Automatic measurement method for antenatal ultrasonic image

Abstract

The invention discloses an automatic measurement method for an antenatal ultrasonic image, and the method comprises the steps: obtaining a to-be-measured ultrasonic image, and extracting all to-be-measured target objects carried by the to-be-measured ultrasonic image; and measuring all the extracted biomass of all the to-be-measured target objects, and determining a measurement result of each to-be-measured target object according to all the measured biomass. The measurement result of the target object is automatically measured, the workload of obtaining the measurement result is reduced, and the working efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of prenatal ultrasonic image processing, and more particularly to an automatic measurement method for prenatal ultrasonic images. Background Art

[0002] Prenatal ultrasound diagnosis is one of the important means to discover congenital defects. Ultrasound diagnosis has the advantages of no radiation, convenience and real-time, and is widely used in clinical detection of prenatal fetuses. At present, in the parameter measurement process of conventional ultrasound images, doctors need to rely on their rich clinical experience to operate the equipment and instruments to find the biomass to be measured in the ultrasound image, and then manually draw the measurement line and finally perform numerical calculations by the system. Therefore, in the traditional parameter measurement process of ultrasound images, doctors need to perform a lot of tedious manual operations, which seriously affects the doctor's diagnostic efficiency, especially when a frame of ultrasound image contains multiple biomass to be measured. This situation is more serious. In addition, due to different operating habits between different doctors, it is possible to artificially introduce certain measurement errors during the measurement process and affect the final measurement results. In addition, fetal congenital heart disease is the most common and most serious congenital malformation in the world, and it is also the first factor in neonatal death. Although there have been great advances in the diagnosis and management of fetal congenital heart disease, fetal congenital heart disease is still the most common cause of death in the first year of newborns. Therefore, effective prenatal diagnosis and prevention will affect the outcome of newborns with congenital heart disease and reduce birth defects and mortality. In actual clinical medicine, ultrasound is generally used to obtain a cross-sectional image of the chest at the level of the fetal four-chamber heart, and then the fetal cardiac axis is obtained by manual tracing measurement to assist in the screening and diagnosis of fetal congenital heart disease. However, the above-mentioned method of measuring the cardiac axis is easily affected by the doctor's clinical experience and other subjective factors, resulting in large errors in the manually measured data, leading to missed or misdiagnosed cases. Summary of the invention

[0003] 1. Technical issues to be solved

[0004] In view of the problems existing in the prior art, the present invention provides an automatic measurement method for prenatal ultrasound images to solve the technical problems mentioned in the background technology.

[0005] (II) Technical solution

[0006] To achieve the above object, the present invention provides the following technical solution: a method for automatically measuring prenatal ultrasound images, comprising the following steps:

[0007] Step 1: Acquire an ultrasound image to be measured, extract all target objects to be measured carried by the ultrasound image to be measured, acquire an ultrasound four-chamber heart section image of the fetal heart to be measured, obtain an image to be segmented, and perform target segmentation processing on the target four-chamber heart ultrasound section image based on a preset target segmentation model to obtain atrial septum contour data, ventricular septum contour data, spine contour data, and chest cavity contour data;

[0008] Step 2: measuring all biomasses of all extracted target objects to be measured, and determining the measurement results of each target object to be measured according to all measured biomasses, inputting the image to be segmented into the trained segmentation model to obtain a segmentation result image; the segmentation result image includes segmented areas of various structures of the fetal heart, and performing axis fitting based on the atrial septum contour data and the ventricular septum contour data to obtain a first heart axis;

[0009] Step 3: The segmentation model that has been trained is obtained by training a preset segmentation model, and the measurement of all biomasses of all target objects to be measured and the determination of the measurement results of each target object to be measured based on all the measured biomasses specifically include: measuring all biomasses of all target objects to be measured and extracted;

[0010] Step 4: for each target object to be measured, determine the object type of the target object to be measured; screen the biomass of the target object to be measured from all biomasses according to the object type to obtain the measurement result of the target object to be measured; the object type of the target object to be measured is determined according to a preset neural network, and when the preset neural network fails, re-determine the measurement result of the target object to be measured based on all the acquired biomasses.

[0011] The present invention is further configured such that after extracting all target objects to be measured carried by the ultrasound image to be measured, the step further includes: obtaining the number of target objects to be measured extracted; when the number is greater than or equal to 1, executing the step of measuring each target object to be measured; when the number is less than 1, prompting that the target object to be measured is not identified, performing axis fitting based on the spinal contour data and the thoracic contour data to obtain a second heart axis; and obtaining the heart axis according to the angle between the first heart axis and the second heart axis.

[0012] The present invention is further configured such that the size of each segmented area in the segmentation result image is measured respectively, including: determining the outer boundary information of each segmented area in the segmentation result image respectively; determining the minimum enclosing rectangle of each segmented area based on the outer boundary information; and measuring the minimum enclosing rectangle of each segmented area respectively to obtain the size of each segmented area.

[0013] The present invention is further configured such that after measuring and extracting all biomasses of all target objects to be measured and determining the measurement results of each target object to be measured based on all the measured biomasses, the method further includes: for each target object to be measured, drawing a corresponding measurement line according to the updated measurement result of the target object to be measured, and drawing the measurement line in the ultrasonic image.

[0014] The present invention is further configured such that the measurement result of the biomass includes position information of the biomass and structural information of the biomass.

[0015] The present invention is further configured such that the acquiring of the ultrasonic image to be measured specifically includes: acquiring a number of continuous ultrasonic image frames; and screening the number of continuous ultrasonic image frames to obtain the ultrasonic image to be measured.

[0016] The present invention is further configured such that the extraction of all target objects to be measured carried by the ultrasound image to be measured specifically includes: inputting the ultrasound image to be measured into a preset recognition model, wherein the preset recognition model is trained based on a training sample set, and each training sample in the training sample set includes an ultrasound image and all target objects to be measured carried by the ultrasound image; and outputting all target objects to be measured carried by the ultrasound image to be measured through the preset recognition model.

[0017] The present invention is further configured such that respectively determining the outer boundary information of each segmented area in the segmentation result image includes: denoising the segmentation result image to obtain a denoised segmentation result image; performing edge detection on the denoised segmentation result image to obtain an edge detection image; and binarizing the edge detection image to obtain the outer boundary information of each segmented area.

[0018] (III) Beneficial effects

[0019] Compared with the prior art, the present invention provides an automatic measurement method for prenatal ultrasound images, which has the following beneficial effects:

[0020] The present invention provides an automatic measurement method, storage medium and ultrasonic equipment for prenatal ultrasonic images. The method includes obtaining an ultrasonic image to be measured, extracting all target objects to be measured carried by the ultrasonic image to be measured, measuring all biomasses of all extracted target objects to be measured, and determining the measurement results of each target object to be measured according to all measured biomasses. The measurement results of the target objects are automatically measured, the workload of obtaining the measurement results is reduced, and the work efficiency is improved; on the other hand, the measurement error or failure of the measurement result caused by the wrong judgment of the object type can be avoided, thereby improving the accuracy of the measurement result of the target object. The application can realize the accurate segmentation of each structure of the fetal heart in the segmented image, so that the accuracy of the segmentation result is higher, and the size of each segmented area in the segmentation result image is measured respectively to realize automatic measurement, thereby avoiding the problem of low accuracy of the measurement result caused by the cumbersome measurement process and the existence of non-standardization in the prior art. Furthermore, the application can ensure the accuracy of the measurement result, so that the expert only needs to make a diagnosis based on the measurement result, thereby providing conditions for the expert's ability to benefit more patients and regions, thereby improving the prenatal detection rate of fetal congenital heart disease. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is an overall flow chart of an automatic measurement method of prenatal ultrasound images in the present invention; DETAILED DESCRIPTION

[0022] Embodiment 1:

[0023] like Figure 1 As shown, the present invention discloses a method for automatically measuring prenatal ultrasound images, comprising the following steps:

[0024] Step 1: Acquire an ultrasound image to be measured, extract all target objects to be measured carried by the ultrasound image to be measured, acquire an ultrasound four-chamber heart section image of the fetal heart to be measured, obtain an image to be segmented, and perform target segmentation processing on the target four-chamber heart ultrasound section image based on a preset target segmentation model to obtain atrial septum contour data, ventricular septum contour data, spine contour data, and chest cavity contour data;

[0025] Step 2: measuring all biomasses of all extracted target objects to be measured, and determining the measurement results of each target object to be measured according to all measured biomasses, inputting the image to be segmented into the trained segmentation model to obtain a segmentation result image; the segmentation result image includes segmented areas of various structures of the fetal heart, and performing axis fitting based on the atrial septum contour data and the ventricular septum contour data to obtain a first heart axis;

[0026] Step 3: The segmentation model that has been trained is obtained by training a preset segmentation model, and the measurement of all biomasses of all target objects to be measured and the determination of the measurement results of each target object to be measured based on all the measured biomasses specifically include: measuring all biomasses of all target objects to be measured and extracted;

[0027] Step 4: for each target object to be measured, determine the object type of the target object to be measured; screen the biomass of the target object to be measured from all biomasses according to the object type to obtain the measurement result of the target object to be measured; the object type of the target object to be measured is determined according to a preset neural network, and when the preset neural network fails, re-determine the measurement result of the target object to be measured based on all the acquired biomasses.

[0028] In a further embodiment of the present invention, after extracting all the target objects to be measured carried by the ultrasound image to be measured, the method further includes: obtaining the number of the extracted target objects to be measured; when the number is greater than or equal to 1, executing the step of measuring each target object to be measured; when the number is less than 1, indicating that the target object to be measured is not identified, performing axis fitting based on the spinal contour data and the thoracic contour data to obtain a second heart axis; and obtaining the heart axis according to the angle between the first heart axis and the second heart axis.

[0029] The present invention is further configured such that the size of each segmented area in the segmentation result image is measured respectively, including: determining the outer boundary information of each segmented area in the segmentation result image respectively; determining the minimum enclosing rectangle of each segmented area based on the outer boundary information; and measuring the minimum enclosing rectangle of each segmented area respectively to obtain the size of each segmented area.

[0030] Embodiment 2:

[0031] An automatic measurement method for prenatal ultrasound images comprises the following steps:

[0032] Step 1: Acquire an ultrasound image to be measured, extract all target objects to be measured carried by the ultrasound image to be measured, acquire an ultrasound four-chamber heart section image of the fetal heart to be measured, obtain an image to be segmented, and perform target segmentation processing on the target four-chamber heart ultrasound section image based on a preset target segmentation model to obtain atrial septum contour data, ventricular septum contour data, spine contour data, and chest cavity contour data;

[0033] Step 2: measuring all biomasses of all extracted target objects to be measured, and determining the measurement results of each target object to be measured according to all measured biomasses, inputting the image to be segmented into the trained segmentation model to obtain a segmentation result image; the segmentation result image includes segmented areas of various structures of the fetal heart, and performing axis fitting based on the atrial septum contour data and the ventricular septum contour data to obtain a first heart axis;

[0034] Step three: the segmentation model that has completed training is obtained by training a preset segmentation model, and the measurement and extraction of all biomasses of all target objects to be measured, and determining the measurement results of each target object to be measured based on all the measured biomasses specifically include: measuring and extracting all biomasses of all target objects to be measured; for each target object to be measured, determining the object type of the target object to be measured; screening the biomass of the target object to be measured from all biomasses based on the object type to obtain the measurement result of the target object to be measured; the object type of the target object to be measured is determined based on a preset neural network, and when the preset neural network fails, re-determining the measurement result of the target object to be measured based on all the acquired biomasses.

[0035] In a further embodiment of the present invention, the measuring and extracting of all biomasses of all target objects to be measured, and determining the measurement results of each target object to be measured based on all the measured biomasses, further comprises: for each target object to be measured, drawing its corresponding measurement line according to the updated measurement result of the target object to be measured, and drawing the measurement line in the ultrasonic image; the measurement result of the biomass includes the location information of the biomass and the structural information of the biomass; the obtaining of the ultrasonic image to be measured specifically comprises: obtaining a plurality of continuous ultrasonic image frames; screening the plurality of continuous ultrasonic image frames to obtain the ultrasonic image to be measured; and extracting all target objects to be measured carried by the ultrasonic image to be measured. Specifically, the method comprises: inputting the ultrasound image to be measured into a preset recognition model, wherein the preset recognition model is trained based on a training sample set, and each training sample in the training sample set comprises an ultrasound image and all target objects to be measured carried by the ultrasound image; outputting all target objects to be measured carried by the ultrasound image to be measured through the preset recognition model; and separately determining the outer boundary information of each segmented area in the segmentation result image, comprising: denoising the segmentation result image to obtain a denoised segmentation result image; performing edge detection on the denoised segmentation result image to obtain an edge detection image; and binarizing the edge detection image to obtain the outer boundary information of each segmented area.

[0036] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be listed as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for automatic measurement of prenatal ultrasound images, characterized in that: The following steps are involved: Step 1: Acquire an ultrasound image to be measured, extract all target objects to be measured carried by the ultrasound image to be measured, acquire an ultrasound four-chamber heart section image of the fetal heart to be measured, obtain an image to be segmented, and perform target segmentation processing on the target four-chamber heart ultrasound section image based on a preset target segmentation model to obtain atrial septum contour data, ventricular septum contour data, spine contour data, and chest cavity contour data; Step 2: measuring all biomasses of all extracted target objects to be measured, and determining the measurement results of each target object to be measured according to all measured biomasses, inputting the image to be segmented into the trained segmentation model to obtain a segmentation result image; the segmentation result image includes segmented areas of various structures of the fetal heart, and performing axis fitting based on the atrial septum contour data and the ventricular septum contour data to obtain a first heart axis; Step 3: The segmentation model that has been trained is obtained by training a preset segmentation model, and the measurement of all biomasses of all target objects to be measured and the determination of the measurement results of each target object to be measured based on all the measured biomasses specifically include: measuring all biomasses of all target objects to be measured and extracted; Step 4: for each target object to be measured, determine the object type of the target object to be measured; screen the biomass of the target object to be measured from all biomasses according to the object type to obtain the measurement result of the target object to be measured; the object type of the target object to be measured is determined according to a preset neural network, and when the preset neural network fails, re-determine the measurement result of the target object to be measured based on all the acquired biomasses.

2. The method for automatically measuring prenatal ultrasound images according to claim 1, characterized in that: After extracting all the target objects to be measured carried by the ultrasound image to be measured, the method also includes: obtaining the number of the extracted target objects to be measured; when the number is greater than or equal to 1, executing the step of measuring each target object to be measured; when the number is less than 1, indicating that the target object to be measured is not identified, performing axis fitting based on the spinal contour data and the thoracic contour data to obtain a second heart axis; and obtaining the heart axis according to the angle between the first heart axis and the second heart axis.

3. The method for automatically measuring prenatal ultrasound images according to claim 1, characterized in that: The measuring of the size of each segmented area in the segmentation result image respectively includes: determining the outer boundary information of each segmented area in the segmentation result image respectively; determining the minimum circumscribed rectangle of each segmented area based on the outer boundary information; and measuring the minimum circumscribed rectangle of each segmented area respectively to obtain the size of each segmented area.

4. The method for automatically measuring prenatal ultrasound images according to claim 1, characterized in that: After measuring and extracting all biomasses of all target objects to be measured and determining the measurement results of each target object to be measured based on all the measured biomasses, the method further includes: for each target object to be measured, drawing a corresponding measurement line according to the updated measurement result of the target object to be measured, and drawing the measurement line in the ultrasonic image.

5. The method for automatically measuring prenatal ultrasound images according to claim 1, characterized in that: The measurement result of the biomass includes the position information of the biomass and the structure information of the biomass.

6. The method for automatically measuring prenatal ultrasound images according to claim 5, characterized in that: The acquiring of the ultrasonic image to be measured specifically includes: acquiring a number of continuous ultrasonic image frames; and screening the number of continuous ultrasonic image frames to obtain the ultrasonic image to be measured.

7. The method for automatically measuring prenatal ultrasound images according to claim 6, characterized in that: The extracting of all target objects to be measured carried by the ultrasound image to be measured specifically includes: inputting the ultrasound image to be measured into a preset recognition model, wherein the preset recognition model is trained based on a training sample set, and each training sample in the training sample set includes an ultrasound image and all target objects to be measured carried by the ultrasound image; and outputting all target objects to be measured carried by the ultrasound image to be measured through the preset recognition model.

8. The method for automatically measuring prenatal ultrasound images according to claim 3, characterized in that: The method of separately determining the outer boundary information of each segmented area in the segmentation result image includes: denoising the segmentation result image to obtain a denoised segmentation result image; performing edge detection on the denoised segmentation result image to obtain an edge detection image; and binarizing the edge detection image to obtain the outer boundary information of each segmented area.

Images