CN112120735A - Ultrasound imaging method and device, and storage medium - Google Patents

Abstract

The embodiment of the invention discloses an ultrasonic imaging method, which comprises the following steps: acquiring three-dimensional data of a spine; identifying a spine key anatomical structure from spine three-dimensional volume data; optimizing spine three-dimensional volume data based on a spine key anatomical structure to obtain a spine enhanced image, wherein the spine enhanced image comprises a three-dimensional enhanced image and/or a profile enhanced image; and displaying the spine enhanced image.

Description

Ultrasound imaging method and device, storage medium

technical field

The invention relates to the technical field of ultrasonic imaging, and in particular, to an ultrasonic imaging method and equipment, and a storage medium.

Background technique

Ultrasound examination is to transmit ultrasonic waves to the human body, and obtain images of internal organs by receiving and processing the reflected signals. Because of its advantages of safety, convenience, non-radiation, and low cost, ultrasonography has become the main auxiliary means of clinical diagnosis.

At present, spine examination is an important one in clinical ultrasound examination items for fetuses. The corresponding three-dimensional volume data can be obtained by scanning the fetal spine with ultrasound imaging equipment. Based on the volume data, not only any section of the volume data can be displayed, but also a virtual reality (VR) image can be rendered by rendering, which can be used as a spine examination. in accordance with.

Ultrasound imaging results are usually affected by multiple imaging parameters. For example, when the acquisition angle deviates from the median position of the spine, the signal-to-noise ratio of the spine area will decrease significantly, and the spine structure will not be clearly displayed; when the maternal fat is thick or the fetal position is deep. The contrast between the fetal spine and the background area is not obvious enough, the spine boundary is unclear, and the grayscale information is not rich enough. Usually, doctors can adjust imaging parameters based on actual experience. However, not only is the imaging efficiency low, but also depends on subjective experience, which cannot guarantee a better imaging effect.

SUMMARY OF THE INVENTION

In order to solve the existing technical problems, the embodiments of the present application are expected to provide an ultrasonic imaging method, equipment, and storage medium, by identifying the key anatomical structures of the spine in the three-dimensional volume data of the spine, to assist the spine imaging, not only the imaging efficiency is high. , and improve the imaging effect.

In order to achieve the above purpose, the technical solutions of the embodiments of the present application are implemented as follows:

The embodiment of the present application provides an ultrasonic imaging method, the method includes:

Obtain spine 3D volume data;

from the three-dimensional volume data of the spine, identifying key anatomical structures of the spine;

Performing optimization processing on the three-dimensional volume data of the spine based on the key anatomical structure of the spine to obtain a spine-enhanced image, wherein the spine-enhanced image includes a stereo-enhanced image and/or a section-enhanced image;

The spine-enhanced image is displayed.

In one embodiment, the key anatomical structures of the spine are identified from the three-dimensional volume data of the spine, including:

Obtain marking instructions and/or preset structure detection methods;

According to the marking instruction and/or the preset structure detection method, the key anatomical structure of the spine is identified from the three-dimensional volume data of the spine.

In one embodiment, identifying the key anatomical structure of the spine from the three-dimensional volume data of the spine according to the marking instruction includes:

marking part of the volume data in the three-dimensional volume data of the spine according to the instruction of the marking instruction;

A structure surrounded by the partial volume data is determined as the spine key anatomical structure.

In one embodiment, the preset structure detection method includes a feature detection method, and the key anatomical structure of the spine is identified from the spine three-dimensional volume data according to the preset structure detection method, including:

According to the feature detection method, binarization segmentation and morphological processing are performed on the spine three-dimensional volume data to obtain a plurality of candidate structures;

determining a probability that each of the multiple candidate structures is a spine structure based on the spine structure feature, obtaining multiple probabilities, and determining a maximum probability from the multiple probabilities;

Among the plurality of candidate structures, the structure corresponding to the maximum probability is determined as the key anatomical structure of the spine.

In one embodiment, the preset structure detection method includes a machine learning or deep learning detection method, and the key anatomical structure of the spine is identified from the spine three-dimensional volume data according to the preset structure detection method, including: :

Build a database of key anatomical structures of the spine according to the machine learning or deep learning detection method;

Perform model training based on the database of key anatomical structures of the spine to obtain a recognition model of key anatomical structures of the spine;

The key anatomical structures of the spine are identified from the three-dimensional volume data of the spine using the key anatomical structure identification model of the spine.

In one embodiment, the optimized processing of the three-dimensional volume data of the spine based on the key anatomical structure of the spine to obtain an enhanced spine image includes:

Based on the key anatomical structure of the spine, grayscale adjustment is performed on the three-dimensional volume data of the spine to obtain target three-dimensional volume data;

The spine-enhanced image is obtained according to the target three-dimensional volume data.

In one embodiment, performing grayscale adjustment on the three-dimensional volume data of the spine based on the key anatomical structure of the spine to obtain target three-dimensional volume data, including:

From the three-dimensional volume data of the spine, obtain the first volume data included in the key anatomical structure of the spine;

increasing the gray value of the first volume data to obtain the target three-dimensional volume data;

and / or,

From the three-dimensional volume data of the spine, obtain second volume data different from the first volume data included in the key anatomical structure of the spine;

The grayscale value of the second volume data is reduced to obtain the target three-dimensional volume data.

In one embodiment, the optimized processing of the three-dimensional volume data of the spine based on the key anatomical structure of the spine to obtain an enhanced spine image includes:

determining imaging parameters based on the key anatomical structures of the spine;

The spine three-dimensional volume data is optimized according to the determined imaging parameters to obtain the spine enhancement image.

In one embodiment, the imaging parameter includes a target gain, and the determining the imaging parameter based on the key anatomical structure of the spine includes:

From the three-dimensional volume data of the spine, obtain the first volume data included in the key anatomical structure of the spine;

Counting the grayscale values of the first volume data to obtain a grayscale statistical result of the first volume data;

The target gain is determined according to the grayscale statistical result.

In one embodiment, the imaging parameters include gradient threshold parameters, and the determining of the imaging parameters based on the key anatomical structures of the spine includes:

From the three-dimensional volume data of the spine, obtain the first volume data included in the key anatomical structure of the spine;

Determine the gradient value corresponding to the key anatomical structure of the spine based on the gray value of the first volume data;

According to the gradient value, a first gradient threshold corresponding to the first volume data is determined according to a preset gradient threshold calculation method;

obtaining a second gradient threshold corresponding to the second volume data outside the key anatomical structure of the spine in the three-dimensional volume data of the spine;

The first gradient threshold and the second gradient threshold are determined as the gradient threshold parameters.

In one embodiment, the imaging parameters include profile thickness parameters, and the imaging parameters are determined based on the key anatomical structures of the spine, including:

From the three-dimensional volume data of the spine, obtain the position information of the key anatomical structures of the spine;

determining spine thickness information based on the position information;

The profile thickness parameter is determined based on the spine thickness information.

In one embodiment, obtaining the position information of the key anatomical structures of the spine from the three-dimensional volume data of the spine includes:

From the three-dimensional volume data of the spine, obtain boundary volume data corresponding to the boundaries of the key anatomical structures of the spine;

The bounding volume data is determined as the position information.

The embodiment of the present application provides an ultrasonic imaging method, the method includes:

Obtain two-dimensional slice data of the spine;

Identifying key anatomical regions of the spine from the two-dimensional section data of the spine;

Optimizing the two-dimensional section data of the spine based on the key anatomical regions of the spine to obtain an enhanced spine image;

The spine-enhanced image is displayed.

In one embodiment, the acquiring two-dimensional section data of the spine includes:

Obtain spine 3D volume data;

According to a preset slice selection method or a received selection instruction, the spine two-dimensional slice data is determined from the spine three-dimensional volume data.

In one embodiment, the optimized processing of the two-dimensional section data of the spine based on the key anatomical regions of the spine to obtain an enhanced spine image includes:

Based on the key anatomical regions of the spine, grayscale adjustment is performed on the two-dimensional section data of the spine to obtain the spine-enhanced image.

In one embodiment, performing grayscale adjustment on the two-dimensional section data of the spine based on the key anatomical regions of the spine to obtain the spine-enhanced image, including:

From the two-dimensional section data of the spine, obtain the first section data included in the key anatomical region of the spine;

increasing the gray value of the first section data to obtain the spine-enhanced image;

and / or,

From the two-dimensional section data of the spine, obtain second section data that is different from the first section data included in the key anatomical structure of the spine;

The grayscale value of the second slice data is reduced to obtain the spine-enhanced image.

In one embodiment, the optimized processing of the two-dimensional section data of the spine based on the key anatomical regions of the spine to obtain an enhanced spine image includes:

determining imaging parameters based on the critical anatomical region of the spine;

The spine two-dimensional slice data is optimized according to the determined imaging parameters to obtain the spine enhancement image.

An embodiment of the present application provides an ultrasonic imaging device, and the ultrasonic imaging device includes:

probe;

Transmit/receive selector switch;

a transmitting circuit for exciting the probe to transmit ultrasonic waves to the measured spine;

a receiving circuit, configured to receive the ultrasonic echo returned from the measured spine through the probe to obtain an ultrasonic echo signal;

a beamformer, configured to perform beamformation processing on the ultrasonic echo signal to obtain a beamformed ultrasonic echo signal;

a signal processor, configured to perform signal processing on the beam-synthesized ultrasonic echo signals to obtain processed ultrasonic echo signals;

an imaging processor for: acquiring three-dimensional volume data of the spine according to the processed ultrasonic echo signals; identifying key anatomical structures of the spine from the three-dimensional volume data of the spine; The three-dimensional volume data is optimized to obtain a spine-enhanced image, wherein the spine-enhanced image includes a stereo-enhanced image and/or a section-enhanced image;

a display for displaying the spine-enhanced image.

In one embodiment, the imaging processor is configured to acquire a marking instruction and/or a preset structure detection method, and identify the spine 3D volume data from the spine three-dimensional volume data according to the marking instruction and/or the preset structure detection method. The spine key anatomy.

In one embodiment, the imaging processor is configured to: mark part of the volume data in the three-dimensional volume data of the spine according to the instruction of the marking instruction; and determine the structure surrounded by the part of the volume data as the key of the spine. Anatomy.

In one embodiment, the preset structure detection method includes a feature detection method, and the imaging processor is configured to: perform binarization segmentation and morphological processing on the spine three-dimensional volume data according to the feature detection method, so as to obtain multi-dimensional data. a plurality of candidate structures; based on the features of the spine structure, determine the probability that each of the multiple candidate structures is a spine structure, obtain multiple probabilities, and determine the maximum probability from the multiple probabilities; Among the structures, the structure corresponding to the maximum probability is determined as the key anatomical structure of the spine.

In one embodiment, the preset structure detection method includes a machine learning or deep learning detection method, and the imaging processor is configured to: construct a database of key anatomical structures of the spine according to the machine learning or deep learning detection method; The key anatomical structure database is used for model training to obtain a spinal key anatomical structure identification model; the spinal key anatomical structure is identified from the spinal three-dimensional volume data by using the spinal key anatomical structure identification model.

In one embodiment, the imaging processor is configured to: perform grayscale adjustment on the three-dimensional volume data of the spine based on the key anatomical structure of the spine to obtain target three-dimensional volume data; and obtain the spine enhancement according to the target three-dimensional volume data. image.

In one embodiment, the imaging processor is configured to: obtain the first volume data included in the key anatomical structure of the spine from the three-dimensional volume data of the spine; increase the gray value of the first volume data to obtain the target three-dimensional volume data; and/or, from the spine three-dimensional volume data, obtain second volume data that is different from the first volume data included in the key anatomical structures of the spine; reduce the grayscale of the second volume data value to obtain the target 3D volume data.

In one embodiment, the imaging processor is configured to: determine imaging parameters based on the key anatomical structures of the spine; optimize the three-dimensional volume data of the spine according to the determined imaging parameters to obtain the spine-enhanced image.

In one embodiment, the imaging parameter includes a target gain, and the imaging processor is configured to: obtain first volume data included in the key anatomical structure of the spine from the three-dimensional volume data of the spine; and count the first volume data. The grayscale value of the first volume data is obtained, and the grayscale statistical result of the first volume data is obtained; the target gain is determined according to the grayscale statistical result.

In one embodiment, the imaging parameters include gradient threshold parameters, and the imaging processor is configured to: obtain first volume data included in the key anatomical structures of the spine from the three-dimensional volume data of the spine; The gray value of the data is used to determine the gradient value corresponding to the boundary of the key anatomical structure of the spine; according to the gradient value, the first gradient threshold corresponding to the first volume data is determined according to the preset gradient threshold calculation method; In the three-dimensional volume data of the spine, the second gradient threshold corresponding to the second volume data outside the key anatomical structure of the spine; the first gradient threshold and the second gradient threshold are determined as the gradient threshold parameters.

In one embodiment, the imaging parameters include profile thickness parameters, and the imaging processor is configured to: obtain position information of the key anatomical structures of the spine from the spine three-dimensional volume data; determine spine thickness information based on the position information. ; Determine the profile thickness parameter according to the spine thickness information.

In one embodiment, the imaging processor is configured to: obtain boundary volume data corresponding to the boundary of the key anatomical structure of the spine from the spine three-dimensional volume data; and determine the boundary volume data as the position information.

An embodiment of the present application provides an ultrasonic imaging device, and the ultrasonic imaging device includes:

probe;

Transmit/receive selector switch;

a transmitting circuit for exciting the probe to transmit ultrasonic waves to the measured spine;

a receiving circuit, configured to receive the ultrasonic echo returned from the measured spine through the probe to obtain an ultrasonic echo signal;

a beamformer, configured to perform beamformation processing on the ultrasonic echo signal to obtain a beamformed ultrasonic echo signal;

a signal processor, configured to perform signal processing on the beam-synthesized ultrasonic echo signals to obtain processed ultrasonic echo signals;

An imaging processor, configured to: acquire two-dimensional section data of the spine based on the processed ultrasonic echo signals; identify key anatomical regions of the spine from the two-dimensional section data of the spine; The two-dimensional section data of the spine is optimized to obtain the spine-enhanced image;

a display for displaying the spine-enhanced image.

In one embodiment, the imaging processor is configured to: acquire three-dimensional volume data of the spine according to the processed ultrasonic echo signals; Two-dimensional slice data of the spine.

In one embodiment, the imaging processor is configured to: perform grayscale adjustment on the two-dimensional section data of the spine based on the key anatomical regions of the spine to obtain the spine-enhanced image.

In one embodiment, the imaging processor is specifically configured to obtain, from the two-dimensional data of the spine, a first section included in the key anatomical region of the spine, and a second section outside the key anatomical region of the spine. Slice data; increase the gray value of the first slice data to obtain the spine enhancement image; and/or, from the spine two-dimensional slice data, obtain the first everything related to the key anatomical structure of the spine second slice data with different slice data; reduce the gray value of the second slice data to obtain the spine-enhanced image.

In one embodiment, the imaging processor is specifically configured to determine imaging parameters based on the key anatomical regions of the spine; optimize the two-dimensional section data of the spine according to the determined imaging parameters to obtain the spine-enhanced image.

An embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores an ultrasound imaging program, and the ultrasound imaging program can be executed by a processor to implement the foregoing ultrasound imaging method.

Embodiments of the present invention provide an ultrasonic imaging method and device, the method includes: acquiring three-dimensional volume data of the spine; identifying key anatomical structures of the spine from the three-dimensional volume data of the spine; optimizing the three-dimensional volume data of the spine based on the key anatomical structures of the spine processing, to obtain a spine-enhanced image, wherein the spine-enhanced image includes a stereo-enhanced image and/or a section-enhanced image; and the spine-enhanced image is displayed. The technical solutions provided by the embodiments of the present invention assist spine imaging by identifying key anatomical structures of the spine in three-dimensional volume data of the spine, which not only has high imaging efficiency, but also improves imaging effects.

Description of drawings

FIG. 1 is a schematic structural block diagram of an ultrasonic imaging device in an embodiment of the present invention;

FIG. 2 is a schematic flowchart 1 of an ultrasonic imaging method provided by an embodiment of the present invention;

3 is an exemplary conventional spine image provided by an embodiment of the present invention;

FIG. 4 is an exemplary spine enhancement image provided by an embodiment of the present invention;

FIG. 5 is a second schematic flowchart of an ultrasonic imaging method according to an embodiment of the present invention.

Detailed ways

In order to understand the features and technical contents of the embodiments of the present invention in more detail, the implementation of the embodiments of the present invention will be described in detail below with reference to the accompanying drawings. The accompanying drawings are for reference only and are not used to limit the embodiments of the present invention.

FIG. 1 is a schematic structural block diagram of an ultrasonic imaging device in an embodiment of the present invention. The ultrasound imaging apparatus 10 may include a probe 100 , a transmit/receive selection switch 101 , a transmit circuit 102 , a receive circuit 103 , a beamforming circuit 104 , a signal processor 105 , an imaging processor 106 and a display 107 . The transmitting circuit 102 can excite the probe 100 to transmit ultrasonic waves to the measured spine; the receiving circuit 103 can receive the ultrasonic echoes returned from the measured spine through the probe 100 to obtain ultrasonic echo signals; the ultrasonic echo signals are processed by the beam forming circuit 104. After beamforming processing, it is sent to the signal processor 105 . The signal processor 105 performs signal processing on the ultrasonic echo signal to obtain a processed ultrasonic echo signal. Imaging processor 106 generates spine enhancement images and/or spine slice images based on the processed ultrasound echo signals. These ultrasound spine-enhanced images can be displayed on display 107 .

In the embodiment of the present invention, the display 107 of the ultrasound imaging device 10 may be a touch display screen, a liquid crystal display screen, etc., or may be an independent display device such as a liquid crystal display, a television set, etc. independent of the ultrasound imaging device 10 , or Displays on electronic devices such as mobile phones, tablet computers, etc.

In this embodiment of the present invention, the signal processor 105 and the imaging processor 106 may be integrated into one processor, or may be implemented by two or more processors. The processors (processor, signal processor 105 and/or imaging processor 106, etc.) mentioned here may be application specific integrated circuits (ASICs), digital signal processors (Digital 100 Signal Processors, DSPs) ), Digital Signal Processing Device (DSPD), Programmable Logic Device (Programmable Logic Device, PLD), Field Programmable Gate Array (Field Programmable Gate Array, FPGA), Central Processing Unit (Central Processing Unit, CPU) , at least one of a controller, a microcontroller, and a microprocessor, so that the processor can execute the corresponding steps of the ultrasonic imaging method in each embodiment of the present invention.

In an embodiment of the present invention, the ultrasound imaging apparatus 10 may further include a memory, and the acquired spine-enhanced images may be stored in the memory. The memory may be a volatile memory (volatile memory), such as a random access memory (Random Access Memory, RAM); or a non-volatile memory (non-volatile memory), such as a read-only memory (Read Only Memory, ROM), Flash memory (flash memory), hard disk (Hard Disk Drive, HDD) or solid-state drive (Solid-State Drive, SSD); or a combination of the above types of memory, and provide instructions and data to the processor.

The technical solution of the present invention will be described in detail below based on the above-mentioned ultrasonic imaging device 10 .

Embodiments of the present invention provide an ultrasonic imaging method. FIG. 2 is a schematic flowchart 1 of an ultrasound imaging method according to an embodiment of the present invention. As shown in Figure 2, it mainly includes the following steps:

S201. Acquire three-dimensional volume data of the spine.

In the embodiment of the present invention, the imaging processor 106 of the ultrasonic imaging device 10 may receive the processed ultrasonic echo signal obtained after the signal processing by the signal processor 105, so as to determine the three-dimensional volume data of the spine according to the ultrasonic echo signal, That is, the three-dimensional volume data of the spine is obtained.

Specifically, in the embodiment of the present invention, in the ultrasonic imaging device 10, the transmitting circuit 102 excites the probe 100 to transmit ultrasonic waves to the spine under test, so that the receiving circuit 103 receives the ultrasonic echoes returned from the spine under test through the probe 100, so as to The ultrasonic echo signal is obtained. After that, the beamformer 104 performs beamformation processing on the ultrasonic echo signal to obtain the ultrasonic echo signal after the beamformation. The signal processor 105 performs signal processing on the ultrasonic echo signal after the beamformation to obtain The processed ultrasound echo signals, and finally, the imaging processor 106 acquires three-dimensional volume data of the spine according to the processed ultrasound echo signals. The three-dimensional volume data of the spine is volume data corresponding to the tested spine, and the specific tested spine is not limited in this embodiment of the present invention.

S202. Identify the key anatomical structures of the spine from the three-dimensional volume data of the spine.

In the embodiment of the present invention, after acquiring the three-dimensional volume data of the spine, the imaging processor 106 of the ultrasound imaging device 10 identifies the key anatomical structures of the spine from the three-dimensional volume data of the spine.

It should be noted that, in the embodiment of the present invention, the key anatomical structures of the spine may include the vertebral arch, the vertebral body, the spinal cord vertebra, and the area included in the vertebral arch, vertebral body, etc. in a specific section in the three-dimensional volume data of the spine . For example, in the sagittal plane of the fetal spine in the three-dimensional volume data of the spine, the key anatomical structure of the spine may be the tissue structure surrounded by the vertebral arch and the vertebral body on this plane. The specific key anatomical structure of the spine may be determined according to actual imaging requirements, which is not limited in the embodiment of the present invention.

It should be noted that, in the embodiment of the present invention, the imaging processor 106 may realize the identification of the key anatomical structures of the spine from the three-dimensional volume data of the spine in a manual, automatic, or semi-automatic manner.

In the embodiment of the present invention, the imaging processor 106 identifies the key anatomical structures of the spine from the three-dimensional volume data of the spine, including: obtaining a marking instruction and/or a method for detecting a preset structure; method to identify key anatomical structures of the spine from 3D volume data of the spine.

Specifically, in the embodiment of the present invention, the imaging processor 106 identifies the key anatomical structures of the spine from the three-dimensional volume data of the spine according to the labeling instruction, including: according to the instruction of the labeling instruction, for some volume data in the three-dimensional volume data of the spine Labeling; structures surrounded by partial volume data are identified as critical anatomical structures of the spine.

It should be noted that, in the embodiment of the present invention, the user can use a keyboard, a mouse and other tools to mark the key anatomical structures of the spine for the three-dimensional volume data of the spine with a certain workflow, that is, send a marking instruction to the imaging processor 106, Therefore, the imaging processor 106 delineates the part of the volume data indicated by the user from the three-dimensional volume data of the spine according to the marking instruction, so as to determine the key anatomical structure of the spine. The method of identifying the key anatomical structures of the spine by the marking instruction is actually a manual method, and the specific form of providing the marking instruction is not limited in the embodiment of the present invention.

It should be noted that, in the embodiment of the present invention, the partial volume data may be some volume data on the boundary of the key anatomical structures of the spine, or may be all the volume data corresponding to the key structures of the spine. The specific part of the volume data is not limited in the embodiment of the present invention.

Exemplarily, in the embodiment of the present invention, the user can select the volume data corresponding to the key anatomical structures of the spine by using a box, a circle, an ellipse, and an irregular polygon with a tool such as a keyboard, a mouse, and the like for the three-dimensional volume data of the spine. Several volumes are selected from the three-dimensional volume data of the spine, and these partial volume data are used to represent the positions of the key anatomical structures of the spine.

It should be noted that, in the embodiment of the present invention, a preset structure detection method may be set in the ultrasonic imaging device 10, and the imaging processor 106 can identify the key anatomical structures of the spine from the three-dimensional volume data of the spine according to the detection method, The detection principle of the preset structure detection method is actually to extract the relevant features of the key anatomical structures from the three-dimensional volume data of the spine, so as to use the relevant features to determine the key structures. The preset structure detection method may be a feature detection method based on traditional grayscale and/or morphology, or a detection method based on machine learning or deep learning. Of course, it may also be other types of detection methods. The specific preset structure Detection method is not limited in the embodiment of the present invention.

Specifically, in the embodiment of the present invention, the preset structure detection method includes a feature detection method, and the imaging processor 106 identifies the key anatomical structure of the spine from the three-dimensional volume data of the spine according to the preset structure detection method, including: detecting according to the feature The method is to perform binary segmentation and morphological processing on the spine three-dimensional volume data to obtain multiple candidate structures; based on the spine structure features, determine the probability that each of the multiple candidate structures is a spine structure, obtain multiple probabilities, and determine the probability that each of the multiple candidate structures is a spine structure. The maximum probability is determined from multiple probabilities; among multiple candidate structures, the structure corresponding to the maximum probability is determined as the key anatomical structure of the spine.

It can be understood that, in the embodiment of the present invention, since the echoes of the vertebral arches and the vertebral bodies in the spine are stronger and the echoes of the surrounding tissues are weak after the ultrasonic waves are sent to the tested spine, there are obvious differences between the two. It can be reflected in the grayscale values corresponding to different volume data. Therefore, the preset structure detection method can be selected based on traditional grayscale and/or morphological feature detection methods to realize the detection of the key anatomical structures of the spine. Specifically, based on the traditional grayscale and/or morphological feature detection methods, the imaging processor 106 can first perform binary segmentation on the three-dimensional volume data of the spine, perform some routine and necessary morphological operations to obtain multiple candidate structures, and then , and sequentially determine the probability of each candidate structure as the key anatomical structure of the spine according to the shape and other characteristics, and finally, determine the candidate structure with the highest probability as the key anatomical structure of the spine. Of course, the preset structure detection method may also be other traditional grayscale detection and segmentation methods, such as Otsu threshold, level set, graph cut, and snake algorithm.

Specifically, in the embodiment of the present invention, the preset structure detection method includes a machine learning or deep learning detection method, and the imaging processor 106 identifies the key anatomical structures of the spine from the spine three-dimensional volume data according to the preset structure detection method, including : Build a database of key anatomical structures of the spine according to machine learning or deep learning detection methods; perform model training based on the database of key anatomical structures of the spine to obtain a recognition model of key anatomical structures of the spine; use the recognition model of key anatomical structures of the spine to identify the 3D volume data of the spine key anatomical structures of the spine.

It can be understood that, in the embodiment of the present invention, the preset structure detection method may be a detection method based on machine learning or deep learning, and the method needs to first obtain the features or According to the learned features or rules, the acquired three-dimensional volume data of the spine is automatically identified.

It should be noted that, in the embodiment of the present invention, a database may be constructed in the ultrasound imaging device 10 first, and the database includes a plurality of three-dimensional volume data of the spine, and the key anatomical structures of the spine corresponding to each three-dimensional volume data of the spine The calibration result can be set according to actual needs, or it can be a frame of the key anatomical structures of the spine marked by the user in the three-dimensional volume data of the spine, or it can be a mask for accurate segmentation of the key anatomical structures of the spine, etc. After the database is constructed, a machine learning algorithm or a deep learning algorithm can be designed to learn the characteristics or laws of the key anatomical structures and non-spine key anatomical structures in the database, so as to identify the key spinal anatomical structures from the 3D volume data of the spine. The specific implementation method Including but not limited to the following methods.

Exemplarily, in the embodiment of the present invention, the imaging processor 106 may use a traditional sliding window-based method to realize the recognition. For the three-dimensional volume data of the spine, a sliding window may be inserted to perform traversal recognition in sequence. For the structure in the current sliding window Feature extraction, feature extraction methods can be Principal Components Analysis (PCA), Linear Discriminate Analysis (LDA), Haar (Haar) features, and texture features, etc., or a deep neural network , after that, the extracted features are matched with the database, and the discriminators such as the nearest neighbor algorithm, support vector machine, random forest, and neural network are used for classification to determine whether the structure in the current sliding window is the key anatomical structure of the spine, of course, It is also possible to simultaneously obtain which type of key anatomical structures of the spine the structure belongs to, such as the spine.

Exemplarily, in the embodiment of the present invention, the imaging processor 106 may use a deep learning-based bounding box method to realize recognition, form a specific network by stacking the base layer volume base layer and the fully connected layer, and perform feature learning and analysis on the constructed data block. For the regression of parameters, for the spine 3D volume data, a structural frame can be directly returned through a specific network, and the structure within the frame is the key anatomical structure of the spine. The specific network can be various convolutional neural networks, such as R-CNN, Fast R-CNN, Faster-RCNN, SSD, and YOLO, etc.

Exemplarily, in the embodiment of the present invention, the imaging processor 106 may also use an end-to-end semantic segmentation network method based on deep learning, which is similar to the structure of the specific network of the above-mentioned deep learning-based bounding box method. Layer removal, adding an upsampling layer or a deconvolution layer to make the input and output sizes the same, can directly obtain the key anatomical structures of the spine in the spine 3D volume data.

Exemplarily, in the embodiment of the present invention, the imaging processor 106 may further locate the key anatomical structures of the spine according to the above three methods, and use an additional classifier to perform classification and judgment to determine that the specific key anatomical structure of the spine is the vertebral arch. or vertebral bodies, etc.

It should be noted that, in the embodiment of the present invention, one or more of the above methods may be combined to automatically identify the key anatomical structure of the spine in the three-dimensional volume data of the spine, which is not limited in the embodiment of the present invention.

It should be noted that, in the embodiment of the present invention, the imaging processor 106 may also identify the key anatomical structures of the spine in the three-dimensional volume data of the spine in a semi-automatic manner. After obtaining a larger candidate structure, a preset structure detection method is used to further accurately identify the marked larger candidate structure to obtain the key anatomical structure of the spine, which is not limited in the embodiment of the present invention.

S203 , performing optimization processing on the three-dimensional volume data of the spine based on the key anatomical structure of the spine to obtain a spine-enhanced image; the spine-enhanced image includes a stereo-enhanced image and/or a section-enhanced image.

In the embodiment of the present invention, after obtaining the key anatomical structure of the spine and the three-dimensional volume data of the spine, the imaging processor 106 of the ultrasound imaging device 10 can perform optimization processing on the three-dimensional volume data of the spine based on the key anatomical structure of the spine, so as to obtain the spine Enhance images.

It should be noted that, in the embodiments of the present invention, the spine-enhanced images include stereo-enhanced images and/or cross-section enhanced images, that is, three-dimensional spine-enhanced imaging may be performed for all or a part of the spine three-dimensional volume data, or three-dimensional spine images may be enhanced. Section enhancement imaging is performed for a certain section data in the volume data, and the specific spine enhancement image is not limited in the embodiment of the present invention.

In the embodiment of the present invention, the imaging processor 106 performs optimization processing on the three-dimensional volume data of the spine based on the key anatomical structure of the spine to obtain an enhanced image of the spine. The target 3D volume data; the spine enhancement image is obtained according to the target 3D volume data.

Specifically, in the embodiment of the present invention, the imaging processor 106 performs spine imaging processing based on the key anatomical structure of the spine and the three-dimensional volume data of the spine to obtain an enhanced image of the spine, including: obtaining the key anatomy of the spine from the three-dimensional volume data of the spine The structure includes the first volume data and/or the second volume data that is not the first volume data, and increases the gray value of the first volume data and/or decreases the gray value of the second volume data, so as to obtain the target 3D volume data.

For example, in one embodiment, the imaging processor 106 may obtain, from the three-dimensional volume data of the spine, the first volume data included in the key anatomical structures of the spine, and the second volume data that is not the first volume data; according to the preset grayscale adjustment method , increasing the grayscale value of the first volume data to obtain the first volume data after grayscale adjustment; and obtaining the spine enhancement image based on the grayscale adjusted first volume data and the second volume data.

It can be understood that, in the embodiment of the present invention, after the imaging processor 106 identifies the key anatomical structures of the spine from the three-dimensional volume data of the spine, it actually knows the positions of the key anatomical structures of the spine in the three-dimensional volume data of the spine, and The key anatomical structure of the spine is the part that the user is interested in. Correspondingly, the non-spine key anatomical structure does not need to be highlighted. Therefore, the imaging processor 106 can obtain the first volume data included in the key anatomical structure of the spine from the three-dimensional volume data of the spine. , in order to enhance its gray value, so as to optimize the spine imaging based on the grayscale-adjusted first volume data and second volume data. better.

It should be noted that, in the embodiment of the present invention, the imaging processor 106 performs imaging based on the grayscale-adjusted first volume data and the second volume data, which may include the imaging processor 106 directly performing the imaging based on the grayscale-adjusted first volume data Volume data and second volume data for spine imaging. In addition, the imaging processor 106 may also be provided with a preset grayscale reduction method, and the imaging processor 106 may reduce the grayscale value of the second volume data according to the preset grayscale reduction method, so as to obtain a grayscale adjusted grayscale value. The second volume data, that is, the imaging processor 106 can also perform imaging based on the grayscale-adjusted second volume data and the first volume data, and can also achieve that the part of the key anatomical structures of the spine in the spine-enhanced image is more prominent and the contrast is better. good effect. Of course, the grayscale value of the first volume data can also be increased and the grayscale value of the second volume data can be decreased to perform imaging. A specific grayscale adjustment manner is not limited in this embodiment of the present invention.

It should be noted that, in the embodiment of the present invention, the grayscale adjustment for the first volume data may be performed by using a corresponding preset adjustment value to adjust the grayscale value, and of course, it may also be performed by machine learning or deep learning. The method adjusts the gray value. For example, the imaging processor 106 can first collect a plurality of spine three-dimensional volume data, and identify the key anatomical structures of the spine from each spine three-dimensional volume data, so as to obtain the information included in each spine key anatomical structure. The volume data is used as a training set. After that, the user can manually adjust the gray value of these volume data to an appropriate range, and use the adjusted volume data as a validation set. Then the user can design a neural network independently, using the training set and validation set. The set of training is performed to obtain a trained neural network, and the trained neural network can be used to automatically adjust the gray value of the first volume data included in the key anatomical structures of the spine in the three-dimensional volume data of the spine.

In the embodiment of the present invention, the imaging processor 106 performs optimization processing of spine imaging based on the key anatomical structures of the spine and three-dimensional volume data of the spine to obtain an enhanced image of the spine, and may further include: determining imaging parameters based on the key anatomical structures of the spine; The three-dimensional volume data is optimized for spine imaging according to the imaging parameters to obtain spine-enhanced images.

It should be noted that, in the prior art, the imaging parameters are usually fixed values set by the user based on experience, that is, the imaging parameters used for different spine 3D volume data are generally the same, even if the user targets different spine 3D volume data. Manual parameter adjustment of volume data is also based on experience and self-adjustment, which cannot guarantee a better image effect, and is less efficient. In the embodiment of the present invention, the imaging processor 106 can adaptively determine the imaging parameters based on the key anatomical structures of the spine, which not only has high efficiency, but also determines the imaging parameters based on the key anatomical structures of the spine that the user is interested in, which can better match the imaging parameters. The effect of the obtained spine-enhanced image is also better.

It should be noted that, in the embodiment of the present invention, the imaging processor 106 determines imaging parameters based on the key anatomical structures of the spine, wherein the imaging parameters may be brightness, contrast, overall gain, time gain compensation, or may be a three-dimensional volume of the spine. The parameters used for filtering or smoothing the data, in addition, may also be parameters such as image thickness in cross-sectional image enhancement, and specific imaging parameters can be determined according to actual imaging requirements, which are not limited in the embodiment of the present invention.

It should be noted that there may be differences in the optimal gain required for different objects under test, different tissues under test, and under different imaging modes. In the prior art, the user is required to independently adjust the gain to an appropriate range, the adjustment process depends on the subjective experience of the human, and the efficiency is low, but in the embodiment of the present invention, it can be adjusted adaptively based on the key anatomical structure of the spine gain to the appropriate range.

Specifically, in the embodiment of the present invention, the imaging parameter includes a target gain, and the type of the target gain may actually include the above-mentioned overall gain and/or time gain compensation, and the imaging processor 106 determines the imaging parameter based on the key anatomical structure of the spine, The method includes: obtaining the first volume data included in the key anatomical structure of the spine from the three-dimensional volume data of the spine; counting the gray values of the first volume data to obtain the gray statistics results of the first volume data; adjusting the gain according to the gray statistics results, Get the target gain.

It can be understood that, in the embodiment of the present invention, since the key anatomical structures of the spine have been identified from the three-dimensional volume data of the spine, the imaging processor 106 can directly acquire the first volume data included in the key anatomical structures of the spine, wherein , different first volume data have respective gray values, and the imaging processor 106 can count the distribution of gray values, that is, obtain gray statistics, and adjust the gain according to the gray statistics to obtain the target gain. For example, if the grayscale statistics result is that 90% of the grayscale values are greater than the grayscale value A, and the gain adjustment method is that the grayscale values of more than 60% are greater than A, the preset gain will be reduced according to a specific value. Therefore, the imaging processor 106 That is, the gain is reduced to obtain the target gain.

It should be noted that, in the embodiment of the present invention, in the process of spine imaging, anisotropic filtering can be used to remove speckle noise from the three-dimensional volume data of the spine, while retaining the edges of the key anatomical structures of the spine, so as to obtain a spine with better quality To enhance the image, the gradient threshold parameter needs to be set in anisotropic filtering. When performing anisotropic filtering, the gray value of the volume data whose gradient is greater than the gradient parameter threshold in the volume data can be retained, and the gray value of the volume data can be smoothed according to a certain method. . However, the gradient threshold parameters required for anisotropic filtering are different for different test objects, different test tissues, and different spinal structures. Therefore, the gradient threshold parameters can be determined based on the key anatomical structures of the spine.

Specifically, in the embodiment of the present invention, the imaging parameters include gradient threshold parameters, and the imaging processor 106 determines the imaging parameters based on the key anatomical structures of the spine, including: from the three-dimensional volume data of the spine, obtaining a first volume included in the key anatomical structures of the spine data; based on the gray value of the first volume data, determine the gradient value corresponding to the boundary of the key anatomical structure of the spine; according to the gradient value, determine the first gradient threshold value corresponding to the first volume data according to the preset gradient threshold value calculation method; obtain the three-dimensional spine In the volume data, the second gradient threshold corresponding to the second volume data outside the key anatomical structure of the spine; the first gradient threshold and the second gradient threshold are determined as gradient threshold parameters.

It should be noted that, in the embodiment of the present invention, the key anatomical structure of the spine is a closed structure, and the gradient values corresponding to the boundary are actually multiple gradient values. The preset gradient threshold calculation method may be preset according to actual requirements, for example, a method such as calculating an average value, and the specific preset gradient threshold calculation method is not limited in this embodiment of the present invention.

It should be noted that, in the embodiment of the present invention, the user can set a larger gradient threshold for the second volume data according to actual needs, that is, the second gradient threshold, which is used to remove speckle noise and enhance the imaging of key anatomical structures of the spine Effect. The imaging processor 106 may directly acquire the second gradient threshold, and the specific second gradient threshold is not limited in this embodiment of the present invention.

It should be noted that, in the embodiment of the present invention, the imaging processor 106 may also use some volume data in the three-dimensional volume data of the spine to image a certain section, and increase the thickness of the section to obtain a section enhanced image.

Specifically, in the embodiment of the present invention, the imaging parameters include profile thickness parameters, and the imaging processor 106 determines the imaging parameters based on the key anatomical structures of the spine, including: obtaining position information of the key anatomical structures of the spine from the three-dimensional volume data of the spine; The position information determines the spine thickness information; the profile thickness parameter is determined according to the spine thickness information.

It should be noted that, in the embodiment of the present invention, the imaging processor 106 obtains the position information of the key anatomical structures of the spine from the three-dimensional volume data of the spine, including: obtaining the boundary correspondence of the key anatomical structures of the spine from the three-dimensional volume data of the spine The bounding volume data of ; determine the bounding volume data as location information.

It should be noted that, in the embodiment of the present invention, in the three-dimensional volume data of the spine, each individual data corresponds to the x, y, and z-axis coordinates of the space. Therefore, the imaging processor 106 obtains the boundary corresponding to the key anatomical structure of the spine. The thickness of the spine can be calculated according to the relevant coordinate information, and the thickness parameter of the profile is determined according to the thickness of the spine, that is, the thickness that needs to be superimposed on the profile to obtain the enhanced profile image. The imaging processor 106 determines the thickness of the spine according to the information. The section thickness parameter may follow a certain rule, or may be preset with a certain algorithm or a preset calculation method, which is not limited in the embodiment of the present invention.

It can be understood that, in the embodiment of the present invention, in the process of generating the stereo enhanced image based on the key anatomical structure of the spine and the three-dimensional volume data of the spine, the imaging processor 106 may actually include using a three-dimensional rendering algorithm such as ray tracing to render the stereoscopic image. Other conventional processing such as rendering will not be repeated here.

S204, displaying the spine-enhanced image.

In the embodiment of the present invention, after the imaging processor 106 of the ultrasound imaging apparatus 10 obtains the spine-enhanced image, the display 107 can display the spine-enhanced image.

FIG. 3 is an exemplary conventional spine image provided by an embodiment of the present invention. As shown in Figure 3, when the existing technology is used to generate a spine image, the key anatomical structures of the spine are relatively blurred, and the imaging effect is poor.

FIG. 4 is an exemplary spine enhancement image provided by an embodiment of the present invention. As shown in FIG. 4 , using the ultrasonic imaging method of the present invention for imaging, the contrast between the spine and the non-spine is obvious, the structure is more obvious, and the imaging effect is better.

An ultrasonic imaging method provided by an embodiment of the present invention obtains three-dimensional volume data of the spine; identifies key anatomical structures of the spine from the three-dimensional volume data of the spine; performs spinal imaging processing based on the key anatomical structures of the spine and the three-dimensional volume data of the spine to obtain the spine Enhanced images; spine enhanced images include stereo enhanced images and/or cross-sectional enhanced images; the spine enhanced images are displayed. The technical solutions provided by the embodiments of the present invention assist spine imaging by identifying key anatomical structures of the spine in three-dimensional volume data of the spine, which not only has high imaging efficiency, but also improves imaging effects.

The embodiment of the present invention also provides an ultrasonic imaging method for two-dimensional data. FIG. 5 is a second schematic flowchart of an ultrasonic imaging method according to an embodiment of the present invention. As shown in Figure 5, it mainly includes the following steps:

S501. Acquire two-dimensional section data of the spine.

In the embodiment of the present invention, the imaging processor 106 may first acquire the two-dimensional slice data of the spine, so as to perform subsequent ultrasound imaging processing.

Specifically, in the embodiment of the present invention, the imaging processor 106 acquires two-dimensional spine slice data, including: acquiring spine three-dimensional volume data; selecting from the spine three-dimensional volume data according to a preset slice selection method or a received selection instruction Two-dimensional slice data of the spine.

It should be noted that, in the embodiment of the present invention, the process of acquiring the three-dimensional volume data of the spine by the imaging processor 106 is exactly the same as the above-mentioned step S201 , that is, the process of acquiring the three-dimensional volume data of the spine through the ultrasonic imaging device 10 is the transmission circuit 102 , the ultrasonic probe 100 , the receiving The circuit 103, the beamformer 104, and the signal processor 105 obtain the processed ultrasound echo signals, and the imaging processor 106 can obtain three-dimensional volume data of the spine according to the processed ultrasound echo signals. Refer to step S201 for a specific manner of acquiring the three-dimensional volume data of the spine, which will not be repeated here.

It can be understood that, in the embodiment of the present invention, the user can directly send a selection instruction to the imaging processor 106 through a keyboard, a mouse and other tools, and the imaging processor 106 can directly select from the three-dimensional volume data of the spine according to the selection instruction. Two-dimensional slice data of the spine. The two-dimensional spine slice data is all data representing a certain spine slice in the spine three-dimensional volume data. The specific spine two-dimensional slice data can be determined according to actual imaging requirements, which is not limited in the embodiment of the present invention.

It can be understood that, in the embodiment of the present invention, the imaging processor 106 can also automatically select the two-dimensional slice data of the spine according to a preset slice selection method. For example, according to a certain data detection method, the imaging processor 106 can use the data of the plane containing the most specific type of data as the two-dimensional section data of the spine, and can also select all the data whose z-coordinate is a specific value, as 2D slice data of the spine. The specific preset section selection method can be customized according to actual needs, which is not limited in the embodiment of the present invention.

S502. Identify the key anatomical regions of the spine from the two-dimensional section data of the spine.

In the embodiment of the present invention, after acquiring the two-dimensional slice data of the spine, the imaging processor 106 can further identify the key anatomical regions of the spine from the two-dimensional slice data of the spine.

It should be noted that, in the embodiment of the present invention, the manner in which the imaging processor 106 identifies the key anatomical regions of the spine from the two-dimensional section data of the spine is the same as the above step S202. The imaging processor 106 can perform identification of key anatomical regions of the spine with marking instructions and/or a preset structure detection method, wherein the preset structure detection method can also be a feature detection method, and a machine learning or deep learning detection method, etc., and the above steps. The difference in S202 is only the difference in the recognition objects. The recognition in this step is aimed at the two-dimensional section data of the spine. However, the recognition process and principle are completely the same, and will not be repeated here.

S503 , performing optimization processing on the two-dimensional section data of the spine based on the key anatomical regions of the spine to obtain an enhanced spine image.

In the embodiment of the present invention, after identifying the key anatomical regions of the spine from the two-dimensional data of the spine, the imaging processor 106 can perform optimization processing on the two-dimensional data of the spine based on the key anatomical regions of the spine to obtain an enhanced image of the spine.

In the embodiment of the present invention, the imaging processor 106 performs optimization processing on the two-dimensional section data of the spine based on the key anatomical regions of the spine to obtain an enhanced image of the spine, including: performing grayscale adjustment on the two-dimensional section data of the spine based on the key anatomical regions of the spine , obtain the target two-dimensional slice data; optimize the spine imaging according to the target two-dimensional slice data, and obtain the spine enhancement image.

Specifically, in the embodiment of the present invention, the imaging processor 106 performs grayscale adjustment on the two-dimensional section data of the spine based on the key anatomical regions of the spine, and obtains the target two-dimensional section data, including: from the two-dimensional section data of the spine, obtaining The first section data included in the key anatomical region of the spine and/or the second section data outside the key anatomical region of the spine; increase the gray value of the first section data and/or the second section data to obtain the gray-scale adjusted first section data. Section data and/or second section data after grayscale adjustment; determine the first section data and/or second section data after grayscale adjustment as target two-dimensional section data.

It should be noted that, in the embodiment of the present invention, the imaging processor 106 may also only reduce the data of the second slice, similar to the manner in which the grayscale adjustment is performed on the three-dimensional volume data of the spine based on the key anatomical structure of the spine in the above step S203. The gray value of the first section data and the second section data after the grayscale adjustment are determined as the target two-dimensional section data, or, the gray value of the first section data is increased, and the second section data is decreased. The grayscale value of , and the first section data after the grayscale adjustment and the second section data after the grayscale adjustment are determined as the target two-dimensional section data. A specific grayscale adjustment manner is not limited in this embodiment of the present invention.

In the embodiment of the present invention, the imaging processor 106 performs optimization processing of spine imaging based on the key anatomical regions of the spine and the two-dimensional data of the spine, and obtains an enhanced spine image, and may further include: determining imaging parameters based on the key anatomical regions of the spine; Using the two-dimensional slice data of the spine, the spine imaging is optimized according to the imaging parameters, and the spine enhancement image is obtained.

It should be noted that, in the embodiment of the present invention, the manner in which the imaging processor 106 determines the imaging parameters based on the key anatomical regions of the spine is similar to that in the above step S203, where the imaging parameters are determined based on the key anatomical structures of the spine, and the difference is only in that this step is aimed at is the relevant parameter of two-dimensional imaging. Compared with step S203, parameters such as gain and gradient threshold can also be determined. However, for two-dimensional imaging, the section thickness parameter cannot actually be determined, because the current spine two-dimensional body section data is only For a data on a spine surface, it is a plane data and cannot measure the thickness. The specific manner of determining the imaging parameter is detailed in step S203, and details are not described herein again.

S504, displaying the spine-enhanced image.

In the embodiment of the present invention, the imaging processor 106 performs optimal processing of spine imaging based on the key anatomical regions of the spine and the two-dimensional data of the spine, and after obtaining the spine enhancement image, the display 107 can display the spine enhancement image.

It should be noted that, in the embodiment of the present invention, the imaging processor 106 performs optimal processing of spine imaging based on the key anatomical regions of the spine and the data of the two-dimensional slices of the spine. Therefore, the obtained enhanced images of the spine are actually two-dimensional slices image.

The embodiment of the present invention provides an ultrasonic imaging method, which acquires two-dimensional section data of the spine; identifies the key anatomical regions of the spine from the two-dimensional section data of the spine; and performs spine imaging based on the key anatomical regions of the spine and the two-dimensional section data of the spine. Optimize processing to obtain spine-enhanced images; display spine-enhanced images. The technical solutions provided by the embodiments of the present invention assist spine imaging by identifying key anatomical regions of the spine in the two-dimensional section data of the spine, which not only has high imaging efficiency, but also improves imaging effects.

As shown in FIG. 1 , an embodiment of the present invention provides an ultrasonic imaging device, and the ultrasonic imaging device includes:

Probe 100;

Transmit/receive selection switch 101;

a transmitting circuit 102, configured to excite the probe 100 to transmit ultrasonic waves to the measured spine;

a receiving circuit 103, configured to receive ultrasonic echoes returned from the measured spine through the probe 100 to obtain ultrasonic echo signals;

a beamformer 104, configured to perform beamformation processing on the ultrasonic echo signal to obtain a beamformed ultrasonic echo signal;

a signal processor 105, configured to perform signal processing on the beam-synthesized ultrasonic echo signals to obtain processed ultrasonic echo signals;

The imaging processor 106 is configured to obtain three-dimensional volume data of the spine according to the processed ultrasonic echo signals; from the three-dimensional volume data of the spine, identify key anatomical structures of the spine; The three-dimensional volume data is optimized to obtain a spine-enhanced image, wherein the spine-enhanced image includes a stereo-enhanced image and/or a section-enhanced image;

The display 107 is used for displaying the spine-enhanced image.

Optionally, the imaging processor 106 is specifically configured to obtain a marking instruction and/or a preset structure detection method; according to the marking instruction and/or the preset structure detection method, from the spine three-dimensional volume data The key anatomical structures of the spine are identified.

Optionally, the imaging processor 106 is specifically configured to mark part of the volume data in the three-dimensional volume data of the spine according to the instruction of the marking instruction; and determine the structure surrounded by the part of the volume data as the The key anatomy of the spine.

Optionally, the preset structure detection method includes a feature detection method, and the imaging processor 106 is specifically configured to perform binarization segmentation and morphological processing on the spine three-dimensional volume data according to the feature detection method, Obtain multiple candidate structures; determine the probability that each of the multiple candidate structures is a spine structure based on the spine structure feature, obtain multiple probabilities, and determine a maximum probability from the multiple probabilities; Among the candidate structures, the structure corresponding to the maximum probability is determined as the key anatomical structure of the spine.

Optionally, the preset structure detection method includes a machine learning or deep learning detection method, and the imaging processor 106 is specifically configured to construct a database of spinal key anatomical structures according to the machine learning or deep learning detection method; Perform model training on the spine key anatomical structure database to obtain a spine key anatomical structure identification model; use the spine key anatomical structure identification model to identify the spine key anatomical structure from the spine three-dimensional volume data.

Optionally, the imaging processor 106 is specifically configured to perform grayscale adjustment on the three-dimensional volume data of the spine based on the key anatomical structure of the spine to obtain target three-dimensional volume data; obtain the three-dimensional volume data according to the target. Spine-enhanced image.

Optionally, the imaging processor 106 is specifically configured to obtain the first volume data included in the key anatomical structure of the spine from the three-dimensional volume data of the spine; increase the gray value of the first volume data to obtain the target three-dimensional volume data; and/or, from the spine three-dimensional volume data, obtain second volume data that is different from the first volume data included in the key anatomical structures of the spine; reduce the amount of the second volume data gray value to obtain the target three-dimensional volume data.

Optionally, the imaging processor 106 is specifically configured to determine an imaging parameter based on the key anatomical structure of the spine; optimize the three-dimensional volume data of the spine according to the determined imaging parameter to obtain the spine-enhanced image.

Optionally, the imaging parameters include target gain,

The imaging processor 106 is specifically configured to acquire, from the three-dimensional volume data of the spine, the first volume data included in the key anatomical structures of the spine; The grayscale statistical result of the volume data; the gain is adjusted according to the grayscale statistical result to obtain the target gain.

In the above ultrasonic imaging device, the imaging parameters include gradient threshold parameters,

The imaging processor 106 is specifically configured to acquire first volume data included in the key anatomical structure of the spine from the three-dimensional volume data of the spine; and determine the key spine based on the gray value of the first volume data the gradient value corresponding to the boundary of the anatomical structure; according to the gradient value, determine the first gradient threshold value corresponding to the first volume data according to the preset gradient threshold value calculation method; in the acquisition of the spine three-dimensional volume data, the spine key anatomy a second gradient threshold corresponding to the second volume data outside the structure; the first gradient threshold and the second gradient threshold are determined as the gradient threshold parameters.

In the above ultrasonic imaging device, the imaging parameter includes a profile thickness parameter,

The imaging processor 106 is specifically configured to obtain the position information of the key anatomical structures of the spine from the three-dimensional volume data of the spine; determine the spine thickness information based on the position information; determine the profile according to the spine thickness information Thickness parameter.

In the above ultrasonic imaging device, the imaging processor 106 is specifically configured to obtain boundary volume data corresponding to the boundary of the key anatomical structures of the spine from the three-dimensional volume data of the spine; and determine the boundary volume data as the the location information.

As shown in FIG. 1 , an embodiment of the present application provides an ultrasonic imaging device, and the ultrasonic imaging device includes:

Probe 100;

Transmit/receive selection switch 101;

a transmitting circuit 102, configured to excite the probe 100 to transmit ultrasonic waves to the measured spine;

a receiving circuit 103, configured to receive ultrasonic echoes returned from the measured spine through the probe 100 to obtain ultrasonic echo signals;

a beamformer 104, configured to perform beamformation processing on the ultrasonic echo signal to obtain a beamformed ultrasonic echo signal;

a signal processor 105, configured to perform signal processing on the beam-synthesized ultrasonic echo signals to obtain processed ultrasonic echo signals;

The imaging processor 106 is configured to obtain two-dimensional section data of the spine based on the processed ultrasonic echo signals; from the two-dimensional section data of the spine, identify key anatomical regions of the spine; The two-dimensional section data of the spine is optimized to obtain the spine-enhanced image;

The display 107 is used for displaying the spine-enhanced image.

Optionally, the imaging processor 106 is specifically configured to acquire three-dimensional volume data of the spine according to the processed ultrasonic echo signals; Select the two-dimensional slice data of the spine.

Optionally, the imaging processor 106 is specifically configured to perform grayscale adjustment on the two-dimensional section data of the spine based on the key anatomical regions of the spine to obtain the spine-enhanced image.

Optionally, the imaging processor 106 is specifically configured to obtain the first section data included in the key anatomical region of the spine, and the second section outside the key anatomical region of the spine from the two-dimensional section data of the spine. Slice data; increase the gray value of the first slice data to obtain the spine enhancement image; and/or, from the spine two-dimensional slice data, obtain the first everything related to the key anatomical structure of the spine second slice data with different slice data; reduce the gray value of the second slice data to obtain the spine-enhanced image.

Optionally, the imaging processor 106 is specifically configured to determine imaging parameters based on the key anatomical regions of the spine; optimize the two-dimensional section data of the spine according to the determined imaging parameters to obtain the spine-enhanced image.

An embodiment of the present application provides a computer-readable storage medium, where an ultrasound imaging program is stored in the computer-readable storage medium, and the ultrasound imaging program can be executed by a processor to implement the foregoing ultrasound imaging method. The computer-readable storage medium may be a volatile memory (volatile memory), such as a random-access memory (Random-Access Memory, RAM); or a non-volatile memory (non-volatile memory), such as a read-only memory (Read- OnlyMemory, ROM), flash memory (flash memory), hard disk (Hard Disk Drive, HDD) or solid-state drive (Solid-State Drive, SSD); it can also be a respective device including one or any combination of the above memories, such as mobile Phones, computers, tablet devices, personal digital assistants, etc.

As will be appreciated by those skilled in the art, the embodiments of the present application may be provided as a method, a system, or a computer program product. Accordingly, the application may take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media having computer-usable program code embodied therein, including but not limited to disk storage, optical storage, and the like.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present application. It will be understood that each flow and/or block in the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to the processor of a general purpose computer, special purpose computer, embedded processor or other programmable signal processing device to produce a machine such that the instructions executed by the processor of the computer or other programmable signal processing device produce Means for implementing the functions specified in a flow or flow of a flowchart and/or a block or blocks of a block diagram.

These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable signal processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory result in an article of manufacture comprising instruction means, the instructions The apparatus implements the functions specified in the flow or flow of the flowcharts and/or the block or blocks of the block diagrams.

These computer program instructions can also be loaded on a computer or other programmable signal processing device to cause a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process such that The instructions provide steps for implementing the functions specified in the flow or blocks of the flowcharts and/or the block or blocks of the block diagrams.

The above descriptions are only preferred embodiments of the present application, and are not intended to limit the protection scope of the present application.

Claims (29)

1. an ultrasonic imaging method, is characterized in that, described method comprises: Obtain spine 3D volume data; from the three-dimensional volume data of the spine, identifying key anatomical structures of the spine; Performing optimization processing on the three-dimensional volume data of the spine based on the key anatomical structure of the spine to obtain a spine-enhanced image, wherein the spine-enhanced image includes a stereo-enhanced image and/or a section-enhanced image; The spine-enhanced image is displayed. 2. The method according to claim 1, wherein, identifying the key anatomical structures of the spine from the three-dimensional volume data of the spine, comprising: Obtain marking instructions and/or preset structure detection methods; According to the marking instruction and/or the preset structure detection method, the key anatomical structure of the spine is identified from the three-dimensional volume data of the spine. 3. The method according to claim 1, wherein the optimizing the three-dimensional volume data of the spine based on the key anatomical structure of the spine to obtain an enhanced image of the spine, comprising: Based on the key anatomical structure of the spine, grayscale adjustment is performed on the three-dimensional volume data of the spine to obtain target three-dimensional volume data; The spine-enhanced image is obtained according to the target three-dimensional volume data. 4. The method according to claim 3, wherein, based on the key anatomical structure of the spine, performing grayscale adjustment on the three-dimensional volume data of the spine to obtain target three-dimensional volume data, comprising: From the three-dimensional volume data of the spine, obtain the first volume data included in the key anatomical structure of the spine; increasing the gray value of the first volume data to obtain the target three-dimensional volume data; and / or, From the three-dimensional volume data of the spine, obtain second volume data different from the first volume data included in the key anatomical structure of the spine; The grayscale value of the second volume data is reduced to obtain the target three-dimensional volume data. 5. The method according to claim 1, characterized in that, performing optimization processing on the three-dimensional volume data of the spine based on the key anatomical structure of the spine to obtain an enhanced image of the spine, comprising: determining imaging parameters based on the key anatomical structures of the spine; The spine three-dimensional volume data is optimized according to the determined imaging parameters to obtain the spine enhancement image. 6. The method of claim 5, wherein the imaging parameter comprises a target gain, and the determining the imaging parameter based on the key anatomical structure of the spine comprises: From the three-dimensional volume data of the spine, obtain the first volume data included in the key anatomical structure of the spine; Counting the grayscale values of the first volume data to obtain a grayscale statistical result of the first volume data; The target gain is determined according to the grayscale statistical result. 7. The method according to claim 5, wherein the imaging parameters comprise gradient threshold parameters, and the determining of the imaging parameters based on the key anatomical structures of the spine comprises: From the three-dimensional volume data of the spine, obtain the first volume data included in the key anatomical structure of the spine; Determine the gradient value corresponding to the key anatomical structure of the spine based on the gray value of the first volume data; According to the gradient value, a first gradient threshold corresponding to the first volume data is determined according to a preset gradient threshold calculation method; obtaining a second gradient threshold corresponding to the second volume data outside the key anatomical structure of the spine in the three-dimensional volume data of the spine; The first gradient threshold and the second gradient threshold are determined as the gradient threshold parameters. 8. The method of claim 5, wherein the imaging parameter comprises a profile thickness parameter, and the determining the imaging parameter based on the key anatomical structure of the spine comprises: From the three-dimensional volume data of the spine, obtain the position information of the key anatomical structures of the spine; determining spine thickness information based on the position information; The profile thickness parameter is determined based on the spine thickness information. 9. The method according to claim 8, wherein obtaining the position information of the key anatomical structures of the spine from the three-dimensional volume data of the spine, comprising: From the three-dimensional volume data of the spine, obtain boundary volume data corresponding to the boundaries of the key anatomical structures of the spine; The bounding volume data is determined as the position information. 10. An ultrasound imaging method, characterized in that the method comprises: Obtain two-dimensional slice data of the spine; Identifying key anatomical regions of the spine from the two-dimensional section data of the spine; Optimizing the two-dimensional section data of the spine based on the key anatomical regions of the spine to obtain an enhanced spine image; The spine-enhanced image is displayed. 11. The method according to claim 10, wherein the acquiring two-dimensional slice data of the spine comprises: Obtain spine 3D volume data; According to a preset slice selection method or a received selection instruction, the spine two-dimensional slice data is determined from the spine three-dimensional volume data. 12. The method according to claim 10, characterized in that, performing optimization processing on the two-dimensional section data of the spine based on the key anatomical regions of the spine to obtain an enhanced spine image, comprising: Based on the key anatomical regions of the spine, grayscale adjustment is performed on the two-dimensional section data of the spine to obtain the spine-enhanced image. 13. The method according to claim 12, wherein, based on the key anatomical regions of the spine, performing grayscale adjustment on the two-dimensional section data of the spine to obtain the spine-enhanced image, comprising: From the two-dimensional section data of the spine, obtain the first section data included in the key anatomical region of the spine; increasing the gray value of the first section data to obtain the spine-enhanced image; and / or, From the two-dimensional section data of the spine, obtain second section data that is different from the first section data included in the key anatomical structure of the spine; The grayscale value of the second slice data is reduced to obtain the spine-enhanced image. 14. The method according to claim 10, wherein, based on the key anatomical region of the spine, performing grayscale adjustment on the two-dimensional section data of the spine to obtain the spine-enhanced image, comprising: determining imaging parameters based on the critical anatomical region of the spine; The spine two-dimensional slice data is optimized according to the determined imaging parameters to obtain the spine enhancement image. 15. An ultrasonic imaging device, wherein the ultrasonic imaging device comprises: probe; a transmitting circuit for exciting the probe to transmit ultrasonic waves to the measured spine; a receiving circuit, configured to receive the ultrasonic echo returned from the measured spine through the probe to obtain an ultrasonic echo signal; a beamformer, configured to perform beamformation processing on the ultrasonic echo signal to obtain a beamformed ultrasonic echo signal; a signal processor, configured to perform signal processing on the beam-synthesized ultrasonic echo signals to obtain processed ultrasonic echo signals; an imaging processor for: acquiring three-dimensional volume data of the spine according to the processed ultrasonic echo signals; identifying key anatomical structures of the spine from the three-dimensional volume data of the spine; The three-dimensional volume data is optimized to obtain a spine-enhanced image, wherein the spine-enhanced image includes a stereo-enhanced image and/or a section-enhanced image; a display for displaying the spine-enhanced image. 16. The ultrasound imaging apparatus of claim 15, wherein The imaging processor is configured to acquire a marking instruction and/or a preset structure detection method, and identify the spine key anatomy from the spine three-dimensional volume data according to the marking instruction and/or the preset structure detection method structure. 17. The ultrasound imaging apparatus of claim 15, wherein The imaging processor is configured to perform grayscale adjustment on the three-dimensional volume data of the spine based on the key anatomical structure of the spine to obtain target three-dimensional volume data, and obtain the enhanced spine image according to the target three-dimensional volume data. 18. The ultrasound imaging apparatus of claim 17, wherein The imaging processor is configured to obtain the first volume data included in the key anatomical structure of the spine from the spine three-dimensional volume data, and increase the gray value of the first volume data to obtain the target three-dimensional volume data; And/or, from the three-dimensional volume data of the spine, obtain second volume data that is different from the first volume data included in the key anatomical structure of the spine; reduce the gray value of the second volume data to obtain the Target volume data. 19. The ultrasound imaging apparatus of claim 15, wherein The imaging processor is configured to determine imaging parameters based on the key anatomical structures of the spine, and optimize the three-dimensional volume data of the spine according to the determined imaging parameters to obtain the spine-enhanced image. 20. The ultrasound imaging apparatus of claim 19, wherein the imaging parameter comprises a target gain, The imaging processor is configured to obtain the first volume data included in the key anatomical structure of the spine from the three-dimensional volume data of the spine, count the gray value of the first volume data, and obtain the gray value of the first volume data. The grayscale statistical result is obtained, and the target gain is determined according to the grayscale statistical result. 21. The ultrasound imaging apparatus of claim 19, wherein the imaging parameter comprises a gradient threshold parameter, The imaging processor is configured to: obtain the first volume data included in the key anatomical structure of the spine from the three-dimensional volume data of the spine; and determine the corresponding key anatomical structure of the spine based on the gray value of the first volume data According to the gradient value, according to the preset gradient threshold calculation method, determine the first gradient threshold corresponding to the first volume data; in the acquisition of the spine three-dimensional volume data, the second spine outside the key anatomical structure of the spine is obtained. The second gradient threshold corresponding to the volume data; the first gradient threshold and the second gradient threshold are determined as the gradient threshold parameters. 22. The ultrasound imaging apparatus of claim 19, wherein the imaging parameter comprises a profile thickness parameter, The imaging processor is configured to: obtain position information of the key anatomical structures of the spine from the three-dimensional volume data of the spine; determine spine thickness information based on the position information; and determine the profile thickness parameter according to the spine thickness information. 23. The ultrasound imaging apparatus of claim 22, wherein The imaging processor is configured to: obtain boundary volume data corresponding to the boundary of the key anatomical structure of the spine from the spine three-dimensional volume data; and determine the boundary volume data as the position information. 24. An ultrasound imaging device, wherein the ultrasound imaging device comprises: probe; a transmitting circuit for exciting the probe to transmit ultrasonic waves to the measured spine; a receiving circuit, configured to receive the ultrasonic echo returned from the measured spine through the probe to obtain an ultrasonic echo signal; a beamformer, configured to perform beamformation processing on the ultrasonic echo signal to obtain a beamformed ultrasonic echo signal; a signal processor, configured to perform signal processing on the beam-synthesized ultrasonic echo signals to obtain processed ultrasonic echo signals; An imaging processor, configured to: acquire two-dimensional section data of the spine based on the processed ultrasonic echo signals; identify key anatomical regions of the spine from the two-dimensional section data of the spine; The two-dimensional section data of the spine is optimized to obtain the spine-enhanced image; a display for displaying the spine-enhanced image. 25. The ultrasound imaging apparatus of claim 24, wherein The imaging processor is used for: acquiring three-dimensional volume data of the spine according to the processed ultrasonic echo signals; slice data. 26. The ultrasound imaging apparatus of claim 24, wherein The imaging processor is configured to: perform grayscale adjustment on the two-dimensional section data of the spine based on the key anatomical regions of the spine to obtain the spine-enhanced image. 27. The ultrasound imaging apparatus of claim 26, wherein The imaging processor is configured to: obtain the first section data included in the key anatomical region of the spine from the two-dimensional section data of the spine; increase the gray value of the first section data to obtain the spine enhancement and/or, from the two-dimensional section data of the spine, obtain second section data different from the first section data included in the key anatomical structure of the spine; reduce the gray value of the second section data , to obtain the spine-enhanced image. 28. The ultrasound imaging apparatus of claim 24, wherein The imaging processor is configured to: determine imaging parameters based on the key anatomical regions of the spine; optimize the two-dimensional section data of the spine according to the determined imaging parameters to obtain the spine-enhanced image. 29. A computer-readable storage medium, characterized in that, the computer-readable storage medium stores an ultrasound imaging program, and the ultrasound imaging program can be executed by a processor to implement any one of claims 1-14. method of ultrasound imaging.

Images