Guided transcranial ultrasound imaging using neural networks and associated devices, systems and methods

Abstract

Ultrasound imaging devices, systems and methods are provided. A medical ultrasound imaging system includes: an interface that communicates with an ultrasound imaging component and is configured to receive a first image representing blood vessels of a patient's brain when the ultrasound imaging component is positioned at a first imaging position relative to the patient; and a processing component that communicates with the interface and is configured to apply a convolutional neural network (CNN) to the first image to generate a motion control configuration for repositioning the ultrasound imaging component from the first imaging position to a second imaging position associated with a transcranial examination, the CNN being trained based on at least a known vascular topology map.

Description

Technical Field

[0001] The present disclosure relates generally to ultrasound imaging and, more particularly, to the application of a neural network to guide a user in aligning an ultrasound imaging component with a desired imaging plane during a transcranial examination. Background Art

[0002] Cerebrovascular hemodynamic measurement results can be used to diagnose and monitor various cerebrovascular conditions in adults and pediatric populations. Based on radiopaque computed tomography (CT) tracer technology and based on magnetic resonance imaging (MRI) contrast agent technology are usually used to obtain cerebrovascular hemodynamic measurement results. However, based on radiopaque CT tracer and based on MRI contrast agent technology may not provide a sufficiently high temporal resolution for assessing hemodynamics. In addition, based on radiopaque CT tracer and based on MRI contrast agent technology may require a large amount of equipment and facilities and may be expensive.

[0003] Another scheme for measuring the blood flow in the intracranial artery is to use transcranial Doppler (TCD) ultrasound. TCD ultrasound is a non-invasive ultrasound imaging technique that can be used for point-of-care testing and diagnosis. During TCD monitoring, ultrasound waves are transmitted through the patient's skull and reflect the blood flow in the brain. The frequency shift in the echo signal allows the estimation of blood flow and the detection of various cerebrovascular conditions. For example, TCD can detect and monitor microemboli in intracranial aneurysms, patent foramen ovale, vasospasm, stenosis, brain death, shunts, and surgery or flow settings without generating radiation.

[0004] Although TCD ultrasound can provide cerebrovascular hemodynamic measurements with sufficiently high temporal resolution at a relatively low cost, consistent TCD measurements are difficult to obtain. The attenuation and aberration of the skull and the variability and tortuosity of the perforating cerebral vessels require highly trained or experienced operators. For example, accurate TCD examinations may require operators to have knowledge and understanding of cerebrovascular topology, cerebrovascular patterns, cerebrovascular changes, and / or Doppler ultrasound techniques. A challenge in transcranial ultrasound imaging is the blurring and signal absorption that occurs due to the skull. These acoustic effects bend the ultrasound beam, which makes it difficult to identify vascular flow patterns. In these cases, expert users can rely on landmark topology and vascular branches near the Circle of Willis (CoW) to obtain blood flow measurements. However, poor image quality may hinder accurate TCD-based inspections. Therefore, the scope of using TCD as a clinical tool may be limited. In addition, TCD measurements may be limited to inter-operator variability even between experienced operators.

[0005] One approach to assisting a user in performing TCD ultrasound imaging is to provide imaging feedback by displaying a depth projection of a power Doppler signal while the user searches for the middle cerebral artery (MCA) in the patient's brain. Although imaging feedback may allow a user to monitor hemodynamics with greater consistency or accuracy, feedback-based approaches are limited to proximal MCA examinations and may be subject to variations in the tortuosity of the M1 and M2 branches of the MCA.

[0006] US2002 / 0103436 discloses a system for automatic manipulation of a transcranial Doppler probe device, comprising: a cylindrical probe housing with internal electrical wiring; an ultrasonic transducer having a cable attached to a probe cylindrical base with a coil, the probe cylindrical base being placed within the cylindrical probe housing; a rolling ball system; a system software program and a microprocessor for controlling the probe position; and a removable handle attached to the probe cylindrical base. The system software "learns" the probe angle after initial manual manipulation and then performs cerebrovascular sonication using operator-independent electromotive force. Summary of the invention

[0007] Although existing processes for assessing cerebrovascular conditions using TCD ultrasound imaging have proven useful for clinical processes, there is still a clinical need for improved systems and techniques for providing efficient, accurate and automatic processes for aligning ultrasound imaging components with desired imaging planes for transcranial examinations. Embodiments of the present disclosure provide mechanisms for guiding users during TCD examinations using deep learning networks. For example, an ultrasound imaging component can capture images of flow within cerebral vessels (e.g., in color Doppler images). The captured image can be fed into a convolutional neural network (CNN), which is trained to identify the current imaging plane of the ultrasound imaging component within a cerebrovascular atlas (e.g., a known cerebrovascular topology) or the current vascular position captured by the image. The target vascular position or target imaging plane for transcranial examinations in the cerebrovascular atlas can be known (e.g., the position of the MCA for MCA examinations can be predetermined). Therefore, a set of motion control parameters for aligning the ultrasound imaging component with the target imaging plane can be calculated based on a geometric distance or angle calculation between the current imaging plane and the target imaging plane. The disclosed embodiments can provide instructions to direct alignment of the ultrasound imaging component with the target imaging plane based on the motion control parameters. The disclosed embodiments can provide a graphical view including superimposing the imaged vessel, the target imaging plane, and / or the current imaging plane on a cerebral vascular atlas. The disclosed embodiments can provide a graphical view including a virtual view of the target vessel region, which can be outside the current field of view of the ultrasound imaging component.

[0008] In one embodiment, a medical ultrasound imaging system is provided. The system includes: an interface that communicates with an ultrasound imaging component and is configured to receive a first image representing blood vessels of a patient's brain when the ultrasound imaging component is positioned at a first imaging position relative to the patient; and a processing component that communicates with the interface and is configured to apply a convolutional neural network (CNN) to the first image to generate a motion control configuration for repositioning the ultrasound imaging component from the first imaging position to a second imaging position associated with a transcranial examination, the CNN being trained based on at least a known vascular topology map.

[0009] In some embodiments, the processing component is further configured to determine Doppler information representing blood flow within a blood vessel of the patient's brain based on data associated with the first image, and wherein the CNN is applied to the Doppler information. In some embodiments, the processing component is further configured to determine connectivity information associated with the blood vessel of the patient's brain based on the Doppler information, and to determine a covariance matrix based on the connectivity information, and wherein the CNN is applied to the covariance matrix. In some embodiments, the connectivity information includes coordinates corresponding to vascular locations along the blood vessel of the patient's brain. In some embodiments, the processing component is further configured to apply the CNN to the Doppler information to determine an imaging plane corresponding to a first imaging position within a known vascular topology map; and determine the motion control configuration based on the imaging plane and a target imaging plane associated with a transcranial examination within the known vascular topology map. In some embodiments, the processing component is further configured to apply the CNN to the Doppler information to determine a feature vector representing the blood vessel of the patient's brain; and determine an imaging plane within the known vascular topology map based on a comparison of the feature vector with the known vascular topology map. In some embodiments, the CNN is further trained based on at least a covariance matrix determined based on connectivity information of the known vascular topology map and a weighting function associated with the transcranial examination, and wherein the connectivity information includes coordinates corresponding to vascular positions along the blood vessels indicated in the known vascular topology map. In some embodiments, the motion control configuration includes at least one of a translation or a rotation of the ultrasound imaging component. In some embodiments, the system further includes: a user interface that communicates with the processing component, the user interface being configured to receive a selection of at least one of a type of the transcranial examination or a target blood vessel position associated with the transcranial examination, wherein the processing component is further configured to determine the second imaging position based on the selection. In some embodiments, the system further includes: a display that communicates with the processing component, the display being configured to display instructions based on the motion control configuration for operating the ultrasound imaging component so that the ultrasound imaging component is repositioned to the second imaging position. In some embodiments, the system further comprises a display in communication with the processing unit, the display being configured to display a graphical view comprising superimposing at least one of a blood vessel of the patient's brain, a first imaging plane associated with the first imaging position, or a second imaging plane associated with the second imaging position on the known vascular topology map.In some embodiments, the system further comprises a display in communication with the processing component, the display being configured to display a graphical view comprising superimposing an expected view of blood vessels of the patient's brain associated with the second imaging position on top of the known vascular topology map.

[0010] In one embodiment, a method of medical ultrasound imaging is provided. The method includes: receiving a first image representing blood vessels of a patient's brain from an ultrasound imaging component when the ultrasound imaging component is positioned at a first imaging position relative to the patient; and applying a convolutional neural network (CNN) to the first image to generate a motion control configuration for repositioning the ultrasound imaging component from the first imaging position to a second imaging position associated with a transcranial examination, the CNN being trained based on at least a known vascular topology map.

[0011] In some embodiments, the method further comprises: determining Doppler information representing blood flow within a blood vessel of the patient's brain based on data associated with the first image, wherein the CNN is applied to the Doppler information. In some embodiments, the method further comprises: determining connectivity information associated with a blood vessel of the patient's brain based on the Doppler information, the connectivity information comprising coordinates corresponding to vascular locations along the blood vessel of the patient's brain; and determining a covariance matrix based on the connectivity information, wherein the CNN is applied to the covariance matrix. In some embodiments, the method further comprises: applying the CNN to the Doppler information to determine an eigenvector representing a blood vessel of the patient's brain; determining an imaging plane within the known vascular topology map based on a comparison of the eigenvector with the known vascular topology map; and determining the motion control configuration based on the imaging plane and a target imaging plane associated with a transcranial examination within the known vascular topology map, the motion control configuration comprising at least one of a translation or a rotation for operating the ultrasound imaging component. In some embodiments, the method further comprises: receiving a selection of at least one of a type of the transcranial examination or a target vascular location associated with the transcranial examination; and determining the second imaging position based on the selection. In some embodiments, the method further comprises: based on the motion control configuration, transmitting instructions to at least one of a display or a robotic system for operating the ultrasound imaging component so that the ultrasound imaging component is repositioned to the second imaging position, the instructions comprising at least one of a translation or a rotation of the ultrasound imaging component. In some embodiments, the method further comprises: displaying a graphical view comprising superimposing at least one of a first imaging plane associated with the first imaging position, a second imaging plane associated with the second imaging position, or a blood vessel of the patient's brain on the known vascular topology map. In some embodiments, the method further comprises: displaying a graphical view comprising superimposing an expected view of the blood vessels of the patient's brain associated with the second imaging position on the known vascular topology map.

[0012] Additional aspects, features, and advantages of the present disclosure will become apparent from the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Illustrative embodiments of the present disclosure will be described with reference to the accompanying drawings, in which:

[0014] Figure 1 is a schematic diagram of a medical ultrasound imaging system for transcranial examination according to aspects of the present disclosure.

[0015] Figure 2is a schematic diagram illustrating the vasculature of a patient's brain in accordance with aspects of the present disclosure.

[0016] Figure 3 is a schematic diagram illustrating a graphical representation of a portion of a blood vessel of a patient's brain in accordance with aspects of the present disclosure.

[0017] Figure 4 is a schematic diagram illustrating a scheme for guiding an ultrasound imaging component to a desired imaging plane for a transcranial examination in accordance with aspects of the present disclosure.

[0018] Figure 5 An example of two-dimensional (2D) Doppler imaging in accordance with aspects of the present disclosure is illustrated.

[0019] Figure 6 An example of three-dimensional (3D) Doppler imaging in accordance with aspects of the present disclosure is illustrated.

[0020] Figure 7 is a schematic diagram illustrating a configuration for a convolutional neural network (CNN) according to aspects of the present disclosure.

[0021] Figure 8 is a schematic diagram illustrating a scheme for generating a covariance matrix from a cerebral vascular atlas in accordance with aspects of the present disclosure.

[0022] Fig. 9 is a schematic diagram illustrating a display for guiding transcranial ultrasound imaging in accordance with aspects of the present disclosure.

[0023] Fig.10 is a flow chart of a method for applying a CNN to guide an ultrasound imaging component to a desired imaging plane for a transcranial examination in accordance with aspects of the present disclosure. DETAILED DESCRIPTION

[0024] For the purpose of promoting understanding of the principles of the present disclosure, reference will now be made to the embodiments illustrated in the accompanying drawings and specific language will be used to describe the same. However, it should be understood that limitations on the scope of the present disclosure are not contemplated. Any variations and other modifications of the described devices, systems and methods and any other applications of the principles of the present disclosure are fully contemplated and included in the present disclosure, as those skilled in the art to which the present disclosure relates will generally think of. Specifically, it is fully contemplated that the features, components and / or steps described with respect to one embodiment can be combined with the features, components and / or steps described with respect to other embodiments of the present disclosure. For the sake of brevity, however, many iterations of these combinations will not be described separately.

[0025] Figure 11 is a schematic diagram of a medical imaging system for transcranial examination according to various aspects of the present disclosure. The system 100 includes a host 130 and an imaging probe 120 that communicate with each other. At a high level, the probe 120 can be arranged in contact with the patient's head 110 to capture the patient's brain 102 and / or the image of the blood vessels 104 in the patient's brain 102, and the host 130 can provide instructions to the user to reposition the probe 102 to the desired position for imaging the region of interest for a specific transcranial examination. For example, a transcranial examination can measure the blood flow in the blood vessels around the Circle of Willis (CoW), such as the anterior cerebral artery (ACA), the internal carotid artery (ICA), the middle cerebral artery (MCA), the posterior cerebral artery (PCA), the posterior communicating artery, the basilar artery (BA) and / or the blood vessels fed by the corresponding arteries. The arteries associated with CoW are described in more detail herein. In some instances, a transcranial examination can also measure the blood flow in other arteries and / or veins in any position of the patient's head 110 or the patient's brain 102. System 100 may be an ultrasound imaging system, and probe 120 may be an external ultrasound imaging probe.

[0026] The probe 120 may include an imaging component 122, which includes one or more ultrasonic sensors or transducer elements. The transducer element may emit ultrasonic energy toward the patient's anatomical structure (e.g., head 110). The ultrasonic energy is reflected by the vessels and tissues of the patient's brain and the patient's skull. The ultrasonic transducer element may receive the reflected ultrasonic signal. In some embodiments, the probe 120 may include an internal or integrated processing component that can locally process the ultrasonic echo signal to generate an image signal representing the anatomical structure of the patient under imaging. (One or more) ultrasonic transducer elements may be arranged to provide a 2D image or 3D image of the patient's anatomical structure. The probe 120 may be configured to perform duplex mode ultrasonic imaging using B-mode imaging and color Doppler flow measurement results. The user may place the probe 120 at various locations on the external surface of the patient's head 110 to perform a transcranial examination.

[0027] For example, the probe 120 can be placed at a specific position on the patient's head 119. The position is selected based on the bone thickness and bone composition of the patient's head 110. For example, compact bone has less air, and therefore, can reflect less sound waves than cancellous bone. The position that allows efficient ultrasound transmission for transcranial examination can be called a transcranial window or an acoustic window. The transcranial window that is usually used to examine the six main cerebral arteries (e.g., ACA, ICA, MAC, PCA, posterior communicating artery, and BA) can include a temporal bone window, a submandibular window, and a suboccipital window. For example, the temporal bone window can be used to examine the ACA, MCA, ICA, PCA, posterior communicating artery, and the adjacent blood vessels fed by the corresponding artery. Alternatively, the suboccipital transcranial window can be used to examine the BA or the submandibular transcranial window can be used to examine the ICA. The user can manipulate the probe 120 by translation and / or rotation of the probe 120 to reach the target position corresponding to the transcranial window of the selected transcranial examination. Host 130 can provide guidance to the user during TCD imaging to facilitate proper acquisition of fluid dynamics within vessels within patient's head 110 or patient's brain 102. For example, TCD imaging can be transcranial color Doppler (TCCD) imaging.

[0028] Host 130 may include memory 132, display 134, processing component 136, and communication interface 138. Processing component 136 may be coupled to and in communication with memory 132, display 134, and communication interface 138. Host 130 may be a computer workstation, a mobile phone, a tablet computer, or any suitable computing device.

[0029] The memory 132 may include cache memory (e.g., cache memory of the processing component 136), random access memory (RAM), magnetoresistive RAM (MRAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), flash memory, solid-state memory devices, hard disk drives, solid-state drives, other forms of volatile and non-volatile memory, or a combination of different types of memory. The memory 132 may be configured to store the cerebral vascular atlas 140 and one or more CNNs 142.

[0030] The cerebrovascular atlas 140 can be a 3D model that describes a generalized or personalized arrangement of blood vessels within the human brain. Although the variability and tortuosity of blood vessels in the lateral regions of the human brain can make vessel identification difficult, CoW is bilateral and symmetrical, which exhibits landmark morphology, connectivity patterns, and fluid dynamics. Therefore, a predictable portion of the blood vessels within the human brain (e.g., CoW) can be used to construct the cerebrovascular atlas 140. For example, the cerebrovascular atlas 140 can include major cerebral blood vessels (e.g., MCA, ICA, ACA, PCA, and BA) and connectivity, topology, and position information associated with the blood vessels relative to each other and relative to bone structures in the human skull. The location and / or topology of the major blood vessels are relatively predictable and similar across patients, however the location and / or topology of smaller blood vessels (e.g., away from the major blood vessels) may vary between different patients. For example, Figure 1 A portion 108 of vessels within a patient's head 110 including an ACA (where the location or arrangement of the ACA may be relatively predictable across patients) and a portion 106 including vessels distal to the ACA (where the location or arrangement may vary for different patients) are illustrated. In some embodiments, the memory 132 may store a plurality of cerebral vascular atlases 140. For example, the vascular topologies in the cerebral vascular atlases 140 may be generated based on imaging data of the patient's brain collected from a clinical study and / or from previously captured imaging data of a corresponding patient under examination. The atlas 140 can be configured to store empirically known data about vessels in the patient's head, brain, neck, and / or other anatomical structures, including vessel structures, relationships, connections, blood flow patterns, geometries, locations from one or more imaging windows, etc. The patient undergoing the current examination may or may not be part of the plurality of patients on which the atlas 140 is based. In some embodiments, patient-specific vascular data associated with the patient undergoing the current examination is used as the atlas 140, for example, from earlier imaging of the patient's brain. The patient-specific vascular data can be used to replace or supplement reference data from a plurality of other patients.

[0031] The CNN 142 may be trained to identify the location of the blood vessels imaged by the probe 120 relative to the cerebral vascular atlas 140. In some embodiments, the memory 132 may store a plurality of CNNs 142. For example, the CNNs 142 may include a predictive CNN for identifying the current imaging plane to provide guidance to the user in aligning the probe 120 to reach the target or desired imaging plane and a limiting CNN for limiting the identification provided by the predictive CNN.

[0032] The processing component 136 may include a central processing unit (CPU), a digital signal processor (DSP), an application specific integrated circuit (ASIC), a controller, a field programmable gate array (FPGA) device, another hardware device, a firmware device, or any combination thereof, which is configured to perform the operations described herein. The processing component 136 may also be implemented as a combination of computing devices (e.g., a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration).

[0033] In an embodiment, the processing component 136 is configured to receive an image of the patient's head 110 from the probe 120. The processing component 136 can determine Doppler information (e.g., blood flow within the blood vessel 104) from the image and determine a graphical representation of the blood vessel 104 based on the Doppler information. For example, the graphical representation can include spatial coordinates describing the location of segments along the blood vessel and the connectivity of the blood vessel. In some instances, the graphical representation can be in the form of a connectivity graph or a tree graph. The processing component 136 can determine the current position of the probe 120 relative to the patient's vasculature based on the connectivity information, and determine a control configuration (e.g., including translation and / or rotation parameters) for repositioning the probe 120 to a target position or target imaging plane for obtaining images for a specific transcranial examination.

[0034] In an embodiment, the processing component 136 may apply the CNN 142 to the Doppler information, and the CNN 142 may identify the current position of the probe 120 within the cerebral vascular atlas 140. The processing component 136 may determine the translation and / or rotation that may be required to reposition the probe 120 to the target location.

[0035] In an embodiment, the processing component 136 is configured to train a CNN 142 for aligning the imaging component 122 with a target image plane based on one or more cerebral vascular atlases 140. In an embodiment, the processing component 136 is configured to apply the CNN 142 in a clinical setting to determine motion control parameters to align the probe 120 with a patient for a particular transcranial examination. For example, the imaging component 122 is aligned to obtain an image of the ACA of a patient for a transcranial examination. Mechanisms for mapping Doppler information into a graphical representation, training the CNN 142, and applying the CNN 142 are described in more detail herein.

[0036] In some embodiments, memory 132 may include a non-transitory computer-readable medium. Memory 132 may store instructions that, when executed by processing component 136, cause processing component 136 to perform the operations described herein with reference to CNN training and / or CNN application in conjunction with embodiments of the present disclosure. Instructions may also be referred to as code. The terms "instructions" and "code" should be broadly interpreted as including any type of (one or more) computer-readable statements. For example, the terms "instructions" and "code" may refer to one or more programs, routines, subroutines, functions, processes, etc. "Instructions" and "code" may include a single computer-readable medium.

[0037] The display 134 may include a computer screen or any suitable display for displaying a user interface (UI) 144. The UI 144 may include a graphical representation or view of the probe 120. The UI 144 may include a visual indicator indicating the translation and / or rotation of the probe 120. The UI 144 may include a graphical view including superimposing a current image acquired by the probe 120 on the cerebral vascular atlas 140. The graphical view may additionally include superimposing a desired view of the patient's vessels or blood vessels 104 at the target location on the cerebral vascular atlas 140. Although the display 134 is shown as an integral component of the host 130, in some embodiments, the display 134 may be external to the host 130 and communicate with the host 130 via the communication interface 138. For example, the display 134 may include a stand-alone display, augmented reality glasses, or a mobile phone.

[0038] The communication interface 138 may be configured to communicate with the imaging component 122 of the probe 120 via a communication link 150. For example, the host 130 may send controls to control the transmission and reception of the ultrasound transducer elements (e.g., for beamforming) and may receive acquired images from the probe 120 via the communication link 150. The communication link 150 may include a wireless link and / or a wired link. Examples of wireless links may include low power A wireless link, an Institute of Electrical and Electronics Engineers (IEEE) 802.11 (WiFi) link, or any suitable wireless link. Examples of wired links may include a Universal Serial Bus (USB) link or any suitable wired link.

[0039] In some embodiments, the communication interface 138 may also be configured to receive user input, for example, via a keyboard, mouse, or touch screen. The UI 144 may update a particular display or view based on the user input. The UI 144 is described in more detail herein.

[0040] In some embodiments, the system 100 may also include a robotic system 160 in communication with the communication interface 138 and the probe 120. The robotic system 160 may include electrical and / or mechanical components, such as motors, rollers, and gears, configured to reposition the probe 120. In such embodiments, the processing component 136 can be configured to send motion control parameters to the robotic system 150, for example, via the communication interface 138. The robotic system 160 can automatically align the probe 120 with the patient for a particular transcranial examination based on the motion control parameters. For example, the robotic system 150 can automatically align the probe without manual repositioning by the user.

[0041] Although the system 100 is illustrated with an ultrasound imaging probe 120, the system 100 can be configured to automatically align any suitable imaging component 122 with the patient for a clinical procedure. The imaging component 122 can provide any suitable imaging modality. Examples of imaging modalities can include optical imaging, optical coherence tomography (OCT), radiographic imaging, X-ray imaging, angiography, fluoroscopy, computed tomography (CT), magnetic resonance imaging (MRI), elastography, etc.

[0042] In some other embodiments, system 100 may include any suitable sensing components, including pressure sensors, flow sensors, temperature sensors, optical fibers, reflectors, mirrors, prisms, ablation elements, radio frequency (RF) electrodes, conductors, and / or combinations thereof for performing a clinical or therapeutic procedure, wherein images of the anatomical structures of a patient receiving the procedure may be acquired by imaging component 122 before, during, and / or after the procedure.

[0043] In general, the system 100, the probe 120, and / or other devices described herein can be used to examine any suitable anatomical structure of the patient's body. In some instances, the probe 120 can be positioned outside the patient's body to examine anatomical structures and / or lumens within the patient's body. For anatomical structures and / or lumens, natural and artificial fluid-filled or surrounding structures can be represented. For example, the probe of the present disclosure can be positioned on the surface of the patient's head to obtain blood flow measurements within the patient's brain. In some embodiments, the probe of the present disclosure can be used to examine any number of anatomical locations and tissue types, including but not limited to organs including the liver, heart, kidney, gallbladder, pancreas, lungs; ducts; intestines; nervous system structures, including the brain, dural sac, spinal cord and peripheral nerves; urinary tract; and valves, chambers or other parts of the heart, and / or other systems of the body. The anatomical structure and / or lumen within the patient's body can be a blood vessel, such as an artery or vein of the patient's vascular system, including cerebral blood vessels, cardiac vascular system, peripheral vascular system, neural vascular system, renal vascular system, and / or any other suitable lumen within the body.

[0044] Figure 2 2 is a schematic diagram of a vessel 200 of a patient's brain (such as brain 102) according to various aspects of the present disclosure. The vessel 200 can be imaged by an ultrasound imaging probe (such as probe 120). The blood flow (e.g., velocity and direction) within the vessel 200 can be determined based on color Doppler measurements, as described in more detail herein. As shown, the vessel 200 includes a ring arterial structure 210, which can be referred to as a CoW. The vessel 200 is located at the base of the patient's brain. The vessel 200 includes a network of blood vessels. The vessel 200 can supply blood to the brain and surrounding tissues and structures. As shown, the vessel 200 includes six major arteries, including ACA 214, ICA 216, MCA 212, PCA 218, posterior communicating artery 220, and BA 222. Each of the arteries 212, 214, 216, 218, 220, and 222 can branch into smaller vessels. As described above, the vascular arrangement around the CoW can be substantially similar for all patients. Therefore, the structural arrangement (e.g., connectivity, topology, and / or location) of the arteries 212, 214, 216, 218, 220, and 222 can be described in a graphical representation and used to construct the cerebral vascular atlas 140, as described in more detail herein.

[0045] Figure 3 3 is a schematic diagram of a graphical representation 300 of a portion of a vasculature of a patient's brain according to aspects of the present disclosure. For example, the graphical representation 300 corresponds to a portion of the vasculature 200 (e.g., around the structure 210). The graphical representation 300 represents the structural arrangement of blood vessels within the vasculature. The graphical representation 300 includes nodes 310 connected by edges 312, representing the geometric topology of the vasculature, such as the vessel 104 and arteries 212, 214, 216, 218, 220, and 222 in space. The nodes 310 can correspond to vascular branches, endpoints of the vessels, and / or vascular locations along segments or flow paths of the vessels. Each edge 312 can connect two or more than two nodes 310. For example, the vessel can be divided into multiple segments represented by a series of nodes 310 connected to each other by edges 312. The graphical representation 300 can be referred to as a connectivity graph, a vascular tree, or a node-edge graph.

[0046] As an example, Figure 2 The intersection of the arteries 212, 214, 216, 218, 220, and 222 shown by the dashed circle in FIG. 3 is represented by a node 310 and the segments of the arteries 212, 214, 216, 218, 220, and 222 connected to the intersection are represented by edges 312. Figure 3Nodes 310 are used to represent vessel branches, but in some embodiments, a vessel (e.g., MCA 212) can be represented by multiple nodes 310 interconnected by multiple edges 312 corresponding to segments of the vessel. Thus, the graphical representation 300 can include any suitable number of nodes 310 interconnected by any suitable number of edges 312. In addition, the graphical representation 300 can include nodes 310 and edges 312 representing smaller vessels fed by the main arteries 212, 214, 216, 218, 220, and 222.

[0047] In an embodiment, nodes 310 and / or edges 312 are represented by spatial Cartesian coordinates and / or flow vectors, as described in more detail herein. In an embodiment, cerebral vascular atlas 140 uses a graphical representation 300 to describe the connectivity, topology, and / or location information of blood vessels within the human brain. In an embodiment, CNN 142 operates on the graphical representation 300 of a Doppler image, as described in more detail herein.

[0048] Figure 4-6 Collectively, a transcranial examination using system 100 is illustrated. Figure 4 is a schematic diagram illustrating a scheme 400 for guiding an ultrasound imaging probe to a desired imaging plane for a transcranial examination in accordance with aspects of the present disclosure. Figure 5 An example of 2D Doppler imaging 500 is illustrated in accordance with aspects of the present disclosure. Figure 6 An example of 3D Doppler imaging 600 according to aspects of the present disclosure is illustrated. The scheme 400 can be implemented by the system 100. As illustrated, the scheme 400 includes a plurality of enumerated steps, but embodiments of the scheme 400 can include additional steps before, after, and between the enumerated steps. In some embodiments, one or more of the enumerated steps can be omitted or performed in a different order.

[0049] At step 410, the user can select a transcranial examination for the patient, for example, based on an underlying pathology or a clinician-directed protocol. For example, the user can determine to examine the MCA (e.g., the left M1 segment of MAC 212), the PCA (e.g., the right segment of PCA 218), the ICA (e.g., ICA 216), the ACA (e.g., ACA 214), the posterior communicating artery (e.g., the posterior communicating artery 220), the area near the BA (e.g., BA 222), or any area of ​​interest within the patient's brain (e.g., the brain 102). The user can place the ultrasound imaging probe 120 adjacent to or in contact with the patient's head (e.g., the head 110) at an initial position near the transcranial window for the selected transcranial examination. For example, the transcranial window can be a temporal transcranial window, a submandibular transcranial window, a suboccipital transcranial window, or any other suitable transcranial window.

[0050] In an embodiment, scheme 400 may employ a UI (e.g., UI 144) to guide a user in locating a suitable transcranial window for a selected transcranial examination. For example, the UI may display a brain map (e.g., cerebral vascular atlas 140), and the user may select a desired vascular location for a transcranial examination from the brain map. Alternatively, the user may select a type of transcranial examination (e.g., MCA examination). Processing component 136 may determine a transcranial window suitable for the desired transcranial examination based on the user's selection. The UI may provide instructions and / or instructions to guide the user to the corresponding transcranial window. In addition, processing component 136 may determine a target imaging plane for a selected vascular location or a selected transcranial examination.

[0051] At step 420, the user may use the probe 120 to acquire an initial Doppler image of the patient's head when the probe 120 is in the initial position. The initial Doppler image may include blood flow measurements of blood vessels within the patient's head. In one embodiment, the initial Doppler image may be a 2D color Doppler image 510, such as in Figure 5 In another embodiment, for 3D color Doppler imaging, multiple 2D Doppler images, such as those obtained from X-plane imaging, can be used. X-plane refers to a high frame rate 3D imaging strategy in which two 2D planes are obtained at different angles, typically 90 degrees, about the axis of acoustic propagation. As an example, the initial Doppler images may include images corresponding to different views of the 3D image volume, such as those in Figure 6 4 and 5. In general, the scheme 400 can be applied to 2D input data, X-plane input data, multi-plane input data (which may or may not be separated by 90 degrees), or full 3D imaging data. As the size of the input data increases, the computational cost will increase, but the ability of the system to identify the appropriate location in the atlas also increases.

[0052] The probe 120 can transmit ultrasound waves toward the patient's head, which then bounce back from structures (e.g., brain tissue and blood vessels) within the patient's head and are received by the probe 120 as echo signals. The probe 120 can be configured to transmit ultrasound signals at a specific frequency (e.g., between about 1 MHz and about 3 MHz) depending on the desired imaging resolution and / or the absorption of energy by the skull. The velocity of the blood relative to the probe causes a phase shift (i.e., the Doppler effect) as the frequency increases or decreases. For example, a processing component 136 at the host 130 can receive the echo signals, determine the change in frequency, and calculate the velocity of the scatterer.

[0053] In an embodiment, the processing component 136 can employ the following Doppler equation:

[0054] Δf=(2×f0×V×cosθ)) / C (1)

[0055] Where Δf is the frequency shift, f0 is the frequency of the transmitted wave, V is the velocity of the reflector (e.g., red blood cells), θ is the angle between the incident wave and the direction of movement of the reflector (i.e., the angle of incidence), and C is the speed of sound in the medium. The frequency shift is maximum when the transducer is oriented parallel to the direction of blood flow and θ is zero degrees (cos 0=1). The frequency shift is nonexistent when the transducer is oriented perpendicular to the direction of blood flow and θ is 90 degrees (cos 90=0). Higher Doppler shifts are obtained as the velocity increases, the incident wave is more aligned with the direction of blood flow, and / or when higher frequencies are transmitted.

[0056] At step 430, the processing component 136 may determine a graphical representation of the blood vessel captured by the acquired Doppler image. For example, the Cartesian coordinates of the blood vessel may be represented by the coordinates of the blood vessel as described above with respect to Figure 3 300. For example, the processing component 136 can convert the color Doppler image 510 into a graphical representation 520 including nodes 522 (e.g., nodes 310) connected by edges 524 (e.g., edges 312). For example, edge 524u can represent upstream blood flow and can be color-coded in red or indicated by a red arrow, while edge 524d can represent downstream blood flow and can be color-coded in blue or indicated by a blue arrow. The interconnection of nodes 522 and edges 524 in the graphical representation 520 can be expressed as a set of flow vectors. The orientation of the flow vectors in space can be expressed as follows:

[0057]

[0058] Among them, V i represents the flow vector i, x i and i denote the x-coordinate and y-coordinate in the 2D ultrasound imaging plane, θ i represents the tilt angle, and φ i Indicates the azimuth.

[0059] Similarly, when the Doppler images correspond to 3D color Doppler images 610 and 612, the processing component 136 can convert the 3D color Doppler images 610 and 612 into a graphical representation 620 including nodes 622 (e.g., nodes 310) connected by edges 624 (e.g., edges 312). Edges 624u can represent upstream blood flow and edges 624d can represent downstream blood flow. The interconnections of nodes 622 and edges 624 in the graphical representation 620 can be expressed as a set of flow vectors. The orientation of the flow vectors in space can be expressed as follows:

[0060]

[0061] Among them, V i represents the flow vector {i}, x i ,y i 、z i denote the x-coordinate, y-coordinate and z-coordinate in the 3D ultrasound imaging volume, θ i represents the tilt angle, φ i represents the azimuth, and ψ i Represents connectivity parameters.

[0062] For example, the graphical representation of blood vessels in the Doppler image may be divided into subsets of coordinates expressed as follows:

[0063]

[0064] The matrix M comprises a vectorized representation of the graphical representation of the blood vessels.

[0065] At step 440, the processing component 136 may determine the covariance matrix, denoted as C, as follows:

[0066] C=M T ×W×M, (5)

[0067] Among them, M T denotes the transpose of the matrix M, and W denotes a weighting matrix, including weighting factors for the coordinates. In other words, the covariance matrix C comprises the weighted inner products of the N subsets of the coordinates. The matrix M can be used in N subsets of coordinates (e.g., ∈R N ) dataset R N , and the covariance matrix C can be found in the dataset R N×N Within (for example, C∈R N×N ).

[0068] In an embodiment, the weighting factors may be determined empirically and can be different for each coordinate (e.g., between pairs (x, y) and pairs (θ, φ)). In an embodiment, the weighting factors in the matrix W may be configured such that nodes corresponding to the aorta (e.g., nodes 310, 522, and 622) are given higher weights (e.g., larger values) and nodes corresponding to vessel branches are given smaller weights (e.g., smaller values). The weighting factors may be manually determined for a transcranial examination. For example, for a MAC examination, nodes associated with the MCA may be given higher weights than other vessels. In some embodiments, the weighting matrix W may be excluded from the calculation of the covariance matrix C (e.g., all weighting factors are set to a value of one).

[0069] At step 450, the processing component 136 may apply the CNN 142 to the covariance matrix C to identify the current imaging plane of the probe 120 relative to the cerebral vascular atlas 140. The internal architecture, training, and application of the CNN 142 are described in more detail herein.

[0070] At step 460, the processing component 136 can determine a motion control configuration (e.g., including translation and rotation parameters) for repositioning the probe 120 to a target imaging plane for the selected transcranial examination. For example, the target imaging plane for the selected transcranial examination is predetermined relative to the cerebral vascular atlas 140. After the current imaging plane of the probe 120 is identified relative to the cerebral vascular atlas 140, the motion control configuration to reach the target imaging plane can be determined based on geometric distance (e.g., translation) and / or angle (e.g., rotation) calculations.

[0071] In an embodiment, the processing component 136 may calculate a rotation matrix between the current imaging plane and the target imaging plane to obtain an angle or rotation parameter (denoted as (θ, φ)) for repositioning the probe 120 to point to the target imaging plane or target field of view. If the rotation is not sufficient to reach the target imaging plane, the processing component 136 may additionally calculate a translation vector between the current imaging plane and the target imaging plane, which may be outside the current field of view.

[0072] At step 470, the display 134 may provide user guidance for repositioning the probe 120 to the target imaging plane. For example, the display 134 may display a graphical view of the probe 120 indicating the amount or direction of translation and / or rotation for repositioning the probe 120. The graphical display may include an animated movement of the probe 120 to reach the target imaging plane. The display 134 may display a graphical view including superimposing the current imaging plane and / or the target imaging plane on top of the cerebral vascular atlas 140. The graphical display is described in more detail herein.

[0073] At step 475, the user may reposition the probe 120 to the next position according to the user guidance. At step 480, the next Doppler image may be acquired when the probe 120 is at the new position. In some embodiments, steps 430-480 may be repeated for the probe 120 to reach the target imaging plane.

[0074] In some embodiments, the motion control configuration can be sent to a mechanical actuation unit (e.g., robotic system 160) to automatically control or reposition the probe 120, as shown in step 490, rather than providing user guidance and having the user reposition the probe 120, as shown in steps 470 and 475.

[0075] After the probe 120 is aligned with the target imaging plane, the user can proceed with the selected transcranial examination. In some embodiments, the scheme 400 can also use spectral Doppler to further classify the blood vessels under examination and provide further guidance to the user with higher accuracy when reaching the target imaging plane. The coordinates of the desired or target blood vessel obtained from the CNN 142 can be input into the Doppler beamforming unit so that continuous Doppler traces and blood flow velocities can be generated. In some embodiments, after the transcranial examination is completed, the user can update the CNN 142 and / or the cerebral vascular atlas 140 with information associated with the target blood vessel (e.g., coordinates).

[0076] Figure 7-8 The mechanisms in employing CNN 142 and cerebrovascular atlas 140 for transcranial examinations are collectively illustrated. Figure 7 is a schematic diagram illustrating a configuration 700 for CNN 142 in accordance with aspects of the present disclosure. Figure 8 8 is a schematic diagram illustrating a scheme 800 for generating a covariance matrix from a cerebral vascular atlas according to aspects of the present disclosure. Given an input image, a CNN 142 is trained using one or more cerebral vascular atlases 140 to identify vascular locations on the cerebral vascular atlases 140. After the CNN 142 is trained, the CNN 142 is applied to a covariance image 702 (e.g., a covariance matrix C) calculated in real time from live imaging data during a transcranial examination, for example, as described in step 450 of the scheme 400.

[0077] CNN 142 may include a set of N convolutional layers 712 followed by a set of K fully connected layers 714, where N and K may be any positive integers. The values ​​of N and K may vary depending on the embodiment. In some embodiments, N may be between about 3 and about 200, and K may be between about 1 and about 5. Each convolutional layer 712 may include a set of filters 720 configured to extract imaging features (e.g., one-dimensional (1D) feature maps) from an input image. The fully connected layers 714 may be nonlinear and may be used to filter the last convolutional layer 712. (N) The high-dimensional output of gradually shrinks to a length corresponding to the number of classification layers (e.g., various vascular locations on the cerebral vascular atlas 140) at the output 716 of the CNN 142. Figure 7 , but in some embodiments, the convolutional layers 712 may be interleaved with pooling layers, each of which includes a set of downsampling operations that can reduce the dimensionality of the extracted imaging features. In addition, the convolutional layers 712 may include a nonlinear function configured to extract a rectified feature map (e.g., including a rectified nonlinear (ReLU) operation).

[0078] During training of the CNN 142, the cerebral vascular atlas 140 may be converted to coordinates or flow vectors represented by the matrix M as shown above in equation (4). In some embodiments, the coordinates and / or flow vectors may be stored in a 3D node file. The file may include additional information at each vertex or node (e.g., nodes 310, 522, and 622), including artery class, flow direction, artery diameter range, flow range (e.g., for end-diastolic volume (EDV) and / or for end-systolic volume (ESV)), and / or connectivity information (e.g., faces and vertices).

[0079] CNN 142 can be trained based on the weighted covariance C of the matrix M calculated as shown in equation (5) above. As described above, the cerebrovascular atlas 140 may include cerebrovascular topology determined based on real patient data obtained from clinical studies and / or live clinical data. The coordinates in the atlas 140 can be divided into subsets and labeled according to different locations of the brain, for example, including a subset 742 corresponding to the ICA region, a subset 744 corresponding to the PCA region, and a subset 746 corresponding to the MCA region. Each subset 742, 744, and 746 of the coordinates can be labeled according to the corresponding vascular location (e.g., the M1 segment of the MCA). The covariance matrix 740 can be calculated for each subset 742, 744, and 746.

[0080] In an embodiment, a covariance matrix 740 may be generated, such as in Figure 8 As shown in Figure 8 As shown in , the portion of PCA812 in atlas 810 (e.g., atlas 140) is represented by a node graph 820 (e.g., representation 300) in space, including nodes 822 (e.g., nodes 310, 522, and 622) connected by edges 824 (e.g., edges 312, 524, and 624). A covariance matrix 830 (e.g., covariance matrix 740) is calculated based on the node graph 820.

[0081] CNN 142 is trained on the covariance matrix 740 of each subset 742, 744, and 746 retrieved from the atlas, for example, using forward propagation and back propagation. The coefficients of filter 720 can be adjusted, for example, by using back propagation to minimize the classification error (e.g., between the vessel location indicated by output 716 and the label for the corresponding subset 742, 744, or 746). For example, the last convolutional layer 712 (N) A feature vector 718 having coordinates representing a particular vessel location may be output, and output 716 may indicate a classification corresponding to the vessel location.

[0082] In an embodiment, CNN 142 is trained to identify m vessel locations (e.g., classifiers), where m is a positive integer. Thus, CNN 142 may generate output 716 indicating one of the m classes. For example, when CNN 142 operates on covariance matrix 740 of subset 742, CNN 142 may generate output 716 indicating one of the m classes. (N) Output feature vector 718 (1) And outputs a classifier indicating ICA at output 716. Alternatively, when CNN 142 operates on the covariance matrix 740 of subset 746, CNN 142 may perform the following operations at the last convolutional layer 712: (N) Output feature vector 718 (m-2) And a classifier indicating MCA is output at output 716. The training of the CNN 142 may be repeated using multiple cerebrovascular atlases 140 constructed from real patient data obtained via clinical studies and / or vital data from clinical settings.

[0083] During a transcranial examination, a covariance image 702 is inferred in real time using CNN 142 based on ultrasound data obtained from imaging the patient's head (e.g., head 110) (e.g., calculated as shown in equation (5)). Covariance image 702 is matched to corresponding labeled cerebral vessels in cerebral vascular atlas 140 to estimate the possible vessel location within the patient's brain represented by the current frame of color Doppler imaging. For example, the last convolutional layer 712 (N) A feature vector 730 may be output to represent the input covariance image 702. The feature vector 730 may then be matched to a set of m feature vectors 718 that were previously generated by feeding the covariance matrix of the m labeled cerebral vascular atlas into the same CNN 142. The CNN 142 may indicate the classification of the feature vector 730 at the output 716 based on matching the feature vector 730 to the set of m labeled feature maps 718 as indicated by the dashed curved arrows. As an example, the feature vector 730 may match the feature vector 718 as indicated by the solid curved arrows. (m-2) Thus, output 716 may indicate a feature vector 718 that matches (m-2) The corresponding classifier (e.g., MCA) matches feature vector 730 to feature vector 718. (m-2) The vessel location of the current imaging plane on the cerebrovascular atlas 140 corresponding to the covariance image 702 is then identified. The vessel location relative to the current imaging plane of the cerebrovascular atlas 140 may be used to provide user guidance as described in more detail herein.

[0084] In some embodiments, CNN 142 may provide two possible matches for the acquired Doppler image at output 716. For example, CNN 142 may output an approximately 50% match for the MCA and an approximately 50% match for the ICA. In such an embodiment, the user may switch to configure the probe 120 to measure spectral Doppler to obtain a velocity profile to define the classification output of CNN 142. For example, an additional CNN or other waveform matching technique may be used to determine whether the acquired Doppler image corresponds to an image of the MCA or an image of the ICA. When an additional CNN is used, the additional CNN may be trained based on the velocity profiles of various blood vessels obtained from spectral Doppler. The additional CNN may have a substantially similar architecture to CNN 142.

[0085] Fig. 9 is a schematic diagram illustrating a display view 900 for guiding transcranial ultrasound imaging according to aspects of the present disclosure. View 900 may correspond to a display view on display 134 in system 100. View 700 includes three sub-views 910, 920, and 930. Sub-views 910, 920, and 930 may be displayed side by side, as in Fig. 9 , or alternatively, configured in any suitable display configuration to provide similar functionality.

[0086] Subview 910 shows a current image (e.g., live color Doppler images 510, 610, and 612) of a patient's blood vessels under examination using system 100. The current image may be captured in real time by probe 120 at a current imaging plane 922. The current image may correspond to an input to CNN 142 for use in the above-described Figure 7 The described configuration 700 computes an image of the covariance image 702. The subview 910 may include labels marking the vessels captured by the current image. As shown, the subview 910 includes labels marking the MCA (e.g., MCA 212), the ACA (e.g., 214), and the PCA (e.g., 218).

[0087] Subview 920 shows a current imaging plane 922 and a target imaging plane 944 (e.g., with partial transparency) based on the selected transcranial examination superimposed on a cerebrovascular topology map (e.g., cerebrovascular atlas 140). The superimposition of the current imaging plane 922 can be based on a comparison of the feature vector 730 extracted from the current image with the set of m feature vectors 718 extracted from the cerebrovascular atlas 140. In some embodiments, the display of the cerebrovascular atlas 140 can be 3D. The blood vessels in the current imaging and / or the target blood vessels for the transcranial examination can be highlighted on the cerebrovascular atlas 140.

[0088] The sub-view 930 provides instructions to the user to reposition the probe 120 from the current imaging plane 922 to the target imaging plane 924 (e.g., as determined in step 460 of the protocol 400). As shown, the sub-view 930 may include a visual indicator 932 that may illustrate the required translation (e.g., based on the calculated translation (x, y)) and a visual indicator 934 that may illustrate the required rotation (e.g., based on the calculated rotation (θ, φ)) for manipulating the probe 120 to reach the target imaging plane 924. In some embodiments, the sub-view 930 may also display an animated view of the visual indicator 932 and 934 illustrating the suggested movement of the probe 120 to reach the target imaging plane 924.

[0089] In some embodiments, UI 144 may also include a user interface portion 940, for example, including a dial 944. The user may configure the subview 920 by manipulating the dial 944. For example, when the target vessel for transcranial examination is not within the current field of view, the user may manipulate the dial 944 to increase the thickness of the imaging volume beyond the current field of view to obtain a desired view of the target vessel or predict a virtual view. Therefore, when the target vessel may not be in the current field of view, the subview 920 may allow the user to visualize the position of the target vessel relative to the current imaging plane 922. In an embodiment, the virtual target vessel may be displayed in the subview 920 in a transparent mode. The virtual vessel corresponds to the vascular position predicted by the CNN 142 based on the atlas 140. For example, the subview 920 may provide 3D position information of the target vessel when imaging is performed using 2D imaging. In some embodiments, the user interface portion 940 may include other buttons, sliders, and / or any suitable user interface components that can accept user input.

[0090] Fig.10 1 is a flow chart of a method 1000 for applying a CNN to guide an ultrasound imaging component to a desired imaging plane for a transcranial examination according to aspects of the present disclosure. The steps of the method 1000 can be performed by the system 100. The method 1000 can be respectively implemented as in Figure 3 , Figure 4 ,and Figure 7 Similar mechanisms in the described graphical representation 300, scheme 400, and CNN configuration 700. As illustrated, method 1000 includes a plurality of enumerated steps, but embodiments of method 1000 may include additional steps before, after, and during the enumerated steps. In some implementations, one or more of the enumerated steps may be omitted or performed in a different order.

[0091] At step 1010, method 1000 includes receiving a first image (e.g., images 510, 610, and 612) from an ultrasound imaging component (e.g., imaging component 122) when the ultrasound imaging component is positioned at a first imaging position relative to the patient. The first image may represent a blood vessel (e.g., blood vessel 104 or arteries 212, 214, 216, 218, 220, and 222) of a brain (e.g., brain 102) of the patient. The first imaging position may be any suitable position of the patient's head (e.g., head 110). In some embodiments, the first imaging position may correspond to an imaging plane (e.g., imaging plane 922).

[0092] At step 1020, method 1000 includes determining Doppler information based on data associated with the first image. The Doppler information may represent blood flow within a blood vessel of the patient's brain. For example, the Doppler information may be calculated using equation (1) described above.

[0093] At step 1030, method 1000 includes applying a CNN (e.g., CNN 142) to the Doppler information to generate a motion control configuration for repositioning an ultrasound imaging component for the selected transcranial examination. The CNN can be trained based at least on a known vascular topology map (e.g., cerebral vascular atlas 140) within the brains of multiple patients. In some embodiments, the known vascular topology map can be determined based on a previous scan of the brain of the patient under examination. The motion control configuration can include translation and / or rotation parameters for aligning the ultrasound imaging component with a target imaging plane (e.g., target imaging plane 924) for the selected transcranial examination.

[0094] In some embodiments, method 1000 may also include: determining connectivity information (e.g., matrix M) associated with blood vessels of the patient's brain based on the Doppler information; determining a covariance matrix (e.g., matrix C) based on the connectivity information and a weighting function (e.g., matrix W); and applying a CNN to the covariance matrix. The connectivity information may be associated with a structural arrangement of blood vessels and / or flow paths of blood flow through the blood vessels, for example, as shown in graphical representation 300. The connectivity information may include coordinates corresponding to vascular locations and flow paths of blood vessels along the patient's brain (e.g., as shown in equation (2) and the equation (3) ). The weighting function may be associated with the relevance of the vessel location or flow path relative to the transcranial examination.

[0095] In some embodiments, method 1000 may also include: applying CNN to the Doppler information to determine an imaging plane (e.g., initial imaging plane 922) within a known vascular topology map corresponding to the first imaging position: and determining the motion control configuration based on the imaging plane and a target imaging plane (e.g., target imaging plane 924) associated with the transcranial examination within the known vascular topology map.

[0096] In some embodiments, method 1000 may also include: applying a CNN to the Doppler information to determine a feature vector (e.g., feature vector 730) representing a blood vessel of the patient's brain; and determining an imaging plane within a known vascular topology map based on a comparison of the feature vector with a feature vector of a known vascular topology map (e.g., feature map 718).

[0097] At step 1040, method 1000 includes providing user guidance based on the motion control configuration. In some embodiments, the user guidance may include displaying instructions based on the motion control configuration for operating the ultrasound imaging component such that the ultrasound imaging component is repositioned to the target imaging position, the instructions including at least one of a translation or a rotation of the ultrasound imaging component, for example, as shown by visual indicators 932 and 934 in subview 930. In some embodiments, the user guidance may include displaying a graphical view including superimposing at least one of a first imaging plane associated with the first imaging position, a second imaging plane associated with the second imaging position, or a blood vessel of the patient's brain on a known blood vessel topology map, for example, as shown in subview 920. In some embodiments, the user guidance may include displaying a graphical view including superimposing a desired view (e.g., a virtual out-of-plane view) of blood vessels of the patient's brain associated with the second imaging position on a known blood vessel topology map.

[0098] Aspects of the present application can provide several benefits. For example, using deep learning to automatically identify the current imaging plane in real time based on the currently captured image and provide user guidance can eliminate the need for experienced operators to perform TCD ultrasound, and therefore, the use of TCD ultrasound in medical diagnostic processes can be expanded. In addition, automatic identification and user guidance can eliminate the inter-operator variability in TCD ultrasound, and therefore, more consistent and accurate results can be provided for TCD ultrasound-based examinations. For example, the disclosed embodiments can enable TCD ultrasound to be performed in settings such as emergency rooms, rural medical centers, battlefields, and ambulances for continuous monitoring, screening residues, and evidence-based applications for the treatment of conditions containing cerebrovascular diseases. The display of live Doppler images together with the imaged blood vessels or the superposition of the current imaging plane and the target imaging plane on the cerebrovascular map can provide further assistance when guiding the user to the target imaging plane. The real-time or live display of virtual blood vessels around the target blood vessel area outside the current field of view can provide further guidance to the user when searching for or reaching the target blood vessel. In addition, real-time automatic identification enables continuous blood flow measurement results without the need for the user to select a position for measurement within the field of view. Although the disclosed embodiments are described in the context of training and applying a predictive network for guiding an ultrasound imaging probe, the disclosed embodiments can be applied to provide automatic alignment for any imaging component of any imaging modality.

[0099] Those skilled in the art will recognize that the apparatus, system and method described above can be modified in various ways. Therefore, those of ordinary skill in the art will appreciate that the embodiments covered by the present disclosure are not limited to the specific exemplary embodiments described above. In this regard, although illustrative embodiments have been shown and described, various modifications, changes and substitutions are contemplated in the foregoing disclosure. It should be understood that such variations may be made to the foregoing without departing from the scope of the present disclosure. Therefore, it is appropriate that the appended claims be understood broadly and in a manner consistent with the present disclosure.

Claims

1. A medical ultrasound imaging system, comprising: an interface in communication with the ultrasound imaging component and configured to receive a first image representing a blood vessel of a brain of the patient when the ultrasound imaging component is positioned at a first imaging position relative to the patient; as well as a processing component in communication with the interface and configured to: determining Doppler information representative of blood flow within the blood vessel based on data associated with the first image; determining a graphical representation of the blood vessel based on the Doppler information, the graphical representation comprising spatial coordinates describing locations of segments along the blood vessel and connectivity information of the blood vessel; applying a convolutional neural network (CNN) to the graphical representation to identify a first imaging plane of the ultrasound imaging component relative to a known vascular topology map, the first imaging plane corresponding to the first imaging position, the CNN being trained based on at least the known vascular topology map; and A motion control configuration is generated for repositioning the ultrasound imaging component from the first imaging position to a second imaging position associated with a transcranial examination.

2. The system according to claim 1, wherein: The processing unit is further configured to: determining a covariance matrix based on the connectivity information, Therein, the CNN is applied to the covariance matrix.

3. The system according to claim 1, wherein: The processing unit is further configured to: The motion control configuration is determined based on the first imaging plane and a target imaging plane associated with the transcranial examination within the known vascular topology.

4. The system according to claim 3, wherein: The processing unit is further configured to: applying the CNN to the graphical representation to determine feature vectors representing the blood vessels of the patient's brain; and The first imaging plane within the known vascular topology is determined based on a comparison of the feature vector with the known vascular topology.

5. The system according to claim 1, wherein: The CNN is further trained based on at least a covariance matrix determined based on connectivity information of the known vascular topology.

6. The system according to claim 1, wherein: The motion control configuration includes at least one of a translation or a rotation of the ultrasound imaging component.

7. The system of claim 1, further comprising: A user interface in communication with the processing component, the user interface being configured to receive a selection of at least one of a type of the transcranial examination or a target vessel location associated with the transcranial examination, wherein the processing component is further configured to determine the second imaging position based on the selection.

8. The system of claim 1, further comprising: A display is in communication with the processing component, the display being configured to display instructions based on the motion control configuration for operating the ultrasound imaging component such that the ultrasound imaging component is repositioned to the second imaging position.

9. The system of claim 1, further comprising: a display in communication with the processing component, the display being configured to: displaying a graphical view including superimposing at least one of a first imaging plane associated with the first imaging position, a second imaging plane associated with the second imaging position, or the blood vessels of the patient's brain on the known blood vessel topology, and / or A graphical view is displayed that includes superimposing a desired view of blood vessels of the patient's brain associated with the second imaging position over the known vascular topology map.

10. A method for medical ultrasound imaging, comprising: receiving a first image representing a blood vessel of a brain of a patient from the ultrasound imaging component when the ultrasound imaging component is positioned at a first imaging position relative to the patient; determining Doppler information representative of blood flow within the blood vessel based on data associated with the first image; determining a graphical representation of the blood vessel based on the Doppler information, the graphical representation comprising spatial coordinates describing locations of segments along the blood vessel and connectivity information of the blood vessel; applying a convolutional neural network (CNN) to the graphical representation to identify a first imaging plane of the ultrasound imaging component relative to a known vascular topology map, the first imaging plane corresponding to the first imaging position, the CNN being trained based on at least the known vascular topology map; and A motion control configuration is generated for repositioning the ultrasound imaging component from the first imaging position to a second imaging position associated with a transcranial examination.

11. The method according to claim 10, further comprising: determining a covariance matrix based on the connectivity information, Therein, the CNN is applied to the covariance matrix.

12. The method according to claim 10, further comprising: Based on the motion control configuration, instructions are transmitted to at least one of a display or a robotic system for operating the ultrasound imaging component so that the ultrasound imaging component is repositioned to the second imaging position, and the instructions include at least one of a translation or a rotation of the ultrasound imaging component.

Images