Method for determining area, ultrasonic device and computer storage medium

Abstract

Disclosed are a method for determining an area, an ultrasonic device, and a computer storage medium. The method for determining an area based on an ultrasonic image includes: emitting an ultrasonic beam to a target object including a gastric antrum region by stimulating an ultrasonic probe, and receiving an ultrasonic echo returned from the target object by the ultrasonic probe to obtain an ultrasonic echo signal; a processor processes the ultrasonic echo signal to obtain an ultrasonic image of the target object; the processor determines a measurement frame image based on the ultrasonic image; the processor determines the gastric antrum region of the target object in the measurement frame image, and further determines the area of ​​the gastric antrum region. It can be seen that in the embodiment of the present invention, the processor processes and analyzes the ultrasonic image to obtain the area of ​​the gastric antrum region of the target object, eliminating the need for the user to manually select based on experience. On the one hand, this method saves time and effort and is more efficient; on the other hand, it does not rely on user behavior, and the determined area is more accurate.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of ultrasound, and more particularly, to a method for determining an area based on an ultrasound image, an ultrasound device, and a computer storage medium. Background Art

[0002] Ultrasound examination is a radiation-free, convenient, and low-cost imaging tool. Reflux aspiration of gastric contents is a serious complication of anesthesia, potentially leading to airway obstruction, aspiration pneumonia, and, in severe cases, death. The gastric antrum is a common site for inflammation, ulcers, and tumors. Ultrasound measurement of the cross-sectional area of ​​the gastric antrum is an important tool for assessing the risk of reflux aspiration.

[0003] Currently, manual methods are commonly used in clinical practice to measure the cross-sectional area of ​​the gastric antrum. This involves the doctor manually marking the major and minor axes of the antrum and approximating them using an ellipse, or manually outlining the antrum's boundaries. However, this method is demanding for the doctor, requiring precise markings and prone to errors. Summary of the Invention

[0004] Embodiments of the present invention provide a method for determining an area based on an ultrasound image, an ultrasound device, and a computer storage medium.

[0005] In a first aspect, a method for determining an area based on an ultrasound image is provided, comprising:

[0006] transmitting an ultrasonic beam toward a target object including a gastric antrum region by stimulating the ultrasonic probe, and receiving an ultrasonic echo returned from the target object by the ultrasonic probe to obtain an ultrasonic echo signal;

[0007] The processor processes the ultrasonic echo signal to obtain an ultrasonic image of the target object;

[0008] The processor determines a measurement frame image based on the ultrasound image;

[0009] The processor determines a gastric antrum region of the target object in the measurement frame image and further determines an area of ​​the gastric antrum region.

[0010] In a second aspect, a method for determining an area based on an ultrasound image is provided, comprising:

[0011] transmitting an ultrasonic beam toward a target object including a gastric antrum region by stimulating the ultrasonic probe, and receiving an ultrasonic echo returned from the target object by the ultrasonic probe to obtain an ultrasonic echo signal;

[0012] The processor processes the ultrasonic echo signal to obtain an ultrasonic image of the target object;

[0013] The processor determines a measurement frame image based on the ultrasound image;

[0014] The processor inputs the measurement frame image into a pre-trained deep learning model to obtain an output of the deep learning model, which includes the area of ​​the gastric antrum region in the measurement frame image.

[0015] In a third aspect, a method for determining an area based on an ultrasound image is provided, comprising:

[0016] The ultrasonic probe is excited to transmit an ultrasonic beam toward the target object, and the ultrasonic probe receives an ultrasonic echo returned from the target object to obtain an ultrasonic echo signal;

[0017] The processor processes the ultrasonic echo signal to obtain an ultrasonic image of the target object;

[0018] The processor determines a measurement frame image based on the ultrasound image;

[0019] The processor determines a target region of a target to be measured of a target object in a measurement frame image, and further determines an area of ​​the target region, wherein the target to be measured includes a target organ or tissue, and the target region includes a section or a region of interest in the section.

[0020] In a fourth aspect, an ultrasound device is provided, comprising:

[0021] Ultrasound probe;

[0022] a transmitting / receiving circuit, configured to excite the ultrasonic probe to transmit an ultrasonic beam toward a target object, and control the ultrasonic probe to receive ultrasonic echoes returned from the target object to obtain ultrasonic echo signals;

[0023] a memory, configured to store a program executed by the processor;

[0024] A processor, configured to implement any one of the methods described in the first to third aspects above.

[0025] In a fifth aspect, a computer storage medium is provided, on which a computer program is stored. When the computer program is executed by a computer or a processor, the steps of any one of the methods described in the first to third aspects are implemented.

[0026] It can be seen that in the embodiment of the present invention, the processor processes and analyzes the ultrasound image to obtain the area of ​​the gastric antrum region of the target object, eliminating the need for the user to manually select based on experience. On the one hand, this method saves time and effort and is more efficient; on the other hand, it does not rely on user behavior, and the determined area is more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments or descriptions of the prior art. 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 paying any creative work.

[0028] Figure 1 is a schematic block diagram of an ultrasound device;

[0029] Figure 2 is a schematic flow chart of a method for determining an area based on an ultrasound image according to an embodiment of the present invention;

[0030] Figure 3 is a schematic diagram of a curve showing changes in the area of ​​the gastric antrum region over time according to an embodiment of the present invention;

[0031] Figure 4 is another schematic diagram of a curve showing changes in the area of ​​the gastric antrum region over time according to an embodiment of the present invention;

[0032] Figure 5 is another schematic diagram of a curve showing changes in the area of ​​the gastric antrum region over time according to an embodiment of the present invention;

[0033] Figure 6 is another schematic diagram of a curve showing changes in the area of ​​the gastric antrum region over time according to an embodiment of the present invention;

[0034] Figure 7 is another schematic diagram of a curve showing changes in the area of ​​the gastric antrum region over time according to an embodiment of the present invention;

[0035] Figure 8 is a schematic diagram of a gastric antrum cycle and a single gastric antrum contraction shown on an example of a curve of changes in the area of ​​the gastric antrum region over time according to an embodiment of the present invention;

[0036] Figure 9 is another schematic flow chart of a method for determining an area based on an ultrasound image according to an embodiment of the present invention;

[0037] Figure 10 is another schematic flowchart of a method for determining an area based on an ultrasound image according to an embodiment of the present invention. DETAILED DESCRIPTION

[0038] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0039] Reflux aspiration of gastric contents is a serious complication of anesthesia that can lead to airway obstruction or aspiration pneumonia, and in severe cases, death. Measuring the cross-sectional area of ​​the gastric antrum through ultrasound examination is an important means of assessing the risk of reflux aspiration. Currently, manual methods are commonly used in clinical practice, that is, manually marking the major and minor axes of the gastric antrum area and using elliptical area approximation or manually outlining the gastric antrum boundary. During the examination, the doctor must first save a video of the gastric antrum section and select a frame from it for measurement. This operation requires relatively precise marking and is prone to errors.

[0040] In intensive care, in order to assess the tolerance of enteral nutrition in patients with feeding tubes and formulate nutritional treatment plans, it is necessary to measure parameters related to gastric motility, such as gastric emptying time and antral motility index. In the measurement of gastric emptying time, the cross-sectional area of ​​the antrum needs to be measured multiple times at certain intervals. After each measurement, the examination time needs to be recorded and the measured results compared with the cross-sectional area of ​​the antrum in the fasting state. When measuring the antral motility index, the doctor needs to record a long video of the antral section, identify each maximum relaxation and minimum contraction, measure the antral area, and calculate the average antral contraction frequency and amplitude. This operation requires multiple manual identifications, manual markings, and certain calculations. The operation is cumbersome, time-consuming, and prone to errors.

[0041] This problem is further amplified because physicians performing bedside ultrasound in anesthesia and intensive care settings are often not specialized sonographers and have limited understanding and proficiency in ultrasound examinations.

[0042] In view of the above problems, this application provides a method for determining area based on ultrasound images. It should be noted that although the above problems are posed for the gastric antrum, the method described in this application can be applied to a variety of different organs or tissues, including but not limited to: the gastric antrum, cardiac cavities, blood vessels, kidneys, bladders, follicles, tumors, etc.

[0043] The present application provides a method for determining an area based on an ultrasound image. The method can be implemented by an ultrasound device, which can also be called an ultrasound device or an ultrasound system, such as Figure 1 As shown, the ultrasound device 10 includes an ultrasound probe 110 , a transmit / receive selection switch 120 , a transmitting circuit 160 , a receiving circuit 170 , a memory 130 , a processor 140 , and a display 150 .

[0044] The transmit / receive selector switch 120 can activate the ultrasonic probe 110 via the transmit circuit 160 to transmit an ultrasonic beam toward the target object, and control the ultrasonic probe 110 via the receive circuit 170 to receive ultrasonic echoes returned from the target object to obtain ultrasonic echo signals. The processor 140 can process the ultrasonic echo signals to obtain an ultrasonic image of the target object; further, the processor 140 can determine the area of ​​the target to be measured in the ultrasonic image.

[0045] For example, the ultrasound image obtained by the processor 140 may be stored in the memory 130. Optionally, the ultrasound image may be displayed on the display 150.

[0046] Optionally, the display 150 in the ultrasound device 10 may be a touch screen display, a liquid crystal display, or the like; or the display 150 may be an independent display device such as a liquid crystal display or a television set that is independent of the ultrasound device 10; or the display 150 may be a display screen of an electronic device such as a smartphone or a tablet computer, etc. The number of displays 150 may be one or more.

[0047] Optionally, the memory 130 in the ultrasound device 10 may be a volatile memory and / or a non-volatile memory, a removable memory and / or a non-removable memory, etc., for example, it may be a flash memory card, a solid-state memory, a hard disk, etc.

[0048] Optionally, the processor 140 in the ultrasound device 10 can be implemented by software, hardware, firmware, or any combination thereof, and can use a circuit, a single or multiple application specific integrated circuits (ASICs), a single or multiple general-purpose integrated circuits, a single or multiple microprocessors, a single or multiple programmable logic devices, or any combination of the aforementioned circuits and / or devices, or other suitable circuits or devices, so that the processor 140 can execute the corresponding steps of the methods in the various embodiments of this specification.

[0049] It should be understood that Figure 1 The components shown in the ultrasound device 10 are merely illustrative, and the device may include more or fewer components. For example, the ultrasound device 10 may also include input devices such as a keyboard, a mouse, a scroll wheel, a trackball, and / or an output device such as a printer. The corresponding external input / output ports may be wireless communication modules, wired communication modules, or a combination of the two. The external input / output ports may also be implemented based on USB, bus protocols such as CAN, and / or wired network protocols. The present invention is not limited to this.

[0050] The following will be combined Figures 2 to 10A method for determining an area based on an ultrasound image according to an embodiment of the present invention is described.

[0051] Figure 2 FIG. 4 is a schematic flowchart of a method for determining an area based on an ultrasound image according to an embodiment of the present invention. Figure 2 The method shown includes the following steps:

[0052] S110, emitting an ultrasonic beam toward a target object including a gastric antrum region by stimulating an ultrasonic probe, and receiving an ultrasonic echo returned from the target object by the ultrasonic probe to obtain an ultrasonic echo signal;

[0053] S120, the processor processes the ultrasonic echo signal to obtain an ultrasonic image of the target object;

[0054] S130, the processor determines a measurement frame image based on the ultrasound image;

[0055] S140: The processor determines a gastric antrum region of the target object in the measurement frame image, and further determines an area of ​​the gastric antrum region.

[0056] In this way, in this application, the processor can process and analyze the ultrasound image to obtain the area of ​​the gastric antrum region of the target object, eliminating the need for the user to manually select based on experience. On the one hand, this method saves time and effort and is more efficient; on the other hand, it does not rely on user behavior, and the determined area is more accurate.

[0057] In one embodiment, the ultrasound image determined in S120 is a frame image. Accordingly, in S130, the frame image can be used as the measurement frame image.

[0058] In another embodiment, the ultrasound image determined in S120 includes multiple frames of images, for example, the ultrasound image is video data. Accordingly, in S130, one frame from the multiple frames of images may be selected as the measurement frame image.

[0059] As an implementation manner, in S130 , the image quality of each frame of the multiple frames of images may be determined; and the image with the best image quality among the multiple frames of images may be determined as the measurement frame image.

[0060] Exemplarily, image quality can be expressed as a quality score, and the quality score includes a weighted sum of multiple sub-scores. The sub-scores of an image can include at least one of the following: the intersection-over-union ratio of the gastric antrum region in the image to the gastric antrum region in the previous image; the reciprocal of the Hausdorff distance between the gastric antrum region in the image and the gastric antrum region in the previous image; the reciprocal of the area difference between the gastric antrum region in the image and the gastric antrum region in the previous image; the reciprocal of the length of the longest straight line segment on the boundary of the gastric antrum region in the image; the reciprocal of the number of acute angles on the boundary of the gastric antrum region in the image; the reciprocal of the area of ​​inward depression on the boundary of the gastric antrum region in the image; or one or more scores of an image given by one or more machine learning models.

[0061] It can be understood that each sub-score is directly proportional to the image quality.

[0062] Taking the t-th frame image in the multi-frame image as an example, the sub-score of the t-th frame image may include:

[0063] (1) Gastric antrum region R in the t-th frame image t and the gastric antrum region R in the t-1 frame image t-1 The intersection and union ratio.

[0064] Specifically, the gastric antrum region R in the t-1 frame image can be determined t-1 The area S t-1 , determine the gastric antrum area R in the t-th frame image t The area S t . Further, the two regions R can be calculated t-1 and R t The intersection area S1 and the union area S2 are obtained, and the ratio of the intersection area S1 to the union area S2 is used as the intersection-union ratio.

[0065] Optionally, the sub-score of the first frame of image may be defined as 1.

[0066] (2) Gastric antrum region R in the t-th frame image t and the gastric antrum region R in the t-1 frame image t-1 The reciprocal of the Hausdorff distance.

[0067] Specifically, the gastric antrum region R in the t-1 frame image can be determined t-1 The image feature F t-1 , determine the gastric antrum area R in the t-th frame image t The image feature F t . Further, the two image features F can be calculated t-1 and F t The distance between the two images is inversely proportional to the image quality, i.e., the smaller the distance, the higher the image quality. Therefore, the reciprocal of the distance is used as the sub-score here.

[0068] Optionally, the sub-score of the first frame of image may be defined as 1.

[0069] (3) Gastric antrum region R in the t-th frame image t and the gastric antrum region R in the t-1 frame image t-1 The reciprocal of the area difference.

[0070] Specifically, the gastric antrum region R in the t-1 frame image can be determined t-1 The area S t-1 , determine the gastric antrum area R in the t-th frame image t The area S t . Further, the two areas S can be calculated t-1 and S t The absolute value of the difference between the two values ​​is used to obtain the area difference. The area difference is inversely proportional to image quality: the smaller the area difference, the higher the image quality. Therefore, the reciprocal of the area difference is used as the sub-score.

[0071] Optionally, the sub-score of the first frame of image may be defined as 1.

[0072] (4) Gastric antrum region R in the t-th frame image t The reciprocal of the length of the longest straight line segment on the boundary.

[0073] Specifically, it is determined whether there are multiple consecutive pixels on the boundary of the gastric antrum region located on the same straight line. If not, the length is 0. If so, the longest straight line segment is determined, assumed to be L. The unit can be millimeters (mm) or other dimensions. The length of the longest straight line segment is inversely proportional to image quality: the shorter the longest straight line segment, the higher the image quality. Therefore, the reciprocal of the longest straight line segment length is used as the sub-score here.

[0074] (5) Gastric antrum region R in the t-th frame image t The reciprocal of the number of acute angles on the boundary.

[0075] Specifically, for every three adjacent pixels on the border of the gastric antrum, we determine whether they form an acute angle. If so, the number of acute angles is incremented by 1 until all points on the border have been determined. The fewer the number of acute angles, the higher the image quality. Therefore, the reciprocal of the number of acute angles is used as the sub-score.

[0076] (6) Gastric antrum region R in the t-th frame image t The reciprocal of the inward concave area on the boundary.

[0077] Specifically, the gastric antrum region R in the t-th frame image can be determined t The area S t, and determine the gastric antrum area R in the t-th frame image t The area of ​​the circumscribed circle or circumscribed ellipse is S, and the difference between the two is SS t As the area of ​​the inward depression. Among them, the smaller the inward depression area, the higher the image quality. Therefore, the reciprocal of the inward depression area is used as the sub-score here.

[0078] (7) One or more scores for the t-th frame image given by one or more machine learning models.

[0079] Machine learning models can be traditional machine learning methods, deep learning methods, or methods based on ReverseClassification Accuracy. Deep learning detection methods include CNN models based on AlexNet, VGG, Inception, ResNet, DenseNet, or multilayer perceptrons composed of fully connected layers. Traditional machine learning methods include feature extraction and classification methods. Feature extraction methods can be CNN models or traditional methods such as PCA, HOG, and LDA. Classification methods can be methods such as KNN, SVM, and random forest.

[0080] In an embodiment where the ultrasound image determined in S120 includes multiple frames of images, optionally, after S120, the following steps may be performed: determining, for the multiple frames of images, a gastric antrum region in each frame of image and calculating the area of ​​the gastric antrum region in each frame of image; determining a measurement frame image from the multiple frames of image based on the gastric antrum region in each frame of image in the multiple frames of image; and further using the area of ​​the gastric antrum region in the measurement frame image as the area of ​​the gastric antrum region of the target object.

[0081] The method of determining the gastric antrum region and calculating the area will be described below in conjunction with S140. If an image frame has an error, the image quality of the image frame may be determined to be zero.

[0082] The errors mentioned here may include any of the following: the area of ​​the gastric antrum region is smaller than the set minimum threshold, or the result determined by the binary classification discriminator based on any of the machine learning models in (7) above is an error.

[0083] In this way, the measurement frame image may be determined based on the image quality in S130 , and then the area of ​​the gastric antrum region in the measurement frame image may be used as the area of ​​the gastric antrum region of the target object in S140 .

[0084] As another implementation method, S130 can determine the image quality of each frame of the multiple frames; obtain at least two frames of images whose image quality is greater than a quality threshold in the multiple frames of images; and determine the image with the largest gastric antrum area in the at least two frames of images as the measurement frame image.

[0085] It can be understood that the method of determining image quality in this implementation can refer to the description of the above implementation, for example, the image quality score is the weighted sum of multiple sub-scores.

[0086] In this implementation, a quality threshold may be preset, assuming it is expressed as Mth, and it may be determined which frames of the multiple frames have an image quality score greater than Mth.

[0087] Assuming that the number of multiple image frames is N1, the quality of N1 images corresponding to the N1 frames can be determined. Assuming that among the N1 frames, the image quality of N2 frames (N2 ≤ N1) is greater than the quality threshold, that is, the image quality score of the N2 frames is greater than Mth, then the measurement frame image can be selected from the N2 frames.

[0088] For example, if N2 = 1, then this image frame is the measurement frame image. For another example, if N2 > 1, then the area of ​​the gastric antrum region in each of the N2 image frames can be further determined. For example, the corresponding N2 gastric antrum region areas can be obtained, and then the image frame with the largest gastric antrum region area can be used as the measurement frame image.

[0089] As another implementation, the area of ​​the gastric antrum region in each frame of the multiple frames may be determined in S130; and the image with the largest gastric antrum region area among the multiple frames is determined as the measurement frame image.

[0090] Assuming that the number of the multiple frames of images is N1, then N1 areas corresponding to the N1 frames of images can be determined, and further, the frame of image with the largest area can be used as the measurement frame image.

[0091] In addition, it should be noted that although the method of determining the measurement frame image is described above in a variety of different implementation methods, the present application is not limited thereto. For example, a frame with suboptimal image quality can be used as the measurement frame image, for example, a frame with the second largest gastric antrum area can be used as the measurement frame image, and so on.

[0092] For example, the present application does not specifically limit the implementation method of determining the gastric antrum region in S140.

[0093] In one embodiment, the measurement frame image may be segmented to determine the gastric antrum region in the measurement frame image. The image segmentation method may be a segmentation method based on deep learning or traditional machine learning.

[0094] The deep learning detection method may be a CNN model based on FCN, U-Net, RCNN, YOLO, Inception, ResNet, DenseNet, etc. For example, the measurement frame image may be input into the CNN model to obtain an output, which includes the gastric antrum area.

[0095] Among them, traditional machine learning methods include feature extraction methods and segmentation methods. The feature extraction method can be a CNN model or traditional methods such as PCA, HOG, LDA, and the segmentation method can be FCM, SVM, ACM and other methods.

[0096] In another embodiment, the measurement frame image can be identified to determine multiple points on the boundary of the gastric antrum area; coordinate interpolation is performed based on the multiple points to determine the boundary of the gastric antrum area where the multiple points are located, and then determine the gastric antrum area.

[0097] Exemplarily, a deep learning model may be used to identify the measurement frame image, thereby regressing multiple points on the boundary.

[0098] The coordinate interpolation method may be, for example, linear interpolation, spline curve interpolation, polynomial interpolation, etc., which is not limited in this application.

[0099] In another embodiment, the measurement frame image may be identified to determine a plurality of points on the boundary of the gastric antrum region; and curve fitting may be performed on the plurality of points to determine the gastric antrum region.

[0100] Exemplarily, a deep learning model may be used to identify the measurement frame image, thereby regressing multiple points on the boundary.

[0101] The curve fitting method may be a quadratic curve fitting method. As an example, the gastric antrum region may be determined by an ellipse fitting method. That is, the gastric antrum region determined by the curve fitting is in an ellipse shape.

[0102] Furthermore, in S140 , the area of ​​the gastric antrum region may be determined.

[0103] In one embodiment, if the determined gastric antrum region is elliptical, the area of ​​the ellipse may be calculated and used as the area of ​​the gastric antrum region.

[0104] Assuming that the major axis of the ellipse is a and the minor axis is b, the area of ​​the ellipse is π×a×b / 4.

[0105] In another embodiment, a long axis and a short axis perpendicular to each other may be determined based on the gastric antrum region, and the area may be determined based on the long axis and the short axis perpendicular to each other.

[0106] For example, the longest line segment between any two points on the boundary of the gastric antrum region can be determined as the long axis; the short axis is determined based on the long axis, wherein the short axis and the long axis are perpendicular to each other and the two endpoints of the line segment where the short axis is located are located on the boundary of the gastric antrum region.

[0107] For example, the short axis may be determined as a line segment connecting two points on the boundary of the gastric antrum region and coinciding with the perpendicular bisector of the long axis. Alternatively, the short axis may be determined as the longest of the line segments connecting two points on the boundary of the gastric antrum region and perpendicular to the long axis.

[0108] As can be seen, the two ends of the major axis and the minor axis are on the boundaries of the antrum region. The major and minor axes are perpendicular to each other, thus dividing the antrum region into four subregions. Further, the area of ​​each of the four subregions can be calculated, and the sum of the areas of the four subregions is taken as the area of ​​the antrum region.

[0109] As an example, a quarter ellipse may be fitted based on the semi-major axis and the semi-minor axis associated with the sub-region, and the area of ​​the quarter ellipse may be used as the area of ​​the sub-region.

[0110] Assume the major axis is a and the minor axis is b. The minor axis divides the major axis into two segments, with lengths a1 and a2, where a1 + a2 = a. The major axis divides the minor axis into two segments, with lengths b1 and b2, where b1 + b2 = b. The areas of the four subregions are π × a1 × b1 / 4, π × a1 × b2 / 4, π × a2 × b1 / 4, and π × a2 × b2 / 4, respectively. The sum of these subregions yields the area of ​​the gastric antrum.

[0111] In another embodiment, the area of ​​the gastric antrum region can be determined based on the number of pixels contained in the gastric antrum region. Specifically, the total number of pixels in the gastric antrum region can be counted; and the area of ​​the gastric antrum region can be determined based on the total number and the actual area represented by each pixel.

[0112] The actual area represented by a single pixel depends on the properties of the ultrasound device, such as the size (width and height) and resolution of the ultrasound image obtained by the device. Assuming the total number of pixels obtained is N0 and the actual area represented by a single pixel is A0, the area of ​​the gastric antrum can be determined by multiplying the two to obtain N0 × A0.

[0113] It can be seen that through the above combination Figure 2 The method shown can determine the area of ​​the gastric antrum in the measurement frame image and use it as the area of ​​the gastric antrum of the target object. Furthermore, this process does not rely on user experience and can be automatically performed by a processor. This is not only time-efficient and efficient, but also produces more accurate results, providing a reliable basis for further analysis of the target object's gastric antrum.

[0114] Illustratively, after S140 , the method may further include: displaying the measurement frame image, and marking the gastric antrum region in the measurement frame image.

[0115] Optionally, the measurement frame image may be displayed on a display, and the gastric antrum region and the area of ​​the gastric antrum region may be marked in the measurement frame image.

[0116] As an example, the boundary of the gastric antrum region can be marked with a line. For example, the outline of the gastric antrum region in the measurement frame image can be outlined with a line. The line can be a solid line, a dashed line, or another line type. This application does not limit the width, color, or transparency of the line. For example, the user can customize it by selecting a checkbox.

[0117] As another example, the gastric antrum region can be covered with a mask. Alternatively, the gastric antrum region can be approximated as an ellipse, and the ellipse can be covered with a mask. For example, the gastric antrum region in the measurement frame image can be covered with a color block. The present application does not limit the color, transparency, feathering, etc. of the color block, and users can customize it, for example.

[0118] As another example, the gastric antrum region can be marked with an ellipse, with the major and minor axes of the ellipse marked. For example, the gastric antrum region can be approximated as an ellipse, and the outline of the ellipse can be marked with a first line. The first line can be a solid line or a dashed line, and the width, color, and transparency of the first line can be customized by the user. For example, the major and minor axes of the ellipse can be marked with a second line. The second line can be a solid line or a dashed line, and the width, color, and transparency of the second line can be customized by the user. It is understood that the first and second lines can have the same or different forms, and this application does not limit this.

[0119] When marking the antrum region, the area of ​​the antrum region can also be marked. For example, the area can be marked inside or outside the antrum region, such as A1cm 2 For example, the area (such as A1cm 2 ) and point to the gastric antrum area by an arrow or other means. Exemplarily, the area can be displayed on the measurement frame image, or can also be displayed outside the imaging area, which is not limited in this application.

[0120] It can be understood that, although the above embodiment describes the process of determining the area of ​​the gastric antrum region in the measurement frame image, in fact, the area of ​​the gastric antrum region in any frame image can be determined through a similar process for any frame image.

[0121] For example, the area of ​​the gastric antrum region obtained through continuous measurement may be acquired, and a curve showing a change in the area of ​​the gastric antrum region obtained through continuous measurement over time may be displayed.

[0122] For the time interval from time t1 to time t2, the area of ​​the gastric antrum region in each frame of the image can be determined, and further: a curve showing the change of the area of ​​the gastric antrum region over time in the time interval from time t1 to time t2 can be displayed, such as Figure 3 shown.

[0123] The change curve may correspond to the ultrasound image obtained in real time. For example, the ultrasound image includes N1 frame images, then the area of ​​the gastric antrum region may be determined for each frame image, and the area of ​​the gastric antrum region may be determined by Figure 3 The broken line shown is displayed. Figure 3 The area-time curve of the gastric antrum region is represented by a broken line. Optionally, the curve can be updated in real time during the ultrasound measurement, for example Figure 3 A vertical line can be displayed at time t2 to indicate the current moment, and when the next frame of image is acquired, the area of ​​the gastric antrum region in the next frame of image can be determined and updated in the curve in real time. Accordingly, the vertical line corresponding to the current moment will move backward.

[0124] Optionally, the area-time curve of the gastric antrum region in the broken line form can be further fitted into a smooth curve, such as Figure 4 Shown is a smoothed area-time curve after fitting between time t01 and time t02.

[0125] Optionally, the area-time curves of multiple different time periods can be obtained and displayed on the same coordinate axis, wherein different time periods can be displayed discontinuously. Specifically, the change curve can include multiple curve segments corresponding to multiple time periods. Figure 5 As shown, it includes three time periods, namely t11 to t12, t21 to t22, and t31 to t32. Among them, multiple time periods are not continuous, and are represented by discontinuous time axes between different time periods. Figure 5 In the figure, t12 and t21 are discontinuous, t22 and t31 are discontinuous, and two parallel oblique lines represent the time axis interruption.

[0126] exist Figure 5 In the figure, t32 may be the current time, and the curve may be updated in real time as ultrasound measurements are taken.

[0127] In addition, Figure 3 or Figure 5 In the example, the singular point on the curve can be deleted and re-determined using other adjacent points through interpolation or other methods. Alternatively, if the image quality of a frame image is lower than the quality threshold, the area of ​​the gastric antrum region of the frame image can be deleted from the curve and re-determined using other adjacent points through interpolation or other methods.

[0128] Alternatively, the area-time curve can be low-pass filtered. Low-pass filtering can be achieved by performing a Fourier transform on the area-time curve of the gastric antrum region to obtain components at various frequencies, multiplying each frequency component by a coefficient, and then performing an inverse Fourier transform on the result. Low-pass filtering indicates that the coefficients of low-frequency components are larger, while the coefficients of high-frequency components are smaller. Alternatively, the coefficients are inverse Fourier transformed and convolved with the gastric antrum area-time curve.

[0129] It should be noted that Figure 5 In the example, the curve segments of each time period can also be fitted into smooth curves before displaying them. Figures 3 to 5 This is only for illustration, for example, a specific area value may be shown on the vertical axis, but this application does not limit this.

[0130] Illustratively, after S140 , the method may further include: calculating a gastric antrum dynamic parameter according to the area of ​​the gastric antrum region.

[0131] As an implementation, the antral dynamic parameter is gastric emptying time. Calculating the antral dynamic parameter may include: obtaining a first area of ​​the gastric antrum region in an empty stomach state; obtaining a second area of ​​the gastric antrum region at a first moment in a filled stomach state; obtaining a third area of ​​the gastric antrum region after the stomach is filled through intermittent measurement, and recording the second moment of obtaining the third area when the difference between the third area and the first area is less than an area threshold; and determining the gastric emptying time based on the first and second moments.

[0132] When the target object is in an empty stomach, Figure 2 The method shown in FIG4 is used to determine the area of ​​the gastric antrum region, assuming that the empty stomach area is S0. After the stomach of the target object is filled, it is marked as the first time t41, which can be determined by Figure 2 The method shown in FIG. 1 determines the area of ​​the gastric antrum region, assuming it is S1. Subsequently, the area of ​​the gastric antrum region can be determined at intervals of time (e.g., half an hour or 5 minutes or other values) by Figure 2 The method shown determines the area of ​​the gastric antrum region and determines a second time t42, and further determines the gastric emptying time as t42-t41.

[0133] In one example, if the absolute value of the difference between the third area of ​​the gastric antrum region determined at the second moment and the first area is smaller than the area threshold, the second moment may be recorded.

[0134] In another example, if the area of ​​the gastric antrum region determined at the third moment is larger than the first area, and the area of ​​the gastric antrum region determined at a fourth moment after the third moment is smaller than the first area, then a second moment at which the area of ​​the gastric antrum region between the third moment and the fourth moment is approximately equal to the first area can be further determined.

[0135] The second time t42 may be determined by interpolating data based on the area of ​​the gastric antrum region measured intermittently.

[0136] Combine Figure 6 Assume that the area of ​​the gastric antrum determined at the first moment t41 is S1. Then, measurements are performed at t43, t44, t45, and t46 to determine the area of ​​the gastric antrum at each moment. Since the area determined at time t45 is greater than S0, and the area determined at time t46 is less than S0, the area between t45 and t46 can be determined by interpolation to be approximately S0 ( Figure 6 The second time t42 of the fasting area S0 is shown by a black dotted line.

[0137] It should be noted that this application does not limit the specific method of interpolation. For example, Figure 6 The method can be linear interpolation, polynomial interpolation, spline interpolation, or other methods. The polynomial interpolation method can be: assuming that the area-time curve satisfies a certain polynomial, then the coordinates of each known point can be substituted into the polynomial to obtain the parameters of the polynomial. The spline interpolation method can be: assuming that the coordinates of each point on the area-time curve are the weighted sum of the coordinates of the known points, with closer known points having greater weights.

[0138] Optionally, the areas of multiple gastric antral regions determined during the measurement of gastric emptying time can be displayed. For example, a curve showing the change in the area of ​​the gastric antral region over time within the time interval from a first moment (t41) to a second moment (t42) can be displayed.

[0139] As an example, the variation curve may be a broken line marked by multiple points of intermittent measurement, such as Figure 6 As shown in FIG, by connecting two adjacent points with a straight line, an area-time broken line is formed.

[0140] As another example, the variation curve may be a smooth curve obtained by interpolating multiple points of intermittent measurement, such as Figure 7 shown.

[0141] It is understandable that the second moment determined by using different interpolation methods may be different, such as Figure 6 The second time t42 determined in Figure 7 The second time t42 determined in Figure 6 The gastric emptying time obtained in Figure 7 However, the difference in the second moment time obtained by using different interpolation methods is very small compared to the gastric emptying time, and can even be ignored. Therefore, different interpolation methods have little effect on the gastric emptying time.

[0142] Optionally, the gastric emptying time and / or the area of ​​the gastric antrum when fasting can also be displayed while displaying the change curve. Figure 7 , also shows "gastric emptying time: t42-t41=20min" and "fasting area: S0=2cm 2 ”.

[0143] As another implementation, the antral dynamic parameter includes an average antral contraction amplitude (ΔA / A) and / or an average antral contraction frequency (F). Calculating the antral dynamic parameter may include determining the average antral contraction amplitude and the average antral contraction frequency based on a change in the area of ​​the antral region over time within a preset time period.

[0144] As another implementation, the antral dynamic parameter includes an antral motility index (MI). Calculating the antral dynamic parameter may include: determining an average antral contraction amplitude and an average antral contraction frequency based on a change in the area of ​​the antral region over time within a preset time period; and calculating the antral motility index based on the average antral contraction amplitude and the average antral contraction frequency.

[0145] Specifically, a curve showing the change in area of ​​the antral region over time obtained through continuous measurement within a preset time period can be obtained; multiple antral movement cycles can be determined based on the change curve; and the average antral contraction amplitude and average antral contraction frequency can be determined based on the contraction amplitude and contraction frequency of each antral movement cycle in the multiple antral movement cycles.

[0146] For example, the curve of area change over time can be as follows: Figure 3 The initial change curve shown, or it can be as Figure 4 Alternatively, the area variation curve over time may also be a smooth curve obtained by smoothing the initial variation curve. Figure 3 or as Figure 4 The curve shown is the curve obtained after low-pass filtering.

[0147] For example, the preset time period may be Figure 3 From time t1 to time t2 as shown in FIG. , or it may be as shown in FIG. Figure 4 In general, the preset time period should include multiple gastric antral motility cycles. One gastric antral cycle can be the time between one peak (or trough) of the gastric antral area and the next peak (or trough).

[0148] Reference Figure 8 The duration between two adjacent peaks is one antral cycle, such as T1; the duration between two adjacent troughs is one antral cycle, such as T2. It can be understood that the frame corresponding to the area peak is the antral diastolic frame, and the frame corresponding to the area trough is the antral contraction frame.

[0149] The amplitude of antral contraction can be determined based on a single antral contraction, referring to Figure 8 As shown in the dashed rectangle in the figure, the area from the peak to the next trough can be considered a single antral contraction. The peak represents the maximum area, while the trough represents the minimum area. The antral contraction amplitude for this single antral contraction can be calculated as (maximum area - minimum area) / maximum area.

[0150] Then, the average gastric antrum contraction amplitude can be obtained by calculating multiple gastric antrum contraction amplitudes corresponding to multiple single gastric antrum contractions and taking the average value.

[0151] Still refer to Figure 8 , the contraction duration of a single antral contraction can be determined as t52 - t51. Similarly, multiple contraction durations corresponding to multiple single antral contractions can be determined. Furthermore, the average of the multiple contraction durations can be calculated to obtain the average contraction duration, and the reciprocal of the average contraction duration can be used as the average antral contraction frequency.

[0152] In addition, the product of the above-obtained average antral contraction amplitude and the average antral contraction frequency can also be determined as the antral motility index.

[0153] Therefore, this application is based on Figure 2 The obtained area of ​​the gastric antrum region is further used to determine the gastric antrum dynamic parameters, which can ensure that the obtained gastric antrum dynamic parameters are more accurate and can better reflect the actual condition of the gastric antrum.

[0154] Illustratively, after S140, the method may further include determining the gastric volume based on the area of ​​the gastric antrum region. Specifically, the gastric volume may be determined based on the area of ​​the gastric antrum region in multiple different cross-sections of the stomach. It is understood that the method for determining the gastric volume based on the area of ​​the gastric antrum region can refer to existing technologies and will not be further described here.

[0155] Furthermore, a gastric volume time curve showing the change of gastric volume over time may be determined and displayed based on the gastric volumes determined at multiple different moments.

[0156] The gastric volumes corresponding to the multiple different moments can be determined based on the areas of the gastric antrum regions determined at the multiple different moments. The multiple different moments can be discrete moments, i.e., multiple moments measured at intervals, for example, with an interval (e.g., 5 minutes or other value) between two adjacent moments. Optionally, the gastric volume-time curve can be a broken line; or alternatively, the gastric volume-time curve can be smoothed by curve fitting or interpolation, thereby forming a smooth curve.

[0157] Exemplarily, the method may further include predicting or calculating the gastric emptying time based on the gastric volume-time curve. For example, a first gastric volume when the stomach is empty may be determined, a second gastric volume when the stomach is full may be determined at a first moment, and a second moment when the gastric volume decreases to the first gastric volume over time may be detected. The gastric emptying time may then be calculated as the duration between the first moment and the second moment.

[0158] Thus, through the above combination Figures 2 to 8 In this embodiment, the processor can automatically determine the area of ​​the gastric antrum. This process requires no manual intervention and is independent of user experience, saving user effort and improving processing efficiency. Furthermore, a more accurate gastric antrum area can be obtained. Furthermore, the area of ​​the gastric antrum can be dynamically displayed, allowing the user to intuitively see changes in the gastric antrum area over time through an area-time curve.

[0159] It should be noted that although the above combined Figures 2 to 8 A method for determining the area of ​​the gastric antrum region is described, but the present application is not limited thereto, for example Figure 9 As shown, the area can be directly obtained based on the deep learning model. The method includes:

[0160] S110, emitting an ultrasonic beam toward a target object including a gastric antrum region by stimulating an ultrasonic probe, and receiving an ultrasonic echo returned from the target object by the ultrasonic probe to obtain an ultrasonic echo signal;

[0161] S120, the processor processes the ultrasonic echo signal to obtain an ultrasonic image of the target object;

[0162] S130, the processor determines a measurement frame image based on the ultrasound image;

[0163] S240: The processor inputs the measurement frame image into a pre-trained deep learning model to obtain an output of the deep learning model, which includes the area of ​​the gastric antrum region in the measurement frame image.

[0164] The deep learning model can be obtained by training based on a training data set. The training data set may include multiple samples, each of which is a labeled ultrasound image. Specifically, the ultrasound image includes a gastric antrum region, and the area of ​​the gastric antrum region is marked.

[0165] The input of this deep learning model is an ultrasound image, and the output is the area of ​​the gastric antrum. This deep learning model may include convolutional layers, deconvolutional layers, pooling layers, fully connected layers, etc., which are not limited in this application.

[0166] It is understandable that in Figure 9In the method shown, S130 may include: using a deep learning model to determine the area of ​​the gastric antrum region in each frame image, and further determining the measurement frame image. The process of determining the measurement frame image can refer to the above-mentioned combination Figure 2 The described part will not be repeated here.

[0167] In addition, it is understood that although the above embodiment provides a method for determining the area of ​​the gastric antrum, the present application is not limited thereto, and the method in the present application can be applied to various organs or tissues, including but not limited to: gastric antrum, heart cavity, blood vessels, kidneys, bladder, follicles, tumors, etc. Figure 10 is a flow chart of another method for determining area in this application, Figure 10 The methods shown include:

[0168] S10, emitting an ultrasonic beam toward a target object by stimulating the ultrasonic probe, and receiving an ultrasonic echo returned from the target object by the ultrasonic probe to obtain an ultrasonic echo signal;

[0169] S20, the processor processes the ultrasonic echo signal to obtain an ultrasonic image of the target object;

[0170] S30, the processor determines a measurement frame image based on the ultrasound image;

[0171] S40, the processor determines a target region of a target to be measured of the target object in the measurement frame image, and further determines an area of ​​the target region, wherein the target to be measured includes a target organ or tissue, and the target region includes a section or a region of interest in the section.

[0172] Illustratively, the target to be measured may be a target organ or tissue, including but not limited to: a gastric antrum, a cardiac cavity, a blood vessel, a kidney, a bladder, an ovarian follicle, a tumor, etc. Illustratively, the target region of the target to be measured may be a section or a portion of a section (e.g., a region of interest) of the target organ or tissue, and the section may be a cross section or a profile.

[0173] For example, the area of ​​the target region to be measured may be: a gastric antrum cross-sectional area, a cardiac cavity cross-sectional area, a blood vessel cross-sectional area, an ovarian follicle cross-sectional area, a tumor cross-sectional area, and the like.

[0174] In one embodiment, the ultrasound image determined in S20 is a frame image. Accordingly, in S30, the frame image can be used as the measurement frame image.

[0175] In another embodiment, the ultrasound image determined in S20 includes multiple frames of images, for example, the ultrasound image is video data. Accordingly, in S30, one frame from the multiple frames of images may be selected as the measurement frame image.

[0176] As an implementation manner, in S30 , the image quality of each frame of the multiple frames of images may be determined; and the image with the best image quality among the multiple frames of images may be determined as the measurement frame image.

[0177] Exemplarily, image quality can be expressed as a quality score, and the quality score includes a weighted sum of multiple sub-scores. The sub-scores of an image can include at least one of the following: the intersection-over-union ratio of a target area in an image to a target area in a previous image; the reciprocal of the Hausdorff distance between the target area in an image and the target area in a previous image; the reciprocal of the area difference between the target area in an image and the target area in a previous image; the reciprocal of the length of the longest straight line segment on the boundary of the target area in an image; the reciprocal of the number of acute angles on the boundary of the target area in an image; the reciprocal of the area of ​​inward depression on the boundary of the target area in an image; or one or more scores of an image given by one or more machine learning models.

[0178] It can be understood that each sub-score is directly proportional to the image quality.

[0179] Taking the t-th frame image in the multi-frame image as an example, the sub-score of the t-th frame image may include:

[0180] (1) Target region R in the t-th frame image t and the target area R in the t-1 frame image t-1 The intersection and union ratio.

[0181] Specifically, the target area R in the t-1 frame image can be determined t-1 The area S t-1 , determine the target area R in the t-th frame image t The area S t . Further, the two regions R can be calculated t-1 and R t The intersection area S1 and the union area S2 are obtained, and the ratio of the intersection area S1 to the union area S2 is used as the intersection-union ratio.

[0182] Optionally, the sub-score of the first frame of image may be defined as 1.

[0183] (2) Target region R in the t-th frame image t and the target area R in the t-1 frame image t-1 The reciprocal of the Hausdorff distance.

[0184] Specifically, the target area R in the t-1 frame image can be determined t-1 The image feature F t-1 , determine the target area R in the t-th frame image t The image feature F t . Further, the two image features F can be calculatedt-1 and F t The distance between the two images is inversely proportional to the image quality, i.e., the smaller the distance, the higher the image quality. Therefore, the reciprocal of the distance is used as the sub-score here.

[0185] Optionally, the sub-score of the first frame of image may be defined as 1.

[0186] (3) Target region R in the t-th frame image t and the target area R in the t-1 frame image t-1 The reciprocal of the area difference.

[0187] Specifically, the target area R in the t-1 frame image can be determined t-1 The area S t-1 , determine the target area R in the t-th frame image t The area S t . Further, the two areas S can be calculated t-1 and S t The absolute value of the difference between the two values ​​is used to obtain the area difference. The area difference is inversely proportional to image quality: the smaller the area difference, the higher the image quality. Therefore, the reciprocal of the area difference is used as the sub-score.

[0188] Optionally, the sub-score of the first frame of image may be defined as 1.

[0189] (4) Target area R in the t-th frame image t The reciprocal of the length of the longest straight line segment on the boundary.

[0190] Specifically, it can be determined whether there are multiple consecutive pixels on the boundary of the target area located on the same straight line. If not, the length is 0; if so, the longest straight line segment is determined, assumed to be L. The unit can be millimeters (mm) or other dimensions. The length of the longest straight line segment is inversely proportional to image quality: the shorter the longest straight line segment, the higher the image quality. Therefore, the reciprocal of the longest straight line segment length is used as the sub-score here.

[0191] (5) Target area R in the t-th frame image t The reciprocal of the number of acute angles on the boundary.

[0192] Specifically, for every three adjacent pixels on the boundary of the target area, we determine whether they form an acute angle. If so, the number of acute angles is increased by 1 until all points on the boundary have been determined. The fewer the number of acute angles, the higher the image quality. Therefore, the reciprocal of the number of acute angles is used as the sub-score.

[0193] (6) Target area R in the t-th frame image tThe reciprocal of the inward concave area on the boundary.

[0194] Specifically, the target area R in the t-th frame image can be determined t The area S t , and determine the target area R in the t-th frame image t The area of ​​the circumscribed circle or circumscribed ellipse is S, and the difference between the two is SS t As the area of ​​the inward depression. Among them, the smaller the inward depression area, the higher the image quality. Therefore, the reciprocal of the inward depression area is used as the sub-score here.

[0195] (7) One or more scores for the t-th frame image given by one or more machine learning models.

[0196] Machine learning models can be traditional machine learning methods, deep learning methods, or methods based on ReverseClassification Accuracy. Deep learning detection methods include CNN models based on AlexNet, VGG, Inception, ResNet, DenseNet, or multilayer perceptrons composed of fully connected layers. Traditional machine learning methods include feature extraction and classification methods. Feature extraction methods can be CNN models or traditional methods such as PCA, HOG, and LDA. Classification methods can be methods such as KNN, SVM, and random forest.

[0197] In an embodiment where the ultrasound image determined in S20 includes multiple frame images, optionally, S30 may include: for the multiple frame images, determining the target area in each frame image and calculating the area of ​​the target area in each frame image; further calculating the image quality of each frame image; thereby determining the measurement frame image based on the image quality.

[0198] As another implementation method, in S30, the image quality of each frame of the multiple frames can be determined; at least two frames of images whose image quality is greater than a quality threshold in the multiple frames of images are obtained; and the image with the largest target area in the at least two frames of images is determined as the measurement frame image.

[0199] As another implementation, the target region area in each frame of the multiple frames may be determined in S30; and the image with the largest target region area among the multiple frames is determined as the measurement frame image.

[0200] Understandably, Figure 10 The S30 in the above Figure 2 The S130 process in is similar, and Figure 10 The S40 in the above Figure 2 The S140 process in is similar and therefore will not be described in detail here.

[0201] Illustratively, after S40 , the method may further include: displaying the measurement frame image, and marking the target region of the target to be measured and the area of ​​the target region in the measurement frame image.

[0202] As an example, the boundary of the target area can be marked with a line. For example, the outline of the target area in the measurement frame image can be outlined with a line, wherein the line can be a solid line, a dashed line, or other line type, and this application does not limit the width, color, transparency, etc. of the line, for example, the user can customize it through a selection box.

[0203] As another example, the target area can be covered in a mask form. For example, the target area in the measurement frame image can be covered with a color block, and the present application does not limit the color, transparency, feathering degree, etc. of the color block, for example, the user can customize it.

[0204] When marking the target area, the area of ​​the target area can also be marked. For example, the area can be marked inside or outside the target area, such as A11cm 2 For example, the area (such as A11cm 2 ) and point to the target area by an arrow or other means. Exemplarily, the value of the area can be displayed on the measurement frame image, or can also be displayed outside the imaging area, which is not limited in this application.

[0205] Illustratively, after S40 , the method may further include: displaying a curve showing changes in the area of ​​the target area over time based on the areas of the target area at multiple different moments.

[0206] As an example, for the time interval from time t13 to time t14, the area of ​​the target region in each frame of the image can be determined, and further, a curve showing the change of the area of ​​the target region over time can be displayed. Optionally, the curve is a broken line, and the curve can be updated in real time during the ultrasound measurement process. Figure 3 As shown, the current moment is represented by displaying a vertical line. Alternatively, the curve is a smooth curve, which can be obtained, for example, by filtering and fitting the broken line.

[0207] As another example, the area-time curves of multiple different time periods can be obtained and displayed on the same coordinate axis, wherein different time periods can be displayed discontinuously. Specifically, the change curve can include multiple curve segments corresponding to multiple time periods. Figure 5 As shown, two parallel slashes represent timeline interruptions.

[0208] For example, in one embodiment, the target to be measured includes a cardiac cavity. Accordingly, S40 may determine the cross-sectional area of ​​the cardiac cavity. Following S40, the process may also include determining the volume of the cardiac cavity based on the area of ​​the cardiac cavity. Optionally, a curve of the change in the volume of the cardiac cavity over time may be displayed, allowing the user to more intuitively view the data.

[0209] For example, in one embodiment, the target to be measured includes an ovarian follicle. Accordingly, S40 may determine the cross-sectional area of ​​the ovarian follicle. Following S40, the process may further include determining the volume of the ovarian follicle based on the area of ​​the follicle. Optionally, a curve of the follicle volume change over time may be displayed, providing a more intuitive user experience.

[0210] It can be seen that in this application, the processor can automatically determine the cross-sectional area of ​​each target to be measured, which simplifies user operations and does not rely on user experience. Therefore, the method of determining the area in this application is fast, efficient and accurate, and can provide a more accurate reference for subsequent further diagnosis of the target object.

[0211] Now return to Figure 1 The ultrasound device 10 is shown.

[0212] In one implementation, the transmit / receive selector switch 120 can activate the ultrasound probe 110 via the transmit circuit 160 to transmit an ultrasonic beam toward a target object including the gastric antrum region, and control the ultrasound probe 110 via the receive circuit 170 to receive an ultrasonic echo of the ultrasonic beam returned from the target object, thereby acquiring an ultrasonic echo signal. The processor 140 can process the ultrasonic echo signal to obtain an ultrasonic image of the target object; determine a measurement frame image based on the ultrasonic image; determine the gastric antrum region of the target object in the measurement frame image; and further determine the area of ​​the gastric antrum region.

[0213] The display 150 may display the measurement frame image and mark the gastric antrum region and the area of ​​the gastric antrum region. For example, the display 150 may also display the above-mentioned area-time curve.

[0214] In one implementation, the transmit / receive selector switch 120 can activate the ultrasound probe 110 via the transmit circuit 160 to transmit an ultrasonic beam toward a target object including the gastric antrum region, and control the ultrasound probe to receive ultrasound echoes returned from the target object to obtain an ultrasound echo signal. The processor 140 can process the ultrasound echo signal to obtain an ultrasound image of the target object; determine a measurement frame image based on the ultrasound image; and input the measurement frame image into a pre-trained deep learning model to obtain an output of the deep learning model, which includes the area of ​​the gastric antrum region in the measurement frame image.

[0215] In one implementation, the transmit / receive selector switch 120 can activate the ultrasound probe 110 via the transmit circuit 160 to transmit an ultrasonic beam toward the target object, and control the ultrasound probe to receive ultrasonic echoes returned from the target object to obtain an ultrasonic echo signal. The processor 140 can process the ultrasonic echo signal to obtain an ultrasonic image of the target object; determine a measurement frame image based on the ultrasonic image; determine a target region of the target object to be measured in the measurement frame image, and further determine the area of ​​the target region, wherein the target to be measured includes a target organ or tissue, and the target region includes a section or a region of interest within the section.

[0216] In addition, an embodiment of the present invention further provides a computer storage medium having a computer program stored thereon. When the computer program is executed by a computer or processor, the steps of any of the aforementioned methods for determining area based on ultrasound images can be implemented. For example, the computer storage medium is a computer-readable storage medium.

[0217] In one embodiment, the computer program instructions, when executed by a computer or processor, cause the computer or processor to perform the following steps: activating an ultrasound probe to transmit an ultrasound beam toward a target object including the gastric antrum region, and receiving, via the ultrasound probe, ultrasound echoes returned from the target object to obtain ultrasound echo signals; processing the ultrasound echo signals to obtain an ultrasound image of the target object; determining a measurement frame image based on the ultrasound image; determining the gastric antrum region of the target object in the measurement frame image, and further determining the area of ​​the gastric antrum region.

[0218] In one embodiment, the computer program instructions, when executed by a computer or processor, cause the computer or processor to perform the following steps: stimulating an ultrasound probe to transmit an ultrasound beam toward a target object including the gastric antrum region, and receiving ultrasound echoes returned from the target object via the ultrasound probe to obtain ultrasound echo signals; processing the ultrasound echo signals to obtain an ultrasound image of the target object; determining a measurement frame image based on the ultrasound image; and inputting the measurement frame image into a pre-trained deep learning model to obtain an output of the deep learning model, wherein the output includes the area of ​​the gastric antrum region in the measurement frame image.

[0219] In one embodiment, the computer program instructions, when executed by a computer or processor, cause the computer or processor to perform the following steps: stimulating an ultrasonic probe to transmit an ultrasonic beam toward a target object, and receiving ultrasonic echoes returned from the target object via the ultrasonic probe to obtain ultrasonic echo signals. Processing the ultrasonic echo signals to obtain an ultrasonic image of the target object; determining a measurement frame image based on the ultrasonic image; determining a target region of the target object to be measured in the measurement frame image, and further determining the area of ​​the target region, wherein the target to be measured includes a target organ or tissue, and the target region includes a section or a region of interest within the section.

[0220] Computer storage media may include, for example, a memory card of a smartphone, a storage component of a tablet computer, a hard disk of a personal computer, a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a portable compact disk read-only memory (CD-ROM), a USB memory, or any combination thereof. A computer-readable storage medium may be any combination of one or more computer-readable storage media.

[0221] In addition, an embodiment of the present invention further provides a computer program product, which includes instructions. When the instructions are executed by a computer, the computer executes the steps of any of the above methods for determining an area based on an ultrasound image.

[0222] It can be seen that in the embodiment of the present invention, the processor processes and analyzes the ultrasound image to obtain the area of ​​the gastric antrum region of the target object, eliminating the need for the user to manually select based on experience. On the one hand, this method saves time and effort and is more efficient; on the other hand, it does not rely on user behavior, and the determined area is more accurate.

[0223] Although example embodiments have been described herein with reference to the accompanying drawings, it should be understood that the above example embodiments are merely illustrative and are not intended to limit the scope of the present invention. Various changes and modifications may be made therein by those skilled in the art without departing from the scope and spirit of the present invention. All such changes and modifications are intended to be included within the scope of the present invention as claimed in the appended claims.

[0224] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0225] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units described is merely a logical function division. In actual implementation, other division methods may be used, such as combining or integrating multiple units or components into another device, or ignoring or not performing some features.

[0226] In the description provided herein, numerous specific details are described. However, it is understood that embodiments of the present invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques are not shown in detail so as not to obscure the understanding of this description.

[0227] Similarly, it should be understood that in order to streamline the present invention and aid in understanding one or more of the various inventive aspects, in the description of exemplary embodiments of the present invention, the various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this approach to the present invention should not be interpreted as reflecting the intention that the claimed invention requires more features than those explicitly recited in each claim. More precisely, as reflected in the corresponding claims, the inventive point is that the corresponding technical problem can be solved with fewer features than all the features of a single disclosed embodiment. Therefore, the claims following the detailed description are hereby expressly incorporated into the detailed description, with each claim itself serving as a separate embodiment of the present invention.

[0228] It will be understood by those skilled in the art that, except where mutually exclusive, all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or units of any method or apparatus disclosed herein may be combined in any combination. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature providing the same, equivalent, or similar purpose.

[0229] Furthermore, those skilled in the art will appreciate that although some embodiments described herein include certain features included in other embodiments but not other features, combinations of features from different embodiments are intended to be within the scope of the present invention and to form different embodiments. For example, in the claims, any of the claimed embodiments may be used in any combination.

[0230] The various component embodiments of the present invention can be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art will appreciate that in practice, a microprocessor or a digital signal processor (DSP) can be used to implement some or all of the functions of some modules in the ultrasound device according to the embodiments of the present invention. The present invention can also be implemented as a device program (e.g., a computer program and a computer program product) for performing part or all of the methods described herein. Such a program for implementing the present invention can be stored on a computer-readable medium, or can be in the form of one or more signals. Such a signal can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0231] It should be noted that the above embodiments illustrate rather than limit the invention, and that those skilled in the art may devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between brackets should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention may be implemented by means of hardware comprising several different elements and by means of appropriately programmed computers. In a unit claim enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third etc. does not indicate any order. These words may be interpreted as names.

[0232] The foregoing description is merely a specific embodiment of the present invention or an illustration of a specific embodiment. The scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be readily conceived by a person skilled in the art within the technical scope disclosed in the present invention are intended to be encompassed by the scope of protection of the present invention. The scope of protection of the present invention shall be subject to the scope of protection of the claims.

Claims

1. A method for determining an area based on an ultrasound image, characterized in that: include: transmitting an ultrasonic beam to a target object including a gastric antrum region by stimulating an ultrasonic probe, and receiving an ultrasonic echo returned from the target object by the ultrasonic probe to obtain an ultrasonic echo signal; The processor processes the ultrasonic echo signal to obtain an ultrasonic image of the target object; The processor determines a measurement frame image based on the ultrasound image; The processor determines a gastric antrum region of the target object in the measurement frame image, and further determines an area of ​​the gastric antrum region; The ultrasound image includes multiple frame images, and determining the measurement frame image includes: Determining the image quality of each frame of the plurality of frames of images; Based on the image quality, selecting a measurement frame image from the multiple frames of images; The image quality is represented as a quality score, the quality score includes a weighted sum of multiple sub-scores, and the sub-scores of the image include at least two of the following: The intersection-over-union ratio of the gastric antrum region in the image to the gastric antrum region in the previous frame of image; The reciprocal of the Hausdorff distance between the gastric antrum region in the image and the gastric antrum region in the previous frame of image; The reciprocal of the difference in area between the gastric antrum region in the image and the gastric antrum region in the previous frame of image; The reciprocal of the length of the longest straight line segment on the boundary of the gastric antrum region in the image; The inverse of the number of sharp corners on the border of the gastric antrum region in the image; The reciprocal of the inward concave area on the border of the gastric antrum region in the image; One or more scores for the image given by one or more machine learning models.

2. The method according to claim 1, characterized in that Determining the gastric antrum region of the target object in the measurement frame image includes: The measurement frame image is segmented to determine the gastric antrum area.

3. The method according to claim 1, characterized in that Determining the gastric antrum region of the target object in the measurement frame image includes: Identifying the measurement frame image to determine a plurality of points on the boundary of the gastric antrum region; Coordinate interpolation is performed based on the multiple points to determine the boundary of the gastric antrum region where the multiple points are located, and then determine the gastric antrum region.

4. The method according to claim 1, characterized in that Determining the gastric antrum region of the target object in the measurement frame image includes: Identifying the measurement frame image to determine a plurality of points on the boundary of the gastric antrum region; The multiple points are subjected to curve fitting to determine the gastric antrum region.

5. The method according to claim 4, characterized in that The shape of the gastric antrum region after curve fitting is an ellipse, and determining the area of ​​the gastric antrum region includes: The area of ​​the ellipse is calculated as the area of ​​the gastric antrum region.

6. The method according to any one of claims 2 to 4, characterized in that Determine the area of ​​the gastric antrum region, including: Determine the longest line segment between any two points on the boundary of the gastric antrum region as the long axis; Determine a short axis based on the long axis, wherein the short axis is perpendicular to the long axis and two endpoints of a line segment where the short axis is located are located on a boundary of the gastric antrum region; The area of ​​the gastric antrum region is determined according to the long axis and the short axis which are perpendicular to each other.

7. The method according to claim 6, characterized in that Determining the minor axis based on the major axis includes: The short axis is determined as a line segment connecting two points on the boundary of the gastric antrum region and coinciding with the perpendicular bisector of the long axis.

8. The method according to claim 6, characterized in that Determining the minor axis based on the major axis includes: The longest of the line segments connecting two points on the boundary of the gastric antrum region and perpendicular to the long axis is determined as the short axis.

9. The method according to claim 6, characterized in that The major axis and the minor axis divide the gastric antrum region into four sub-regions. Determining the area of ​​the gastric antrum region according to the long axis and the short axis that are perpendicular to each other includes: The area of ​​each of the four sub-regions is calculated, and the sum of the areas of the four sub-regions is taken as the area of ​​the gastric antrum region.

10. The method according to claim 9, characterized in that A quarter ellipse is fitted according to the semi-major axis and the semi-minor axis associated with the sub-region, and the area of ​​the quarter ellipse is used as the area of ​​the sub-region.

11. The method according to any one of claims 2 to 4, characterized in that Determine the area of ​​the gastric antrum region, including: Counting the total number of pixels in the gastric antrum area; The area of ​​the gastric antrum region is determined according to the total number and the actual area represented by the single pixel point.

12. The method according to claim 1, characterized in that Selecting a measurement frame image from the multiple frames of images based on the image quality includes: The image with the best image quality among the multiple frames of images is determined as the measurement frame image.

13. The method according to claim 1, characterized in that Selecting a measurement frame image from the multiple frames of images based on the image quality includes: Acquire at least two frames of images having image quality greater than a quality threshold from among the multiple frames of images; The image with the largest gastric antrum region area among the at least two frames of images is determined as the measurement frame image.

14. The method according to any one of claims 1 to 5, characterized in that Also includes: The measurement frame image is displayed on a display, and the gastric antrum region and the area of ​​the gastric antrum region are marked in the measurement frame image.

15. The method according to claim 14, characterized in that The boundary of the gastric antrum region is marked in the form of a line, or the gastric antrum region is covered in the form of a mask.

16. The method according to claim 14, characterized in that The gastric antrum region is marked with an ellipse, and the major and minor axes of the ellipse are labeled.

17. The method according to any one of claims 1 to 5, characterized in that Also includes: Gastric antrum dynamic parameters are calculated according to the area of ​​the gastric antrum region.

18. The method according to claim 17, characterized in that The gastric antrum dynamic parameters include gastric emptying time, and the calculation of the gastric antrum dynamic parameters includes: acquiring a first area of ​​a gastric antrum region in a fasting state of the stomach; Acquire a second area of ​​the gastric antrum region in a gastric filling state at a first moment; acquiring, by intermittent measurement, a third area of ​​the gastric antrum region after the stomach is in a full state, and recording a second time of acquiring the third area when a difference between the third area and the first area is less than an area threshold; The gastric emptying time is determined according to the first time and the second time.

19. The method according to claim 18, characterized in that Also includes: A curve showing the change of the area of ​​the gastric antrum region over time in the time interval from the first moment to the second moment is displayed.

20. The method according to claim 19, characterized in that The variation curve is a broken line marked by a plurality of points measured intermittently.

21. The method according to claim 19, characterized in that The variation curve is a smooth curve obtained by interpolating a plurality of points measured discontinuously.

22. The method according to claim 17, characterized in that The gastric antrum dynamic parameters include average gastric antrum contraction amplitude and average gastric antrum contraction frequency, and the calculation of the gastric antrum dynamic parameters includes: The average gastric antrum contraction amplitude and the average gastric antrum contraction frequency are determined according to the change in the area of ​​the gastric antrum region over time within a preset time period.

23. The method according to claim 17, characterized in that The gastric antrum dynamic parameter includes a gastric antrum motility index, and the calculation of the gastric antrum dynamic parameter includes: Determine the average gastric antrum contraction amplitude and the average gastric antrum contraction frequency according to the change of the area of ​​the gastric antrum region over time within a preset time period; The gastric antrum motility index is calculated according to the average gastric antrum contraction amplitude and the average gastric antrum contraction frequency.

24. The method according to claim 22 or 23, characterized in that According to the change of the area of ​​the gastric antrum region over time in a preset time period, the average gastric antrum contraction amplitude and the average gastric antrum contraction frequency are determined, including: Obtaining a curve of the area change of the gastric antrum region over time obtained by continuous measurement within a preset time period; determining a plurality of gastric antral motion cycles according to the change curve; The average gastric antrum contraction amplitude and the average gastric antrum contraction frequency are determined according to the contraction amplitude and the contraction frequency of each gastric antrum movement cycle in a plurality of gastric antrum movement cycles.

25. The method according to claim 24, characterized in that The change curve is obtained by smoothing and / or low-pass filtering the initial change curve.

26. The method of claim 22, further comprising: The curve showing the change in the area of ​​the gastric antrum region over time obtained by continuous measurement.

27. The method according to claim 26, characterized in that The change curve corresponds to the ultrasonic image obtained in real time.

28. The method according to claim 26, characterized in that The variation curve includes a plurality of curve segments corresponding to a plurality of time periods. Among them, multiple time periods are not continuous, and are represented by discontinuous time axes between different time periods.

29. The method according to any one of claims 1 to 5, characterized in that Also includes: Based on the area of ​​the gastric antrum region, the gastric volume was determined.

30. The method according to claim 29, characterized in that The determining of the gastric volume comprises: The stomach volume is determined based on the area of ​​the gastric antrum region of the stomach in a plurality of different sections.

31. The method according to claim 29, characterized in that Also includes: Based on the stomach volumes determined at a plurality of different time points, a stomach volume time curve showing the change of the stomach volume over time is determined and displayed.

32. The method according to claim 31, characterized in that Also includes: The multiple different moments are multiple moments of interval measurement, and also include: The gastric emptying time is predicted or calculated according to the gastric volume-time curve.

33. A method for determining an area based on an ultrasound image, characterized in that: include: transmitting an ultrasonic beam to a target object including a gastric antrum region by stimulating an ultrasonic probe, and receiving an ultrasonic echo returned from the target object by the ultrasonic probe to obtain an ultrasonic echo signal; The processor processes the ultrasonic echo signal to obtain an ultrasonic image of the target object; The processor determines a measurement frame image based on the ultrasound image; The processor inputs the measurement frame image into a pre-trained deep learning model to obtain an output of the deep learning model, wherein the output includes the area of ​​the gastric antrum region in the measurement frame image; The ultrasound image includes multiple frame images, and determining the measurement frame image includes: Determining the image quality of each frame of the plurality of frames of images; Based on the image quality, selecting a measurement frame image from the multiple frames of images; The image quality is represented as a quality score, the quality score includes a weighted sum of multiple sub-scores, and the sub-scores of the image include at least two of the following: The intersection-over-union ratio of the gastric antrum region in the image to the gastric antrum region in the previous frame of image; The reciprocal of the Hausdorff distance between the gastric antrum region in the image and the gastric antrum region in the previous frame of image; The reciprocal of the difference in area between the gastric antrum region in the image and the gastric antrum region in the previous frame of image; The reciprocal of the length of the longest straight line segment on the boundary of the gastric antrum region in the image; The inverse of the number of sharp corners on the border of the gastric antrum region in the image; The reciprocal of the inward concave area on the border of the gastric antrum region in the image; One or more scores for the image given by one or more machine learning models.

34. An ultrasonic device, characterized in that: include: Ultrasound probe; a transmitting / receiving circuit, used for stimulating the ultrasonic probe to transmit an ultrasonic beam to a target object, and controlling the ultrasonic probe to receive an ultrasonic echo returned from the target object to obtain an ultrasonic echo signal; A memory, used for storing a program executed by the processor; A processor for implementing the method according to any one of claims 1 to 33.

35. A computer storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a computer or a processor, the steps of the method according to any one of claims 1 to 33 are implemented.

Images