image can be determined by drawing a line segment perpendicular to the extension direction.

[0143] It can be seen that through this optional implementation, the midpoint line of the target neck transparent layer image can be determined according to the multiple pixel point pairs of the image edge of the target neck transparent layer image in the vertical direction of the image, so that the determined midpoint line can be better used to characterize the center position of the transparent layer image, and then the thickness of the neck transparent layer finally determined can achieve a more accurate thickness measurement effect when targeting a curved neck transparent layer image.

[0144] As an optional implementation, the specific manner in which the midline determination module 302 determines the vertical direction of the target nuchal translucency image includes:

[0145] Calculate the minimum enclosing rectangle corresponding to the target neck transparent layer image;

[0146] The extension direction of the wide side of the minimum circumscribed rectangle is determined as the image vertical direction of the target nuchal transparent layer image.

[0147] Optionally, the minimum circumscribed rectangle of the image edge of the target neck transparent layer image can also be calculated to obtain a more accurate effect. Optionally, the wide side of the minimum circumscribed rectangle should be the shorter pair of sides between the two pairs of sides of the rectangle. Through the above settings, the long side of the minimum circumscribed rectangle, that is, the longer pair of sides of the two pairs of sides, can be used to characterize the approximate extension direction of the target neck transparent layer image, so that the wide side of the minimum circumscribed rectangle perpendicular to it can be used to characterize the vertical direction of the image of the target neck transparent layer image.

[0148] It can be seen that through this optional implementation, the extension direction of the wide side of the minimum circumscribed rectangle can be determined as the image vertical direction of the target neck transparent layer image, so that the determined image vertical direction can be better used to characterize the vertical direction of the transparent layer image, and then the subsequently determined midpoint connection line can be better used to characterize the center position of the transparent layer image.

[0149] As an optional implementation, the midline determination module 302 determines the specific method of connecting the midpoints of the target nuchal translucency image according to a plurality of pixel pairs of the image edge of the target nuchal translucency image in the vertical direction of the image, including:

[0150] Determine a plurality of pixel point pairs at the image edge of the target nuchal transparent layer image in the vertical direction of the image;

[0151] For any pixel pair, calculate the midpoint of the virtual line connecting the two pixels included in the pixel pair;

[0152] According to the midpoints of all pixel pairs, the midpoint connecting line of the target nuchal transparent layer image is determined.

[0153] In the embodiment of the present invention, each pixel point pair includes two pixel points where the image edge intersects with a virtual line extending in a direction parallel to the image vertical direction. Specifically, multiple virtual lines extending in a direction parallel to the image vertical direction may be drawn along the image edge to obtain multiple pixel point pairs of the image edge in the image vertical direction.

[0154] In the embodiment of the present invention, the midpoint line of the target nuchal transparent layer image is determined based on the midpoints of all pixel pairs. The midpoint line can be obtained by directly connecting the midpoints of all pixel pairs, or by fitting the midpoints of all pixel pairs using a fitting algorithm to obtain a fitting curve of all midpoints to obtain the midpoint line. The former is more suitable for the case where there are a large number of midpoints, and the latter is suitable for the case where there are a small number of midpoints. The technician can make a choice according to the actual situation, and the present invention does not limit it.

[0155] It can be seen that through this optional implementation, the midpoint line of the target neck transparent layer image can be determined according to the midpoints of all pixel pairs, so that the determined midpoint line can be better used to characterize the center position of the transparent layer image, and then the thickness of the neck transparent layer finally determined can achieve a more accurate thickness measurement effect when targeting a curved neck transparent layer image.

[0156] As an optional implementation, the thickness determination module 303 determines the thickness of the neck transparent layer corresponding to the target neck transparent layer image according to the distance between all pixels on the midpoint line and the image edge in the corresponding normal direction, including:

[0157] Generate normal lines corresponding to all pixels on the midpoint line;

[0158] Calculate the intersection distance between the two intersection points of each normal line and the image edge;

[0159] The maximum value among the intersection distances of all normal lines is determined as the thickness of the nuchal translucency layer corresponding to the target nuchal translucency layer image.

[0160] Specifically, the schematic diagram for determining the thickness of the nuchal translucency layer can refer to Figure 6 .

[0161] It can be seen that through this optional implementation, the maximum value of the intersection distances of all normal lines can be determined as the thickness of the nuchal translucency layer corresponding to the target nuchal translucency layer image, so that when targeting a curved nuchal translucency layer image, accurate thickness information can be calculated, thereby providing an accurate data basis for subsequent fetal clinical diagnosis.

[0162] As an optional implementation, the target ultrasound image includes a plurality of ultrasound image frames, which may be ultrasound videos acquired by an ultrasound diagnostic system. The specific manner in which the image determination module 301 determines the target nuchal translucency image from the target ultrasound image includes:

[0163] Inputting the target ultrasound image into the feature detection model;

[0164] Acquire the target nuchal translucency image determined by the feature detection model.

[0165] In an embodiment of the present invention, a feature detection model is used to filter out target ultrasound image frames that meet preset median sagittal plane feature conditions from multiple ultrasound image frames, and determine a target nuchal translucency image from the target ultrasound image frames. Optionally, the feature detection model can be a trained neural network model, which can adopt a convolutional network model and be trained using a pre-set training image set. Preferably, the feature detection model can include an image frame selection network and an image segmentation network, and these two networks can be trained together until convergence, or trained separately until convergence and then connected for use, which is not limited by the present invention. Among them, the image frame selection network is used to filter out target ultrasound image frames that meet preset median sagittal plane feature conditions from multiple ultrasound image frames, and the image segmentation network is used to determine the target nuchal translucency image from the target ultrasound image frame.

[0166] Optionally, the preset median sagittal plane feature condition may be that the ultrasound image of the fetus displayed by the target ultrasound image frame is the median sagittal section of the fetus, and preferably, the median sagittal section is the median sagittal section of the cranial brain. Specifically, the image frame selection network may determine whether the current ultrasound image frame is the target ultrasound image frame by determining whether the current ultrasound image frame includes one or more of the fetal nose tip image, nasal bone image, maxillary bone image, diencephalon image, and nuchal translucency image.

[0167] It can be seen that through this optional implementation, the target ultrasound image can be input into the feature detection model to obtain the target nuchal translucency image determined by the feature detection model, thereby obtaining an accurate target nuchal translucency image, so that accurate thickness information can be calculated subsequently, thereby providing an accurate data basis for subsequent fetal clinical diagnosis.

[0168] As an optional implementation, Figure 4As shown, the device further includes a curvature determination module 304, which is used to determine the curvature of the target nuchal transparent layer image, and determine whether the curvature is less than a preset curvature threshold, and when the judgment result is yes, enable the centerline determination module 302 and / or the thickness determination module 303 to perform their functions. Optionally, when the judgment result of the curvature determination module 304 is no, the thickness determination method can be directly terminated, that is, the thickness measurement process can be directly terminated to select other images with smaller curvature for thickness measurement.

[0169] In the embodiment of the present invention, the curvature of the target nuchal translucency image is used to characterize the curvature of the image, and can also be further used to characterize the curvature of the fetal body. Through the above steps, the fetal irradiation state corresponding to the fetal ultrasound image can be judged in advance, and the subsequent steps are performed only when it is judged that the curvature is small, thereby avoiding the inaccurate thickness information measured by the image with a large curvature. This is because the thickness of the nuchal translucency will become thinner due to the curvature when the curvature is large, so this situation needs to be avoided. Optionally, the execution order of steps 202-203 is not necessarily after step 201, and it can also be executed after step 204, which can also play a certain avoidance effect, and the present invention does not limit it.

[0170] It can be seen that by implementing this optional implementation, the subsequent thickness measurement step can be performed only when it is determined that the curvature of the target nuchal translucency image is less than the preset curvature threshold, thereby screening out nuchal translucency images that are too curved, so as to improve the accuracy of the thickness information obtained by subsequent calculation, thereby providing an accurate data basis for subsequent fetal clinical diagnosis.

[0171] As an optional implementation, the curvature determination module 304 determines the curvature of the target nuchal transparent layer image and determines whether the curvature is less than a preset curvature threshold in a specific manner including:

[0172] Calculate the curvature of the image edge of the target nuchal translucency image;

[0173] Determine whether the arc is less than a preset arc threshold.

[0174] Optionally, the method for determining the image edge can refer to the method for determining the image edge of the target neck transparent layer image in the first embodiment, which will not be described in detail here. Optionally, the curvature of the image edge can be the curvature of any edge portion of the image edge or the average value of the curvatures of multiple edge portions of the image edge. For example, the image edge of the target neck transparent layer can include two upper and lower edges, and its curvature can be determined by calculating the curvature of the upper edge or the lower edge. Specifically, the determination of the curvature of the upper edge or the lower edge can be determined based on a curvature solution algorithm.

[0175] Optionally, the image edge of the target neck transparent layer is divided into the upper and lower edges by scanning the image edge of the target neck transparent layer column by column, and selecting the boundary positions where the pixel values ​​change as the upper and lower boundaries. Optionally, the upper and lower boundaries can be corrected respectively using a Gaussian smoothing window.

[0176] Optionally, the method of dividing the image edge of the target neck transparent layer into two upper and lower edges can be: calculating the minimum circumscribed rectangle of the image edge, calculating a dividing line that passes through the wide side and divides the minimum circumscribed rectangle into two equal rectangles, and using the dividing line to divide the image edge into two upper and lower edges.

[0177] As an optional implementation, the curvature determination module 304 determines the curvature of the target nuchal transparent layer image and determines whether the curvature is less than a preset curvature threshold in a specific manner including:

[0178] Determine the midpoint line of the target nuchal transparent layer image and calculate the arc of the midpoint line;

[0179] Determine whether the arc of the midpoint connection line is less than the preset arc threshold.

[0180] Optionally, the method for determining the midpoint line can refer to the method for determining the midpoint line of the target nuchal transparent layer image in Example 1, which will not be repeated here. Optionally, the radian of the midpoint line can be the radian of any line segment of the midpoint line or the average radian of multiple line segment portions of the midpoint line. Specifically, the radian of the midpoint line can be determined based on a radian solution algorithm.

[0181] Embodiment 4

[0182] See also Figure 5 , Figure 5 It is another device for determining the thickness of the neck transparent layer disclosed in an embodiment of the present invention. Figure 5 The apparatus for determining the thickness of the nuchal translucency layer described is applied to an image processing system / image processing device / image processing server (wherein the image processing server includes a local image processing server or a cloud image processing server). Figure 5 As shown, the device for determining the thickness of the nuchal translucency layer may include:

[0183] A memory 401 storing executable program codes;

[0184] a processor 402 coupled to the memory 401;

[0185] The processor 402 calls the executable program code stored in the memory 401 to execute the steps of the method for determining the thickness of the nuchal transparent layer described in the first or second embodiment.

[0186] Embodiment 5

[0187] An embodiment of the present invention discloses a computer-readable storage medium storing a computer program for electronic data exchange, wherein the computer program enables a computer to execute the steps of the method for determining the thickness of the nuchal transparent layer described in the first or second embodiment.

[0188] Embodiment 6

[0189] An embodiment of the present invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to enable a computer to execute the steps of the method for determining the thickness of the nuchal transparent layer described in Embodiment 1 or Embodiment 2.

[0190] The device embodiments described above are only illustrative, wherein the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, that is, they may be located in one place, or they may be distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without creative work.

[0191] Through the specific description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the above technical solution can be essentially or partly contributed to the prior art in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, and the storage medium includes a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable rewritable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, a magnetic disk storage, a magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0192] Finally, it should be noted that the method and device for determining the thickness of the neck transparent layer disclosed in the embodiment of the present invention disclose only the preferred embodiments of the present invention, which are only used to illustrate the technical scheme of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical schemes described in the aforementioned embodiments can still be modified, or some of the technical features therein can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical schemes from the spirit and scope of the technical schemes of the various embodiments of the present invention.

Claims

1. A method for determining the thickness of the nuchal translucency layer, It is characterized in that The method comprises: Acquire a target ultrasound image, and determine a target nuchal translucency image from the target ultrasound image; the target ultrasound image includes a plurality of ultrasound image frames; Determine the image edge of the target nuchal transparent layer image, and determine the midpoint connection line of the target nuchal transparent layer image according to the image edge; Determine the thickness of the neck transparent layer corresponding to the target neck transparent layer image according to the distance between all pixel points on the midpoint line and the edge of the image in the corresponding normal direction; The image edge of the target nuchal transparent layer image is obtained by an edge detection algorithm or a neural network model, and the edge detection algorithm includes one of phase asymmetric edge detection algorithms that are relatively robust to noise and grayscale non-uniformity; The phase-asymmetric edge detection algorithm is obtained in the following manner: Constructing a multi-scale, multi-directional filter, convolving the target neck clear layer image with the multi-scale, multi-directional filter group respectively, obtaining a group of even filter responses and odd filter responses for each target neck clear layer image pixel, combining the filter responses of each target neck clear layer image pixel by using a phase asymmetry measure, thereby generating an image edge of the target neck clear layer image; wherein the filter includes Gabor or LogGabor; Wherein, determining a target nuchal translucency image from the target ultrasound image comprises: The target ultrasound image is input into a feature detection model; the feature detection model is used to select a target ultrasound image frame that meets a preset midsagittal plane feature condition from the multiple ultrasound image frames, and determine a target nuchal translucency image from the target ultrasound image frame; Acquire the target nuchal translucency image determined by the feature detection model; And, after determining the target nuchal translucency image from the target ultrasound image, the method further includes: Determining the curvature of the target nuchal translucency image, and judging whether the curvature is less than a preset curvature threshold; When the judgment result is yes, proceed to the next step.

2. The method for determining the thickness of the nuchal translucency layer according to claim 1, It is characterized in that Determining the midpoint connection line of the target nuchal translucency image according to the image edge includes: Determining the image vertical direction of the target nuchal transparent layer image; the image vertical direction is substantially perpendicular to the extension direction of the target nuchal transparent layer image; A midpoint line of the target nuchal transparent layer image is determined according to a plurality of pixel point pairs of the image edge of the target nuchal transparent layer image in a vertical direction of the image.

3. The method for determining the thickness of the nuchal translucency layer according to claim 2, It is characterized in that Determining the vertical direction of the target nuchal translucency image includes: Calculate the minimum circumscribed rectangle corresponding to the target neck transparent layer image; The extension direction of the wide side of the minimum circumscribed rectangle is determined as the image vertical direction of the target nuchal transparent layer image.

4. The method for determining the thickness of the nuchal translucency layer according to claim 2, It is characterized in that Determining a midpoint connection line of the target nuchal transparent layer image according to a plurality of pixel point pairs of the image edge of the target nuchal transparent layer image in a vertical direction of the image comprises: Determine a plurality of pixel pairs of the image edge of the target nuchal translucency image in the vertical direction of the image; each of the pixel pairs includes two pixel points where the image edge intersects with a virtual line extending in a direction parallel to the vertical direction of the image; For any of the pixel point pairs, calculate the midpoint of a virtual line connecting two pixel points included in the pixel point pair; A midpoint line of the target nuchal transparent layer image is determined according to the midpoints of all the pixel pairs.

5. The method for determining the thickness of the nuchal translucency layer according to claim 1, It is characterized in that Determining the thickness of the neck transparent layer corresponding to the target neck transparent layer image according to the distance between all pixel points on the midpoint line and the image edge in the corresponding normal direction includes: Generate normal lines corresponding to all pixel points on the midpoint line; Calculating the intersection distance between two intersection points of each normal line and the edge of the image; The maximum value among the intersection distances of all the normal lines is determined as the thickness of the neck translucency layer corresponding to the target neck translucency layer image.

6. A device for determining the thickness of the nuchal translucency layer, It is characterized in that The device comprises: An image determination module, used to acquire a target ultrasound image, and determine a target nuchal translucency image from the target ultrasound image; the target ultrasound image includes a plurality of ultrasound image frames; A midline determination module, used to determine the image edge of the target nuchal transparent layer image, and determine the midpoint connection line of the target nuchal transparent layer image according to the image edge; A thickness determination module, for determining the thickness of the neck transparent layer corresponding to the target neck transparent layer image according to the distances between all the pixel points on the midpoint line and the edge of the image in the corresponding normal direction; The image edge of the target nuchal transparent layer image is obtained by an edge detection algorithm or a neural network model, and the edge detection algorithm includes one of phase asymmetric edge detection algorithms that are relatively robust to noise and grayscale non-uniformity; The phase-asymmetric edge detection algorithm is obtained in the following manner: Constructing a multi-scale, multi-directional filter, convolving the target neck clear layer image with the multi-scale, multi-directional filter group respectively, obtaining a group of even filter responses and odd filter responses for each target neck clear layer image pixel, combining the filter responses of each target neck clear layer image pixel by using a phase asymmetry measure, thereby generating an image edge of the target neck clear layer image; wherein the filter includes Gabor or LogGabor; The specific method in which the image determination module determines the target nuchal translucency image from the target ultrasound image includes: The target ultrasound image is input into a feature detection model; the feature detection model is used to select a target ultrasound image frame that meets a preset midsagittal plane feature condition from the multiple ultrasound image frames, and determine a target nuchal translucency image from the target ultrasound image frame; Acquire the target nuchal translucency image determined by the feature detection model; Furthermore, the device also includes a curvature determination module, which is used to determine the curvature of the target nuchal translucency image after the image determination module determines the target nuchal translucency image from the target ultrasound image, and to determine whether the curvature is less than a preset curvature threshold, and when the judgment result is yes, enable the midline determination module and / or the thickness determination module to perform their functions.

7. A device for determining the thickness of a nuchal translucent layer, It is characterized in that The device comprises: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the method for determining the thickness of the nuchal translucency layer as described in any one of claims 1-5.

8. A computer storage medium, It is characterized in that The computer storage medium stores computer instructions, and when the computer instructions are called, they are used to execute the method for determining the thickness of the neck transparent layer according to any one of claims 1-5.

Images