Ultrasonic imaging method and ultrasonic equipment

Abstract

According to the ultrasonic imaging method and the ultrasonic equipment, bubbles in a target area of an ultrasonic image are detected, a bubble detection result is determined, whether an aeroembolism risk exists or not is judged according to the bubble detection result, and when it is judged that the aeroembolism risk exists, a user is prompted; the bubble level in the target area can be automatically given, and the user is prompted when the air embolism risk exists, so that automatic monitoring of the bubbles in the target area is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ultrasonic imaging, in particular to an ultrasonic imaging method and an ultrasonic device. BACKGROUND

[0002] In laparoscopic surgery, the abdominal cavity needs to be inflated, and gas may enter the right heart cavity through the abdominal vessel wound. When the amount of gas exceeds the gas exchange capacity of the lungs, the gas will enter the left heart cavity and the systemic circulation, which may cause serious sequelae or death. Therefore, it is necessary to continuously monitor whether there is gas bubble in the heart cavity during laparoscopic surgery, and the gas bubble should be handled in time when it is found.

[0003] At present, such as Figure 1 Transesophageal echocardiography (TEE) is the most sensitive method for intraoperative identification of gas in the heart cavity. However, manual monitoring requires an ultrasonic physician to monitor the TEE image throughout the process, which occupies a large amount of medical resources, and there is a risk of missed detection in long-term manual monitoring. In addition, when the image is three-dimensional, it is difficult to monitor all positions of the gas bubble through the human eye. SUMMARY

[0004] In view of the above problems, the present application provides an ultrasonic imaging method and an ultrasonic device, which will be described in detail below.

[0005] According to a first aspect, an ultrasonic device is provided in an embodiment, comprising:

[0006] an ultrasonic probe, configured to emit ultrasonic waves to a region of interest and receive corresponding echo signals;

[0007] a transmission and reception control circuit, configured to control the ultrasonic probe to emit the ultrasonic waves and receive the corresponding echo signals;

[0008] a processor, configured to:

[0009] generate an ultrasonic image based on the echo signals;

[0010] detect a gas bubble in a target region in the ultrasonic image to determine a gas bubble detection result; the target region includes a heart chamber region and / or a blood vessel lumen region;

[0011] determine whether there is a gas embolism risk according to the gas bubble detection result, and prompt a user when it is determined that there is a gas embolism risk.

[0012] In an embodiment, the gas bubble detection result includes one or more of the following: the number of gas bubbles in the target region, the area distribution of the gas bubbles in the target region, the volume distribution of the gas bubbles in the target region, the ratio of the total area of the gas bubbles to the area of the target region, and the ratio of the total volume of the gas bubbles to the volume of the target region.

[0013] In an embodiment, the target region in the ultrasound image is segmented to obtain the target region of the ultrasound image, and the method further comprises:

[0014] In an embodiment, the target region in the ultrasound image is segmented to obtain the target region of the ultrasound image, and the method further comprises:

[0015] In an embodiment, the target region in the ultrasound image is segmented to obtain the target region of the ultrasound image, and the method further comprises:

[0016] In an embodiment, the target region in the ultrasound image is segmented to obtain the target region of the ultrasound image, and the method further comprises:

[0017] In an embodiment, the target region in the ultrasound image is segmented to obtain the target region of the ultrasound image, and the method further comprises:

[0018] In an embodiment, the target region in the ultrasound image is segmented to obtain the target region of the ultrasound image, and the method further comprises:

[0019] In an embodiment, the target region in the ultrasound image is segmented to obtain the target region of the ultrasound image, and the method further comprises:

[0020] In an embodiment, the target region in the ultrasound image is segmented to obtain the target region of the ultrasound image, and the method further comprises:

[0021] In an embodiment, the target region in the ultrasound image is segmented to obtain the target region of the ultrasound image, and the method further comprises:

[0022] In an embodiment, the target region in the ultrasound image is segmented to obtain the target region of the ultrasound image, and the method further comprises:

[0023] In an embodiment, the target region in the ultrasound image is segmented to obtain the target region of the ultrasound image, and the method further comprises:

[0024] In an embodiment, the target region in the ultrasound image is segmented to obtain the target region of the ultrasound image, and the method further comprises:

[0025] In an embodiment, the target region in the ultrasound image is segmented to obtain the target region of the ultrasound image, and the method further comprises:

[0026] input the ultrasound image into a first deep learning model of a pre-constructed bubble-ultrasound image correspondence relationship, to obtain a bubble detection result of a target region in the ultrasound image.

[0027] In an embodiment, the detecting the bubble in the target region in the ultrasound image to determine the bubble detection result comprises:

[0028] inputting the ultrasound image into a second deep learning model of a pre-constructed target region-bubble-ultrasound image correspondence relationship, to obtain each target region in the ultrasound image, the bubble in the ultrasound image, and the target region to which the bubble belongs, so as to determine the bubble detection result.

[0029] In an embodiment, the judging whether there is a gas embolism risk according to the bubble detection result comprises:

[0030] if the bubble detection result is greater than or equal to a preset threshold, it is judged that there is a gas embolism risk; otherwise, it is judged that there is no gas embolism risk.

[0031] In an embodiment, the method further comprises:

[0032] displaying the ultrasound image and marking the detected bubble in the ultrasound image.

[0033] According to a second aspect, an embodiment provides an ultrasound device, comprising:

[0034] an ultrasound probe configured to emit ultrasound waves to a region of interest and receive corresponding echo signals;

[0035] transmission and reception control circuitry configured to control the ultrasound probe to emit the ultrasound waves and receive the corresponding echo signals;

[0036] a processor configured to:

[0037] generate an ultrasound image based on the echo signals;

[0038] judge whether there is a gas embolism risk based on the ultrasound image, and prompt a user when it is judged that there is a gas embolism risk.

[0039] In an embodiment, the judging whether there is a gas embolism risk based on the ultrasound image comprises:

[0040] inputting the ultrasound image into a third deep learning model of a pre-constructed gas embolism risk-ultrasound image correspondence relationship, to output a result of whether there is a gas embolism risk.

[0041] According to a third aspect, an embodiment provides an ultrasound device, comprising:

[0042] An ultrasonic probe configured to emit ultrasonic waves to a region of interest and receive corresponding echo signals;

[0043] A transmit and receive control circuit configured to control the ultrasonic probe to emit the ultrasonic waves and receive the corresponding echo signals;

[0044] A processor configured to:

[0045] generate an ultrasonic image based on the echo signals;

[0046] detect bubbles in the ultrasonic image and determine a bubble detection result;

[0047] display the bubble detection result.

[0048] In an embodiment, the bubble detection result comprises at least one or more of a number of bubbles, an area distribution of bubbles, a volume distribution of bubbles, a ratio of a total area of bubbles to an area of an imaging region of the ultrasonic image, and a ratio of a total volume of bubbles to a volume of the imaging region of the ultrasonic image.

[0049] According to a fourth aspect, an embodiment provides an ultrasonic imaging method, comprising:

[0050] controlling to emit ultrasonic waves to a region of interest and receive corresponding echo signals, and generating an ultrasonic image based on the echo signals;

[0051] detecting bubbles in a target region of the ultrasonic image and determining a bubble detection result; the target region comprises a cardiac chamber region and / or a blood vessel lumen region;

[0052] judging whether there is a gas embolism risk according to the bubble detection result, and prompting a user when it is judged that there is a gas embolism risk.

[0053] In an embodiment, the bubble detection result comprises at least one or more of a number of bubbles in the target region, an area distribution of bubbles in the target region, a volume distribution of bubbles in the target region, a ratio of a total area of bubbles to an area of the target region, and a ratio of a total volume of bubbles to a volume of the target region.

[0054] In an embodiment, detecting bubbles in a target region of the ultrasonic image and determining a bubble detection result comprises:

[0055] segmenting the target region in the ultrasonic image to obtain a target region of the ultrasonic image;

[0056] segmenting bubbles in the target region of the ultrasonic image to obtain a bubble segmentation result in the target region;

[0057] According to the bubble segmentation result in the target region, the bubble detection result is determined.

[0058] In an embodiment, the detecting the bubbles in the target region of the ultrasound image to determine the bubble detection result comprises:

[0059] The ultrasound image is input into a first deep learning model of a pre-constructed bubble-ultrasound image correspondence relationship to obtain the bubble detection result of the target region in the ultrasound image.

[0060] In an embodiment, the detecting the bubbles in the target region of the ultrasound image to determine the bubble detection result comprises:

[0061] The ultrasound image is input into a second deep learning model of a pre-constructed target region-bubble-ultrasound image correspondence relationship to obtain each target region in the ultrasound image, the bubbles in the ultrasound image, and the target region to which the bubbles belong, so as to determine the bubble detection result.

[0062] According to a fifth aspect, an embodiment provides an ultrasound imaging method, comprising:

[0063] Controlling to emit ultrasound waves to a region of interest and receive corresponding echo signals, and generating an ultrasound image based on the echo signals;

[0064] Based on the ultrasound image, it is judged whether there is a gas embolism risk, and when it is judged that there is a gas embolism risk, the user is prompted.

[0065] In an embodiment, the judging whether there is a gas embolism risk based on the ultrasound image comprises:

[0066] The ultrasound image is input into a third deep learning model of a pre-constructed gas embolism risk-ultrasound image correspondence relationship to output the result of whether there is a gas embolism risk.

[0067] According to a sixth aspect, an embodiment provides an ultrasound imaging method, comprising:

[0068] Controlling to emit ultrasound waves to a region of interest and receive corresponding echo signals, and generating an ultrasound image based on the echo signals;

[0069] Detecting bubbles in the ultrasound image to determine a bubble detection result;

[0070] Displaying the bubble detection result.

[0071] In an embodiment, the bubble detection result comprises one or more of a number of bubbles, an area distribution of bubbles, a volume distribution of bubbles, a ratio of a total area of bubbles to an area of an imaging region of the ultrasound image, and a ratio of a total volume of bubbles to a volume of the imaging region of the ultrasound image.

[0072] According to a seventh aspect, in an embodiment there is provided a computer readable storage medium having stored thereon a program, the program being loadable into a processing unit and adapted to cause the processing unit to execute the method according to any of the embodiments described above when the program is run by the processing unit.

[0073] According to the ultrasound imaging method and the ultrasound device of the above embodiments, the bubbles in the target region of the ultrasound image are detected, the bubble detection result is determined, whether there is a gas embolism risk is judged according to the bubble detection result, and the user is prompted when it is judged that there is a gas embolism risk; since the bubble level in the target region can be automatically given, and the user is prompted when there is a gas embolism risk, the automatic monitoring and prompting of the bubbles in the target region are realized. BRIEF DESCRIPTION OF DRAWINGS

[0074] Figure 1 Structure diagram of an ultrasound device according to an embodiment; Figure 1 Structure diagram of an ultrasound device according to an embodiment;

[0075] Figure 2 Structure diagram of an ultrasound device according to an embodiment; Structure diagram of an ultrasound device according to an embodiment;

[0076] Structure diagram of an ultrasound device according to an embodiment; Figure 3 Flowchart of an ultrasound imaging method according to an embodiment;

[0077] Figure 4 Flowchart of a method for detecting a bubble detection result in an ultrasound image according to an embodiment;

[0078] Figure 5 Flowchart of an ultrasound imaging method according to another embodiment;

[0079] Figure 6 Flowchart of an ultrasound imaging method according to still another embodiment. DETAILED DESCRIPTION

[0080] The application will be described in further detail below with reference to the drawings. Like elements in different embodiments are denoted by like reference numerals. In the following embodiments, many specific details are described in order to provide a thorough understanding of the application. However, it will be apparent to those skilled in the art that the application can be practiced without these specific details. In other instances, well-known methods, procedures, components, and circuits have not been described in detail so as not to unnecessarily obscure aspects of the application. Some operations are not shown or described in the specification in order to avoid obscuring the subject matter of the present application, which are well known to those skilled in the art.

[0081] In addition, the features, operations or characteristics described in the specification can be combined in any appropriate manner to form various embodiments. Meanwhile, the steps or actions in the method description can also be adjusted or changed in sequence as long as it is obvious to those skilled in the art. Therefore, the various sequences in the specification and the drawings are only for the purpose of clearly describing a certain embodiment, and do not mean that the sequence is necessary, unless otherwise stated that a certain sequence must be followed.

[0082] The serial numbers of the components in this paper, such as "first", "second", etc., are only used to distinguish the described objects, and do not have any sequence or technical meaning. The "connection" and "coupling" in this application include direct and indirect connection (coupling) unless otherwise specified.

[0083] Please refer to Figure 2 The ultrasonic device provided by the application comprises an ultrasonic probe 10, a transmitting and receiving control circuit 20, a processor 40, and in some embodiments, an echo processing module 30 and / or a human-computer interaction device. The components will be described below.

[0084] The ultrasound probe 10 is used to transmit ultrasound waves to the region of interest and receive corresponding ultrasound echo signals to obtain ultrasound data, such as two-dimensional ultrasound data or three-dimensional ultrasound data. In some embodiments, the ultrasound probe 10 includes a plurality of array elements for converting electrical signals and ultrasound waves to each other, so as to transmit ultrasound waves to the region of interest and receive corresponding ultrasound echo signals. The array elements can transmit ultrasound waves according to the excitation electrical signals, or convert the received ultrasound waves into electrical signals. Therefore, each array element can be used to transmit ultrasound waves to the biological tissue of the region of interest, and also can be used to receive the ultrasound echo waves returned by the tissue. When performing ultrasound detection, it can be controlled by a transmission sequence and a reception sequence which array elements are used to transmit ultrasound waves and which array elements are used to receive ultrasound waves, or it can be controlled by time slots in which the array elements are used to transmit ultrasound waves or receive ultrasound echoes. All array elements participating in the transmission of ultrasound waves can be excited by electrical signals at the same time, so as to transmit ultrasound waves at the same time; or the array elements participating in the transmission of ultrasound waves can also be excited by several electrical signals with a certain time interval, so as to continuously transmit ultrasound waves with a certain time interval.

[0085] The transmission and reception control circuit 20 is used to control the ultrasound probe 10 to perform the transmission of ultrasound waves and the reception of ultrasound echo signals. For example, the transmission and reception control circuit 20 is used to control the ultrasound probe 10 to transmit ultrasound waves to the region of interest, and is also used to control the ultrasound probe 10 to receive the ultrasound echo signals reflected by the region of interest. In some embodiments, the transmission and reception control circuit 20 is used to generate a transmission sequence and a reception sequence, and outputs the transmission sequence and the reception sequence to the ultrasound probe 10. The transmission sequence is used to control part or all of the plurality of array elements in the ultrasound probe 10 to transmit ultrasound waves to the biological tissue 60, and the parameters of the transmission sequence include the number of array elements used for transmission and the parameters of the transmitted ultrasound waves (such as amplitude, frequency domain, number of transmitted waves, transmission interval, transmission angle, wave type and / or focus position, etc.). The reception sequence is used to control part or all of the plurality of array elements to receive the echoes of the ultrasound waves after passing through the tissue, and the parameters of the reception sequence include the number of array elements used for reception and the reception parameters of the echoes (such as reception angle, depth, etc.). The parameters of the transmitted ultrasound waves in the transmission sequence and the parameters of the echoes in the reception sequence are different according to different uses of the ultrasound echoes or different images generated according to the ultrasound echoes.

[0086] The echo processing module 30 is configured to process the ultrasound echo signals received by the ultrasound probe 10, for example, to perform filtering, amplification, beamforming, etc. on the ultrasound echo signals to obtain ultrasound echo data. In some embodiments, the echo processing module 30 can output the ultrasound echo data to the processor 40, or the ultrasound echo data can be first stored in a memory, and the processor 40 reads the ultrasound echo data from the memory when needed to perform operations based on the ultrasound echo data. It should be understood by those skilled in the art that in some embodiments, the echo processing module 30 can be omitted when the ultrasound echo signals do not need to be processed by filtering, amplification, beamforming, etc.

[0087] The processor 40 is configured to obtain the ultrasound echo data or echo signals, and obtain the required parameters or images by using a correlation algorithm. The processor 40 in some embodiments of the present application includes, but is not limited to, a central processing unit (CPU), a micro controller unit (MCU), a field-programmable gate array (FPGA), a digital signal processing (DSP), and other devices for interpreting computer instructions and processing data in computer software.

[0088] The human-computer interaction device is configured to perform human-computer interaction, i.e., to receive the input of the user and output visualized information. The input of the user can be received by using a keyboard, an operation button, a mouse, a trackball, etc., or by using a touch screen integrated with a display. The output visualized information is output by using the display component 50. The display component 50 can be configured to display information, for example, to display the parameters and images calculated by the processor 40, etc. It should be understood by those skilled in the art that in some embodiments, the ultrasound device itself can not be integrated with a display module, but can be connected to a computer device (for example, a computer), and the information can be displayed by using the display module (for example, a display screen) of the computer device.

[0089] It should be noted that, Figure 2 the structure shown in FIG. 1 is only schematic, and can include more or fewer components than those shown in FIG. 1, or have a different configuration from that shown in FIG. 1. Figure 2 The components shown in FIG. 1 can be implemented by using hardware and / or software. Figure 2 Figure 2 The components shown in FIG. 1 can be implemented by using hardware and / or software.

[0090] Please refer to Figure 3 In some embodiments, an ultrasound imaging method, the processor 40 is configured to perform one step, multiple steps, or all steps of the ultrasound imaging method, including the following steps:

[0091] ​Step 101: Emitting an ultrasound wave to a region of interest, receiving a corresponding echo signal, and generating an ultrasound image based on the echo signal. In some embodiments, the region of interest can be a heart region of a target object, or can be other tissue regions, such as a blood vessel region, etc.

[0092] In some embodiments, the ultrasound image can be a two-dimensional or three-dimensional image, and the type of the ultrasound image can be a B image, a transesophageal echocardiogram (TEE), or other types of ultrasound images, which are not specifically limited in the present embodiment.

[0093] Step 102: Detecting a bubble in a target region in the ultrasound image, and determining a bubble detection result. In an embodiment, the target region can be a heart chamber region, such as a right heart chamber region or a left heart chamber region, or can be a blood vessel lumen region, such as an arterial blood vessel lumen region, etc.

[0094] In some embodiments, the bubble detection result is used to reflect a bubble level in the target region, and can be some parameters of the bubble itself, such as a number of bubbles, an area distribution of the bubbles, a volume distribution of the bubbles, etc., or can be relationship parameters between the bubbles and the target region, such as a ratio of a total area of the bubbles to an area of the target region, a ratio of a total volume of the bubbles to a volume of the target region, etc. In the present embodiment, the bubble detection result can include one or all of the above parameters according to a system predetermined mode or a user customized mode, or can include one of the above parameters. The user customized mode can be that a user inputs or selects a required bubble detection result through a mouse, a keyboard, or other input devices, and then determines the user customized mode.

[0095] The bubble in the ultrasound image can be detected in various ways, and the following describes a process of detecting the bubble in the target region in the ultrasound image.

[0096] Please refer to Figure 4 In some embodiments, detecting the bubble in the target region in the ultrasound image to determine the bubble detection result includes the following steps:

[0097] Step 201: Segmenting the target region in the ultrasound image to obtain the target region of the ultrasound image.

[0098] In some embodiments, a semantic or instance segmentation method based on deep learning or machine learning can be used to segment the target region in the ultrasound image. In an example, a deep learning-based segmentation method can include the following steps: inputting one or more frames of ultrasound images into a pre-trained deep learning model, and outputting the target region in the ultrasound image. In an embodiment, the deep learning model can be a convolutional neural network (CNN), a neural network based on an attention mechanism (Transformer), a neural network based on a full connection (MLP), a spiking neural network (SNN), or a hybrid model of the above network structures. In another example, a machine learning-based segmentation method can include the following steps: using a machine learning feature extraction and segmentation method to extract and segment the target region in the ultrasound image. In an embodiment, the machine learning feature extraction method can be the deep learning network described above, or an existing feature extraction method such as PCA, HOG, LDA, etc. The machine learning segmentation method can be a threshold method, a Snakes method, a clustering method, a watershed method, a level set method, etc.

[0099] In some embodiments, a deep learning method can also be used to obtain the bounding rectangle of the target region in the ultrasound image, and then the target region in the ultrasound image can be obtained according to the pixel features in the bounding rectangle. In an embodiment, the deep learning method can be based on the deep learning model described above.

[0100] Step 202: Segmenting the bubbles in the target region of the ultrasound image to obtain the bubble segmentation result in the target region.

[0101] In some embodiments, a deep learning or machine learning based semantic or instance segmentation method can be used to segment the bubbles in the target region. For example, a deep learning based segmentation method can include the following steps: inputting one or more ultrasound images into a pre-trained deep learning model to segment the bubbles in the ultrasound images, and obtaining the bubbles in the target region according to the positions of the segmented bubbles, for example, assigning the bubbles to the corresponding cardiac chamber region or blood vessel lumen region. In an embodiment, the deep learning model can be a convolutional neural network (CNN), a neural network based on an attention mechanism (Transformer), a neural network based on full connection (MLP), a spiking neural network (SNN), or a hybrid model of the above network structures. For another example, a machine learning based segmentation method can include the following steps: using a machine learning based feature extraction and segmentation method to extract and segment the bubbles in the ultrasound images, and obtaining the bubbles in the target region according to the positions of the segmented bubbles, for example, assigning the bubbles to the corresponding cardiac chamber region or blood vessel lumen region. In an embodiment, the machine learning based feature extraction method can be the above deep learning network, or a conventional feature extraction method such as PCA, HOG, LDA, etc. The machine learning based segmentation method can be a threshold method, a Snakes method, a clustering method, a watershed method, a level set method, etc.

[0102] In some embodiments, a deep learning method is used to obtain a bubble circumscribed rectangle of the target region of the ultrasound image, and then a bubble segmentation result in the target region is obtained according to the pixel features in the circumscribed rectangle. In an embodiment, the deep learning method can be based on the above deep learning model.

[0103] Step 203: determining a bubble detection result according to the bubble segmentation result in the target region.

[0104] In some embodiments, the segmented bubbles in the target region are counted and calculated to obtain the statistical information of the bubbles in the target region, i.e., the bubble detection result. In an embodiment, the bubble detection result can include one or more of the number of bubbles in the target region, the area distribution of the bubbles in the target region, the volume distribution of the bubbles in the target region, the ratio of the total area of the bubbles to the area of the target region, and the ratio of the total volume of the bubbles to the volume of the target region.

[0105] Step 103: determining whether there is a gas embolism risk according to the bubble detection result, and prompting the user when it is determined that there is a gas embolism risk.

[0106] In some embodiments, the air embolism risk can be determined based on whether the preset condition is met based on the bubble detection result. For example, when the ratio of the total area (volume) of the bubbles to the area (volume) of the target region is greater than or equal to a preset threshold, it is determined that there is an air embolism risk, and when the ratio is less than the preset threshold, it is determined that there is no air embolism risk. The preset condition can be the preset threshold, or other conditions set by the doctor.

[0107] In other embodiments, the method based on machine learning, deep learning, or a combination of the two can also be used to determine whether there is an air embolism risk. The deep learning method can be a classification / regression method to output a binary classification of whether there is a risk or a probability of the risk. The specific network structure can be the deep learning network model mentioned above. The machine learning method can be a logistic regression, SVM, or the like.

[0108] In some embodiments, when it is determined that there is an air embolism risk, a prompt message can be displayed on the display interface in the ultrasound device, for example, a symbol, text, or the like representing the air embolism risk can be displayed in or beside the ultrasound image. The user can also be prompted through voice, vibration, indicator light, or the like in a predetermined manner. The voice component, vibration component, or indicator light component in the ultrasound device can be used to execute the prompt.

[0109] In some embodiments, the bubbles in the target region of the ultrasound image are detected, and the bubble detection result can also be obtained by inputting the ultrasound image into a first deep learning model pre-constructed for the relationship between the bubbles and the ultrasound image to obtain the bubble detection result of the target region in the ultrasound image. That is, the first deep learning model is used to directly output the statistical information of the bubbles in the target region of the ultrasound image, without first segmenting the bubbles and then obtaining the statistical information of the bubbles. In an embodiment, the first deep learning model can be some of the deep learning networks mentioned above. After the first deep learning model is constructed, the first deep learning model needs to be trained by a training set of ultrasound images with labeled bubbles. After the training is completed, the ultrasound image can be input to obtain the bubble detection result.

[0110] In some embodiments, the detecting the bubbles in the target region in the ultrasound image, and determining the bubble detection result can further include: inputting the ultrasound image into a second deep learning model of a pre-constructed target region-bubble and ultrasound image correspondence relationship, to obtain each target region in the ultrasound image, the bubbles in the ultrasound image, and the target region to which the bubbles belong, so as to determine the bubble detection result. That is, the second deep learning model is used to directly output the target region in the ultrasound image, the bubbles in the ultrasound image, and the target region to which each bubble belongs, so that the bubble detection result of the target region can be obtained. In an embodiment, the second deep learning model can be some of the deep learning networks mentioned above. After the second deep learning model is constructed, the second deep learning model needs to be trained by a training set of ultrasound images that have been labeled with target region-bubbles. After the training is completed, the ultrasound image can be inputted to obtain the bubble detection result.

[0111] In some embodiments, the ultrasound imaging method can further include: displaying the ultrasound image, and marking the detected bubbles in the ultrasound image. In other embodiments, the ultrasound image can be displayed, and the bubble detection result can also be displayed at the same time. In this way, the user can view the real-time situation of the bubbles in the ultrasound image, so that the user can know the situation of the bubbles in the ultrasound image even in the case of no risk of air embolism.

[0112] Please refer to Figure 5 In some embodiments, another scheme of an ultrasound imaging method is also provided, which can include the following steps:

[0113] Step 301: transmitting an ultrasound wave to a region of interest, receiving a corresponding echo signal, and generating an ultrasound image based on the echo signal. In some embodiments, the region of interest can be a heart region of a target object, or can be other tissue regions, such as a blood vessel region, etc.

[0114] In some embodiments, the ultrasound image can be a two-dimensional or three-dimensional image, and the type of the ultrasound image can be a B image, a transesophageal echocardiogram (TEE), or other types of ultrasound images, which are not limited in the present embodiment.

[0115] Step 302: determining whether there is a risk of air embolism based on the ultrasound image, and prompting the user when it is determined that there is a risk of air embolism.

[0116] In some embodiments, the process of step 302 can be realized in the following way to directly determine whether there is a risk of air embolism based on the ultrasound image:

[0117] The ultrasound image is input into a third deep learning model of a pre-constructed air embolus risk and ultrasound image correspondence relationship, and a result of whether there is an air embolus risk is output. That is, the third deep learning model is used to directly output whether the ultrasound image has an air embolus risk, without first obtaining a bubble detection result and then judging whether there is an air embolus risk based on the bubble detection result. In an embodiment, the third deep learning model can be some of the deep learning networks mentioned above. After the third deep learning model is constructed, the third deep learning model needs to be trained by a training set of ultrasound images that have been labeled for air embolus risk. After the training is completed, the ultrasound image can be input to obtain the result of whether there is an air embolus risk.

[0118] Please refer to Figure 6 In some embodiments, another scheme of an ultrasound imaging method is also provided, which can include the following steps:

[0119] Step 401: Emit an ultrasound wave to a region of interest and receive a corresponding echo signal, and generate an ultrasound image based on the echo signal. In some embodiments, the region of interest can be a heart region of a target object, or can be another tissue region, such as a blood vessel region, etc.

[0120] Step 402: Detect bubbles in the ultrasound image and determine a bubble detection result.

[0121] In some embodiments, the bubble detection result can be some parameters of the bubbles themselves, such as the number of bubbles, the area distribution of the bubbles, the volume distribution of the bubbles, etc., and can also be relationship parameters between the bubbles and the imaging region or the target region, such as the ratio of the total area of the bubbles to the area of the imaging region, the ratio of the total volume of the bubbles to the volume of the imaging region, etc. In this embodiment, according to a system predetermined mode or a user customized mode, the bubble detection result can include the above multiple or all parameters, or can include one of the above parameters; wherein the user customized mode can be that the user inputs or selects the required bubble detection result through a mouse, a keyboard, or other input devices, and then determines the user customized mode.

[0122] The bubbles in the ultrasound image can be detected in various ways, and the process of detecting the bubbles in the target region of the ultrasound image is described below.

[0123] In some embodiments, the bubbles in the target region of the ultrasound image can be detected in the following way:

[0124] First, the target region in the ultrasound image is segmented to obtain the target region of the ultrasound image.

[0125] In some embodiments, a semantic or instance segmentation method based on deep learning or machine learning can be used to segment the target region in the ultrasound image. In an example, a deep learning-based segmentation method can include the following steps: inputting one or more frames of ultrasound images into a pre-trained deep learning model, and outputting the target region in the ultrasound image. In an embodiment, the deep learning model can be a convolutional neural network (CNN), a neural network based on an attention mechanism (Transformer), a neural network based on a full connection (MLP), a spiking neural network (SNN), or a hybrid model of the above network structures. In another example, a machine learning-based segmentation method can include the following steps: using a machine learning feature extraction and segmentation method to extract and segment the target region in the ultrasound image. In an embodiment, the machine learning feature extraction method can be the deep learning network described above, or an existing feature extraction method such as PCA, HOG, LDA, etc. The machine learning segmentation method can be a threshold method, a Snakes method, a clustering method, a watershed method, a level set method, etc.

[0126] In some embodiments, a deep learning method can also be used to obtain the bounding rectangle of the target region in the ultrasound image, and then the target region in the ultrasound image can be obtained according to the pixel features in the bounding rectangle. In an embodiment, the deep learning method can be based on the deep learning model described above.

[0127] Secondly, the bubbles in the target region of the ultrasound image are segmented to obtain the bubble segmentation result in the target region.

[0128] In some embodiments, a deep learning or machine learning based semantic or instance segmentation method can be used to segment the bubbles in the target region. An exemplary deep learning based segmentation method includes the following steps: inputting one or more ultrasound images into a pre-trained deep learning model to segment the bubbles in the ultrasound images, and obtaining the bubbles in the target region according to the positions of the segmented bubbles, for example, assigning the bubbles to the corresponding heart chamber region or blood vessel lumen region. In an embodiment, the deep learning model can be a convolutional neural network (CNN), a neural network based on attention mechanism (Transformer), a neural network based on full connection (MLP), a spiking neural network (SNN), or a hybrid model of the above network structures. An exemplary machine learning based segmentation method includes the following steps: using a machine learning based feature extraction and segmentation method to extract and segment the bubbles in the ultrasound images, and obtaining the bubbles in the target region according to the positions of the segmented bubbles, for example, assigning the bubbles to the corresponding heart chamber region or blood vessel lumen region. In an embodiment, the machine learning based feature extraction method can be the deep learning network described above, or a conventional feature extraction method such as PCA, HOG, LDA, etc. The machine learning based segmentation method can be a threshold method, a Snakes method, a clustering method, a watershed method, a level set method, etc.

[0129] In some embodiments, a deep learning method is used to obtain a bubble circumscribed rectangle of the target region of the ultrasound image, and then a bubble segmentation result in the target region is obtained according to the pixel features in the circumscribed rectangle. In an embodiment, the deep learning method can be based on the method described above.

[0130] Finally, a bubble detection result is determined according to the bubble segmentation result in the target region.

[0131] In some embodiments, the segmented bubbles in the target region are counted and calculated to obtain the statistical information of the bubbles in the target region, i.e., the bubble detection result.

[0132] Step 403: display the bubble detection result.

[0133] In some embodiments, displaying the bubble detection result can include displaying the ultrasound image and displaying the bubble detection result in or beside the ultrasound image, or separately displaying the bubble detection result. In this embodiment, the display form of the bubble detection result is not limited, which can be displayed in a comment box or in a table form, etc.

[0134] It should be noted that the above bubble detection based on the ultrasound image can be detecting bubbles in one or more target regions in the ultrasound image, the target region can be a region of a target tissue structure in the ultrasound image, such as a blood vessel lumen region, a heart chamber region, etc., the target region can also be the entire imaging region in the ultrasound image, that is, detecting all bubbles in the ultrasound image.

[0135] The principles herein are described with reference to various exemplary embodiments. However, a person of ordinary skill in the art will recognize that changes and modifications can be made to the exemplary embodiments without departing from the scope hereof. For example, various operational steps and components for carrying out the operational steps can be implemented in different sequences and / or omitted, combined, or combined as desired in a particular implementation or to achieve a variety of cost functions associated with operation of the system.

[0136] In the above embodiments, all or part can be implemented by software, hardware, firmware, or any combination thereof. In addition, as understood by those skilled in the art, the principles herein can be reflected in a computer program product on a computer readable storage medium preloaded with computer readable program code. Any tangible, non-transitory computer readable storage medium can be used, including magnetic storage devices (hard disk, floppy disk, etc.), optical storage devices (CD-ROM, DVD, Blu Ray disc, etc.), flash memory, and / or the like. These computer program instructions can be loaded onto a general purpose computer, a special purpose computer, or other programmable data processing apparatus to form a machine, so that these instructions executed on the computer or other programmable data processing apparatus can generate a device that implements the specified functions. These computer program instructions can also be stored in a computer readable storage medium, which can instruct the computer or other programmable data processing apparatus to operate in a specific way, so that the instructions stored in the computer readable storage medium can form a manufactured item, including an implementation device that implements the specified functions. Computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so as to execute a series of operational steps on the computer or other programmable data processing apparatus to generate a computer-implemented process, so that the instructions executed on the computer or other programmable data processing apparatus can provide steps for implementing the specified functions.

[0137] Although the principles herein have been shown in various embodiments, many modifications in structure, arrangement, proportions, elements, materials, and components specially adapted to specific environments and operational requirements can be used without departing from the principles and scope of the present disclosure. The above modifications and other changes or modifications will be included within the scope of the principles herein.

[0138] The foregoing detailed description has been presented for purposes of illustration and description. However, various modifications and changes are possible in the implementation of the disclosure. Accordingly, the disclosure is intended to embrace all modifications and alterations within the scope and spirit of the disclosure. Thus, the scope of the disclosure is not intended to be limited to the particular form set forth herein, but includes all features that might be provided within the scope and spirit of the disclosure. Likewise, the benefits and advantages of the various embodiments, other advantages, and solutions to problems have been presented in the foregoing detailed description. However, the scope of the disclosure should be determined, not with reference to the above description, but should instead be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled. The claims are not intended to include only the disclosure of a preferred embodiment. This is in order to avoid the drawing of an undue breadth of interpretation of the claims and so that suitable modifications can be suggested to those skilled in the art that do not depart from the scope of the claims. It is not intended to be limited in scope or spirit to the exact details shown and described. The terms "comprising," "including," and any other variation thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. Additionally, the term "coupled" and any other variation thereof are intended to cover a physical connection, an electrical connection, a magnetic connection, an optical connection, a communicative connection, a functional connection, and / or any other connection.

[0139] Those skilled in the art will recognize that many modifications can be made to the details of the above-described embodiments without departing from the underlying principles of the present application. The scope of the present application should, therefore, be determined not with reference to the above description, but should instead be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled. The claims are not intended to include only the disclosure of a preferred embodiment. This is in order to avoid drawing an undue breadth of interpretation of the claims and so that suitable modifications can be suggested to those skilled in the art that do not depart from the scope of the claims. It is not intended to be limited in scope or spirit to the exact details shown and described. The terms "comprising," "including," and any other variation thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. Additionally, the term "coupled" and any other variation thereof are intended to cover a physical connection, an electrical connection, a magnetic connection, an optical connection, a communicative connection, a functional connection, and / or any other connection.

Claims

1. An ultrasound apparatus, characterized by, The method comprises: an ultrasonic probe configured to emit ultrasonic waves to a region of interest and receive corresponding echo signals; transmission and reception control circuitry configured to control the ultrasonic probe to emit the ultrasonic waves and receive the corresponding echo signals; a processor configured to: generate an ultrasonic image based on the echo signals; detect bubbles in a target region in the ultrasonic image to determine a bubble detection result; the target region comprises a cardiac chamber region and / or a blood vessel lumen region; determine whether there is a risk of gas embolism according to the bubble detection result, and prompt a user when it is determined that there is a risk of gas embolism.

2. The ultrasonic device of claim 1, wherein, The bubble detection result comprises one or more of the following: the number of bubbles in the target region, the area distribution of bubbles in the target region, the volume distribution of bubbles in the target region, the ratio of the total area of bubbles to the area of the target region, and the ratio of the total volume of bubbles to the volume of the target region.

3. Ultrasonic apparatus according to claim 1 or 2, characterized in that The detection of bubbles in the target region in the ultrasonic image to determine a bubble detection result comprises: segmenting the target region in the ultrasonic image to obtain a target region of the ultrasonic image; segmenting bubbles in the target region of the ultrasonic image to obtain a bubble segmentation result in the target region; determining the bubble detection result according to the bubble segmentation result in the target region.

4. The ultrasonic device of claim 3, wherein, The segmentation of the target region in the ultrasonic image to obtain a target region of the ultrasonic image comprises: segmenting the target region in the ultrasonic image using a semantic segmentation method based on deep learning or machine learning to obtain a target region of the ultrasonic image; or, segmenting the target region in the ultrasonic image using an instance segmentation method based on deep learning or machine learning to obtain a target region of the ultrasonic image; or, obtaining a bounding rectangle of the target region in the ultrasonic image using a deep learning method, and then obtaining the target region of the ultrasonic image according to the pixel features within the bounding rectangle.

5. The ultrasonic device of claim 3, wherein, The segmentation of bubbles in the target region of the ultrasonic image to obtain a bubble segmentation result in the target region comprises: segmenting bubbles in the target region of the ultrasonic image using a semantic segmentation method based on deep learning or machine learning to obtain a bubble segmentation result in the target region; or, segmenting bubbles in the target region of the ultrasonic image using an instance segmentation method based on deep learning or machine learning to obtain a bubble segmentation result in the target region; or, obtaining a bounding rectangle of bubbles in the target region of the ultrasonic image using a deep learning method, and then obtaining the bubble segmentation result in the target region according to the pixel features within the bounding rectangle.

6. The ultrasonic apparatus of claim 1 or 2, wherein, The detection of bubbles in the target region in the ultrasonic image to determine a bubble detection result comprises: inputting the ultrasonic image into a first deep learning model of a pre-constructed bubble-ultrasound image correspondence relationship to obtain a bubble detection result of the target region in the ultrasonic image.

7. The ultrasonic apparatus of claim 1 or 2, wherein, The detection of bubbles in the target region in the ultrasonic image to determine a bubble detection result comprises: inputting the ultrasound image into a second deep learning model of a pre-constructed target region-bubble and ultrasound image correspondence relationship, to obtain each target region in the ultrasound image, a bubble in the ultrasound image, and a target region to which the bubble belongs, so as to determine the bubble detection result.

8. The ultrasonic device of claim 1 or 2, wherein, The determining whether there is a gas embolism risk according to the bubble detection result comprises: If the bubble detection result is greater than or equal to a preset threshold, it is determined that there is a gas embolism risk; otherwise, it is determined that there is no gas embolism risk.

9. The ultrasonic device of claim 1 or 2, wherein, Further comprising: displaying the ultrasound image and marking the detected bubble in the ultrasound image.

10. An ultrasound apparatus, characterized by Comprise: an ultrasound probe configured to emit ultrasound waves to a region of interest and receive corresponding echo signals; transmission and reception control circuitry configured to control the ultrasound probe to emit the ultrasound waves and receive the corresponding echo signals; a processor configured to: generate an ultrasound image based on the echo signals; determine whether there is a gas embolism risk based on the ultrasound image, and prompt a user when it is determined that there is a gas embolism risk.

11. The ultrasonic device of claim 10, wherein, The determining whether there is a gas embolism risk based on the ultrasound image comprises: inputting the ultrasound image into a third deep learning model of a pre-constructed gas embolism risk and ultrasound image correspondence relationship, and outputting a result of whether there is a gas embolism risk.

12. An ultrasound apparatus, characterized by Comprise: an ultrasound probe configured to emit ultrasound waves to a region of interest and receive corresponding echo signals; transmission and reception control circuitry configured to control the ultrasound probe to emit the ultrasound waves and receive the corresponding echo signals; a processor configured to: generate an ultrasound image based on the echo signals; detect bubbles in the ultrasound image to determine a bubble detection result; display the bubble detection result.

13. The ultrasonic device of claim 12, wherein, The bubble detection result at least comprises one or more of the number of bubbles, the area distribution of bubbles, the volume distribution of bubbles, the ratio of the total area of bubbles to the area of the imaging region of the ultrasound image, and the ratio of the total volume of bubbles to the volume of the imaging region of the ultrasound image.

14. An ultrasound imaging method, characterized by, Comprise: controlling the emission of ultrasound waves to a region of interest and the reception of corresponding echo signals, and generating an ultrasound image based on the echo signals; detecting bubbles in a target region of the ultrasound image to determine a bubble detection result; the target region comprises a heart chamber region and / or a blood vessel lumen region; determining whether there is a gas embolism risk according to the bubble detection result, and prompting a user when it is determined that there is a gas embolism risk.

15. The method of claim 14, wherein, The bubble detection result at least comprises one or more of the number of bubbles in the target region, the area distribution of bubbles in the target region, the volume distribution of bubbles in the target region, the ratio of the total area of bubbles to the area of the target region, and the ratio of the total volume of bubbles to the volume of the target region.

16. The method of claim 14 or 15, wherein, Detecting bubbles in a target region of the ultrasound image to determine a bubble detection result comprises: segmenting the target region in the ultrasound image to obtain a target region of the ultrasound image; segmenting bubbles in the target region of the ultrasound image to obtain a bubble segmentation result in the target region; determining the bubble detection result according to the bubble segmentation result in the target region.

17. The method of claim 14 or 15, wherein, The detecting of the bubbles in the target region in the ultrasound image includes: The ultrasound image is input into a first deep learning model of a pre-constructed bubble-ultrasound image correspondence relationship to obtain a bubble detection result of the target region in the ultrasound image.

18. The method of claim 14 or 15, wherein, The detecting of the bubbles in the target region in the ultrasound image includes: The ultrasound image is input into a second deep learning model of a pre-constructed target region-bubble-ultrasound image correspondence relationship to obtain each target region in the ultrasound image, the bubbles in the ultrasound image and the target region to which the bubbles belong, so as to determine the bubble detection result.

19. An ultrasound imaging method, characterized by, It includes: controlling to emit ultrasonic waves to a region of interest and receive corresponding echo signals, and generating an ultrasound image based on the echo signals; judging whether there is a gas embolism risk based on the ultrasound image, and prompting the user when it is judged that there is a gas embolism risk.

20. The method of claim 19, wherein, The judging whether there is a gas embolism risk based on the ultrasound image includes: The ultrasound image is input into a third deep learning model of a pre-constructed gas embolism risk-ultrasound image correspondence relationship to output a result of whether there is a gas embolism risk.

21. An ultrasound imaging method, characterized by, It includes: controlling to emit ultrasonic waves to a region of interest and receive corresponding echo signals, and generating an ultrasound image based on the echo signals; detecting bubbles in the ultrasound image to determine a bubble detection result; displaying the bubble detection result.

22. The method of claim 21, wherein, The bubble detection result at least includes one or more of the number of bubbles, the area distribution of bubbles, the volume distribution of bubbles, the ratio of the total area of bubbles to the area of the imaging region of the ultrasound image, and the ratio of the total volume of bubbles to the volume of the imaging region of the ultrasound image.

23. A computer-readable storage medium, characterized in that, The medium has a program stored thereon, and the program can be executed by the processor to implement the method of any one of claims 14-22.

Images