KR20190022185A - Method for measuring fetal body and device for measuring fetal body using the same - Google Patents

Abstract

The present invention provides a method for measuring a fetal body and a device using the same, which have an effect of providing an accurate measurement result on the fetal body. According to the present invention, the method for measuring a fetal body comprises the following steps: receiving an ultrasound image on a fetus; determining a section of a landmark inside the ultrasound image based on the ultrasound image and a process direction of ultrasound by using a prediction model configured to predict the landmark on a part of the fetus; determining the section on the part of the fetus inside the ultrasound image based on the section of the landmark; and measuring a size on the section.

Description

TECHNICAL FIELD [0001] The present invention relates to a fetal body measuring method and a fetal body measuring apparatus using the fetal body measuring method.

The present invention relates to a method of measuring an embryo and a device using the same, and more particularly, to a method of measuring a fetal body by determining a region of a fetus to be measured in an ultrasonic image and measuring a size thereof, .

Due to factors such as elderly birth, electromagnetic waves, and environmental hormones, birth rates of malformed or premature infants are increasing. Therefore, it is very important to obtain information on the growth state of the fetus quickly and accurately. On the other hand, measurement of the fetal body provides important information on the normal growth of the fetus.

Ultrasonic systems have non-invasive and non-destructive properties and are used for fetal body measurements. Such an ultrasound system has an advantage in that it can provide an image of a specific region of a fetus in real time without the need for a surgical operation. Specifically, the ultrasound system transmits an ultrasound signal to a specific part of a fetal body and receives an ultrasound echo signal reflected therefrom to form an ultrasound image for a specific part. (AC), abdominal circumference, biparietal diameter (BDP), occipitofrontal diameter (OFD), femoral diaphysis (FDL) and femur diaphysis (AC) abdominal circumference) of the fetus can be measured.

On the other hand, the method of measuring the fetal body using the conventional ultrasound image has a low error rate due to the high contrast in the ultrasound image of the head of the fetus, the head of the head, the head of the femur, and the femur. However, in the case of the abdomen of the fetus, the contrast between the skin of the fetus and the amniotic fluid in the ultrasound image is not clear, so that the conventional fetal anthropometric method provides low accuracy measurement results.

Therefore, development of fetal body measurement method and physical measurement device which can provide accurate and rapid analysis result in fetal body measurement, especially fetal abdominal measurement, and can provide accurate information on fetal growth state. Is continuously required.

BACKGROUND OF THE INVENTION [0002] Techniques as a background of the invention have been made in order to facilitate understanding of the present invention. And should not be construed as an admission that the matters described in the technical background of the invention are present in the prior art.

On the other hand, the inventors of the present invention have found that each of the fetal parts to be measured has a specific anatomical landmark. Particularly, the inventors of the present invention have confirmed that the abdomen, which is one of the parts having difficulty in measurement, is located inside the amniotic fluid region of the mother, and it can be recognized that a positive number can be a landmark for the fetal abdomen there was.

In addition, the inventors of the present invention have noted that conventional techniques do not consider the presence of shadowing artifacts present in ultrasound images. As a result, the inventors of the present invention can recognize that the accuracy of the body measurement of the fetus can be enhanced when the direction of the ultrasonic wave is parallel to the image pattern of the shadow artifact and the direction of the ultrasonic wave is considered.

As a result, the inventors of the present invention have developed a new fetal body measuring method and a device using the new fetal body that can accurately measure the body of a fetus based on landmarks and ultrasonic wave propagation directions of the fetal region to be measured .

Accordingly, a problem to be solved by the present invention is to provide a fetal body measuring method for measuring a fetal body based on a landmark and an ultrasonic wave propagation direction of a fetal region, and a device using the same.

Furthermore, the inventors of the present invention have recognized that a predictive or validated model can be used that is trained by information related to fetal anatomy, in order to increase accuracy of measurement and provide rapid analysis.

Another problem to be solved by the present invention is to provide a prediction model configured to predict a landmark in an ultrasound image based on a landmark and an ultrasonic wave propagation direction of a fetus region and a fetus region determined in the ultrasound image And to provide a device using the same.

Another object to be solved by the present invention is to provide a method and apparatus for receiving a plurality of ultrasound images and measuring ultrasound waves on the basis of the landmarks and the direction of the ultrasound waves for each image, And a device using the same.

The problems of the present invention are not limited to the above-mentioned problems, and other problems not mentioned can be clearly understood by those skilled in the art from the following description.

According to an aspect of the present invention, there is provided a method for measuring an anatomy of a fetus according to an embodiment of the present invention. The method includes receiving an ultrasound image of a fetus, using a predictive model configured to predict a landmark for a region of the fetus, and determining a region of the landmark within the ultrasound image based on the direction of the ultrasound image and the ultrasound wave Determining an area for a region of the fetus within the ultrasound image based on the area of the landmark, and measuring the size for the region.

According to an aspect of the present invention, the region of the fetus is abdomen, and the landmark may include a positive number. The step of determining an area of the landmark further includes the step of determining an amniotic area in the ultrasound image using the prediction model, wherein the step of determining the area for the fetal region comprises: To form an ellipse, and determining the ellipse to be the abdominal region of the fetus.

 According to another feature of the invention, the region of the fetus is abdomen, and the landmark may comprise at least one and a positive number selected from the vertebrae, the stomach, the liver portal vein, the umbilical vein and the ribs. The step of determining an area of the landmark may also include determining a region of positive numbers and at least one region selected from the vertebral, stomach, liver, umbilical vein and ribs in the ultrasound image using a prediction model . The step of determining the region for the region of the fetus includes determining the region of the spine, the stomach, the portal vein, the umbilical vein, and the rib, within the ultrasound image, Adjusting the contrast of the area for at least one selected from the stomach, liver, portal vein, umbilical vein and ribs, and the area for amniotic fluid; at least one selected from the vertebral, gastric, hepatic portal vein, Forming an ellipse based on a boundary of the region with respect to the region and the region with respect to the positive number, and determining the ellipse as the abdominal region of the fetus. The step of measuring the size for the area may also include measuring at least one of a circumference, a length of the major axis, a length of the minor axis, and an area with respect to the abdominal region of the fetus.

According to another aspect of the invention, determining the region for a region of the fetus comprises determining a boundary for a region of positive numbers, forming an ellipse based on a boundary for the amniotic region, And determining the ellipse as the abdominal region of the fetus when at least one region selected from the stomach, liver, portal vein, umbilical vein and rib is included.

According to another aspect of the present invention, the step of determining a region for a region of the fetus comprises determining a region of the fetus by determining a region of the fetus within at least one region selected from the group consisting of vertebra, stomach, liver portal vein, Determining a region of the abdomen candidate region of the fetus as an abdominal region candidate region of the fetus, filtering the abdomen candidate region of the fetus having a positive region below a predetermined level among the candidate regions, Determining the abdominal region of the fetus based on at least one value of the length of the major axis, the length of the minor axis, the angle of the major axis, and the position of the center.

According to another aspect of the present invention, there is provided an ultrasonic imaging apparatus including an ultrasonic imaging apparatus for providing an indication of a request for re-reception of an ultrasonic image including an abdomen of a fetus, ≪ / RTI >

According to another aspect of the present invention, the ultrasound image may further include shadowing artifacts. The method of claim 1, further comprising determining a shadow artifact region in the ultrasound image based on the ultrasound image and the direction of ultrasound propagation, wherein determining the region for the region of the embryo comprises, based on the artifact region and the region of the landmark, And determining an area for a region of the fetus within the ultrasound image.

According to another aspect of the present invention, a method for measuring an anatomy of a fetus in accordance with an embodiment of the present invention includes the steps of: detecting a target including a region for a determined fetal region using a verification model configured to verify a region for a fetal region; And verifying the region for the region of the fetus based on the position of the region ultrasonic image and the predetermined landmark.

According to another aspect of the present invention, verifying the region for a region of the fetus comprises obtaining an image patch comprising a portion of the target region ultrasound image, and acquiring an image patch of the acquired image and the location of the predetermined landmark As a basis, the verification model may be used to verify the region for the fetal region.

According to another aspect of the present invention, a method of measuring an anatomy of a fetus according to an embodiment of the present invention includes displaying an indication of a request for re-reception of a target site ultrasound image when a region of a fetus region is not verified by a verification model Or providing ultrasound measurement guideline for the site of the fetus based on the landmark.

According to another aspect of the present invention, the step of receiving the ultrasound image may further comprise acquiring an image patch including a part of the ultrasound image. The step of determining the area of the landmark may further comprise determining a region of the landmark within the ultrasound image using a prediction model configured to predict the landmark for the fetal region based on the image patch and the direction of travel of the ultrasound Step < / RTI >

According to another aspect of the present invention, receiving an ultrasound image for a fetus includes receiving a plurality of ultrasound images for a fetus, wherein determining an area of the landmark within the ultrasound image comprises: And determining an area of the landmark for each of the plurality of ultrasound images using the model. Further, determining the region for the fetal region comprises determining an area for the fetal region for each of the plurality of ultrasound images based on the region of the landmark, the step of measuring the size for the region May include measuring the size of the region for the fetal region included in each of the plurality of ultrasound images.

According to another feature of the present invention, the part of the fetus may include at least one selected from the head, the two limbs, the rib cage, the ventricle, the atrium, the kidney and the pelvis.

According to another aspect of the present invention, there is provided a device for measuring an anatomy of a fetus. The device includes an ultrasound probe configured to acquire an ultrasound image of the fetus, a processor operatively coupled to the ultrasound probe, wherein the processor is configured to use a predictive model configured to predict a landmark for a region of the fetus, Determining an area of the landmark in the ultrasound image based on the obtained ultrasound image and the direction of the progress of the ultrasound image and determining an area for the part of the fetus in the ultrasound image based on the area of the landmark, Size.

According to an aspect of the present invention, the region of the fetus is the abdomen, and the landmark may include at least one and a positive number selected from the vertebrae, the stomach, the liver portal vein, the umbilical vein and the ribs. The processor can also use the prediction model to determine at least one region and a positive region selected from the vertebral, stomach, liver, umbilical vein, and ribs in the ultrasound image. The processor is then configured to determine, within the ultrasound image, the vertebral body, the stomach, the portal vein, the umbilical vein, and the umbilical vein within the ultrasound image so as to determine the borderline for the at least one region and the positive region selected from the spine, stomach, liver portal vein, And at least one region selected from the ribs and the positive region, and the region for at least one selected from the vertebrae, the stomach, the liver portal vein, the umbilical vein and the ribs in the ultrasound image, And determine the ellipse to be the abdomen region of the fetus. Further, the processor may be further configured to measure at least one of a circumference, a length of the major axis, a length of the minor axis, and an area for the determined abdominal region of the fetus.

The present invention provides accurate information on fetal growth by providing a fetal body measurement method that considers a predetermined landmark and an ultrasonic wave propagation direction for a fetal part to be measured and a device using the same.

More particularly, the present invention relates to a method and system for estimating a landmark in an ultrasound image, using a predictive model configured to predict a landmark in an ultrasound image and a verification model configured to verify a region of the fetus determined in the ultrasound image, , A method of measuring an anatomy of a fetus, and a device using the same.

Particularly, the present invention has an effect of providing a fetal body measuring method capable of measuring the fetal abdomen based on a predetermined amniotic area determined as a landmark for the abdomen of the fetus, and a device using the fetal body.

Accordingly, the present invention has the effect of providing measurement results with high accuracy for fetal abdomen with a high error rate of analysis among various body parts.

Further, the present invention provides a method of measuring an embryo's body and a device using the same, which can measure a fetal body simultaneously with ultrasonic analysis based on a landmark and an ultrasonic wave propagation direction for each image by receiving a plurality of ultrasonic images There is an effect that can be done.

Accordingly, the present invention has the effect of simultaneously acquiring an image of a body part and simultaneously measuring a part of the fetus determined in the ultrasound image.

Further, according to the present invention, when an image including a specific part of a fetus is not determined or verification of a determined image is not performed before the measurement is performed, it is possible to inform the user of the image or provide an ultrasonic measurement guideline .

Accordingly, the present invention provides an accurate measurement result for the body of a fetus irrespective of the skill of the user using the ultrasonic system.

The effects according to the present invention are not limited by the contents exemplified above, and more various effects are included in the specification.

FIG. 1 illustrates a configuration of a fetal body measuring device according to an embodiment of the present invention.
FIG. 2A shows a procedure of a method for measuring an embryo in accordance with an embodiment of the present invention.
FIG. 2B is a view illustrating an example of determining a region of a landmark in an ultrasound image in a fetal body measurement method according to an embodiment of the present invention.
FIG. 2c is a graph showing the body measurement of a fetus according to an embodiment of the present invention The prediction model used in the method is exemplarily shown.
FIG. 2D illustrates shading artifacts in a received ultrasound image according to a method of measuring an embryo in accordance with an embodiment of the present invention.
FIG. 2E is a view illustrating an example of determining a region of a fetus region in an ultrasound image in a fetal body measurement method according to an embodiment of the present invention.
FIG. 2F is a view illustrating an example of a step of verifying a region of a fetus in a fetal body measurement method according to an embodiment of the present invention.
FIG. 3 illustrates a method of measuring an anatomy of a fetus according to an embodiment of the present invention and a result of evaluation of a device using the same.

BRIEF DESCRIPTION OF THE DRAWINGS The advantages of the invention, and how to achieve them, will become apparent with reference to the embodiments described in detail below with reference to the accompanying drawings. The present invention may, however, be embodied in many different forms and should not be construed as being limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. Is provided to fully convey the scope of the invention to those skilled in the art, and the invention is only defined by the scope of the claims.

The shapes, sizes, ratios, angles, numbers, and the like disclosed in the drawings for describing the embodiments of the present invention are illustrative, and thus the present invention is not limited thereto. In the following description, well-known functions or constructions are not described in detail since they would obscure the invention in unnecessary detail. Where the terms 'comprises', 'having', 'done', and the like are used herein, other parts may be added as long as '~ only' is not used. Unless the context clearly dictates otherwise, including the plural unless the context clearly dictates otherwise.

In interpreting the constituent elements, it is construed to include the error range even if there is no separate description.

It is to be understood that each of the features of the various embodiments of the present invention may be combined or combined with each other partially or entirely and technically various interlocking and driving is possible as will be appreciated by those skilled in the art, It may be possible to cooperate with each other in association.

For clarity of interpretation of the present specification, the terms used in this specification will be defined below.

As used herein, the term "ultrasound image" may refer to an image formed on the basis of an echo signal received by an ultrasound system, such as an ultrasound probe. Further, ultrasound images can be used interchangeably with "ultrasound images" in this specification. Thus, the ultrasound image of the fetus may mean a two-dimensional image including a fetal body part, a three-dimensional image, a still image of one cut, and a moving image composed of a plurality of cuts. For example, when the ultrasound image is a moving image composed of a plurality of cuts, the fetal part of each of a plurality of ultrasound images can be measured according to the fetal body measurement method according to an embodiment of the present invention. As a result, the fetal body measurement method and the device using the fetal body according to the embodiment of the present invention can provide the result of the measurement substantially simultaneously with the reception of the fetal ultrasound image by the ultrasonic probe.

The ultrasound images of the fetus can show various information about the body of the fetus. For example, the ultrasound image of the fetus can be based on the judgment of the fetal growth state, the position of the fetus in the uterus, the confirmation of the physical abnormality, the fetal movement and the confirmation of the heartbeat, and the gender judgment of the fetus.

On the other hand, fetal body measurement based on ultrasound images of the fetus can be an important test for determining whether the fetus is growing normally. Specifically, it is possible to determine whether or not the fetus normally grows by measuring the head circumference of the fetus, the amputation diameter, the frontal bone frontal diameter, the femur diameter, and the abdomen circumference. Accordingly, the term "fetal region ", as used herein, may be a specific body region of the fetus in which the size is to be measured. For example, the region of the fetus may include the head, amniocentesis, hindlimb bone, femur, and abdomen, and may further include the thoracic cavity, ventricle, atrium, kidney, and pelvis. However, not limited to the above site, the site of the fetus can be any site that can be an indicator of fetal growth. At this time, the error rate of the measurement may be low as the head of the fetus, the head of the head, the head of the head, and the femur show high contrast in the ultrasound image. However, in the case of the abdomen of the fetus, the contrast of the fetal skin and amniotic fluid in the ultrasound image is not clear, so the accuracy of the analysis may be lower than that of other parts in the body measurement.

In the fetal body measurement using ultrasound images, the accuracy of measurement can be improved when the fetal body site specific landmarks are taken into consideration. At this time, the term "landmark" may refer to a predetermined anatomical symbol region to identify a fetal body part. For example, the abdomen of the fetus may be located within the amniotic region of the mother within the ultrasound image. Accordingly, when the positive number is set as a landmark for the abdomen of the fetus, the abdomen of the fetus can be determined with high accuracy in the ultrasound image. As a result, the accuracy of measurement for the fetal abdomen can also be enhanced. In addition, vertebrae, stomach, liver portal vein, umbilical vein and ribs can be set as landmarks for the abdomen of the fetus. Therefore, when determining the abdominal region of the fetus based on the positive region in the ultrasound image and the region of the vertebra, the stomach, the liver portal vein, the umbilical vein, and the ribs, the determined region may be the actual abdomen of the fetus.

The ultrasound image may include shadow artifacts. As used herein, the term "shading artifact" may refer to an artifact not associated with a region of the fetus within the ultrasound image. Thus, considering the presence of shadow artifacts in fetal body measurement may be important in lowering the error. On the other hand, in the ultrasound image, the azimuth pattern of the shadow artifact may appear parallel to the direction of the ultrasonic wave propagation. Accordingly, when the direction of the ultrasonic wave is taken into consideration, the accuracy of measurement of the fetal body can be high.

As used herein, the term "prediction model" may refer to a predictive model trained to determine a landmark for a fetal body part. For example, the predictive model can be configured to discriminate and determine the region of each landmark within the ultrasound image as a model trained using an ultrasound image containing the data set for a predetermined landmark as an input. More specifically, the predictive model analyzes the ultrasound image using a composite neural network and analyzes the area of the amniotic fluid which is set as a landmark specific to the abdomen of the fetus, and furthermore, the region of the vertebra, the stomach, the liver portal vein, the umbilical vein, You can decide. In addition, the prediction model can analyze the direction of the ultrasonic wave propagation and determine the area of the shadow artifact in the ultrasonic image.

Based on the landmark determined by the prediction model, the region for the fetal region can be determined. For example, the area of the fetal abdomen can be determined on the basis of the boundary line of each of the regions of the landmark, that is, the landmark, and also the regions of the vertebra, stomach, liver portal vein, umbilical vein, and rib. Specifically, the region of the fetal abdomen within the ultrasound image may be an elliptical region formed by a plurality of boundary lines for a positive region. Accordingly, the elliptical region formed based on the landmark in the ultrasound image can be determined as the candidate region or the abdominal region of the abdomen. However, the region for the fetal region is not limited to the elliptical shape, and may have various shapes depending on the type of the target region to be measured.

As used herein, the term "verification model" may refer to a model trained to verify that the region of the fetus determined in the ultrasound image based on the landmark is correct for the fetal region to be actually measured. Since the results of fetal body measurements may vary according to various variables such as fetal position and fetal posture, the use of verification model may be more important for accurate measurement. For example, the verification model may be trained by a data set that includes information of the position for each of the predetermined landmarks, thereby performing verification of the determined target region of the fetal target region. More specifically, in the case of the fetal abdomen, it may be present in an internal region formed by the boundary line of the amniotic region in the ultrasonic image. Thus, the verification model can perform verification on the region determined by the abdomen of the fetus based on the level of the positive number. At this time, the verification model can receive the ultrasound image including the abdomen region of the fetus based on the composite neural network, and read out whether the abdomen region in the ultrasound image is a standardized image.

The predictive model and verification model disclosed herein may use an image patch that includes the entire region of the received ultrasound image, or a portion thereof. For rapid analysis, it may be desirable to use an image that is scaled to have a constant number of pixels in the form of an image patch. However, it is not limited thereto.

Finally, the region for the region of the fetus that has been verified by the verification model is measured in terms of its circumference, the length of the long axis, the length or area of the short axis, and such measurements can be used as information on fetal body growth.

Hereinafter, a method of measuring an embryo's body and a device using the same will be described with reference to FIGS. 1 to 2F.

First, with reference to FIG. 1, a fetus body measuring device according to an embodiment of the present invention will be described in detail. FIG. 1 illustrates a configuration of a fetal body measuring device according to an embodiment of the present invention.

Referring to FIG. 1, a fetal aneurysm device 100 includes an ultrasonic probe 110, an input unit 120, an output unit 130, a storage unit 140, and a processor 150.

Specifically, the ultrasonic probe 110 may transmit an ultrasonic signal to a body part of a fetus and receive an ultrasonic echo signal reflected therefrom. Further, the ultrasonic probe 110 is configured to further receive an ultrasonic image for a specific part of the fetus formed based on the received echo signal. Further, the ultrasonic probe 110 may receive an ultrasonic echo signal for a landmark of a specific area. Further, the ultrasound image for a specific site received through the ultrasound probe 110 may include a landmark. For example, the ultrasound probe 110 may receive ultrasound images of the abdomen, head, two limbs, thorax, ventricle, atrium, kidney, and pelvis of the fetus. On the other hand, the ultrasound image of the abdomen of the fetus received via the ultrasound probe 110 may include images of amniotic fluid, vertebrae, stomach, liver portal, umbilical vein or ribs predetermined as a landmark for the abdomen .

The input unit 120 is not limited to a keyboard, a mouse, a touch screen panel, and the like. The input unit 120 can set the fetus's body measurement device 100 and can instruct the fetal body measurement device 100 to operate. For example, the input unit 120 may direct the measurement of a specific fetal region of the fetus, such as the abdomen, head, head, thorax, ventricle, atrium, kidney or pelvis. Further, the user can directly determine the region to be measured in the ultrasonic image received by the ultrasonic probe 110 through the input unit 120, and determine the direction of the ultrasonic wave propagation.

Meanwhile, the output unit 130 can visually display the ultrasonic image received by the ultrasonic probe 110. Furthermore, the output unit 130 is configured by the processor 150 to display the landmark region determined in the ultrasound image, the region of the specific region, and the measurement result for these regions. In addition, the output unit 130 can display a notification so that the user can easily recognize the fact if the area of the specific part to be measured is not determined or not verified, and the ultrasonic measurement guideline Can be displayed.

The storage unit 140 may store an image of the fetus region received through the ultrasonic probe 110 and store an instruction of the fetal body measurement device 100 set through the input unit 120. [ Furthermore, the storage unit 140 is configured to store the landmark region, the region of the specific region determined by the processor 150, which will be described later, and store the measurement result for the determined region of the specific region. However, without being limited to the foregoing, the storage unit 140 may store various information determined by the processor 150 for fetal anatomy.

The processor 150 may be a major component for providing accurate measurement results of the fetal anatomic device 110. At this time, for accurate measurement, the processor 150 may be configured to verify a predictive model configured to predict a landmark in the ultrasound image and a fetal region determined in the ultrasound image, based on the landmark and the direction of the ultrasound traveling on the fetal region And configured to use a configured verification model. For example, the processor 150 may use a prediction model configured to predict a landmark for a region of the fetus, based on the ultrasound image acquired through the ultrasound probe 110 and the direction of travel of the ultrasound, To determine the area of the landmark within the ultrasound image. In addition, the processor 150 may be configured to determine an area for a region of the fetus within the ultrasound image based on the region of the determined landmark, and to measure the size for the determined region.

On the other hand, the processor 150 may be configured to adjust the contrast of the ultrasound image to determine the borderline to the landmark, and finally to provide a measurement result for the region of the fetal region. For example, if the region of the fetus to be measured is the abdomen, the processor 150 uses the predictive model to determine at least one region of the spine, stomach, liver portal, umbilical vein, To determine the area of the landmark, and to adjust the contrast of the ultrasound image to determine the boundary line for each landmark area within the ultrasound image. In addition, the processor 150 may form an ellipse based on the boundary line of each landmark region in the ultrasound image, and determine an ellipse as the abdomen region of the fetus. Finally, the processor 150 may measure the size of at least one of the circumference, the length of the major axis, the length of the minor axis, and the area for the determined abdominal region of the fetus.

Next, with reference to FIGS. 2A to 2F, a method of measuring an anatomy of a fetus according to an embodiment of the present invention will be described in detail. In this case, the measurement of the abdomen of the fetus has been described as an example. However, the method of measuring the fetus according to the embodiment of the present invention can be applied to the measurement of more various fetal regions. For example, the various fetal regions can be either the head of the fetus, the two glands, the thoracic cavity, the ventricle, the atrium, the kidney or the pelvis. FIG. 2A shows a procedure of a method for measuring an embryo in accordance with an embodiment of the present invention. FIG. 2B is a view illustrating an example of determining a region of a landmark in an ultrasound image in a fetal body measurement method according to an embodiment of the present invention. FIG. 2C illustrates an example of a prediction model used in a fetal body measurement method according to an embodiment of the present invention. FIG. 2D illustrates shading artifacts in a received ultrasound image according to a method of measuring an embryo in accordance with an embodiment of the present invention. FIG. 2E is a view illustrating an example of determining a region of a fetus region in an ultrasound image in a fetal body measurement method according to an embodiment of the present invention. FIG. 2F is a view illustrating an example of a step of verifying a region of a fetus in a fetal body measurement method according to an embodiment of the present invention.

Referring to FIG. 2A, the procedure of the body measurement method according to an embodiment of the present invention is as follows. First, an ultrasonic image of the fetus is received (S210). Next, a landmark region is determined in the ultrasound image based on the direction of the ultrasound image and the ultrasonic wave using a predictive model configured to predict a landmark for the fetal region (S220). Then, based on the area of the landmark, the region for the fetal region within the ultrasound image is determined (S230). Finally, the size of the region for the determined fetal region is measured (S240).

Specifically, in the step of receiving the ultrasound image (S210), the ultrasound image 212 formed on the basis of the ultrasound echo signal reflected at a specific part of the fetus can be obtained. Alternatively, in step S210 of receiving the ultrasound image, an image patch including a part of the ultrasound image may be obtained. For example, the image patch has the resolution or size required by the prediction model, and the use of the image patch improves the processing speed in the prediction model because the image patch has a smaller resolution or size than the original ultrasound image. At this time, the ultrasound image acquired through the step of receiving the ultrasound image (S210) may include a part of the fetus to be measured, a landmark therefor, and further a shadow artifact.

Next, in step S220 of determining an area of the landmark, based on the ultrasonic image acquired in the step S210 of receiving the ultrasonic image and the traveling direction of the ultrasonic wave, Using the prediction model, the area of the landmark in the ultrasound image can be determined.

For example, in step S220 of determining the area of the landmark, the area of the landmark for the fetal abdomen can be determined. Referring to FIG. 2B, the step S220 of determining the area of the landmark includes receiving the ultrasound image 212 received in the step S210 of receiving the ultrasound image or a part of the ultrasound image 212 Based on a plurality of image patches 214 and 216 and an ultrasonic wave propagation direction 222. [ At this time, a prediction model 244 may be used to determine the area of the landmark for the abdomen. Specifically, prediction model 224 may be configured to determine an area of the landmark for the abdomen as it is trained with a data set for a predetermined landmark for the abdomen of the fetus. Referring to FIG. 2C, a prediction model 224 is illustratively shown. The prediction model 224 may be composed of a data input unit 224a, an image analysis unit 224b, an ultrasonic wave propagation direction analysis unit 224c, an analysis result integration unit 224d and an integrated result output unit 224e have. Specifically, the data input unit 224a may be configured to receive the plurality of image patches 214 and 216 and the ultrasonic propagation direction 222 received in the step S210 of receiving the ultrasonic image. The image analyzing unit 224b may be configured to analyze the plurality of image patches 214 and 216 input through the data input unit 224a using the composite neural network to predict landmarks for the abdomen. For example, the image analyzer 224b may be configured to determine an area of the landmark for the abdomen, such as the amniotic fluid, stomach, umbilical vein, within the ultrasound image. The ultrasonic traveling direction analyzing unit 224 c may be configured to analyze the ultrasonic traveling direction 222 received through the data input unit 224 a using the composite neural network to predict the shadow artifact. Further, the analysis result integration unit 224d may be configured to integrate the analysis results obtained through the image analysis unit 224b and the ultrasonic wave direction analysis unit 224c using the composite neural network. Finally, the final analysis result in which the regions of the landmark for the abdomen in the ultrasound image are separated can be shown through the integrated result output section 224e. 2d, a fetal abdomen region 218 and a shadow artifact region 219 in the ultrasound image 212 are shown. In the case of the fetal abdomen region 218, the ultrasonic propagation direction 222 appears to be different from the image pattern direction 227. In contrast, the shading artifact region 219 ultrasonic propagation direction 222 appears to be parallel to the image pattern direction 227. Accordingly, the prediction model 224 using the ultrasonic propagation direction 222 can determine the area of the shadow artifact with high accuracy, and as a result, can reduce the error of the measurement by the shadow artifact. Referring to the ultrasound image 226 where the landmark region of FIG. 2B has been determined, as a result of the step S220 of determining the landmark region, the area of each of the positive, The area of the artifact can be determined within the ultrasound image.

Next, based on the area of the landmark determined in the step S230 of determining the region for the fetus region and the step S220 for determining the region of the landmark, the region for the region of the fetus in the ultrasound image is determined do.

For example, with reference to FIG. 2E, contrast can be adjusted to determine the borderline for the amniotic area determined as a landmark in the ultrasound image in the case of the abdomen of the fetus (the area of the landmark is determined as the ultrasound image 226 and Contrast adjusted ultrasound image 232). As a result, a boundary line for the positive region can be determined (see the ultrasonic image 234 for which the boundary line is determined). A plurality of ellipses 234 (a) and 234 (b) can be formed based on the determined boundary line in the ultrasonic image 234 in which the boundary line is determined. At this time, in the step S230 of determining the region for the fetal region, the ellipses 234 (a) including a region having a positive number equal to or higher than the predetermined level are excluded and the ellipses 234 (b)) can be filtered and determined as the abdominal region of the fetus. At this time, for determination into the abdominal region of the fetus, the length of the major axis, the length of the minor axis, the angle of the major axis, or the position of the center of the ellipse 234a including a region of positive or more than a predetermined level may be considered. In step S230 of determining the region of the fetus, the ellipsoid of the fetus may be determined as an ellipse in which the spinal cord, the stomach, the liver portal vein, the umbilical vein, or the rib region is included in the ellipse formed. If the region is not determined as a region, a step S230 of determining an area for the fetus region may request the user to re-receive the ultrasound image 212, or may provide an ultrasonic measurement guideline for the fetal abdomen.

Meanwhile, in the body measurement method according to the embodiment of the present invention, the region of the fetus determined in the ultrasound image can be further verified through the step S230 of determining the region of the fetus region.

For example, referring to FIG. 2F, an ellipse 234 (b) containing a positive region below a predetermined level, determined as the abdominal region, can be additionally performed to verify that it is the abdominal region of the normalized fetus ( An ultrasound image with an ellipse 238). At this time, a verification model trained to verify for the determined fetal abdomen can be used. Specifically, the verification model is trained by a data set consisting of an ultrasound image including an ultrasound image of the abdomen of the fetus and a position of each of the predetermined landmarks, thereby performing verification of the determined abdomen region of the fetus . Optionally, the verification model may include an abdominal region image patch 239 comprising an elliptical ultrasound image 238 or an ellipse 234 (b) containing a positive region less than a predetermined level, determined as the abdominal region, Can be used. For example, the verification model may be applied to a region of the liver portal, an umbilical vein, or a rib that includes an elliptical ultrasound image 238 or an ellipse 234 (b) that includes an amniotic region within the abdominal region image patch 239 Based on the location, the region for the determined fetal abdomen can be verified. In addition, if the region of the fetal abdomen determined in the present verification is not verified, the present verification step may require the user to re-receive the ultrasonic image 238 formed with an ellipse, or provide an ultrasonic measurement guideline for the fetal abdomen .

Finally, in step S240 of measuring the size of the region of the fetal region, a region for the fetal region determined in the step S230 of determining the region for the fetal region, or a region for the verified fetal region Is performed.

For example, in step S240 of measuring the size of a region of the fetal region, the length of the ellipse 234 (b) including the amniotic region less than the predetermined level determined as the abdominal region, the length of the long axis, The length or the size of the area can be measured. The measurement of the length of the circumference, the long axis, the length of the short axis or the area of the fetal abdomen measured as a result of the step S240 of measuring the size of the fetal region can be used to provide information on whether the fetus is normally grown have.

Hereinafter, a method of measuring an anatomy of a fetus according to an embodiment of the present invention and a result of a suitability evaluation for a device using the same will be described with reference to the first embodiment.

Example 1: Method of measuring fetal body according to one embodiment of the present invention and evaluation of suitability for devices using the same

In this evaluation, a total of 105 ultrasound images were used, and for the same image, an expert and a fetal body measuring device according to an embodiment of the present invention determine a true abdominal plane (True) and a false abdominal plane (False).

Referring to FIG. 3, a device according to an embodiment of the present invention determines 64 false images among 77 images determined by the expert as false abdominal planes from among 105 images as false abdominal planes in the same manner as that of experts. In addition, the device according to an embodiment of the present invention determines, as the empirical plane, the 17 images among the 28 images determined by the expert on the true abdominal plane among 105 images, as determined by the expert. Accordingly, the accuracy of a device according to one embodiment of the present invention, with respect to the determination of one of 105 images, is about 77%. Specifically, the accuracy of the device according to one embodiment of the present invention is 64 for the false abdominal plane and 64 for the false abdominal plane, The value divided by 105, which is the number of images, is a percentage.

As a result of the first embodiment, the fetal body measuring method and the device using the fetal body according to the embodiment of the present invention can measure the fetal body with high accuracy. Particularly, the fetal body measuring method and the device using the fetal body according to an embodiment of the present invention are applicable to an abdominal region having a high error rate of measurement among various fetal body parts, anatomical landmarks, direction of ultrasonic wave propagation, , It is possible to provide accurate measurement results.

Further, the fetal body measuring method and the device using the fetal body according to an embodiment of the present invention may further include a predictive model configured to predict a landmark for a fetal region in the ultrasound image, and a fetus model configured to verify a fetal region determined in the ultrasound image By using the verification model, the accuracy of the measurement can be improved and the result of the analysis can be provided quickly. In particular, the fetal body measurement method and the device using the fetal body according to an embodiment of the present invention can quickly provide accurate measurement results on the abdomen of the fetus.

Meanwhile, the method of measuring an anatomy of a fetus according to an embodiment of the present invention and the range and effect of use of the device using the method are not limited. For example, a fetal body measurement method and a device using the fetal body according to an embodiment of the present invention may include receiving a plurality of ultrasound images for a fetal body region, determining regions for a body region for each of the plurality of ultrasound images, Since the measurement can be performed, it is possible to simultaneously provide the result of the measurement together with the reception of the ultrasound image of the fetus by the ultrasonic probe. In addition, the fetal body measuring method and the device using the fetal body according to an embodiment of the present invention can be used not only in the abdomen of the fetus but also in the measurement of the head, the head, the thorax, the ventricle, the atrium, the kidney or the pelvis. Furthermore, the method of measuring an anatomy of a fetus according to an embodiment of the present invention and the device using the method can provide accurate measurement results for these regions.

Although the embodiments of the present invention have been described in detail with reference to the accompanying drawings, it is to be understood that the present invention is not limited to those embodiments and various changes and modifications may be made without departing from the scope of the present invention. Therefore, the embodiments disclosed in the present invention are intended to illustrate rather than limit the scope of the present invention, and the scope of the technical idea of the present invention is not limited by these embodiments. Therefore, it should be understood that the above-described embodiments are illustrative in all aspects and not restrictive. The scope of protection of the present invention should be construed according to the following claims, and all technical ideas within the scope of equivalents should be construed as falling within the scope of the present invention.

100: Fetal body measuring device
110: Ultrasonic probe
120: Input unit
130:
140:
S210: receiving ultrasound image for fetus
212: Ultrasound image
214, 216: Patch image
218: fetal abdomen area
219: Shaded Artifact Areas
S220: a step of determining an area of the landmark
222: direction of ultrasonic propagation
224: Predictive model
224a: Data input section
224 b: Image analysis section
224 c: ultrasonic traveling direction analyzing unit
224 d: Analysis result integration department
224 e: Overall result output section
226: Ultrasonic image in which the area of the landmark is determined
S230: determining the region for the fetal region
232: Contrast-adjusted ultrasound image
234: Ultrasonic image with boundaries determined
234 (a): an ellipse containing a positive area above a predetermined level
234 (b): an ellipse containing a positive area less than a predetermined level
238: Ultrasonic image formed with an ellipse
239: patch image of abdominal region
S240: Measuring the size of the region for the fetal region

Claims (15)

Receiving an ultrasound image of the fetus;
Determining an area of the landmark in the ultrasound image based on the direction of the ultrasound image and the direction of the ultrasound using a predictive model configured to predict a landmark for the part of the fetus;
Determining an area for the region of the fetus within the ultrasound image, based on the region of the landmark; and
And measuring a size for the region. The method according to claim 1,
The region of the fetus is the abdomen,
Wherein the landmark comprises a positive number,
Wherein the step of determining an area of the landmark comprises the step of determining the amniotic area in the ultrasound image using the prediction model,
Wherein determining the region for the fetal region comprises forming an ellipse based on the determined boundary of the amniotic region,
And determining the ellipse as the abdominal region of the fetus. The method according to claim 1,
The region of the fetus is the abdomen,
Wherein the landmark comprises at least one and a positive number selected from the vertebra, the stomach, the liver portal vein, the umbilical vein and the rib,
Wherein the step of determining the region of the landmark comprises the step of determining at least one region and a positive region selected from the spine, stomach, liver, umbilical vein and ribs in the ultrasound image using the prediction model Including,
Wherein determining the region for the region of the fetus comprises determining a region for the region of the fetus to determine a boundary line for at least one region and a positive region selected from the vertebral, stomach, liver, umbilical vein, and ribbons within the ultrasound image, Adjusting the contrast of the region with respect to at least one selected from the spinal, stomach, liver, umbilical vein,
Forming an ellipse based on the boundary of the region for at least one selected from the vertebral, gastric, hepatic, umbilical vein and ribs in the ultrasound image and the region for positive numbers; and
Determining the ellipse as the abdomen region of the fetus,
Wherein measuring the size of the area comprises measuring at least one of a circumference, a length of the major axis, a length of the minor axis, and an area with respect to the abdominal region of the fetus. The method of claim 3,
Wherein determining the region for the region of the fetus comprises: determining a boundary for the region of positive numbers;
Forming an ellipse based on a boundary line for the positive number region, and
And determining the ellipse to be the abdomen region of the fetus when the at least one region selected from the vertebrae, the stomach, the liver portal vein, the umbilical vein and the ribs is included in the ellipse. The method of claim 3,
Wherein the step of determining a region for the fetal region comprises determining a region within a plurality of boundaries that passes through a boundary between the vertebra, the at least one region selected from the stomach, the liver portal vein, the umbilical vein vein, Determining a candidate region;
Filtering the abdomen candidate region of the fetus having a positive region below a predetermined level among the candidate regions, and
Determining an abdomen region of the fetus based on at least one value of a length of a major axis, an axis of a minor axis, an angle of a major axis, and a center position with respect to an ellipse formed by the abdomen candidate region of the fetus, Fetal body measurement method. 6. The method of claim 5,
If the abdominal area of the fetus is not determined,
Further comprising providing an indication of a re-receive request for an ultrasound image including the abdomen of the fetus or providing an ultrasound measurement guideline for the fetal abdomen based on the landmark. The method according to claim 1,
Wherein the ultrasound image further comprises a shadowing artifact,
Further comprising determining the shadow artifact region in the ultrasound image based on the ultrasound image and the ultrasound propagation direction,
Wherein determining the region for the fetal region comprises determining an area for the fetal region within the ultrasound image based on the artifact region and the region of the landmark, Way. The method according to claim 1,
Using a verification model configured to verify a region for the fetal region based on a location of the landmark and a target region ultrasound image comprising a region for the determined region of the fetus, Further comprising the step of verifying the region. 9. The method of claim 8,
Wherein verifying the region for the fetal region comprises obtaining an image patch comprising a portion of the target region ultrasound image,
And verifying the region for the region of the fetus using the verification model based on the acquired image patch and the predetermined position of the landmark. 9. The method of claim 8,
If the region for the fetus region is not verified by the verification model,
Further comprising providing an indication of a re-receipt request for the target site ultrasound image or providing an ultrasound measurement guideline for the site of the fetus based on the landmark. The method according to claim 1,
The step of receiving the ultrasound image further comprises acquiring an image patch including a part of the ultrasound image,
Wherein the step of determining the area of the landmark comprises the steps of: using the image patch and a prediction model configured to predict a landmark for the part of the fetus based on the traveling direction of the ultrasonic wave, And determining an area of the fetus. The method according to claim 1,
The step of receiving an ultrasound image for a fetus includes receiving a plurality of ultrasound images for the fetus,
Wherein the step of determining an area of the landmark in the ultrasound image includes the step of determining an area of the landmark for each of the plurality of ultrasound images using the prediction model,
Wherein determining the region for the fetal region comprises determining an area for the fetal region for each of the plurality of ultrasound images based on the region of the landmark,
Wherein the measuring the size of the area comprises measuring a size of a region of the fetal region included in each of the plurality of ultrasound images. The method according to claim 1,
Wherein the fetal region comprises at least one selected from the group consisting of a head, a cervix, a thoracic cavity, a ventricle, an atrium, a kidney and a pelvis. An ultrasound probe configured to acquire ultrasound images of the fetus,
And a processor operably coupled to the ultrasonic probe,
Wherein the processor uses a prediction model configured to predict a landmark for a region of the fetus so as to determine a region of the landmark in the ultrasound image based on the ultrasound image acquired through the ultrasound probe and the direction of travel of the ultrasound wave, And determine a region for the region of the fetus within the ultrasound image based on the region of the landmark and measure the size for the region. 15. The method of claim 14,
The region of the fetus is the abdomen,
Wherein the landmark comprises at least one and a positive number selected from the vertebra, the stomach, the liver portal vein, the umbilical vein and the rib,
Wherein the processor is configured to use the prediction model to determine at least one region and a positive region selected from the spine, stomach, liver portal, umbilical vein, and ribbing within the ultrasound image, At least one region selected from the vertebrae, the stomach, the liver portal vein, the umbilical vein and the ribs in the ultrasound image to determine a boundary line for the at least one region and the positive region selected from the stomach, liver portal vein, umbilical vein vein, And adjusting the contrast of the region with respect to the positive numbers and determining an ellipse based on a boundary of the region for at least one selected from the vertebrae, the stomach, the liver portal vein, the umbilical vein and the ribs in the ultrasound image, The ellipse is determined as the abdomen region of the fetus, and the determined abdomen region of the fetus Circumferential length of the long axis, further configured to measure at least one of the size of the length and the area of the reduced, fetal anthropometric device.

Images