Background multiplanar reconstruction of three-dimensional ultrasound imaging data and associated devices, systems, and methods

Abstract

An ultrasound imaging system includes a processor circuit in communication with an ultrasound transducer, the processor circuit configured to receive three-dimensional ultrasound data of an anatomical structure and generate a target image corresponding to a target image plane of the anatomical structure and a plurality of adjacent images corresponding to image planes adjacent to the target image plane along a simulated motion path. The processor circuit is further configured to display the target image, receive a user input indicative of a direction of motion along the simulated motion path, and display an adjacent image of the plurality of adjacent images corresponding to the direction of motion. Thus, a user can view the target image in its spatial context by scanning through the target image and one or more adjacent images on a display as if the ultrasound transducer were scanning along the simulated motion path.

Description

TECHNICAL FIELD

[0001] The present disclosure relates generally to the acquisition and processing of ultrasound images, and in particular to systems and methods for reconstructing two-dimensional images from three-dimensional ultrasound image data. BACKGROUND

[0002] Ultrasound imaging is often used to obtain images of a patient's internal anatomy. An ultrasound system typically includes an ultrasound transducer probe that includes a transducer array coupled to a probe housing. The transducer array is activated to vibrate at ultrasound frequencies, thereby emitting ultrasound energy into a patient's anatomy, and then receiving ultrasound echoes reflected or backscattered by the patient's anatomy to create an image. Such a transducer array can include various layers, including some layers of piezoelectric material that vibrate in response to an applied voltage to produce a desired pressure wave. These transducers can be used to successively emit and receive a number of ultrasound pressure waves through various tissues of the body. The various ultrasound responses can also be processed by the ultrasound imaging system to display various structures and tissues of the body.

[0003] Doctors and sonographers often desire to obtain certain views of a patient's body such that the imaging plane of the ultrasound transducer is aligned to obtain images of a particular combination and orientation of anatomical structures. For example, in fetal ultrasound biometry, a sonographer obtains various images or views of the fetal anatomy to perform measurements and assessments of the development of the fetus. Examples of these views include transventricular, transcerebellar, and transthalamic image planes. Typically, these target views are achieved manually by an experienced sonographer positioning and orienting the ultrasound transducer while viewing a real-time or near real-time image stream of the field of view of the ultrasound transducer.

[0004] When the sonographer determines that the desired view is shown, the sonographer can freeze or save the image frame to memory, and / or continue to move the probe around the target view to ensure that the desired view is actually achieved. The temporal and spatial information obtained in a conventional two-dimensional ultrasound imaging procedure can be useful in the assessment images, as it provides important context information in addition to the dedicated two-dimensional images to be used in the biometry assessment, for example. Thus, the sonographer has a spatial awareness of the location of the plane they are searching for based on their knowledge of the images of the anatomy in the vicinity or proximity of the plane they want to capture. In essence, the confidence in correctly detecting the target plane is based on the temporal context in the two-dimensional image stream. Nonetheless, this conventional two-dimensional acquisition workflow relies heavily on the individual skills of the sonographer.

[0005] Alternatively, some ultrasound imaging systems allow for automatic acquisition of a target view from a three-dimensional ultrasound data set. Automatic image reconstruction techniques can be beneficial because they are less dependent on the individual skill of the sonographer. For example, an ultrasound transducer array can be used to sweep out a three-dimensional volume and obtain a three-dimensional ultrasound data set representing the volume while the sonographer holds the ultrasound probe body in a fixed position and orientation. A multi-planar reconstruction (MPR) technique can be used to generate a two-dimensional image associated with the desired view. Although MPR imaging techniques advantageously allow for less experienced sonographers to achieve a target view, one drawback of MPR is that the image can not be correctly reconstructed by the system and the user does not have the spatial / temporal context to confirm that the reconstructed image is correct or optimal. In contrast, conventional two-dimensional ultrasound imaging provides real-time feedback and spatial context to the sonographer because the user can make manual adjustments to the probe to evaluate the imaging volume around the target view. SUMMARY

[0006] Aspects of the present disclosure provide ultrasound systems and devices that provide multi-planar reconstruction of images with context visualization to increase confidence in the selection of a plane during an ultrasound imaging procedure. For example, in one embodiment, an ultrasound imaging system includes a processor circuit in communication with an ultrasound transducer configured to obtain a three-dimensional ultrasound data set. The processor circuit is configured to reconstruct a target image or image slice corresponding to a target view or image plane (such as an apical view of a heart, trans cerebellum, trans thalamus, etc.) from the ultrasound data set. In addition, the processor circuit is configured to reconstruct one or more adjacent images corresponding to planes adjacent to the target image plane from the same three-dimensional ultrasound data set. The adjacent images are reconstructed such that the adjacent images and the target image correspond to image planes on a simulated motion path that also corresponds to the target image. For example, the simulated motion path can represent a linear translation, sweep or fan motion, tilt, wobble motion, and / or rotation of the ultrasound probe, where the target image plane extends from a point along the simulated motion path. The reconstructed adjacent images can be output to a display such that a user can view the adjacent image portion of the sequence with the target image to obtain spatial context information about the target image. In this regard, in some embodiments, the user can scan through the target image and the one or more adjacent images on the display as if the ultrasound transducer were scanning along the simulated motion path. Thus, the user can determine that the target image reconstructed from the image data is correctly aligned with the target image plane, or determine that an adjustment can be made to reconstruct the target image to be correctly aligned with the target image plane.

[0007] According to one embodiment of the present disclosure, an ultrasound imaging apparatus includes a processor circuit configured to: receive three-dimensional ultrasound data of an anatomical structure from an ultrasound probe communicatively coupled to the processor circuit; generate a target image corresponding to a target image plane of the anatomical structure from the three-dimensional ultrasound data; generate a plurality of adjacent images corresponding to image planes adjacent to the target image plane along a simulated motion path from the three-dimensional ultrasound data, wherein the target image and the plurality of adjacent images comprise two-dimensional images based on the three-dimensional ultrasound data; output the target image to a display in communication with the processor circuit; receive a user input representing a direction of motion along the simulated motion path; and output an adjacent image of the plurality of adjacent images corresponding to the direction of motion to the display.

[0008] In some embodiments, the apparatus further includes the ultrasound probe. In some embodiments, the processor circuit is configured to: interpolate between a position and orientation of the target image and a position and orientation of the adjacent image to generate an interpolated image; and output the interpolated image to the display. In some embodiments, the processor circuit is configured to: determine a direction of uncertainty relative to the target image plane; and determine the simulated motion path based on the determined direction of uncertainty. In some embodiments, the processor circuit is configured to apply a covariance matrix to the three-dimensional ultrasound data to determine the direction of uncertainty.

[0009] In some embodiments, the processor circuit is configured to: identify a plane of interest different from the target image and the adjacent image from the three-dimensional ultrasound data; and determine the simulated motion path based on the target image, the adjacent image, and the plane of interest. In some embodiments, the processor circuit is configured to output the plane of interest in response to receiving the user input. In some embodiments, the plurality of adjacent images includes a plurality of parallel adjacent images associated with a plurality of parallel adjacent image planes. In some embodiments, the processor circuit is configured to generate the target image to exclude an anatomical feature. In some embodiments, the processor circuit is configured to generate the adjacent image to include the anatomical feature. In some embodiments, the processor circuit is further configured to output a graphical representation of an adjacent image plane associated with the adjacent image to the display, wherein the graphical representation includes: a diagrammatic view of a body portion associated with the target image; and an indicator of the adjacent image plane superimposed on the diagrammatic view of the body portion.

[0010] According to another embodiment, a method for reconstructing an ultrasound image includes receiving three-dimensional ultrasound data of an anatomical structure obtained by an ultrasound probe; generating a target image corresponding to a target image plane of the anatomical structure from the three-dimensional ultrasound data; generating a plurality of adjacent images corresponding to image planes adjacent to the target image plane along a simulated motion path from the three-dimensional ultrasound data, wherein the target image and the plurality of adjacent images comprise two-dimensional images based on the three-dimensional ultrasound data; outputting the target image to a display; receiving a user input representing a motion direction along the simulated motion path; and outputting an adjacent image of the plurality of adjacent images corresponding to the motion direction to the display.

[0011] In some embodiments, generating the plurality of adjacent images includes interpolating between a position and orientation of the target image and a position and orientation of the adjacent images to generate an interpolated image. In some embodiments, the method further includes outputting the interpolated image to the display. In some embodiments, generating the plurality of adjacent images includes determining an uncertainty direction relative to the target image plane; and determining the simulated motion path based on the determined uncertainty direction. In some embodiments, determining the uncertainty direction includes applying a covariance matrix to the three-dimensional ultrasound data.

[0012] In some embodiments, the method further includes identifying a different image of interest from the target image and the adjacent images from the three-dimensional ultrasound data; and determining the simulated motion path based on the target image, the adjacent images, and the image of interest. In some embodiments, the method further includes outputting the image of interest in response to receiving the user input. In some embodiments, generating the plurality of adjacent images includes generating a plurality of parallel adjacent images associated with a plurality of parallel adjacent image planes. In some embodiments, generating the target image includes generating the target image to exclude an anatomical feature. In some embodiments, generating the adjacent images includes generating the adjacent images to include the anatomical feature. In some embodiments, the method further includes outputting a graphical representation of an adjacent image plane associated with the adjacent images to the display. In some embodiments, the graphical representation includes a diagrammatic view of a body part associated with the target image; and an indicator of the adjacent image plane superimposed on the diagrammatic view of the body part.

[0013] Additional aspects, features, and advantages of the disclosure will become apparent from the following detailed description. BRIEF DESCRIPTION OF DRAWINGS

[0014] Illustrative embodiments of the disclosure will be described with reference to the accompanying drawings, of which:

[0015] Figure 1 is a schematic illustration of an ultrasound imaging system according to embodiments of the present disclosure.

[0016] Figure 2 is a schematic illustration of a processor circuit according to embodiments of the present disclosure.

[0017] Figure 3 is a diagrammatic view of image slices reconstructed from a volumetric three-dimensional ultrasound data set according to aspects of the present disclosure.

[0018] Figure 4 is a flowchart illustrating a method for generating and displaying a background multiplanar image reconstruction from a three-dimensional ultrasound data set according to aspects of the present disclosure.

[0019] Figure 5 is a diagrammatic view of a plurality of image planes or image slices corresponding to a simulated linear motion path according to aspects of the present disclosure.

[0020] Figure 6A is a diagrammatic view of a plurality of image planes or image slices corresponding to a simulated linear motion path according to aspects of the present disclosure.

[0021] Figure 6B is a diagrammatic view of a plurality of image planes or image slices corresponding to a simulated curved motion path according to aspects of the present disclosure.

[0022] Figure 6C is a diagrammatic view of a plurality of image planes or image slices corresponding to a simulated curved motion path according to aspects of the present disclosure.

[0023] Figure 7 is a flowchart illustrating a method for generating interpolated image slices for a background multiplanar image reconstruction sequence according to aspects of the present disclosure.

[0024] Figure 8 is a flowchart illustrating a method for reconstructing image slices based on a direction of uncertainty of a background multiplanar image reconstruction sequence according to aspects of the present disclosure.

[0025] Figure 9 is a screen of a graphical user interface including a reconstructed image frame corresponding to a target view according to aspects of the present disclosure. DETAILED DESCRIPTION

[0026] For the purpose of promoting an understanding of the principles of the present disclosure, reference will now be made to the embodiments illustrated in the drawings, and specific language will be used to describe the same. It will nevertheless be understood that no limitation of the scope of the disclosure is intended. Alterations and further modifications of the described devices, systems, and methods, and any additional applications of the principles of the disclosure are fully contemplated as would normally be understood by those skilled in the art to which the disclosure pertains. In particular, 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. However, for the sake of brevity, we will not separately describe all possible combinations of these features, components, and / or steps.

[0027] In Figure 1 An ultrasound system 100 according to embodiments of the present disclosure is shown in block diagram form in FIG. 1. An ultrasound imaging device or ultrasound probe 10 has a transducer array 12 comprising a plurality of ultrasound transducer elements or acoustic elements. In some instances, the array 12 can comprise any number of acoustic elements. For example, the array 12 can comprise between 1 acoustic element and 100,000 acoustic elements, including values such as 2 acoustic elements, 4 acoustic elements, 36 acoustic elements, 64 acoustic elements, 128 acoustic elements, 300 acoustic elements, 812 acoustic elements, 3000 acoustic elements, 9000 acoustic elements, 30,000 acoustic elements, 65,000 acoustic elements, and / or other values that are larger or smaller. In some instances, the acoustic elements of the array 12 can be arranged in any suitable configuration, such as a linear array, a planar array, a curved array, a curvilinear array, a circumferential array, a ring array, a phased array, a matrix array, a one-dimensional (ID) array, a 1.X-dimensional array (e.g., a 1.5D array), or a two-dimensional (2D) array. The array of acoustic elements can be controlled and activated uniformly or independently (e.g., one or more rows, one or more columns, and / or one or more orientations). The array 112 can be configured to obtain one-dimensional, two-dimensional, and / or three-dimensional images of patient anatomy.

[0028] While the present disclosure relates to synthetic aperture external ultrasound imaging using an external ultrasound probe, it will be understood that one or more aspects of the present disclosure can be implemented in any suitable ultrasound imaging probe or system, including external ultrasound probes and intraluminal ultrasound probes. For example, aspects of the present disclosure can be implemented in an ultrasound imaging system using a mechanically scanned external ultrasound imaging probe, an intracardiac (ICE) echocardiogram catheter, and / or a transesophageal echocardiogram (TEE) probe, a rotational intravascular ultrasound (IVUS) imaging catheter, a phased array IVUS imaging catheter, a transthoracic echocardiogram (TTE) imaging device, or any other suitable type of ultrasound imaging device.

[0029] Referring againFigure 1 The acoustic elements of array 12 can include piezoelectric / piezoresistive elements, lead zirconate titanate (PZT), piezoelectric micromachined ultrasonic transducer (PMUT) elements, capacitive micromachined ultrasonic transducer (CMUT) elements, and / or any other suitable type of acoustic element. The acoustic elements of array 12 are in communication with (e.g., electrically coupled to) electronic circuitry 14. In some embodiments, such as the embodiment of FIG. 1, electronic circuitry 14 can include a microbeamformer (pBF). In other embodiments, the electronic circuitry includes multiplexer circuitry (MUX). Electronic circuitry 14 is located in probe 10 and is communicatively coupled to transducer array 12. In some embodiments, one or more components of electronic circuitry 14 can be positioned in probe 10. In some embodiments, one or more components of electronic circuitry 14 can be positioned in computing device or processing system 28. Computing device 28 can be or include a processor, such as one or more processors in communication with memory. As described further below, computing device 28 can include processor circuitry as illustrated in FIG. 2. In some aspects, some components of electronic circuitry 14 are positioned in probe 10 and other components of electronic circuitry 14 are positioned in computing device 28. Electronic circuitry 14 can include one or more electronic switches, transistors, programmable logic devices, or other electronic components configured to combine multiple inputs and / or sequentially switch between multiple inputs to transmit a signal from each of the multiple inputs across one or more common communication channels. Electronic circuitry 14 can be coupled to the elements of array 12 through multiple communication channels. Electronic circuitry 14 is coupled to cable 16, which transmits signals including ultrasonic imaging data to computing device 28. Figure 1 Figure 2 In computing device 28, the signals are digitized and coupled to channels of system beamformer 22, which appropriately delays each signal. The delayed signals are then combined to form a coherent steered and focused receive beam. The system beamformer can include electronic hardware components that perform a beamforming algorithm, hardware controlled by software, or a microprocessor. In this regard, beamformer 22 can be referred to as electronic circuitry. In some embodiments, beamformer 22 can be a system beamformer, such as a

[0030] In computing device 28, the signals are digitized and coupled to channels of system beamformer 22, which appropriately delays each signal. The delayed signals are then combined to form a coherent steered and focused receive beam. The system beamformer can include electronic hardware components that perform a beamforming algorithm, hardware controlled by software, or a microprocessor. In this regard, beamformer 22 can be referred to as electronic circuitry. In some embodiments, beamformer 22 can be a system beamformer, such as a Figure 1 ​The system beamformer 22 can be implemented by a circuit within the ultrasound probe 10, or it can be a beamformer implemented by circuitry within the probe 10. In some embodiments, the system beamformer 22 works in conjunction with a micro-beamformer (e.g., the electronic circuitry 14) disposed within the probe 10. In some embodiments, the beamformer 22 can be an analog beamformer, or in some embodiments, a digital beamformer. In the case of a digital beamformer, the system includes an A / D converter that converts the analog signals from the array 12 into sampled digital echo data. The beamformer 22 will generally include one or more microprocessors, shift registers, and / or digital or analog memory to process the echo data into coherent echo signal data. Delays are implemented by various means, such as by the time of sampling of the received signals, the write / read interval of data temporarily stored in memory, or by the length or clock rate of shift registers, as described in U.S. Patent 4,173,007 to McKeighen et al., which is hereby incorporated by reference in its entirety. Additionally, in some embodiments, the beamformer can apply appropriate weights to each of the signals generated by the array 12. The beamformed signals from the image field are processed by a signal and image processor 24 to produce 2D or 3D images for display on an image display 30. The signal and image processor 24 can include electronic hardware components that perform image processing algorithms, hardware controlled by software, or microprocessors. It will generally also include specialized hardware or software, such as scan converters, that process the received echo data into image data for the desired display format. In some embodiments, the beamforming functionality can be split between different beamforming components. For example, in some embodiments, the system 100 can include a micro-beamformer located within the probe 10 and in communication with the system beamformer 22. The micro-beamformer can perform preliminary beamforming and / or signal processing that can reduce the number of communication channels required to transfer the received signals to the computing device 28.

[0031] Control of the ultrasound system parameters, such as scan mode (e.g., B-mode, M-mode), probe selection, beam steering and focusing, and signal and image processing, is under the control of a system controller 26, which is coupled to the various modules of the system 100. The system controller 26 can be formed by an application specific integrated circuit (ASIC) or microprocessor circuitry and software data storage devices, such as RAM, ROM, or hard drives. In the case of the probe 10, some of this control information can be provided from the computing device 28 to the electronic circuitry 14 over the cable 16, adjusting the electronic circuitry 14 to operate the array for the needs of a particular scan procedure. The user enters these operating parameters by means of the user interface device 20.

[0032] In some embodiments, the image processor 24 is configured to generate different modes of images for further analysis or for output to the display 30. For example, in some embodiments, the image processor can be configured to compile B-mode images of the patient's anatomy, such as live B-mode images. In other embodiments, the image processor 24 is configured to generate or compile M-mode images. M-mode images can be described as images that show the time-varying behavior of the imaged anatomy along a single scan line.

[0033] It will be appreciated that the computing device 28 can include hardware circuitry (such as a computer processor, an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), a capacitor, a resistor, and / or other electronic devices), software, or a combination of hardware and software. In some embodiments, the computing device 28 is a single computing device. In other embodiments, the computing device 28 includes separate computer devices in communication with one another.

[0034] Figure 2 is a schematic diagram of a processor circuit 150 according to embodiments of the present disclosure. The processor circuit 150 can be implemented in the computing device 28, the signal and image processor 24, the controller 26, and / or the probe 10. Figure 1 As shown, the processor circuit 150 can include a processor 160, a memory 164, and a communication module 168. These elements can be in communication with one another, directly or indirectly, for example via one or more buses.

[0035] The processor 160 can include a central processing unit (CPU), a digital signal processor (DSP), an ASIC, a controller, a FPGA, another hardware device that is configured to perform the operations described herein, a firmware device, or any combination thereof. The processor 160 can also be implemented as a combination of computing devices (e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration).

[0036] The memory 164 can include cache memory (e.g., cache memory of the processor 160), random access memory (RAM), magneto resistive 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 drive, other forms of volatile and non-volatile memory, or a combination of different types of memory. In an embodiment, the memory 164 includes a non-transitory computer-readable medium. The memory 164 can store instructions 166. The instructions 166 can include, when executed by the processor 160, cause the processor 160 to perform one or more of the methods described herein with respect to the processor 28 and / or the probe 10. Figure 1instructions 166 can also be referred to as code. The terms "instructions" and "code" should be interpreted broadly to mean any type of computer readable statement(s). For example, the terms "instructions" and "code" can refer to one or more programs, routines, sub-routines, functions, procedures, etc. "Instructions" and "code" can include a single computer readable statement, or many, compressed, or compressed.

[0037] Communication module 168 can include any electronic and / or logic circuitry to facilitate direct or indirect data communication between processor 28, probe 10, and / or display 30. In this regard, communication module 168 can be an input / output (I / O) device. In some instances, communication module 168 facilitates direct or indirect communication between various elements of processor circuitry 150 and / or processing system 106 Figure 1 A).

[0038] As noted above, some ultrasound imaging systems are configured to obtain three-dimensional ultrasound data of a volume of a patient. In some embodiments, the ultrasound imaging system includes an ultrasound transducer comprising an array of one or more transducer elements, where the ultrasound transducer is configured to scan or sweep through a volume of a patient to obtain a three-dimensional ultrasound data set of the volume. This three-dimensional ultrasound data can be used to generate a three-dimensional image or model of the anatomy. Alternatively, two-dimensional images corresponding to various imaging planes intersecting the volume can be reconstructed from the three-dimensional ultrasound data. In this regard, Figure 3 A plurality of image slices 210, 220 forming a three-dimensional ultrasound data set of volume 200 are shown. Slices 210, 220 are obtained using ultrasound transducer array 12 as part of an image sequence. In the illustrated embodiment, array 12 includes a two-dimensional array of ultrasound transducer elements that can be controlled as a phased array to electronically steer an ultrasound beam of acoustic energy, thereby sweeping out image slices 210, 220 in a plurality of corresponding image planes. In other embodiments, ultrasound transducer array 12 can include a single ultrasound transducer element or a one-dimensional array of transducer elements, where transducer 12 is manually scanned in one or more degrees of freedom to acquire the three-dimensional data set of volume 200.

[0039] In Figure 3In particular embodiments, the image slices 210, 220 can correspond to image planes of an image sequence performed by the ultrasound transducer. Thus, the three-dimensional data set of the volume 200 can be formed by acquiring ultrasound data in a step-wise fashion, where the image slices are obtained by steering or sweeping an ultrasound beam or scan line across a corresponding image plane and combining multiple different image slices 210, 220 to form the three-dimensional data set. In some embodiments, the slices 210, 220 can be combined into a three-dimensional image by determining interpolated intensity values from different slices 210, 220, such that an intensity value can be determined for any location in the three-dimensional image (e.g., in a Cartesian coordinate system). Additionally, Figure 3 A reconstructed image 230 is shown that is reconstructed from the three-dimensional data set after the image sequence has completed scanning the volume 200. Thus, in contrast to the image slices 210, 220, the reconstructed image 230 does not directly correspond to a two-dimensional segment of the three-dimensional imaging sequence, and does not represent an image plane intersecting the ultrasound transducer 12. For example, if the ultrasound transducer 12 has been moved to a different position and orientation relative to the volume 200, the reconstructed image 230 can represent an image plane through the volume. However, in some embodiments, the image can be reconstructed from the three-dimensional data set representing an imaging plane intersecting the ultrasound transducer 12.

[0040] In some embodiments, multi-planar reconstruction (MPR) can be used to reconstruct a target image (e.g., a trans-ventricular, trans-cerebellar, trans-thalamic, apical view, etc.) corresponding to a target view or imaging plane of the patient from the three-dimensional ultrasound data. For example, MPR can be used to generate Figure 3The MPR can be performed using input from various image processing techniques, including artificial intelligence (A.I.), machine learning, and / or deep learning techniques, to identify a two-dimensional cross-section in the three-dimensional ultrasound data that includes a combination and / or arrangement of anatomical structures associated with a particular view. In this regard, while the MPR involves generating an image for an identified plane within a three-dimensional data set, the estimation or identification of the plane to be used in the MPR image can be referred to as plane estimation (PE). Methods for PE are described, for example, in U.S. Patent No. 6,443,896, issued September 3, 2002, U.S. Publication No. 2014 / 0155737, filed August 13, 2012, U.S. Publication No. 2010 / 0268085, filed November 13, 2008, and U.S. Publication No. 2019 / 0272667, filed June 12, 2017, and in “You Only Look Once: Unified, Real-Time Object Detection” by Joseph Redmon et al., The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, pp. 779-788, each of which is incorporated herein by reference in its entirety.

[0041] While PE advantageously allows for automatic reconstruction of a target image (i.e., manual adjustment of the probe to achieve a target view) with less input from the sonographer, PE techniques can have some drawbacks. For example, the sonographer can not be able to readily confirm that the image generated and displayed as an MPR using PE is a correct or optimal view. In this regard, because the target image is automatically reconstructed and displayed with little or no interaction from the sonographer, the sonographer can lack the spatial context that would otherwise be obtained if the sonographer were manually scanning the probe around the anatomical structures of the target view. As a result, confidence that the correct image has been acquired can be reduced. Additionally, an incorrect or suboptimal image plane can be selected by the PE algorithm, and the sonographer can not be able to readily identify what adjustments to make to more fully achieve the desired view.

[0042] Accordingly, the present disclosure provides devices, systems, and associated methods for providing a contextual visualization associated with an automatically reconstructed two-dimensional image, such as an image generated from a PE. In particular, the present disclosure provides automated image reconstruction techniques that allow for a target image to be generated and displayed while still providing spatial context information to the sonographer so that the sonographer can confirm that the correct image plane has been selected for the target image to be automatically reconstructed. For example, in an example workflow, a sonographer places an ultrasound probe configured to obtain three-dimensional ultrasound data on the skin of a patient at a location that is typically associated with a target image plane. A three-dimensional data set of the patient's volume is acquired, and the sonographer can put down the probe. A processing system and / or processor circuit analyzes the three-dimensional ultrasound data using, for example, a PE and / or MPR to identify, reconstruct, or otherwise generate a target image associated with a target image plane or view, such as a transver sine cerebri view. The target image can then be displayed to the user using the MPR, which can allow the user to take measurements or assessments of the anatomy in the target image.

[0043] Additionally, the processor circuit is configured to provide spatial context for the target image by generating a plurality of proximate images associated with imaging planes proximate to the target image plane. The proximate images are generated such that the target image and the proximate images all correspond to a common simulated motion path or trajectory. The simulated motion path is calculated such that a scan through the proximate images and the target image simulates a physical scan or pacing of an ultrasound probe around the target image plane. As described below, in example embodiments, the sonographer can scan through the proximate images along the simulated motion path using a keyboard, trackball, or other interface device as if the user were manually moving the probe around the target image plane. Accordingly, the sonographer can have increased confidence that the processor circuit has correctly and / or optimally reconstructed the target image. Additionally, by scanning the reconstructed proximate images, the sonographer can be able to determine whether the target image reconstruction should be modified or re-done.

[0044] Accordingly, embodiments of the present disclosure provide an experience of seeing two-dimensional images that are proximate or close to the target plane. The two-dimensional images can be created or reconstructed from the three-dimensional data set, recreating a conventional experience in order to improve the sonographer's confidence that the desired anatomical plane has been correctly detected by the automated process. In particular, the systems and methods described herein enable a simulation of ultrasound probe movements that allow for a smooth transition from one automatically detected plane to another. These smooth image transitions simulate the transducer movements that a sonographer is familiar with from standard two-dimensional ultrasound examinations. For example, a smooth transition between different view planes can be achieved by interpolating the plane parameters (normal and offset vectors) and thus obtaining a continuous spline trajectory for the two-dimensional plane displacement in three-dimensional space.

[0045] The sequence of image frames shown to the user can be an assembly of neighboring MPRs determined based on geometric spacing. In some embodiments, the sequence of images or image frames shown includes a selection of geometrically neighboring images that have been selected, e.g., by an A.I. algorithm, to slightly resemble the target image plane. For example, in the case of a fetal head biometric assessment, where multiple images are captured at different image planes, the proposed dynamic representation can be used to perform interpolation between the respective image plane parameters. These interpolated images can be used to dynamically transition from one selected plane to another, thereby performing a transition through the three-dimensional volume in which the planes are located.

[0046] Once a target image of a three-dimensional data set is generated, one or more neighboring images can be generated in a variety of ways. In one embodiment, the neighboring image frames correspond to planes that are parallel to the target plane. In applications where more than one plane of interest is within the volume (e.g., abdomen and chest, trans-ventricular (TV), trans-cerebellum (TC), trans-thalamus (TT)), a set of neighboring image frames can be created by interpolating between the planes of interest. To this end, interpolation can be performed in a fixed order (e.g., always from TT to TV), or by finding the two closest planes. In some embodiments, an A.I.-based approach for plane detection provides a way of measuring uncertainty, such that the direction of highest uncertainty can be identified. The images can be interpolated along the direction of highest uncertainty. This will utilize the effective manual correction of the images determined automatically.

[0047] In applications where the target image plane is required to exclude certain anatomical structures or features, the neighboring images can be generated such that at least one of the neighboring images includes the anatomical feature to be excluded, and the other neighboring images are generated to show the anatomical feature that has disappeared from the field of view (e.g., the cerebellum in a TV plane of a fetal brain). For educational or reporting purposes, the set of neighboring images can be displayed with schematic anatomical views, as the three-dimensional volume together with the estimated plane geometry and the positioned anatomical objects provide sufficient information for registration to a three-dimensional anatomical model.

[0048] Figure 4 is a flowchart illustrating a method 300 for providing a spatial context for automatic ultrasound image plane reconstruction. It will be understood that one or more steps of the method 300 can be performed by, for example, the ultrasound imaging system 100 and / or the A.I. system 200 shown in Figure 1 Figure 2 ​The processor circuit 150 shown in FIG. 1 executes. In step 310, the processor circuit receives three-dimensional ultrasound data of a patient's anatomy from an ultrasound probe communicatively coupled to the processor circuit. In some embodiments, the ultrasound data can be obtained during a fetal ultrasound procedure, a transthoracic ultrasound procedure, a cardiac ultrasound procedure, or any other suitable procedure. The ultrasound probe includes an ultrasound transducer having an array of one or more ultrasound transducer elements. In some embodiments, the ultrasound transducer includes a one-dimensional array of ultrasound transducer elements, a 1.5-dimensional array, a 1.X-dimensional array, a two-dimensional array, or any other suitable type of array. In some embodiments, the array is mechanically scanned to obtain a plurality of image slices of a volume. In some embodiments, the array operates as a solid state array or a phased array configured to electronically scan across a volume. In some embodiments, the three-dimensional ultrasound data is comprised of a plurality of two-dimensional images or image slices. In some embodiments, the three-dimensional ultrasound data is comprised of a plurality of individual scan lines of ultrasound data.

[0049] In some embodiments, the ultrasound data received by the processor circuit includes raw ultrasound signals or data. In other embodiments, the ultrasound data received by the processor circuit includes beamformed, partially beamformed, and / or filtered ultrasound data. In some embodiments, the processor circuit receives the ultrasound data directly from the ultrasound probe. In some embodiments, the ultrasound data is first received by a memory that stores the ultrasound data, and then the processor circuit receives the ultrasound data from the memory. Thus, in some embodiments, the processor circuit indirectly receives the ultrasound data from the ultrasound probe after being stored in the memory.

[0050] In some embodiments, step 310 includes generating user instructions to be output to a user output device or interface (e.g., a display, a speaker, etc.) to first place the ultrasound probe in a general position such that the target view is within a three-dimensional field of view of the ultrasound probe. For example, the user instructions can include a diagrammatic illustration of the patient's body, and an indicator showing how to position and orient the probe relative to the patient's body.

[0051] In step 320, the processor circuit generates a target image corresponding to a target image plane of the anatomical structure from the three-dimensional ultrasound data. For example, as described above, the target image plane or view to be achieved is a transventricular, transcerebellar, or transthalamic view of the fetus, an apical view of the heart of the patient, or any other suitable view. In some embodiments, generating the target image includes automatically reconstructing the target image using PE and / or MPR procedures. However, various image processing and recognition techniques can be used, including A.I. techniques, machine learning techniques, deep learning techniques, state machines, etc. In some embodiments, generating the target image can include comparing the three-dimensional image data to a model of the anatomical structure. In some embodiments, generating the target image includes comparing various portions of the three-dimensional ultrasound data representing different cross-sections of the three-dimensional field of view to one or more exemplary images of the anatomical structure.

[0052] In step 330, the processor circuit generates one or more proximate images corresponding to image planes proximate to the target image plane along a simulated motion path from the three-dimensional ultrasound data, wherein the target image and the one or more proximate images include two-dimensional images based on the three-dimensional ultrasound data. The proximate image(s) generated during step 330 can be used to provide background information to the user, which can accompany the display of the target image generated in step 320.

[0053] As described above, the simulated motion path can represent a physical adjustment or movement of the ultrasound probe, such as a translation, linear movement, compression, rotation, sector, sweep, wobble, or any other suitable type of motion. In some embodiments, step 330 includes determining or calculating the simulated motion path. For example, the processor circuit can determine or calculate the simulated motion path based on the target view to be reconstructed. The simulated motion path can be a parameter for reconstructing the proximate images, where the simulated motion path or trajectory defines the spatial relationship between the proximate images to be reconstructed from the three-dimensional ultrasound data of the volume. In some embodiments, the path will be defined by a set of points that are smoothly connected to a curved line, for example based on a spline interpolation technique. The set of points can be a set of anatomical landmarks, such as organs to be examined, or can follow an anatomical structure such as the spine. In another embodiment, the path can be constructed such that, in order to educate, when the transducer is manipulated (such as translating the transducer position on the skin surface, or rotating / tilting the transducer around its three main axes), it shows the degrees of freedom.

[0054] Figure 5Target image 410, neighboring image 420, and interpolated image 430 are shown corresponding to simulated motion path 440. In the illustrated embodiment, the darker rectangles represent image planes 410, 420 reconstructed from ultrasound data. The white rectangle represents an interpolated image plane 430 generated using target image 410 and neighboring image 420. Target image 410, neighboring image 420, and interpolated image 430 are generated to correspond to simulated motion path 440. In this regard, simulated motion path 440 can be determined or calculated by the processor circuit, and can represent a simulated type of movement of the ultrasound probe (i.e., as if the ultrasound probe physically scanned over the volume). In the illustrated embodiment, simulated motion path 440 represents a mixed tilt / rocking movement of the ultrasound probe. Orthogonal axes 442 of images 410, 420, 430 are tangent to simulated motion path 440 at points where images 410, 420, 430 intersect simulated motion path 440.

[0055] It will be appreciated that the present disclosure contemplates a variety of simulated motion paths. The simulated motion path can be predetermined or selected based on a target imaging plane desired to be obtained. In this regard, Figure 6A , 6B and 6C represent reconstructed and interpolated image planes corresponding to three different simulated motion paths. For example, in Figure 6A , image slices include reconstructed image slices 510 represented by dark rectangles and interpolated images 520 represented by white rectangles. Image slices 510, 520 are associated with a simulated linear motion path. Thus, the images correspond to image planes that are parallel to each other and spaced apart from each other. In Figure 6B , image 600 is associated with a simulated curved motion path that includes a combination of simulated tilt and translation of the ultrasound probe. In Figure 6C , image 700 is associated with a simulated curved motion path that includes a simulated tilt of the ultrasound probe. In some embodiments, the simulated motion path is determined or calculated to include more than a single type of motion along its length. For example, the simulated motion path can be calculated such that a plurality of image planes of interest (e.g., Tc, Tv, Tt) will be obtained by simulated movement of the probe along the motion path. Thus, the simulated motion path can be calculated to have different segments associated with different types of simulated movement (e.g., translation, rocking, tilt, rotation, compression, etc.). In some embodiments, the processor circuit is configured to identify an image of interest from the three-dimensional ultrasound data that is different from the target image, and determine the simulated motion path based on the target image plane, the neighboring image plane, and the image plane of interest. The processor circuit can then output the image of interest to the display in response to receiving a user input, as further described below.

[0056] As Figure 5 and 6A- As shown at 6C, the processor circuit can be configured to generate one or more interpolated images (e.g., 430, Figure 5 ) to accompany one or more neighboring images (410, 420, Figure 5 ). In this regard, in some instances, three-dimensional image data of a volume can be relatively sparse. For example, with conventional two-dimensional imaging procedures, an ultrasound probe can be capable of obtaining and displaying approximately 30 frames per second. Thus, an ultrasound physician can be accustomed to the ability to view two-dimensional views with a high degree of temporal and spatial resolution. However, because ultrasound imaging procedures are constrained by the speed of sound, the number of scanlines that can be obtained in a given amount of time is limited. Thus, for three-dimensional ultrasound imaging, temporal and / or spatial resolution can be reduced as compared to two-dimensional ultrasound imaging. Accordingly, ultrasound data points (e.g., voxels) available to reconstruct neighboring images can be limited.

[0057] The present application provides for one or more reconstructed image frames to be used to interpolate one or more images. In this regard, Figure 7 A flowchart illustrating a method 800 for generating and outputting an interpolated image is provided. It will be understood that one or more steps of the method 800 can be performed by, for example, the ultrasound imaging system 100 as shown at Figure 1 and / or the processor circuit 150 as shown at Figure 2 In step 810, the processor circuit generates a neighboring image frame or image slice corresponding to a first image plane that is adjacent to a target image plane. Referring to Figure 5 As described above, a target image and a neighboring image are reconstructed from three-dimensional ultrasound data. In this regard, the neighboring image frame can be generated from a three-dimensional image created by interpolating intensity values of a set of two-dimensional image slices of a volume obtained by the ultrasound imaging system. In some embodiments, the interpolation of the two-dimensional images is performed with respect to a Cartesian coordinate system or grid. The target image and the neighboring image can be generated or reconstructed using MPR.

[0058] Referring to Figure 5 and 8 In step 820, the processor circuit interpolates the image position and orientation between the target image 410 and the neighboring image 420, thereby then generating the image 440 as an MPR at the interpolated position and orientation. In some aspects, as described above, the MPR image 440 is generated based on interpolated intensity values of the three-dimensional image. As Figure 5As shown in FIG. 5, the interpolated image frame 510 is generated between the reconstructed proximal image frame 520 and the reconstructed distal image frame 530. The interpolated image frame 510 is generated to correspond to the simulated motion path 440. Thus, the interpolated image frame 510 is generated to correspond to the simulated motion path 440. The interpolated image frame 510 is generated to correspond to the simulated motion path 440. Thus, the orthogonal axis 442 of the interpolated image frame 510 is tangent to the curved simulated motion path 440. In other words, the target image plane, the proximal image plane, and the interpolated image plane are all orthogonal to the simulated motion path at the point where the planes intersect the simulated motion path. In step 830, the processor circuit outputs the interpolated image to the display. For example, the processor circuit can be configured to generate one or more interpolated images between the target image and the proximal image and / or between the proximal image and the distal image, and to arrange the proximal image and the interpolated image to form an image stack that can be scrolled or scanned through in response to input from a user. For example, in some embodiments, when an ultrasound physician scrolls up or down using a mouse, trackball, or the like, the processor circuit updates the display gradually based on the degree of input. For example, the speed with which the ultrasound physician rotates a trackball or sweeps across a trackpad can determine the speed with which the processor circuit updates the display with successive proximal images and interpolated images. Thus, the ultrasound physician can have a similar amount of control to scan through the proximal image planes as the ultrasound physician would have using conventional two-dimensional imaging methods (i.e., without MPR). This type of visualization can provide a familiar imaging experience while taking advantage of the workflow benefits of automated image reconstruction.

[0059] For example, referring again to FIG. 5, Figure 6A , the interpolated image frame 510 can supplement the reconstructed proximal image frame 520 to provide a smoother or more natural scanning visualization of image planes proximal to the target image plane. Thus, in some embodiments, the processor circuit is configured to interpolate between the target image and the proximal image or between two proximal images to generate an interpolated image. The interpolated image can then be output to the display in response to user input indicating a direction of motion along the simulated motion path. In an exemplary embodiment, the interpolated image is generated by the processor circuit. The interpolated image represents an intermediate image plane in the volume between two reconstructed images generated directly from the three-dimensional ultrasound data. The interpolated image is associated with the same simulated motion path as the reconstructed proximal image(s). Thus, when a user scans through the proximal and interleaved interpolated images, the result is a stream of images that more closely resembles a manual scan over the target region of the imaged volume as performed in a two-dimensional imaging procedure.

[0060] The interpolated images can be obtained by interpolation of the plane normal vectors and the plane offset vectors. The interpolation can be performed based on the position and orientation associated with each of the reconstructed images to obtain interpolated plane positions and normals. The intensity values of the 3D mesh corresponding to the reconstructed images can then be interpolated to generate intensity values for the interpolated planes. In some embodiments, motion vectors can be calculated for individual pixels to generate the interpolated images. Methods and techniques for generating interpolated image frames can be found in, at least, U.S. Patent No. 771354, issued August 10, 2010, which is incorporated by reference herein in its entirety.

[0061] In step 340, the processor circuit outputs the target image slice to a display in communication with the processor circuit. As described below, the processor circuit can output a graphical user interface to the display, where the GUI includes the target image. In step 350, the processor circuit receives user input indicative of a direction of motion along the simulated motion path. For example, the processor circuit can receive the user input via a user input device in communication with the processor circuit. The user input device can include a mouse, a keyboard, a trackball, a microphone, a touch screen display, a trackpad, a physical button, or any other suitable type of user input device. In this regard, the user input can be provided by the sonographer by pressing arrow keys on the keyboard, where the arrow direction corresponds to the direction (e.g., forward, backward) along the simulated motion path. In some embodiments, the user input is received by a scrolling device, such as a mouse, trackball, or trackpad. The direction of the scroll can correspond to the direction along the simulated motion path. However, it should be appreciated that any suitable user input can be used to indicate the direction of motion along the simulated path.

[0062] In step 360, the processor circuit outputs the adjacent images corresponding to the direction of motion from the plurality of adjacent images. For example, in some embodiments, the target image is first output to the display. The sonographer then uses the user input device to scan forward or backward along the simulated direction of motion, and the processor circuit outputs one or more of the adjacent images generated in step 330. Thus, the sonographer can control the display of the reconstructed images to scan forward and backward along the simulated motion path as if the user were manually scanning the ultrasound probe along the motion path, but after the ultrasound image data has been obtained and the ultrasound probe has been put down. Using this procedure, the sonographer is given the benefit of both (1) automatically reconstructing a target image from a three-dimensional data set and (2) the ability to obtain spatial context to confirm that the automatically reconstructed target image is correctly and / or optimally reconstructed.

[0063] In determining or calculating a simulated motion path, it can be beneficial to identify a simulated path that is relevant to a confidence assessment that the target image provided for automated reconstruction is properly determined. In this regard, the simulated motion path can be determined based on a determination of a direction of uncertainty relative to the target image plane. Figure 8 is a flowchart of a method 900 for determining a simulated motion path based on a determination of a direction of uncertainty. It will be understood that one or more steps of the method 900 can be performed by, for example, Figure 1 the ultrasound imaging system 100 shown in FIG. 1 and / or Figure 2 the processor circuit 150 shown in FIG. 1. In step 910, the processor circuit determines a direction of uncertainty relative to the target image plane. In some embodiments, the processor circuit determines the direction of uncertainty relative to the target image plane by applying a covariance matrix to the three-dimensional ultrasound data. In step 920, a simulated motion path is generated by the processor circuit based on the determined direction of uncertainty. In some embodiments, the processor circuit can determine multiple directions of uncertainty. In some embodiments, the processor circuit can determine a direction or vector of highest uncertainty. In other embodiments, the processor circuit determines a direction of median uncertainty, average uncertainty, minimum uncertainty, and / or any other suitable relative measure of uncertainty selected from one or more determined directions or vectors of uncertainty. In step 930, one or more adjacent images are generated by the processor circuit based on the simulated motion path determined in step 920, which is based on the determined direction of uncertainty. With the adjacent images (and / or interpolated images) generated based on the determined direction of uncertainty, the sonographer can be more likely to identify a target image that is improperly aligned, or to determine a probe movement and adjustment to generate a target image such that it is aligned with a target view or target image plane. Further details regarding determining a direction associated with uncertainty in an MPR procedure can be found in Alexander Schmidt-Richberg et al., “Offset regression networks for view plane estimation in 3D fetal ultrasound,” Proc. SPIE, vol. 10949, id. 109493K (15 March 2019), which is incorporated by reference herein in its entirety.

[0064] Figure 9is a graphical user interface (GUI) 1000 of an ultrasound imaging system, where the GUI includes an anatomical illustration 1010 of a body part and an MPR image 1020 of an image plane within the body part. The anatomical illustration 1010 includes a graphical representation of an anatomical structure of a patient (e.g., a head of a fetus) and is annotated with indicators showing how various target image planes intersect with the anatomical structure. In the illustrated embodiment, the body part includes a head of a person. For example, the illustration 1010 can be associated with a head of a fetus for use during a fetal ultrasound procedure. A plurality of line indicators are shown superimposed on the illustration of the anatomical structure 1010. The line indicators represent various image planes or views of interest. In this regard, the views associated with the indicators include a trans-ventricular (TV) image plane, a trans-cerebellum (TC) image plane, and a trans-thalamus (TT) image plane. Another indicator is provided showing a reconstructed frame (MPR). In some embodiments, the MPR indicator can be a dynamic indicator that is responsive to input from a user provided using a user interface device. In the illustrated embodiment, the MPR indicator on the illustration 1010 corresponds to the MPR image 1020 shown next to the illustration 1010. For example, in some embodiments, the GUI 1000 can display a target image corresponding to one of the views shown in the illustration 1010. A user can interact with the GUI using a user interface device such as a mouse, keyboard, trackball, voice input, touch screen input, or any other suitable input. The input from the user indicates a direction of motion along a simulated motion path. For example, the user input can be received in the form of a scroll of a mouse, arrow buttons on a keyboard, and / or a scroll of a trackball. The MPR indicator and the displayed MPR image 1020 can be updated in response to the received input to show adjacent MPR images reconstructed along the simulated motion path. The user can scan / scroll forward and backward along the path, and the MPR indicator can be updated in real-time to show the position and orientation of the image plane associated with the MPR image 1020. The MPR image 1020 is also updated based on the user input to show the image corresponding to the MPR image plane illustrated with respect to the illustration 1010. Thus, an ultrasound physician can control the display of the MPR image as if the user were manually scanning up and down a trajectory of an ultrasound probe, but after the ultrasound image data has already been obtained. For example, the GUI 1000 can be provided after the ultrasound physician has obtained three-dimensional ultrasound data and replaced the ultrasound probe.

[0065] Various modifications to the above-described apparatus, system, and / or methods can be made without departing from the scope of this disclosure. For example, in some embodiments, the processor circuitry is configured to display a loop of target and neighboring images to provide the appearance of periodically scanning forward and / or backward along a simulated motion path. In some embodiments, the processor circuitry is configured to scan multiple neighboring images in response to a single user input (e.g., key press, button press, etc.). In some embodiments, the processor circuitry is configured to receive user input to correct or adjust the reconstructed target image. For example, in some embodiments, the processor circuitry may be configured to receive selections corresponding to neighboring images from an ultrasound physician via a user input device. The processor circuitry may then tag, save, or otherwise identify the selected neighboring image as the target image. In some embodiments, the target image includes an interpolated image generated as described above. In some embodiments, neighboring images are displayed sequentially and one after another. In other embodiments, neighboring images are displayed together with the target image. In some embodiments, multiple neighboring and / or interpolated images are displayed simultaneously. In some embodiments, multiple different simulated motion paths are determined. For example, motion paths corresponding to different directions of uncertainty may be provided. In some embodiments, intersecting simulated motion paths are determined, along with additional neighboring images associated with the intersecting simulated motion paths. Therefore, in some embodiments, different inputs received via the user interface (e.g., up / down, left / right key presses) correspond to neighboring images along different simulated motion paths.

[0066] The embodiments of this disclosure offer numerous benefits for automated image reconstruction workflows. For example, generating and displaying a dynamic neighboring image stream provides a familiar imaging experience, a simple and intuitive way to correct target images, and a streamlined workflow transition from 2D to 3D ultrasound image acquisition. Additionally, embodiments of this disclosure provide a user-friendly way to navigate between multiple planes of interest (e.g., TT, TV, and TC planes for fetal ultrasound) within an anatomical structure, and an efficient way to provide important anatomical background information and mapping directly relevant to clinical guidelines for standard plane selection by using anatomical objects to be included or excluded. Furthermore, embodiments of this disclosure allow for enhanced clinical confidence in the automated reconstruction of images.

[0067] It will be understood that one or more steps in methods 300, 800, and 900 described above can be performed by one or more components of an ultrasound imaging system (such as the system's processor or processor circuitry, multiplexer, beamformer, signal processing unit, image processing unit, or any other suitable component). For example, one or more steps described above can be performed by... Figure 2The processor circuit 150 described is to perform. The processing components of the system can be integrated within the ultrasound imaging device, contained within an external console, or can be separate components.

[0068] Those skilled in the art will realize that the above-described apparatus, systems, and methods can be modified in various ways. Accordingly, persons of ordinary skill in the art will recognize that the embodiments encompassed by the present disclosure are not limited to the particular exemplary embodiments described herein. In this regard, while illustrative embodiments have been shown and described, it is to be understood that various modifications, changes, and substitutions can be made by those skilled in the art without departing from the scope of the present disclosure. It is therefore appropriate that the appended claims be construed broadly and in a manner including all such changes and modifications.

Claims

1. An ultrasound imaging apparatus, comprising: a processor circuit configured to: receive three-dimensional ultrasound data of an anatomical structure from an ultrasound probe communicatively coupled to the processor circuit; generate, from the three-dimensional ultrasound data, a target image corresponding to a target image plane of the anatomical structure, wherein generating the target image includes automatically identifying and reconstructing two-dimensional cross-sections of the three-dimensional ultrasound data associated with the target image plane; generate, from the three-dimensional ultrasound data, a plurality of adjacent images corresponding to image planes adjacent to the target image plane along a simulated motion path, wherein the target image and the plurality of adjacent images include two-dimensional images based on the three-dimensional ultrasound data, wherein the processor circuit is configured to determine a direction of uncertainty relative to the target image plane and determine the simulated motion path based on the determined direction of uncertainty, wherein the uncertainty is associated with a confidence assessment of whether the automatically reconstructed two-dimensional cross-sections represent the target image plane; output the target image to a display in communication with the processor circuit; receive user input representative of a direction of motion along the simulated motion path; and output an adjacent image of the plurality of adjacent images corresponding to the direction of motion to the display.

2. The ultrasound imaging apparatus of claim 1, further comprising the ultrasound probe.

3. The ultrasound imaging apparatus of claim 1, wherein, the processor circuit is configured to: interpolate between a position and orientation of the target image and a position and orientation of the adjacent image to generate an interpolated image; and output the interpolated image to the display.

4. The ultrasound imaging apparatus of any preceding claim, wherein, the processor circuit is configured to apply a covariance matrix to the three-dimensional ultrasound data to determine the direction of uncertainty.

5. The ultrasound imaging apparatus of claim 1, wherein, the processor circuit is configured to: identify, from the three-dimensional ultrasound data, a different image plane of interest than the target image and the adjacent image; and determine the simulated motion path based on the target image, the adjacent image, and the image plane of interest.

6. The ultrasound imaging apparatus of claim 5, wherein, the processor circuit is configured to output the image plane of interest in response to receiving the user input.

7. The ultrasound imaging apparatus of claim 1, wherein, the plurality of adjacent images includes a plurality of parallel adjacent images associated with a plurality of parallel adjacent image planes.

8. The ultrasound imaging apparatus of claim 1, wherein the processor circuit is configured to generate the target image to exclude an anatomical feature, and wherein the processor circuit is configured to generate the adjacent image to include the anatomical feature.

9. The ultrasound imaging apparatus of claim 1, wherein, the processor circuit is further configured to output a graphical representation of an adjacent image plane associated with the adjacent image to the display, wherein the graphical representation includes: a diagrammatic view of a body portion associated with the target image; and an indicator of the adjacent image plane superimposed on the diagrammatic view of the body portion.

10. A method for reconstructing ultrasound images, comprising: receiving three-dimensional ultrasound data of an anatomical structure obtained by an ultrasound probe; generating, from the three-dimensional ultrasound data, a target image corresponding to a target image plane of the anatomical structure; generating, from the three-dimensional ultrasound data, a plurality of adjacent images corresponding to image planes adjacent to the target image plane along a simulated motion path, wherein the target image and the plurality of adjacent images comprise two-dimensional images based on the three-dimensional ultrasound data; outputting the target image to a display; receiving a user input indicative of a direction of motion along the simulated motion path; and outputting, to the display, an adjacent image of the plurality of adjacent images corresponding to the direction of motion; wherein generating the target image comprises automatically identifying and reconstructing a two-dimensional cross-section of the three-dimensional ultrasound data associated with the target image plane, and the method further comprises determining a direction of uncertainty relative to the target image plane and determining the simulated motion path based on the determined direction of uncertainty, wherein the uncertainty is associated with a confidence assessment of whether the automatically reconstructed two-dimensional cross-section represents the target image plane.

11. The method of claim 10, wherein generating the plurality of adjacent images comprises interpolating between a position and orientation of the target image and a position and orientation of the adjacent images to generate an interpolated image, and wherein the method further comprises outputting the interpolated image to the display.

12. The method of claim 10 or 11, wherein, determining the direction of uncertainty comprises applying a covariance matrix to the three-dimensional ultrasound data.

13. The method of claim 10, further comprising: identifying, from the three-dimensional ultrasound data, an image of interest different from the target image and the adjacent images; and determining the simulated motion path based on the target image, the adjacent images, and the image of interest.

14. The method of claim 13, further comprising outputting the image of interest in response to receiving the user input.

15. The method of claim 10, wherein, generating the plurality of adjacent images comprises generating a plurality of parallel adjacent images associated with a plurality of parallel adjacent image planes.

16. The method of claim 10, wherein generating the target image comprises generating the target image to exclude an anatomical feature, and wherein generating the adjacent image comprises generating the adjacent image to include the anatomical feature.

17. The method of claim 10, further comprising: outputting, to the display, a graphical representation of an adjacent image plane associated with the adjacent image, wherein the graphical representation comprises: a diagrammatic view of a body portion associated with the target image; and an indicator of the adjacent image plane superimposed on the diagrammatic view of the body portion.

Images