Method and system for detecting abnormal flow in Doppler ultrasound imaging

The integration of machine learning models in Doppler ultrasound imaging systems for analyzing flow distributions in fetal blood vessels improves diagnostic accuracy by automatically detecting and highlighting abnormal flow patterns, addressing misdiagnosis in obstetric ultrasound examinations.

CN113509203BActive Publication Date: 2024-12-03GE PRECISION HEALTHCARE LLC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202110319797.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2020-04-10
Filing Date
2021-03-25
Publication Date
2024-12-03
Estimated Expiration
2041-03-25

AI Technical Summary

Technical Problem

Existing medical diagnostic ultrasound systems face challenges in accurately detecting abnormal blood flow patterns, particularly in fetal blood vessels, leading to potential misdiagnosis and unnecessary interventions during obstetric ultrasound examinations.

Method used

Implementing a system that utilizes Doppler ultrasound imaging with machine learning models to analyze flow distributions, automatically detecting abnormalities by comparing acquired Doppler measurements against reference flow distributions and highlighting deviations in graphical user interfaces.

Benefits of technology

Enhances the accuracy of diagnosing fetal distress by identifying pathological flow patterns early, reducing the risk of overdiagnosis and unnecessary interventions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113509203B_ABST
    Figure CN113509203B_ABST
Patent Text Reader

Abstract

The present invention is entitled "Methods and Systems for Detecting Abnormal Flow in Doppler Ultrasound Imaging". The present invention provides various methods and systems for ultrasound imaging. In one embodiment, a method includes: acquiring Doppler measurements over a plurality of cardiac cycles via an ultrasound probe; evaluating a flow distribution including the Doppler measurements to detect abnormalities in the flow distribution; and displaying the flow distribution with a reference flow distribution overlaid thereon via a display device. In this way, the attention of a physician evaluating the results from an ultrasound imaging examination can be directed to abnormal flows that are potentially indicative of a pathology, thereby enabling an earlier and more accurate diagnosis of the pathology.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] Embodiments of the subject matter disclosed herein relate to ultrasound imaging, and more particularly to methods and systems for detecting abnormal flow via Doppler ultrasound imaging. Background Art

[0002] Medical diagnostic ultrasound is an imaging modality that uses ultrasound waves to detect the internal structures of a patient's body and produce corresponding images. For example, an ultrasound probe including multiple transducer elements transmits ultrasound pulses, which are reflected or transmitted back, refracted, or absorbed by structures in the body. The ultrasound probe then receives the reflected echoes, which are processed into an image. Such transducer elements of an ultrasound probe typically include electromechanical elements that are capable of converting electrical energy into mechanical energy in order to transmit ultrasound waves into the patient's tissues, and converting the mechanical energy back into electrical energy when the reflected ultrasound waves reach the transducer.

[0003] In addition, during ultrasound of a structure through which blood flows (e.g., the heart, blood vessels, arteries, etc.), Doppler ultrasound can be used to obtain images of blood flow through the blood vessels or structures (color Doppler) and / or blood flow velocity waveforms (continuous or pulsed wave Doppler) for multiple cardiac cycles. The Doppler spectrum from continuous or pulsed wave Doppler can be displayed together with ultrasound images such as B-mode images and color Doppler images via a display device of an ultrasound imaging system. Summary of the invention

[0004] In one embodiment, a method includes: acquiring Doppler measurements over a plurality of cardiac cycles via an ultrasound probe; evaluating a flow distribution including the Doppler measurements to detect abnormalities in the flow distribution; and displaying the flow distribution with a reference flow distribution overlaid thereon via a display device. In this way, the attention of a physician evaluating results from an ultrasound imaging examination can be directed to abnormal flows potentially indicative of a pathology, thereby enabling an earlier and more accurate diagnosis of the pathology.

[0005] It should be understood that the above brief description is provided to introduce in a simplified form selected concepts that are further described in the detailed description. It is not meant to identify key or essential features of the claimed subject matter, the scope of which is solely defined by the claims that follow the detailed description. Furthermore, the claimed subject matter is not limited to implementations that solve any disadvantages noted above or in any part of this disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] The present invention will be better understood by reading the following description of non-limiting embodiments with reference to the accompanying drawings, in which:

[0007] Figure 1A block schematic diagram illustrating an exemplary ultrasound imaging system is shown according to one embodiment;

[0008] Figure 2 is a schematic diagram illustrating an image processing system for detecting and classifying abnormalities in Doppler ultrasound measurements according to an embodiment;

[0009] Figure 3 A high-level flow chart illustrating an exemplary method for ultrasound imaging using automated analysis of Doppler flow patterns is shown, according to one embodiment;

[0010] Figure 4 shows an exemplary display output of a graphical user interface showing Doppler flow distribution of a blood vessel with normal flow according to one embodiment;

[0011] Figure 5 shows an exemplary display output of a graphical user interface showing Doppler flow distribution with a pathological flow vessel according to one embodiment;

[0012] Figure 6 An exemplary display output of a graphical user interface showing a normal flow distribution overlaid thereon is shown according to one embodiment. Figure 5 The pathological Doppler flow distribution;

[0013] Figure 7 An exemplary display output of a graphical user interface according to one embodiment is shown. Figure 5 a pathological Doppler flow profile, wherein a normal flow profile is adjusted to match the pathological flow profile and overlaid thereon;

[0014] Figure 8 An exemplary display output of a graphical user interface according to one embodiment is shown. Figure 5 a pathological Doppler flow distribution of the embodiment, wherein the display is adjusted to highlight differences from a normal flow distribution aligned to match the pathological flow distribution and overlaid thereon;

[0015] Fig. 9 An exemplary display output of a graphical user interface according to one embodiment is shown. Figure 5 a pathological Doppler flow distribution of the invention, wherein a normal flow distribution is aligned and scaled to match the pathological flow distribution and overlaid thereon, and has a visual indicator of a difference from the normal flow distribution overlaid thereon;

[0016] Fig.10shows an exemplary display output of a graphical user interface showing another Doppler flow profile of a vessel having pathological flow according to one embodiment;

[0017] Fig.11 An exemplary display output of a graphical user interface showing a normal flow distribution overlaid thereon is shown according to one embodiment. Fig.10 The pathological Doppler flow distribution; and

[0018] Fig.12 A high-level flow chart illustrating an exemplary method for ultrasound imaging with automated analysis of Doppler flow patterns during follow-up examinations is shown, according to one embodiment. DETAILED DESCRIPTION

[0019] The following description relates to various embodiments of ultrasound imaging. Specifically, the present invention discloses systems and methods for detecting flow anomalies in Doppler ultrasound imaging. Figure 1 Flow abnormalities are detected in the Doppler spectrum generated by Doppler ultrasound data collected by an ultrasound imaging system (such as the ultrasound imaging system shown). In obstetric ultrasound examinations, fetal organ screening is among the most widely used applications. The main part of the examination is the Doppler ultrasound analysis of fetal and maternal blood vessels (such as the maternal uterine artery, fetal umbilical artery, fetal aorta, fetal middle cerebral artery and fetal ductus vein). If the fetal health is disturbed, for example due to insufficient supply of placental oxygen and nutrients, a fetal distress cascade may occur, in which over time, the flow is first disturbed at the maternal uterine artery, then at the fetal umbilical artery, and so on. When the flow is disturbed at the fetal middle cerebral artery, the fetal condition is dangerous. The fetal heart rate is usually monitored, and abnormalities in the fetal heart rate are used to provide a diagnosis of fetal distress. However, potential misinterpretation of the fetal heart rate may lead to overdiagnosis of fetal distress and unnecessary intervention by emergency cesarean section. Each blood vessel has a very specific flow distribution or Doppler spectrum, so specific measurements of velocity and certain indices can be used to classify and detect hemodynamic abnormalities. For example, an exemplary image processing system (such as Figure 2 The image processing system shown in FIG. 1 includes one or more machine learning models that are trained to automatically detect abnormalities in the flow distribution of a vessel of interest. Figure 3 The method shown may include evaluating a Doppler spectrum or flow distribution using a machine learning model during Doppler imaging of a vessel, and displaying an indication of pathological flow when an anomaly is detected by the machine learning model. Pathological flow may be indicated to a user of the ultrasound imaging system, for example, by displaying a nominal or normal flow distribution over the pathological flow distribution. For example, Figure 4 shows normal flow distribution in the uterine artery as displayed in a graphical user interface for ultrasound imaging, and Figure 5 Pathological flow distribution of the uterine artery is shown. When such pathological flow distribution is detected, the abnormality can be highlighted or indicated to the user through the graphical user interface. For example, Figure 6 As shown, the normal flow distribution of the same blood vessel can be displayed in a graphical user interface, for example, by superimposing the normal distribution on the pathological flow distribution. In other examples, the normal flow distribution can be adjusted, for example, by time-aligning the normal flow distribution so that the cycle peak is aligned with the peak of the pathological flow distribution, such as Figure 7 In addition, in some examples, the graphical user interface may specifically highlight or indicate the portion of the pathological flow distribution that deviates from the normal flow distribution, such as Figure 8 In some examples, such as Fig. 9 As shown, the normal flow distribution can be time aligned and scaled to further match the abnormal flow distribution so as to emphasize the differences between the flow distributions. Another example of pathological flow in the uterine artery and an exemplary indication of such pathological flow is shown in Fig.10 and Fig.11 In addition, Doppler measurements from a previous ultrasound scan of a patient can be retrieved during a subsequent ultrasound examination, such as Fig.12 shown.

[0020] See also Figure 1 , a schematic diagram of an ultrasound imaging system 100 according to an embodiment of the present disclosure is shown. The ultrasound imaging system 100 includes a transmit beamformer 101 and a transmitter 102, which drives an element (e.g., a transducer element) 104 within a transducer array (referred to herein as a probe 106) to transmit a pulsed ultrasound signal (referred to herein as a transmit pulse) into a body (not shown). According to one embodiment, the probe 106 can be a one-dimensional transducer array probe. However, in some embodiments, the probe 106 can be a two-dimensional matrix transducer array probe. As further explained below, the transducer element 104 can be made of a piezoelectric material. When a voltage is applied to a piezoelectric crystal, the crystal physically expands and contracts, thereby emitting an ultrasonic ball wave. In this way, the transducer element 104 can convert an electronic transmit signal into an acoustic transmit beam.

[0021] After the elements 104 of the probe 106 transmit the pulsed ultrasound signals into the body (of the patient), the pulsed ultrasound signals are scattered back from structures inside the body (such as blood cells or muscle tissue) to produce echoes that return to the elements 104. The echoes are converted into electrical signals or ultrasound data by the elements 104, and the electrical signals are received by the receiver 108. The electrical signals representing the received echoes pass through the receive beamformer 110 which outputs the ultrasound data. In addition, the transducer elements 104 may generate one or more ultrasound pulses based on the received echoes to form one or more transmit beams.

[0022] According to some embodiments, the probe 106 may include electronic circuits to perform all or part of the transmit beamforming and / or receive beamforming. For example, all or part of the transmit beamformer 101, the transmitter 102, the receiver 108, and the receive beamformer 110 may be located within the probe 106. In the present disclosure, the term "scan" or "scanning" may also be used to refer to the acquisition of data by the process of transmitting and receiving ultrasound signals. In the present disclosure, the term "data" may be used to refer to one or more data sets acquired with an ultrasound imaging system. In one embodiment, the data acquired via the ultrasound system 100 may be used to train a machine learning model. The user interface 115 may be used to control the operation of the ultrasound imaging system 100, including input for controlling patient data (e.g., patient medical history), for changing scanning or display parameters, for starting a probe repolarization sequence, etc. The user interface 115 may include one or more of the following: a rotating element, a mouse, a keyboard, a trackball, hard keys linked to specific actions, soft keys that can be configured to control different functions, and a graphical user interface displayed on a display device 118.

[0023] The ultrasound imaging system 100 also includes a processor 116, which is used to control the transmit beamformer 101, the transmitter 102, the receiver 108, and the receive beamformer 110. The processor 116 is in electronic communication (e.g., communicatively connected) with the probe 106. For the purpose of this disclosure, the term "electronic communication" may be defined as including both wired communication and wireless communication. The processor 116 can control the probe 106 to collect data according to instructions stored in the processor's memory and / or the memory 120. The processor 116 controls which of the elements 104 are active and the shape of the beam emitted from the probe 106. The processor 116 also communicates electronically with the display device 118, and the processor 116 can process data (e.g., ultrasound data) into an image for display on the display device 118. The processor 116 may include a central processing unit (CPU) according to one embodiment. According to other embodiments, the processor 116 may include other electronic components capable of performing processing functions, such as a digital signal processor, a field programmable gate array (FPGA), or a graphics board. According to other embodiments, the processor 116 may include a plurality of electronic components capable of performing processing functions. For example, the processor 116 may include two or more electronic components selected from a list of electronic components, including: a central processing unit, a digital signal processor, a field programmable gate array, and a graphics board. According to another embodiment, the processor 116 may also include a composite demodulator (not shown) that demodulates the RF data and generates raw data. In another embodiment, the demodulation may be performed earlier in the processing chain. The processor 116 is adapted to perform one or more processing operations according to a plurality of optional ultrasound modalities on the data. In one example, the data may be processed in real time during a scanning session because the echo signal is received by the receiver 108 and transmitted to the processor 116. For the purposes of this disclosure, the term "real time" is defined as including a process performed without any intentional delay. For example, an embodiment may acquire images at a real-time rate of 7 to 20 frames / second. The ultrasound imaging system 100 is capable of acquiring 2D data of one or more planes at a significantly faster rate. However, it should be understood that the real-time frame rate may depend on the length of time it takes to acquire each frame of data for display. Therefore, when a relatively large amount of data is acquired, the real-time frame rate may be slower. Thus, some embodiments may have real-time frame rates significantly faster than 20 frames / second, while other embodiments may have real-time frame rates less than 7 frames / second. Data may be temporarily stored in a buffer (not shown) during a scanning session and processed in a less real-time manner in real-time or off-line operation. Some embodiments of the present invention may include multiple processors (not shown) to handle processing tasks handled by processor 116 according to the exemplary embodiments described above.For example, a first processor may be utilized to demodulate and extract the RF signal prior to displaying the image, while a second processor may be utilized to further process the data (eg, by augmenting the data as further described herein). It will be appreciated that other embodiments may utilize different processor arrangements.

[0024] The ultrasound imaging system 100 can continuously acquire data at a frame rate of, for example, 10 Hz to 30 Hz (e.g., 10 to 30 frames per second). The image generated based on the data can be refreshed at a similar frame rate on the display device 118. Other embodiments can acquire and display data at different rates. For example, depending on the size of the frame and the intended application, some embodiments can acquire data at a frame rate less than 10 Hz or greater than 30 Hz. A memory 120 is included for storing frames of processed acquisition data. In an exemplary embodiment, the memory 120 has sufficient capacity to store at least several seconds of ultrasound data frames. The data frames are stored in a manner that facilitates retrieval according to their acquisition order or time. The memory 120 may include any known data storage medium.

[0025] In various embodiments of the present invention, the processor 116 can process data through different mode-related modules (e.g., B-mode, color Doppler, M-mode, color M-mode, spectral Doppler, elastic imaging, TVI, strain, strain rate, etc.) to form 2D or 3D data. For example, one or more modules can generate B-mode, color Doppler, M-mode, color M-mode, spectral Doppler, elastic imaging, TVI, strain, strain rate, and combinations thereof, etc. As an example, one or more modules can process color Doppler data, which may include traditional color flow Doppler, power Doppler, HD flow, slow flow HD, etc. The image lines and / or frames are stored in the memory and may include timing information indicating the time when the image lines and / or frames are stored in the memory. These modules may include, for example, a scan conversion module for performing a scan conversion operation to convert the acquired image from beam space coordinates to display space coordinates. A video processor module may be provided that reads the acquired image from the memory and displays the image in real time while a procedure (e.g., ultrasound imaging) is performed on the patient. The video processor module may include a separate image memory, and the ultrasound image may be written to the image memory for reading and display by the display device 118 .

[0026] In one example, the ultrasound imaging system 100 may be configured to operate in a spectral Doppler mode, wherein the ultrasound imaging system 100 obtains a pulsed wave Doppler spectrum or a continuous wave Doppler spectrum (referred to herein as a Doppler spectrum or flow distribution) including blood flow velocity data for a plurality of cardiac cycles (e.g., beats). An example of a Doppler spectrum displayed via the display device 118 of the ultrasound imaging system 100 is shown in FIG. Figure 4-Figure 11As shown in , and described further below.

[0027] In various embodiments of the present disclosure, one or more components of the ultrasound imaging system 100 may be included in a portable handheld ultrasound imaging device. For example, the display device 118 and the user interface 115 may be integrated into the external surface of the handheld ultrasound imaging device, which may also include a processor 116 and a memory 120. In other examples of the present disclosure, the display device 118 and / or the user interface 115 may be integrated into a remote computing system located locally, such as a viewing workstation or a remotely located remote computing system (e.g., for telemedicine). The probe 106 may include a handheld probe that electronically communicates with the handheld ultrasound imaging device to collect raw ultrasound data. The transmit beamformer 101, the transmitter 102, the receiver 108, and the receive beamformer 110 may be included in the same or different parts of the ultrasound imaging system 100. For example, the transmit beamformer 101, the transmitter 102, the receiver 108, and the receive beamformer 110 may be included in a handheld ultrasound imaging device, a probe, and a combination thereof.

[0028] After performing a two-dimensional ultrasound scan, a data block containing the scan lines and their samples is generated. After applying the back-end filter, a process called scan conversion is performed to transform the two-dimensional data block into a displayable bitmap image with additional scanning information (such as depth, the angle of each scan line, etc.). During scan conversion, interpolation techniques are applied to fill in the missing holes (i.e., pixels) in the resulting image. These missing pixels occur because each element of the two-dimensional block should generally cover many pixels in the resulting image. For example, in current ultrasound imaging systems, bicubic interpolation is applied, which utilizes adjacent elements of the two-dimensional block. Therefore, if the two-dimensional block is relatively small compared to the size of the bitmap image, the scan-converted image will include areas of poor or low resolution, especially for areas with greater depth.

[0029] See also Figure 2 , shows an exemplary medical image processing system 200. In some embodiments, the medical image processing system 200 is incorporated into a medical imaging system such as an ultrasound imaging system (e.g., Figure 1In some embodiments, at least a portion of the medical image processing system 200 is disposed at a device (e.g., an edge device or a server) that is communicatively coupled to the medical imaging system via a wired and / or wireless connection. In some embodiments, the medical image processing system 200 is disposed at a separate device (e.g., a workstation) that can receive images from the medical imaging system or from a storage device that stores images generated by the medical imaging system. The medical image processing system 200 may include an image processor 231, a user input device 232, and a display device 233. For example, the image processor 231 may be operably / communicatively coupled to the user input device 232 and the display device 233.

[0030] The image processor 231 includes a processor 204 configured to execute machine-readable instructions stored in a non-transitory memory 206. The processor 204 may be single-core or multi-core, and the program executed by the processor 204 may be configured for parallel or distributed processing. In some embodiments, the processor 204 may optionally include separate components distributed in two or more devices, which may be remotely located and / or configured for coordinated processing. In some embodiments, one or more aspects of the processor 204 may be virtualized and executed by a networked computing device capable of remote access configured in a cloud computing configuration. In some embodiments, the processor 204 may include other electronic components capable of performing processing functions, such as a digital signal processor, a field programmable gate array (FPGA), or a graphics board. In some embodiments, the processor 204 may include multiple electronic components capable of performing processing functions. For example, the processor 204 may include two or more electronic components selected from a plurality of possible electronic components, including: a central processing unit, a digital signal processor, a field programmable gate array, and a graphics board. In other embodiments, the processor 204 may be configured as a graphics processing unit (GPU), including a parallel computing architecture and parallel processing capabilities.

[0031] exist Figure 2In the illustrated embodiment, the non-transitory memory 206 stores a detection module 212 and medical image data 214. The medical image data 214 may include ultrasound imaging data, including but not limited to B-mode ultrasound image data, color Doppler ultrasound image data, and Doppler spectrum or flow distribution acquired via an ultrasound imaging system (such as ultrasound imaging system 100). The detection module 212 includes one or more artificial intelligence algorithms, including machine learning models, to process input medical images from the medical image data 214. Specifically, the detection module 212 may provide an artificial intelligence system for identifying patient abnormalities within the medical image data 214, and specifically for identifying abnormal flow or pathological flow in the Doppler spectrum or flow distribution of the medical image data 214. For example, the detection module 212 may include one or more deep learning networks, the one or more deep learning networks including multiple weights and biases, activation functions, loss functions, gradient descent algorithms, and instructions for implementing the one or more deep learning networks to process input medical images. Additionally or alternatively, the detection module 212 may store instructions for implementing a neural network, such as a convolutional neural network, for detecting and classifying potential flow anomalies in the Doppler spectrum or flow distribution of the medical image data 214. In such examples, the flow distribution may be input to the neural network as an image. Alternatively, a waveform depicted in the flow distribution may be input to the neural network as a time series. In such examples, the machine learning model or neural network may be constructed and configured to receive a time series instead of an image as input.

[0032] The detection module 212 may include a trained neural network and may also include training routines or parameters (e.g., weights and biases) associated with one or more neural network models stored therein. The detection module 212 may also include image recognition algorithms, shape or edge detection algorithms, anatomical structure segmentation and classification algorithms, etc. In some embodiments, the detection module 212 may evaluate the medical image data 214 as it is acquired in real time. Additionally or alternatively, the detection module 212 may evaluate the medical image data 214 offline rather than in real time. The detection module 212 may also include a trained machine learning model for automatic vessel segmentation and classification, and a trained machine learning model or algorithm for placing an automatic Doppler cursor within a vessel for data acquisition.

[0033] In some embodiments, the patient abnormality may include an abnormal flow or pathology in an anatomical feature imaged in the medical image data 214. For example, when the anatomical feature is a blood vessel, the patient abnormality includes an abnormal flow relative to the nominal flow of the blood vessel. In some examples, the detection module 212 includes multiple machine learning models for multiple blood vessels, where each machine learning model is trained on the flow distribution for a specific blood vessel. For example, the detection module 212 may include a machine learning model trained on the flow distribution of the uterine artery, a machine learning model trained on the flow distribution of the middle cerebral artery, a machine learning model trained on the flow distribution of the umbilical artery, and the like. In other examples, the detection module 212 may include a machine learning model trained on the flow distribution of multiple blood vessels.

[0034] The image processor 231 is communicatively coupled to the training module 210, which includes instructions for training one or more machine learning models stored in the detection module 212. The training module 210 may include instructions that, when executed by the processor, cause the processor to build a model (e.g., a mathematical model) based on sample data to make predictions or decisions about the detection and classification of patient abnormalities without explicit programming of traditional algorithms that do not utilize machine learning. In one example, the training module 210 includes instructions for receiving a training data set from the medical image data 214. The training data set includes a set of Doppler flow distributions, associated ground truth labels, and associated model outputs for training one or more of the machine learning models stored in the detection module 212. The training module 210 may receive Doppler flow distributions, associated ground truth labels, and associated model outputs for training one or more machine learning models from other sources other than the medical image data 214 (such as other image processing systems, the cloud, etc.). In some embodiments, one or more aspects of the training module 210 may include a remotely accessible networked storage device configured in a cloud computing configuration. Furthermore, in some embodiments, the training model 210 is included in the non-transitory memory 206. Additionally or alternatively, in some embodiments, the training model 210 can be used to generate the detection module 212 offline and remote from the image processing system 200. In such embodiments, the training module 210 may not be included in the image processing system 200, but may generate data stored in the image processing system 200.

[0035] The non-transitory memory 206 also stores medical image data 214. The medical image data 214 includes, for example, functional images and / or anatomical images captured by an imaging modality such as an ultrasound imaging system, an MRI system, a CT system, a PET system, etc. As an example, the medical image data 214 may include ultrasound images, such as ultrasound B-mode images, color Doppler image data, and Doppler flow distribution.

[0036] In some embodiments, non-transitory storage 206 may include components disposed on two or more devices that may be remotely located and / or configured for coordinated processing. In some embodiments, one or more aspects of non-transitory storage 206 may include a remotely accessible networked storage device configured with cloud computing. As an example, non-transitory storage 206 may be part of a picture archiving and communication system (PACS) that is configured to store, for example, patient medical history, imaging data, test results, diagnostic information, management information, and / or scheduling information.

[0037] The image processing system 200 may also include a user input device 232. The user input device 232 may include one or more of a touch screen, a keyboard, a mouse, a trackpad, a motion sensing camera, or other devices configured to enable a user to interact with and manipulate data stored within the image processor 231. As an example, the user input device 232 may enable a user to analyze and sort imaged structures and / or respond to notification prompts.

[0038] The display device 233 may include one or more display devices utilizing any type of display technology. In some embodiments, the display device 233 may include a computer monitor and may display unprocessed images, processed images, parameter mapping diagrams, and / or inspection reports. The display device 233 may be combined with the processor 204, the non-transient memory 206, and / or the user input device 232 in a shared housing, or may be a peripheral display device. The display device 233 may include a monitor, a touch screen, a projector, or another type of display device that enables a user to view medical images and / or interact with various data stored in the non-transient memory 206. In some embodiments, the display device 233 may be included in a smart phone, a tablet computer, a smart watch, etc.

[0039] Understandably, Figure 2 The illustrated medical image processing system 200 is one non-limiting embodiment of an image processing system, and other imaging processing systems may include more, fewer, or different components without departing from the scope of the present disclosure. Additionally, in some embodiments, at least a portion of the medical image processing system 200 may include Figure 1The medical image processing system 200 may include at least a portion of the ultrasound imaging system 100, or vice versa (e.g., at least a portion of the ultrasound imaging system 100 may be included in the medical image processing system 200). In addition, although the medical image processing system 200 may include an image processing system separate from a medical imaging system (such as the ultrasound imaging system 100), in some examples, the medical image processing system 200 may be integrated into the ultrasound imaging system 100. For example, the processor 204 and the non-transitory memory 206 may include the processor 116 and the memory 120, respectively, and the user input device 232 and the display device 233 include the user interface 115 and the display device 118, respectively.

[0040] As used herein, the terms "system" and "module" may include hardware and / or software systems that operate to perform one or more functions. For example, a module or system may include, or be included in, a computer processor, a controller, or other logic-based device that performs operations based on instructions stored on a tangible and non-transitory computer-readable storage medium (such as a computer memory). Alternatively, a module or system may include a hard-wired device that performs operations based on the hard-wired logic of the device. The various modules or systems shown in the figures may represent hardware that operates based on software or hard-wired instructions, software that instructs hardware to perform operations, or a combination thereof.

[0041] A "system" or "module" may include or represent hardware and related instructions (e.g., software stored on a tangible and non-transitory computer-readable storage medium (such as a computer hard drive, ROM, RAM, etc.) that perform one or more operations described herein. The hardware may include electronic circuitry that includes and / or is connected to one or more logic-based devices, such as microprocessors, processors, controllers, etc. These devices may be off-the-shelf devices that are appropriately programmed or instructed to perform the operations described herein in accordance with the instructions described above. Additionally or alternatively, one or more of these devices may be hardwired with logic circuitry to perform these operations.

[0042] Figure 3 A high-level flow chart illustrating an exemplary method 300 for ultrasound imaging using automatic analysis of Doppler flow patterns is shown, according to one embodiment. In particular, the method 300 involves evaluating ultrasound Doppler measurements to detect abnormal blood flow in a blood vessel. Figure 1 and Figure 2 The method 300 is described with reference to the systems and components of the present disclosure, but it should be understood that the method 300 can be implemented with other systems and components without departing from the scope of the present disclosure. The method 300 can be implemented, for example, as executable instructions in the non-transitory memory 120 and / or the non-transitory memory 206, and can be executed by the processor 116 and / or the processor 204 to perform the actions described below.

[0043] The method 300 starts at 305. At 305, the method 300 starts ultrasound scanning of a blood vessel. For example, the method 300 controls the transmit beamformer 101 and the transmitter 102 to drive the transducer element 104 of the probe 106 to transmit ultrasound toward the blood vessel into the patient's body, and also controls the receiver 108 and the receive beamformer 110 to detect the echo of the ultrasound via the transducer element 104 of the probe 106, as described above. Therefore, the method 300 can control the components of the ultrasound imaging system 100 according to the Doppler ultrasound imaging protocol. In this way, at 310, the method 300 acquires a B-mode image or a grayscale ultrasound image of the blood vessel.

[0044] At 315, the method 300 receives an indication of an imaged vessel. For example, the method 300 may receive an indication of an imaged vessel via the user interface 115 and / or the user input device 232. That is, a user of the ultrasound imaging system 100 may manually select or otherwise input an imaged vessel, for example, through a menu, and thus the method 300 receives an indication of an imaged vessel. In such an example, the method 300 may receive an indication of an imaged vessel before starting the ultrasound scan at 305. As another example, the method 300 may perform image classification of the B-mode image or ultrasound image acquired at 310 to automatically detect the vessel, and thus the method 300 may receive an indication of the imaged vessel via image classification. In such an example, the method 300 may perform image classification using a trained convolutional neural network or another suitable image classification technique as an illustrative example, such as segmenting the vessel in the ultrasound image and comparing the segmented vessel to a dictionary of vessel segments to identify the vessel. As further described herein, the method 300 also acquires Doppler measurements, and thus in some examples, the method 300 can perform automatic image classification on the color or power Doppler images or on the spectral Doppler images themselves. In addition, the imaged vessels can be identified using both machine learning models trained for segmentation and classification. For example, the vessels can be segmented from the image data using a segmentation machine learning model, and the segmented vessels can be classified using a classification machine learning model, or alternatively, a single machine learning model can both segment and classify the vessels.

[0045] At 320, the method 300 acquires Doppler measurements. For example, the method 300 may control the components of the ultrasound imaging system 100 as described above to operate in a spectral Doppler mode and acquire blood flow velocity data and a Doppler spectrum or flow distribution for at least several heartbeats or cardiac cycles. The Doppler cursor (i.e., the location of the Doppler ROI or the area imaged according to the Doppler imaging mode) may be automatically positioned according to the indication of the imaged vessel. For example, the method 300 may automatically identify the location of the imaged vessel at 315 and automatically position the Doppler cursor at the location of the vessel for Doppler data acquisition. In addition, the method 300 may calculate various measurements or indices based on the acquired Doppler data. For example, the Doppler measurement results may also include calculating a resistance index (RI) and / or a pulsatility index (PI) from the blood flow velocity data and / or the Doppler spectrum, wherein the resistance index comprises the difference between the peak systolic blood flow velocity and the end-diastolic blood flow velocity divided by the peak systolic blood flow velocity, and the pulsatility index comprises the difference between the peak systolic blood flow velocity and the end-diastolic blood flow velocity divided by the time-averaged velocity.

[0046] After acquiring the Doppler measurements, the method 300 continues to 325. At 325, the method 300 displays the ultrasound image with the Doppler measurements. For example, the method 300 may display the blood flow velocity data as color flow image data superimposed on the ultrasound B-mode image. In addition, at 330, the method 300 displays the flow distribution acquired at 320. The flow distribution or Doppler spectrum may be displayed adjacent to the ultrasound B-mode image, with the color flow image data superimposed thereon displayed in a graphical user interface displayed, for example, via the display device 118 or the display device 233. The method 300 may also display additional Doppler measurements, such as the calculated RI and / or PI, in the graphical user interface.

[0047] At 335, the method 300 evaluates the flow distribution. In one example, the method 300 may utilize a machine learning model to evaluate the flow distribution. For example, the method 300 may input one or more of the flow distribution, additional Doppler measurements (e.g., RI and / or PI), and the B-mode image into one or more machine learning models, such as the detection module 212. The machine learning model is trained on a plurality of normal or nominal flow distributions for each vessel. In some examples, the indication of the imaged vessel may be used to evaluate the flow distribution. For example, in some examples, the indication of the imaged vessel may be input together with the flow distribution into a machine learning model, such as one trained on normal flow distributions for a plurality of vessels. In such examples, the machine learning model may also output an indication of the imaged vessel based on the flow distribution. In this way, if the actual imaged vessel is different from the vessel indicated in the indication of the imaged vessel received at 315, the method 300 may identify an error based on a comparison of the indication of the vessel output by the machine learning model with the indication of the vessel received at 315. In such examples, as an example, the method 300 may optionally display a notification requesting confirmation of the imaged vessel. In other examples, method 300 may select a machine learning model trained on a normal flow distribution of an imaged vessel from a plurality of machine learning models, where each machine learning model is trained on a normal flow distribution of a different vessel.

[0048] The machine learning model may be configured to output an indication of whether the traffic distribution is normal, or similarly, to output an indication of a traffic distribution that is abnormal or anomalous relative to a normal traffic distribution. In some examples, the machine learning model may also be configured to output an identification or indication of a portion of the traffic distribution that is abnormal or anomalous relative to a normal traffic distribution.

[0049] In some examples, the machine learning model is trained on pathological flow distributions and normal flow distributions. Therefore, the training data used to train the machine learning model may include a normal flow distribution marked as normal and a pathological flow distribution marked as pathological. In addition, for different types of pathological flow distributions, the training data may also include an indication of the type of pathology presented by the pathological flow distribution. That is, various pathologies may be associated with a given blood vessel and may cause different types of flow, and the machine learning model may be trained on training data that includes multiple abnormal flow distributions associated with different pathologies and labels of pathologies that cause abnormal flow. In such examples, the machine learning model may be configured to output an indication of the type of pathology presented by the pathological flow distribution when the machine learning model outputs an indication that the flow distribution is pathological or abnormal. In addition, the flow distribution may be characterized or graded according to a flow grading system. For example, the flow distribution may be graded as absent (e.g., no flow), minimal, attenuated, suppressed, stenotic, or normal. Abnormal flow distributions with different grades may be labeled with grades used to train the machine learning model so that the machine learning model may output grades for the flow distribution input to the machine learning model.

[0050] In other examples, the method 300 can utilize signal processing methods other than machine learning models for signal analysis to evaluate traffic distribution. For example, the method 300 can perform time series analysis, Fourier analysis, wavelet-based analysis, combinations thereof, etc. to identify anomalies and / or classify traffic distribution.

[0051] After evaluating the flow distribution, the method 300 continues to 340. At 340, the method 300 determines whether an anomaly is detected in the flow distribution. If no anomaly is detected ("No"), the method 300 continues to 355. However, if an anomaly is detected ("Yes"), the method 300 continues to 345. At 345, the method 300 displays the normal flow distribution of the vessel. For example, the method 300 may retrieve the normal flow distribution of the vessel from a library of normal flow distributions (such as, for example, the flow distribution stored in the training module 210). In addition, the normal flow distribution may be selected from a plurality of normal flow distributions of the vessel stored in the library of normal flow distributions, for example, based on acquisition parameters, so that the normal flow distribution corresponds to the flow that can be expected under the given acquisition. In some examples, the normal flow distribution may be displayed adjacent to the flow distribution. In other examples, the normal flow distribution is superimposed on the flow distribution in the graphical user interface. In this way, the user can easily identify the difference between the flow distribution and the normal flow distribution. In some examples, the normal flow distribution may be adjusted based on the flow distribution before superimposing the normal flow distribution. For example, the normal flow distribution may be time aligned such that at least one peak (eg, an R-wave) of the normal flow distribution is aligned with a peak of the flow distribution. Figure 7An example of adjusting the normal flow distribution to be time-aligned with the flow distribution is further described. In addition, the normal flow distribution can be scaled, for example, so that the amplitude of at least one peak that is time-aligned with the peak of the flow distribution is also equal to the amplitude of the peak of the flow distribution. Fig. 9 An example of adjusting normal traffic distribution in this manner is further described.

[0052] At 350, the method 300 displays an indication of a pathology. For example, the method 300 may highlight an abnormal portion of the flow distribution, such as by adjusting the display of the flow distribution at the abnormal portion. As another example, the method 300 may graphically highlight the difference between the flow distribution and the normal flow distribution. For example, the method 300 may automatically highlight the portion of the flow distribution that is different from the normal flow distribution. As another example, the method 300 may color the normal portion of the flow distribution as a first color (e.g., green), and color the different portions with a second color (e.g., red). As yet another example, the method 300 may display an indication of the type of pathology that causes the flow distribution to be abnormal. In an example in which the machine learning model is configured to classify the flow distribution according to a flow grading system, the method 300 may also display a classification or flow grade of the flow distribution (e.g., absent, minimal, weakened, suppressed, narrow, normal, etc.) in a graphical user interface.

[0053] At 355, the method 300 determines whether the scan is complete. For example, the scan may be complete when the user of the ultrasound imaging system 100 indicates via the user interface 115 that the scan is complete. If the scan is not complete ("No"), the method 300 continues to 360 to continue scanning. In order to continue scanning, the method 300 returns to 310, for example, to acquire additional B-mode images and continue to acquire Doppler measurements. In some examples, the method 300 may alternatively return to 320 to acquire additional Doppler measurements without acquiring additional B-mode images. Therefore, the method 300 continues to acquire Doppler measurements and evaluates the flow distribution or Doppler spectrum to determine whether there is pathological flow until the scan is completed. Once the scan is completed at 355 ("Yes"), the method 300 proceeds to 365. At 365, the method 300 ends the scan. Then, the method 300 returns.

[0054] Therefore, a system and method for automatically detecting abnormal flow distribution and indicating abnormal flow distribution is provided. Although the method 300 described above involves evaluating Doppler measurements in real time during ultrasound scanning, it should be understood that the method 300 may be suitable for use as a post-processing technique, for example, after collecting ultrasound data including Doppler measurements. For example, instead of collecting and evaluating ultrasound data in real time during ultrasound scanning, ultrasound data including Doppler measurements may first be collected during ultrasound scanning and then stored in a non-transitory memory (e.g., as medical image data 214 in the non-transitory memory 206 of the medical image processing system 200). The machine learning model as described above may then be used to evaluate the ultrasound data including Doppler measurements to determine whether the flow is pathological or abnormal, and the Doppler spectrum or flow distribution may be displayed together with the normal flow distribution superimposed thereon. In addition, indications of abnormalities in the flow distribution may be highlighted in the display of the flow distribution. Additionally, when a user executes a Scan Assistant protocol or script to perform a patient exam, the user can configure when an image undergoes automatic detection of flow hemodynamic abnormalities and what additional actions, if any, the system may automatically perform in the event of an abnormality (e.g., additional measurements, etc.). The detected abnormal findings can then be automatically populated for the user as comments / annotations and automatically populated into the reporting software.

[0055] As an illustrative and non-limiting example of how abnormal traffic distribution may be indicated to a user, Figure 4-Figure 11 An exemplary graphical user interface for displaying ultrasound data including a Doppler spectrum is shown. The graphical user interface described herein may be displayed, for example, via the display device 118 of the ultrasound imaging system 100 or via the display device 233 of the medical image processing system 200. As an illustrative and non-limiting example, Figure 4 An exemplary display output of a graphical user interface 400 is shown, which shows a Doppler spectrum or flow distribution 410 of a blood vessel with normal flow. The graphical user interface 400 displays an ultrasound image 405 that includes a grayscale or B-mode image with color flow image data (e.g., velocity measurements) superimposed thereon in a color flow-related region (ROI) 409. The graphical user interface 400 may also include a color scale 407 positioned adjacent to the ultrasound image 405 to show the velocity of the color depicted in the color flow image data of the ultrasound image 405. The graphical user interface 400 also displays a Doppler spectrum or flow distribution 410 of a blood vessel positioned in the color flow ROI 409. The graphical user interface 400 also displays a flow distribution trace 420 of the flow distribution 410 that may be automatically generated based on the flow distribution 410.

[0056] The graphical user interface 400 also includes a measurement parameter display 412 that displays one or more Doppler measurements and other measurements. For example, as shown, the measurement parameter display 412 includes measurements of the resistance index (RI) and the heart rate (HR). It should be understood that other measurements can be displayed in the measurement parameter display 412, such as the average maximum blood velocity, the average mean blood velocity, the average maximum pressure gradient of the blood velocity, the average velocity time integral, the average envelope time, the pulsatility index, etc. In some examples, these measurements can be calculated based on the flow distribution trace 420.

[0057] It should be understood that the graphical user interface 400 may include additional components, such as a menu (not shown) that includes options for controlling or adjusting acquisition parameters of the ultrasound imaging system and / or display parameters of the graphical user interface 400 .

[0058] In the depicted example, the imaged vessels include the uterine artery. For a normal uterine artery, the resistance index is typically 0.5 or lower. For pathological flow in the uterine artery, the resistance index may be 0.6 or higher. As shown, the flow distribution is normal,

[0059] Figure 5 An exemplary display output of a graphical user interface 500 is shown, which shows a Doppler flow distribution 510 of a vessel having pathological flow. Specifically, the vessel includes a uterine artery. Similar to graphical user interface 400, graphical user interface 500 includes an ultrasound image 505, including a grayscale or B-mode image with color flow image data superimposed thereon within a color flow ROI 509, a color scale 507, and a measurement parameter display 512. A flow distribution trace 520 corresponding to the upper boundary of flow distribution 510 is displayed on flow distribution 510.

[0060] The first indication of a negative trend in uterine artery flow is a notch behind the systolic slope, which is clearly visible in the flow profile 510, as well as a resistance index above 0.6, as shown. The notch may include an anomaly or abnormality relative to the normal flow profile 420 of the uterine artery.

[0061] When such pathological traffic distribution is detected, the abnormality can be highlighted or indicated to the user through a graphical user interface. Figure 6As shown in the graphical user interface 600 of FIG. 6 , a normal flow distribution 630 of the same blood vessel may be displayed in the graphical user interface 600, for example, by superimposing the normal flow distribution 630 on the pathological flow distribution 510. As shown, the normal flow distribution 630 may include the flow distribution trace 420 of the normal flow distribution 410, but it should be understood that in some examples, the normal flow distribution 410 may be superimposed or otherwise displayed with the pathological flow distribution 510. By displaying the normal flow distribution 630 on the flow distribution 510, a user viewing the graphical user interface 600 may be informed that the flow distribution 510 is abnormal or pathological.

[0062] Additionally, although not shown in graphical user interface 600, in some examples, the resistance index can be highlighted because the value for a given vessel indicates that the flow distribution is pathological. For example, the display of measurement parameter display 512 can be adjusted to highlight (e.g., by adjusting the color of the text or by adding a background color to the text) the resistance index (RI) within measurement parameter display 512.

[0063] In other examples, the normal flow profile 630 may be adjusted, for example, by time aligning the normal flow profile 630 so that the periodic peaks are aligned with the peaks of the pathological flow profile 510. For example, Figure 7 An exemplary graphical user interface 700 is shown in which a normal flow distribution 730 is displayed on the pathological flow distribution 510 , wherein the normal flow distribution 730 corresponds to the normal flow distribution 630 having a first peak time-aligned with a first peak of the flow distribution 510 .

[0064] As another example, the display of an abnormal or abnormal portion of the pathological flow distribution 510 may be adjusted to highlight the abnormality. Figure 8 An exemplary graphical user interface 800 is shown in which flow distribution trace 520 of flow distribution 510 is replaced with flow distribution trace 820. Specifically, flow distribution trace 820 includes flow distribution trace 520 in which the display of portions 841, 842, 843, 844, 845, 846, and 847 of flow distribution trace 820 are adjusted to highlight the anomalies of these portions relative to the normal flow distribution 730. As shown, portions 841, 842, 843, 844, 845, 846 and 847 are dashed rather than solid lines, but it should be understood that portions 841, 842, 843, 844, 845, 846 and 847 can be colored (e.g., displayed in red) while the remainder of the flow distribution trace 820 is displayed in another color (e.g., green) to emphasize or highlight portions 841, 842, 843, 844, 845, 846 and 847.

[0065] In some examples, the normal traffic distribution may be time aligned and scaled to further match the abnormal traffic distribution so as to emphasize the differences between the traffic distributions. Fig. 9 An exemplary graphical user interface 900 is shown in which a normal flow distribution 930 is displayed or superimposed on the flow distribution 510. Specifically, the normal flow distribution 930 corresponds to the normal flow distribution 630, wherein the first peak is time-aligned with the first peak of the flow distribution 510 and scaled so that the amplitudes match, as shown. In this way, a user viewing the graphical user interface 900 via a display device (such as the display device 115 or the display device 233) can easily discern the difference between the flow distributions 930 and 510. In addition, the flow distribution trace 920 of the flow distribution 510 can be adjusted similar to the flow distribution trace 820 to further emphasize the difference between the normal flow distribution 930 and the flow distribution 510. Alternatively, as shown, visual indicator marks 941, 942, 943, 944, 945, 946, and 947 can be superimposed on the flow distribution trace 930 to indicate the location of the anomaly detected by the machine learning model.

[0066] Another example of pathological flow in the uterine artery and an exemplary indication of such pathological flow is shown in Fig.10 and Fig.11 Specifically, Fig.10 An exemplary graphical user interface 1000 is shown displaying a Doppler spectrum or flow profile 1010 of a uterine artery having pathological flow. The flow profile 1010 exhibits pathological flow by substantially low diastolic flow with a resistance index above 0.6, as compared to the pathological flow of the flow profile 510. The graphical user interface 1000 also includes an ultrasound image 1005 having color flow image data superimposed within a color flow ROI 1009, a color scale 1007, and a measurement parameter display 1012.

[0067] As described above, flow profile 1010 may be evaluated by a machine learning model trained on a plurality of normal and / or abnormal flow profiles of the uterine artery, which in turn determines that flow profile 1010 is abnormal or pathological relative to a normal flow profile. To indicate to the user the abnormality of flow profile 1010, a normal flow profile may be superimposed thereon. For example, Fig.11 An exemplary display output of a graphical user interface 1100 is shown showing a normal traffic distribution 1130 overlaid thereon. Fig.10 Pathological Doppler flow distribution 1010. As shown, normal flow distribution 1130 may include flow distribution trace 420 of normal flow distribution 410 without adjusting the timing and proportion of flow distribution trace 420. In other examples, as described above with reference to Figure 7-Figure 9As described above, the normal flow distribution 1130 may be time aligned with the flow distribution 1010 and may be additionally or alternatively scaled so that the amplitudes of the peaks match. In addition, the flow distribution trace 1020 may be adjusted to indicate portions of the flow distribution trace 1020 that are abnormal relative to the normal flow distribution 1130. Additionally or alternatively, a visual indicator may be displayed or superimposed on the flow distribution 1010 to indicate the abnormal portion of the flow distribution 1010.

[0068] For other blood vessels, such as the middle cerebral artery, umbilical artery, ductus vein, etc., the specific shapes of normal and pathological flow distributions may be different from the normal and pathological flow distributions of the uterine artery. Therefore, the resistance index of normal and pathological flow may also be different from that of the uterine artery. For example, for the middle cerebral artery, the normal flow distribution may present a low, flat diastolic flow with a resistance index in the range of 0.8 to 0.85, while the pathological flow distribution may present a significantly increased diastolic flow with a resistance index below 0.7. As another example, for the umbilical artery, the normal flow distribution may present a diastolic flow with a resistance index in the range of 0.6 to 0.8, while the pathological flow distribution may present a too low or missing diastolic flow with a resistance index above 0.8. As yet another example, for the ductus vein, the normal flow distribution may present an M-shaped flow without a complete notch in the end diastole, where the pathological flow distribution presents a notch or even reverse flow down to the zero line.

[0069] Fig.12 A high-level flow chart illustrating an exemplary method 1200 for ultrasound imaging using automatic analysis of Doppler flow patterns during follow-up examinations is shown, according to one embodiment. In particular, the method 1200 involves evaluating ultrasound Doppler measurements to detect abnormal blood flow in a blood vessel. Figure 1 and Figure 2 Method 1200 is described with reference to systems and components of, but it should be understood that method 300 may be implemented with other systems and components without departing from the scope of the present disclosure. Method 1200 may be implemented, for example, as executable instructions in non-volatile memory 120 and / or non-volatile memory 206, and may be executed by processor 116 and / or processor 204 to perform the actions described below.

[0070] The method 1200 starts at 1205. At 1205, the method 1200 loads ultrasound data including Doppler measurements of a blood vessel acquired during a previous ultrasound scan of the patient. Continuing at 1210, the method 1200 starts an ultrasound scan of the blood vessel. For example, the method 1200 controls the transmit beamformer 101 and the transmitter 102 to drive the transducer elements 104 of the probe 106 to transmit ultrasound waves toward the blood vessel into the patient's body, and also controls the receiver 108 and the receive beamformer 110 to detect echoes of the ultrasound waves via the transducer elements 104 of the probe 106, as described above. Thus, the method 1200 can control the components of the ultrasound imaging system 100 according to a Doppler ultrasound imaging protocol. In this way, at 1215, the method 300 acquires a B-mode image or a grayscale ultrasound image of the blood vessel and the Doppler measurements of the blood vessel. The method 1200 may additionally automatically detect the location and type of a blood vessel, as described above, and automatically position a Doppler cursor at the location of the blood vessel to acquire Doppler measurements.

[0071] Continuing at 1220, the method 1200 displays the ultrasound image with the Doppler measurements acquired at 1215, for example, via the display device 118 or the display device 233. In addition, at 1225, the method 1200 displays the flow distribution from the previous ultrasound scan retrieved at 1205 and the flow distribution from the current ultrasound scan acquired at 1215. The flow distributions may be displayed side by side in a graphical user interface displayed via the display device 118 or the display device 233. Since the previous flow distribution and the current flow distribution are acquired at time T1 and time T1+ΔT, respectively, the user may view both flow distributions to monitor disease progression or treatment effects.

[0072] At 1230, the method 1200 evaluates the flow distributions from the previous ultrasound scan and the current ultrasound scan. For example, the method 1200 may input the two flow distributions into a trained machine learning model as discussed above to identify potential anomalies in either flow distribution. If the flow distribution from the previous ultrasound scan was previously evaluated for potential anomalies, the previous results of the analysis of the previous flow distribution may be retrieved from a storage device instead of analyzing the flow distribution during the current scan.

[0073] After evaluating the flow distribution, method 1200 continues to 1235. At 1235, method 1200 determines whether an anomaly is detected in any flow distribution. If no anomaly is detected ("No"), method 1200 continues to 1250. However, if an anomaly is detected ("Yes"), method 1200 continues to 1240. At 1240, method 1200 displays a normal or reference flow distribution for the vessel. For example, method 1200 may retrieve a reference flow distribution for the vessel from a library of normal or reference flow distributions (such as, for example, flow distributions stored in training module 210). In addition, the reference flow distribution may be selected from a plurality of reference flow distributions for the vessel stored in the library of reference flow distributions, for example, based on acquisition parameters, such that the reference flow distribution corresponds to a flow that can be expected under given acquisition conditions. In some examples, the reference flow distribution may be displayed adjacent to the flow distribution that includes the anomaly or anomaly. In other examples, the reference flow distribution is superimposed on the flow distribution characterized by the anomaly in a graphical user interface. In this way, the user can easily identify the difference between the flow distribution and the reference flow distribution. In some examples, the reference flow distribution can be adjusted based on the flow distribution before superimposing the reference flow distribution. For example, the reference flow distribution can be time-aligned so that at least one peak (e.g., R wave) of the reference flow distribution is aligned with the peak of the flow distribution. In addition, the reference flow distribution can be scaled, for example, so that the amplitude of at least one peak that is time-aligned with the peak of the flow distribution is also equal to the amplitude of the peak of the flow distribution.

[0074] At 1245, the method 1200 displays an indication of a pathology. For example, the method 1200 may highlight an abnormal portion of the flow distribution, such as by adjusting the display of the flow distribution at the abnormal portion. As another example, the method 1200 may graphically highlight the difference between the flow distribution and the reference flow distribution, and the difference between the current flow distribution and the previous flow distribution. For example, the method 1200 may automatically highlight the portion of the flow distribution that is different from the reference flow distribution and / or the previous flow distribution. As another example, the method 1200 may color the normal portion of the flow distribution as a first color (e.g., green), and color the different portions with a second color (e.g., red). As yet another example, the method 1200 may display an indication of the type of pathology that causes the flow distribution to be abnormal. In an example in which the machine learning model is configured to classify the flow distribution according to a flow grading system, the method 1200 may also display a classification or flow grade of the flow distribution (e.g., absent, minimal, weakened, suppressed, narrow, normal, etc.) in a graphical user interface.

[0075] At 1250, method 1200 determines whether the scan is complete. For example, the scan may be complete when the user of the ultrasound imaging system 100 indicates via the user interface 115 that the scan is complete. If the scan is not complete ("No"), method 1200 continues to 1255 to continue scanning. To continue scanning, method 1200 returns to 1215, for example, to acquire additional B-mode images and continue to acquire Doppler measurements. Thus, method 1200 continues to acquire Doppler measurements and evaluate the flow distribution or Doppler spectrum to determine whether pathological flow is present until the scan is complete. Once the scan is complete at 1250 ("Yes"), method 1200 proceeds to 1260. At 1260, method 1200 ends the scan. Then, method 1200 returns.

[0076] Thus, during a subsequent examination, the flow distribution or Doppler spectrum from the previous examination and the flow distribution of the current examination can be displayed to the user side by side. In addition, the two flow distributions can be evaluated to detect abnormalities in flow hemodynamics, and both results can be presented to the user for disease / treatment monitoring. In addition, the evaluation of the flow distribution described herein can be performed after the ultrasound examination. For example, the Doppler spectra from one or more previous ultrasound examinations can be retrieved and evaluated for abnormalities, with the results displayed to the user as described above.

[0077] A technical effect of the present disclosure includes displaying a normal flow distribution on a pathological flow distribution. Another technical effect of the present disclosure includes displaying a flow distribution in a graphical user interface, wherein a pathological portion or an abnormal portion of the flow distribution is highlighted in the graphical user interface. Yet another technical effect of the present disclosure includes automatically detecting anomalies in a Doppler spectrum. Another technical effect of the present disclosure includes an indication of a pathological type automatically identified in a Doppler spectrum.

[0078] In one embodiment, a method includes: acquiring Doppler measurements over a plurality of cardiac cycles via an ultrasound probe; evaluating a flow distribution including the Doppler measurements to detect anomalies in the flow distribution; and displaying the flow distribution with a reference flow distribution overlaid thereon via a display device.

[0079] In a first example of the method, the method further includes adjusting the display of the flow distribution to highlight one or more portions of the flow distribution that present an abnormality. In a second example of the method that optionally includes the first example, adjusting the display of the flow distribution to highlight one or more portions of the flow distribution that present an abnormality includes coloring the one or more portions of the flow distribution that present an abnormality as a first color, and coloring the remainder of the flow distribution as a second color. In a third example of the method that optionally includes one or more of the first and second examples, the method further includes superimposing a visual indicator on the flow distribution to highlight the one or more portions of the flow distribution that present an abnormality. In a fourth example of the method that optionally includes one or more of the first to third examples, the method further includes acquiring Doppler measurements of a blood vessel, wherein a reference flow distribution is based on one or more flow distributions of the blood vessel in a non-pathological state. In a fifth example of the method that optionally includes one or more of the first to fourth examples, the method further includes acquiring an ultrasound image of the vessel, and detecting the imaged vessel based on the ultrasound image, wherein detecting the imaged vessel includes detecting one or more of the location of the imaged vessel and the type of the imaged vessel, and wherein detecting the imaged vessel includes one of: automatically detecting the imaged vessel based on the ultrasound image using a machine learning model trained for vessel classification, or receiving a manual selection of the imaged vessel in the ultrasound image via user input. In a sixth example of the method that optionally includes one or more of the first to fifth examples, evaluating the flow distribution includes evaluating the flow distribution using a trained machine learning model, the trained machine learning model being selected from a plurality of trained machine learning models based on the type of the imaged vessel. In a seventh example of the method that optionally includes one or more of the first to sixth examples, the method further includes automatically positioning a Doppler cursor at the location of the imaged vessel for acquiring Doppler measurements. In an eighth example of the method that optionally includes one or more of the first to seventh examples, the method further includes adjusting the display of the reference flow distribution to at least partially match the flow distribution, wherein adjusting the display of the reference flow distribution includes one or more of: aligning the peak of the reference flow distribution with the peak time of the flow distribution, and scaling the reference flow distribution to match the amplitude of the peak of the reference flow distribution with the amplitude of the peak of the flow distribution. In a ninth example of the method that optionally includes one or more of the first to eighth examples, the method further includes determining a type of condition associated with an abnormality in the flow distribution using a trained machine learning model, and displaying an indication of the type of condition via a display device.

[0080] In another embodiment, the method includes: acquiring an ultrasound image of a blood vessel via an ultrasound probe; acquiring Doppler measurements of blood flow in the blood vessel via the ultrasound probe; displaying a graphical user interface via a display device, the graphical user interface including an ultrasound image with color flow data having Doppler measurements superimposed thereon, and a flow distribution constructed from the Doppler measurements; detecting anomalies in the flow distribution using a trained machine learning model; displaying a reference flow distribution of the blood vessel superimposed on the flow distribution in the graphical user interface via the display device; and adjusting the display of the flow distribution in the graphical user interface to indicate anomalies.

[0081] In a first example of the method, detecting an anomaly in the traffic distribution includes inputting the traffic distribution into a trained machine learning model and receiving an indication of the anomaly as an output from the trained machine learning model. In a second example of the method that optionally includes the first example, the traffic distribution is input as an image to the trained machine learning model. In a third example of the method that optionally includes the first example, the traffic distribution is input as a time series to the trained machine learning model. In a fourth example of the method that optionally includes one or more of the first to third examples, the method further includes adjusting the scale and timing of the reference traffic distribution to at least partially match the traffic distribution before superimposing the reference traffic distribution on the traffic distribution in the graphical user interface.

[0082] In another embodiment, an ultrasound imaging system includes an ultrasound probe, a display device, and a processor configured with executable instructions in a non-volatile memory, which executable instructions, when executed, cause the processor to: acquire Doppler measurements over multiple cardiac cycles via the ultrasound probe; evaluate a flow distribution including the Doppler measurements to detect abnormalities in the flow distribution; and display, via the display device, the flow distribution with a reference flow distribution overlaid thereon.

[0083] In a first example of the system, the processor is further configured with executable instructions in a non-transitory memory, which, when executed, causes the processor to adjust the display of the flow distribution to highlight one or more portions of the flow distribution that are abnormal. In a second example of the system that optionally includes the first example, the processor is further configured with executable instructions in a non-transitory memory, which, when executed, causes the processor to adjust the display of the flow distribution to highlight one or more portions of the flow distribution that are abnormal by coloring the one or more portions of the flow distribution that are abnormal as a first color, and coloring the rest of the flow distribution as a second color. In a third example of the system that optionally includes one or more of the first and second examples, the processor is further configured with executable instructions in a non-transitory memory, which, when executed, causes the processor to acquire Doppler measurements of a blood vessel, and select a reference flow distribution from a plurality of reference flow distributions based on the imaged blood vessel. In a fourth example of a system that optionally includes one or more of the first to third examples, the processor is further configured with executable instructions in a non-transitory memory, which, when executed, causes the processor to train a machine learning model with training data including one or more of a plurality of reference flow distributions and a plurality of abnormal flow distributions, and to evaluate the flow distribution to detect anomalies using the trained machine learning model. In a fifth example of a system that optionally includes one or more of the first to fourth examples, the processor is further configured with executable instructions in a non-transitory memory, which, when executed, causes the processor to determine the type of pathology associated with the abnormality of the flow distribution using the trained machine learning model, and to display an indication of the type of pathology via a display device.

[0084] As used herein, the elements or steps listed in the singular and beginning with the word "one" or "a kind of" should be understood as not excluding a plurality of the elements or steps, unless such exclusion is explicitly stated. In addition, the reference to "an embodiment" of the present invention is not intended to be interpreted as excluding the existence of additional embodiments that also include the cited features. In addition, unless explicitly stated to the contrary, the embodiment of "including", "comprising" or "having" an element or multiple elements with a specific characteristic may include additional such elements without the characteristic. The terms "including" and "in..." are used as the concise language equivalents of the corresponding terms "including" and "wherein". In addition, the terms "first", "second" and "third" etc. are only used as marks, and are not intended to impose numerical requirements or specific positional order on their objects.

[0085] This written description uses examples to disclose the invention, including the best mode, and also to enable a person of ordinary skill in the relevant art to practice the invention, including making and using any devices or systems and performing any included methods. The patentable scope of the invention is defined by the claims, and may include other examples that occur to a person of ordinary skill in the art. Such other examples are intended to fall within the scope of the claims if they have structural elements that do not differ from the literal language of the claims, or if they include equivalent structural elements with insignificant differences from the literal language of the claims.

Claims

1. An ultrasound imaging method, performed by a processor, include: Doppler measurements are collected over multiple cardiac cycles using an ultrasound probe; evaluating a flow distribution including said Doppler measurements to detect anomalies in said flow distribution; displaying, via a display device, the flow distribution with a reference flow distribution overlaid thereon; as well as Adjusting the display of the reference flow distribution to at least partially match the flow distribution, wherein adjusting the display of the reference flow distribution comprises one or more of: aligning a peak of the reference flow distribution with a peak time of the flow distribution, and scaling the reference flow distribution to match an amplitude of a peak of the reference flow distribution with an amplitude of a peak of the flow distribution. 2 . The ultrasound imaging method according to claim 1 , further comprising adjusting the display of the flow distribution to highlight one or more portions of the flow distribution exhibiting the abnormality.

3. The ultrasound imaging method according to claim 2, wherein adjusting the display of the flow distribution to highlight the one or more portions of the flow distribution presenting the abnormality includes coloring the one or more portions of the flow distribution presenting the abnormality with a first color, and coloring the rest of the flow distribution with a second color. 4 . The ultrasound imaging method of claim 2 , further comprising superimposing a visual indicator on the flow distribution to highlight the one or more portions of the flow distribution exhibiting the abnormality.

5. The ultrasound imaging method of claim 1, further comprising acquiring the Doppler measurements of a blood vessel, wherein the reference flow profile is based on one or more flow profiles of the blood vessel in the absence of a pathology.

6. The ultrasound imaging method according to claim 5, further comprising acquiring an ultrasound image of the blood vessel, and detecting the imaged blood vessel based on the ultrasound image, wherein detecting the imaged blood vessel comprises detecting one or more of a location of the imaged blood vessel and a type of the imaged blood vessel, wherein detecting the imaged blood vessel comprises one of the following: automatically detecting the imaged blood vessel based on the ultrasound image using a machine learning model trained for blood vessel classification, or receiving a manual selection of the imaged blood vessel in the ultrasound image via user input.

7. The ultrasound imaging method of claim 6, wherein evaluating the flow distribution comprises evaluating the flow distribution using a trained machine learning model, the trained machine learning model being selected from a plurality of trained machine learning models based on the type of the blood vessel being imaged.

8. The ultrasound imaging method of claim 6, further comprising automatically positioning a Doppler cursor at the location of the imaged blood vessel for acquiring the Doppler measurements.

9. The ultrasound imaging method of claim 1, further comprising determining, using a trained machine learning model, a type of pathology associated with the abnormality in the flow distribution, and displaying an indication of the type of pathology via the display device.

10. An ultrasound imaging method, performed by a processor, include: Acquiring ultrasound images of blood vessels via an ultrasound probe; collecting Doppler measurement results of blood flow in the blood vessel via the ultrasound probe; displaying, via a display device, a graphical user interface comprising the ultrasound image with color flow data of the Doppler measurements superimposed thereon and a flow distribution constructed from the Doppler measurements; Detecting anomalies in the traffic distribution using a trained machine learning model; displaying, via the display device, in the graphical user interface, a reference flow distribution of the blood vessel superimposed on the flow distribution; and Adjusting a display of the graphical user interface to indicate the anomaly, wherein adjusting the display includes one or more of: aligning a peak of the reference flow distribution with a peak time of the flow distribution, and scaling the reference flow distribution to match an amplitude of a peak of the reference flow distribution with an amplitude of a peak of the flow distribution.

11. The ultrasound imaging method of claim 10, wherein detecting the anomaly in the flow distribution comprises inputting the flow distribution into the trained machine learning model and receiving an indication of the anomaly as an output from the trained machine learning model.

12. The ultrasound imaging method according to claim 11, wherein the flow distribution is input to the trained machine learning model as an image or as a time series.

13. The ultrasound imaging method of claim 10, further comprising adjusting a scale and a timing of the reference flow distribution to at least partially match the flow distribution before superimposing the reference flow distribution on the flow distribution in the graphical user interface.

14. An ultrasound imaging system, include: Ultrasound probe; Display devices; and A processor configured with executable instructions in a non-volatile memory, wherein the executable instructions, when executed, cause the processor to perform the ultrasound imaging method according to any one of claims 1-9.