Systems and methods for adaptive contrast imaging

Abstract

A system and method for generating adaptive contrast-accumulated imaging images are disclosed. Point spread function refinement / skeletonization techniques can be performed on contrast-enhanced image frames. The aggressiveness parameters of this technique can be adjusted temporally and / or spatially. The aggressiveness parameters can be adjusted based on various factors, including but not limited to time since contrast agent injection, signal intensity, and / or vessel size. Images can be accumulated temporally to generate a final sequence of adaptive contrast-accumulated imaging images.

Description

Technical Field

[0001] This application relates to contrast-enhanced imaging. More specifically, this application relates to generating contrast-accumulated images. Background Technology

[0002] In contrast-enhanced imaging, a contrast agent is delivered to the area or volume to be imaged to provide a higher signal intensity from said area or volume, or to selectively enhance the signal from areas or volumes with high contrast agent concentrations. For example, in contrast-enhanced ultrasound (CEUS), microbubbles can be injected into the blood flow of a subject, and ultrasound images of the subject's vascular system can be acquired. Without microbubbles, blood vessels can provide little or no signal. In contrast-accumulating imaging (CAI), multiple contrast-enhanced images (e.g., multiple image frames) are acquired and combined to form a final image, which can be used to map contrast agent progression and enhance vascular topology and salience. This temporal accumulation imaging of CEUS has been commercialized and is widely used for vascular visualization. Summary of the Invention

[0003] Systems and methods for adaptive contrast-enhanced ultrasound techniques replace the microbubble identification and localization steps in contrast accumulation image processing with adaptive point spread function (PSF) refinement / skeletonization techniques (e.g., refinement techniques). The PSF size can be at least partially adaptive spatially (e.g., for vessel size) and temporally (e.g., for different perfusion times) by adjusting invasive parameters. This can provide enhanced vascular imaging performance for both large branches and microvessels at different contrast agent perfusion stages with reduced processing requirements.

[0004] In some examples, contrast agent-injected tissue images can be acquired for playback. Adaptive PSF thinning can be applied to each frame of the contrast playback, where the PSF size can be adaptive based on the spatial region within the image and / or adaptive over time. After performing adaptive PSF thinning, temporal accumulation of the contrast image frames can be performed to achieve high resolution. In some examples, temporal accumulation can use the maximum intensity of each frame at each pixel or show the average intensity at each pixel.

[0005] According to at least one example disclosed herein, an ultrasound system may include an ultrasound probe and at least one processor, the ultrasound probe being configured to receive ultrasound signals for a plurality of transmission / reception events, the at least one processor being in communication with the ultrasound probe, the at least one processor being configured to: perform an adaptive thinning technique on the ultrasound signals for the plurality of transmission / reception events, wherein the adaptive thinning technique is at least partially based on an aggressive parameter adjusted in at least one of the time or spatial domains; and accumulate the ultrasound signals for the plurality of transmission / reception events on which the adaptive thinning technique is performed in time to generate an adaptive contrast accumulation image.

[0006] According to at least one example disclosed herein, a method may include receiving a plurality of contrast-enhanced ultrasound images; performing an adaptive thinning technique on an individual contrast-enhanced ultrasound image among the plurality of contrast-enhanced ultrasound images, wherein the adaptive thinning technique is based at least in part on an aggressive parameter adjusted in at least one of a time domain or a spatial domain; and accumulating at least two ultrasound images among the individual ultrasound images among the plurality of ultrasound images in time to provide an adaptive contrast-accumulated image.

[0007] According to at least one example disclosed herein, a non-transient computer-readable medium may include instructions that, when executed, cause an ultrasound imaging system to: receive a plurality of contrast-enhanced ultrasound images; perform an adaptive thinning technique on an individual contrast-enhanced ultrasound image among the plurality of contrast-enhanced ultrasound images, wherein the adaptive thinning technique is based at least in part on an aggressive parameter adjusted in at least one of the time or spatial domains; and accumulate at least two ultrasound images among the individual ultrasound images in time to provide an adaptive contrast-accumulated image. Attached Figure Description

[0008] Figure 1 Two example contrast-accumulated images are shown, acquired using two different image processing techniques.

[0009] Figure 2 This is a block diagram of an ultrasound imaging system arranged according to some examples of this disclosure.

[0010] Figure 3 This is a block diagram illustrating an example processor according to some examples of this disclosure.

[0011] Figure 4 This is a flowchart of a method based on some examples of this disclosure.

[0012] Figure 5 yes Figure 4 A more detailed flowchart of a portion of the method shown.

[0013] Figure 6A This is an illustration of an example of spatially adaptive aggressive parameters based on some examples of this disclosure.

[0014] Figure 6B This is an illustration of an example of time-adaptive aggressive parameters based on some examples of this disclosure.

[0015] Figure 7 An example image of dilatation-based refinement with different aggression parameters is shown according to an example of this disclosure.

[0016] Figure 8 Examples of conventional contrast accumulation imaging images and adaptive contrast accumulation imaging images according to this disclosure are shown.

[0017] Figure 9 Examples of conventional contrast accumulation imaging images and adaptive contrast accumulation imaging images according to this disclosure are shown. Detailed Implementation

[0018] The following description of specific exemplary examples is merely illustrative in nature and is in no way intended to limit the invention or its application or use. In the following detailed description of examples of the systems and methods, reference is made to the accompanying drawings, which form a part of the description, and in which specific examples of the described systems and methods are illustrated by way of illustration. These examples are described in sufficient detail to enable those skilled in the art to practice the currently disclosed systems and methods, and it should be understood that other examples may be utilized and structural and logical changes may be made without departing from the spirit and scope of the system. Furthermore, for clarity, detailed descriptions of certain features will not be discussed where they would be obvious to those skilled in the art, so as not to obscure the description of the system. Therefore, the following detailed description should not be regarded in a limiting sense, and the scope of the system is defined only by the claims.

[0019] Accumulated CEUS may have limited spatial resolution due to the large size of the point spread function (PSF) in imaging mode. The PSF is a measure of the blurring or diffusion of a point source by the imaging system. To improve spatial resolution, image processing techniques such as super-resolution imaging (SRI) can be performed. In SRI, individual microbubbles are identified and represented as single pixels (e.g., localized). However, the number of images that need to be accumulated to generate an SRI image can be significant. Furthermore, the processing time and / or power required for microbubble identification and localization can be excessive, especially when real-time or near-real-time visualization is desired.

[0020] Another issue in CEUS imaging is the potential for a wide distribution of vessel sizes within the images. While conventional cumulative CEUS may be sufficient for larger vessels, visualization of smaller vessels may benefit more from other techniques such as SRI. Similarly, a wide distribution of contrast agent concentration over time may exist within a series of images. Applying a single image processing technique may result in enhanced visualization of only one vessel type and / or enhanced visualization only for a specific time period of the angiography scan (e.g., the early perfusion phase, the late accumulation phase).

[0021] Figure 1 Two example contrast-accumulated images acquired using two different image processing techniques are shown. Both images 100 and 102 were acquired from a human thyroid gland during the early perfusion phase of CEUS. Image 100 was generated using a typical contrast-accumulated imaging (CAI) technique. Image 100 shows good sensitivity to contrast agents, but its spatial resolution is limited and shows signs of saturation. Image 102 was acquired using an enhanced CAI technique as described in U.S. Provisional Application US 62 / 787860, filed January 3, 2019. Image 102 shows better spatial resolution than image 100, but it is discontinuous and less sensitive to contrast agent signals than typical CAI techniques. Therefore, while image 102 illustrates an advancement in the art that can allow for greater spatial resolution and / or improved visualization of smaller vascular systems with fewer frames than SRI, adaptive CAI techniques capable of providing improved visualization of both larger and smaller vascular systems over time are desired. Furthermore, adaptive techniques that do not require microbubble identification and localization (as in SRI) would be desirable to reduce processing time and / or power.

[0022] This disclosure relates to systems and methods for performing spatially and temporally adaptive image processing techniques. The techniques described herein may be referred to as "adaptive CAI". These techniques can adjust the aggressive parameters of PSF thinning / skeletonization techniques to provide better visualization of both large and small blood vessels, as well as at all stages of contrast-enhanced imaging scans.

[0023] Figure 2The diagram illustrates a block diagram of an ultrasound imaging system 200 constructed according to the principles of this disclosure. The ultrasound imaging system 200 according to this disclosure may include a transducer array 214, which may be included in an ultrasound probe 212, such as an external or internal probe, for example, an intravascular ultrasound (IVUS) catheter probe. In other examples, the transducer array 214 may be in the form of a flexible array configured to conformally apply to the surface of the object to be imaged (e.g., a patient). The transducer array 214 is configured to emit ultrasound signals (e.g., beams, waves) and receive echoes (e.g., received ultrasound signals) in response to the emitted ultrasound signals. Various transducer arrays may be used, such as linear arrays, curved arrays, or phased arrays. The transducer array 214 may, for example, include a two-dimensional array (as shown) of transducer elements capable of scanning in both elevation and azimuth dimensions for 2D and / or 3D imaging. As is well known, the axis is the direction perpendicular to the array plane (in the case of a curved array, the axis fans out), the azimuth direction is usually defined by the longitudinal dimension of the array, and the elevation direction is transverse to the azimuth direction.

[0024] In some examples, transducer array 214 may be coupled to microwave beamformer 216, which may be located within ultrasonic probe 212, and which may control the transmission and reception of signals by the transducer elements in array 214. In some examples, microwave beamformer 216 may control the transmission and reception of signals by active elements in array 214 (e.g., an active subset of the elements of the array that defines the active aperture at any given time).

[0025] In some examples, the microwave beamformer 216 may be coupled, for example via a probe cable or wirelessly, to a transmit / receive (T / R) switch 218, which switches between transmitting and receiving and protects the main beamformer 222 from high-energy transmitted signals. In some examples, such as in portable ultrasound systems, the T / R switch 218 and other components of the system may be included in the ultrasound probe 212 rather than in the ultrasound system base, which may house image processing electronics. The ultrasound system base typically includes software and hardware components, including circuitry for signal processing and image data generation, and executable instructions for providing a user interface.

[0026] The transmission of ultrasonic signals from transducer array 214, controlled by microwave beamformer 216, is guided by transmit controller 220, which can be coupled to T / R switch 218 and main beamformer 222. Transmit controller 220 can control the direction in which the beam is steered. The beam can be steered directly forward from transducer array 214 (orthogonal to transducer array 214) or at different angles for a wider field of view. Transmit controller 220 can also be coupled to user interface 224 and receive input based on user operations on user controls. User interface 224 may include one or more input devices, such as control panel 252, which may include one or more mechanical controls (e.g., buttons, encoders, etc.), touch-sensitive controls (e.g., touchpad, touchscreen, or similar), and / or other known input devices.

[0027] In some examples, the partially beamformed signal generated by microwave beamformer 216 can be coupled to beamformer 222, where partially beamformed signals from individual facets of transducer elements can be combined into a fully beamformed signal. In some examples, microwave beamformer 216 is omitted, and transducer array 214 is under the control of beamformer 222, with beamformer 222 performing all beamforming of the signal. In examples with and without microwave beamformer 216, the beamformed signal from beamformer 222 is coupled to processing circuitry 250, which may include one or more processors (e.g., signal processor 226, mode-B processor 228, Doppler processor 260, and one or more image generation and processing units 268) configured to generate ultrasound images based on the beamformed signal (i.e., beamformed RF data).

[0028] Signal processor 226 can be configured to process received beamformed RF data in various ways, such as bandpass filtering, decimation, I and Q component separation, and harmonic signal separation. Processor 226 can also perform additional signal enhancement, such as ripple reduction, signal recombination, and electronic noise cancellation. The processed signals (also referred to as I and Q components or IQ signals) can be coupled to additional downstream signal processing circuitry for image generation. IQ signals can be coupled to multiple signal paths within the system, each of which can be associated with a specific arrangement of signal processing components suitable for generating different types of image data (e.g., B-mode image data, Doppler image data). For example, the system may include a B-mode signal path 258 that couples signals from signal processor 226 to B-mode processor 228 to generate B-mode image data.

[0029] The B-mode processor 228 can employ amplitude detection to image structures in the body. According to the principles of this disclosure, the B-mode processor 228 can generate signals for tissue images and / or contrast images. In some embodiments, signals from microbubbles can be extracted from the B-mode signal to form separate contrast images. Similarly, tissue signals can be separated from microbubble signals to generate tissue images. The signals generated by the B-mode processor 228 can be coupled to a scan converter 230 and / or a multi-plane reformer 232. The scan converter 230 can be configured to arrange the echo signals in a desired image format according to the spatial relationship in which the echo signals are received. For example, the scan converter 230 can arrange the echo signals in a two-dimensional sector format, or a three-dimensional (3D) format of a cone or other shape. In another example of this disclosure, the scan converter 230 can arrange the echo signals as a side-by-side contrast-enhanced image and a tissue image.

[0030] The multiplane reformer 232 is capable of converting echoes received from points in a common plane of a volumetric region of the body into an ultrasound image (e.g., a B-mode image) of that plane, such as as described in U.S. Patent US 6,443,896 (Detmer). In some examples, the scan converter 230 and the multiplane reformer 232 may be implemented as one or more processors.

[0031] Volume renderer 234 can generate an image (also known as a projection, drawing, or rendering) of a 3D dataset viewed from a given reference point, for example, as described in U.S. Patent 6530885 (Entrekin et al.). In some examples, volume renderer 234 can be implemented as one or more processors. Volume renderer 234 can generate the rendering using any known or future known techniques such as surface rendering and maximum intensity rendering, such as positive or negative rendering.

[0032] In some examples, the system may include a Doppler signal path 262 that couples the output from signal processor 226 to Doppler processor 260. Doppler processor 260 may be configured to estimate the Doppler frequency shift and generate Doppler image data. The Doppler image data may include color data, which is then overlaid with B-mode (i.e., grayscale) image data for display. Doppler processor 260 may be configured, for example, to use a wall filter to filter out unwanted signals (i.e., noise or clutter associated with non-moving tissue). Doppler processor 260 may also be configured to estimate velocity and power using known techniques. For example, the Doppler processor may include a Doppler estimator such as an autocorrelation function, where the velocity (Doppler frequency) estimate is based on the parameters of a hysteresis-autocorrelation function and the Doppler power estimate is based on the magnitude of a hysteresis-zero autocorrelation function. Motion may also be estimated using known phase-domain (e.g., parametric frequency estimators, such as MUSIC, ESPRIT, etc.) or time-domain (e.g., cross-correlation) signal processing techniques. Other estimators related to the temporal or spatial distribution of velocity, such as estimators of acceleration or temporal and / or spatial velocity derivatives, can be used instead of or as a supplement to the velocity estimator. In some examples, velocity and power estimates may undergo further thresholding to further reduce noise, as well as segmentation and post-processing, such as padding and smoothing. The velocity and power estimates can then be mapped to a desired range of display colors based on the color map. The color data, also known as Doppler image data, can then be coupled to a scan converter 230, where the Doppler image data can be converted to the desired image format and overlaid on a B-mode image of the tissue structure to form a color Doppler or power Doppler image. For example, Doppler image data can be overlaid on a B-mode image of the tissue structure.

[0033] Outputs from the scan converter 230, multiplane reformer 232, and / or volume plotter 334 (e.g., B-mode images, Doppler images) can be coupled to the image processor 236 for further enhancement, buffering, and temporary storage before being displayed on the image display 238. Optionally, in some embodiments, the image processor 236 may receive I / Q data from the signal processor 226 and / or RF data from the beamformer 222 for enhancement, buffering, and temporary storage before display.

[0034] Based on the principles of this disclosure, in some examples, image processor 236 can receive imaging data of image frames corresponding to a sequence of contrast-enhanced images (e.g., multi-frame playback, video playback). Each image frame in the sequence may have been acquired at different times (e.g., image frames may be temporally spaced). In some examples, image processor 236 can perform an adaptive point spread function (PSF) thinning / skeletonization technique (also referred to herein as adaptive thinning technique) on each frame in the sequence. In some examples, the adaptive thinning technique can be performed on each pixel of each image frame, including both separable microbubbles and microbubble clusters (e.g., imaging data corresponding to each pixel). This technique can reshape and / or resize the PSF of system 100. The size of the resized PSF can be based at least in part on the value of an aggression parameter. The larger the value of the aggression parameter, the smaller the size of the resized PSF. The lower the value of the aggression parameter, the closer the resized PSF is to the original PSF of system 100. The aggression parameter can be adapted in the spatial and / or temporal domains. (See reference...) Figure 5 For a more detailed explanation, one or more adaptive refinement techniques can be used.

[0035] In some examples, image processor 236 may perform temporal accumulation after performing adaptive thinning techniques on all images of the sequence. In other words, image processor 236 may combine multiple image frames to create a final image (e.g., high-resolution playback) of an adaptive CAI image sequence, which may include one or more image frames. Various techniques can be used. For example, the temporal accumulation step may be performed on the entire sequence (e.g., an infinite time window) or on a moving window in the time domain (e.g., a finite time window). Reference Figure 4 The techniques used for time accumulation are described in more detail. The final adaptive CAI sequence of image frames can be provided to the display 238 and / or local memory 242.

[0036] Graphics processor 240 can generate graphic overlays for display alongside images. These overlays may include standard identification information such as the patient's name, date and time, imaging parameters, etc. For these purposes, the graphics processor can be configured to receive input from user interface 224, such as typed patient names or other annotations. User interface 244 may also be coupled to multiplane reformer 232 for selecting and controlling the display of multiple multiplane reformatted (MPR) images.

[0037] System 200 may include local memory 242. Local memory 242 may be implemented as any suitable non-transient computer-readable medium (e.g., flash drive, disk drive). Local memory 242 may store data generated by system 200, including B-mode images, contrast images, executable instructions, input provided by the user via user interface 224, or any other information required for operation of system 200.

[0038] As previously described, system 200 includes a user interface 224. User interface 224 may include a display 238 and a control panel 252. Display 238 may include a display device implemented using various known display technologies such as LCD, LED, OLED, or plasma display technologies. In some examples, display 238 may include multiple displays. Control panel 252 may be configured to receive user input (e.g., examination type, contrast agent injection time). Control panel 252 may include one or more hardware controls (e.g., buttons, knobs, dials, encoders, mice, trackballs, or others). In some examples, control panel 252 may additionally or alternatively include software controls (e.g., GUI controls, or simply GUI controls) provided on a touch-sensitive display. In some examples, display 238 may be a touch-sensitive display that includes one or more software controls of control panel 252.

[0039] Based on the principles of this disclosure, in some examples, a user can select an adaptive thinning technique and / or set aggressive parameters to be used to generate an adaptive CAI image via user interface 224. Adjusting (e.g., changing) the aggressive parameters can adjust the final spatial resolution of the adaptive CAI image. In some examples, a user can indicate different aggressive parameters for different regions in an image frame and / or indicate how the aggressive parameters should change over time. In some examples, a user can select an average or initial aggressive parameter to use, and system 100 adjusts the aggressive parameters used spatially on the image frame and / or temporally on multiple image frames. In some examples, the adaptive thinning technique can be based on the type of examination, the type of contrast agent, and / or the nature of the image. In some examples, the aggressive parameters and / or how they are adjusted spatially and / or temporally can be based on the analysis of individual image frames and / or all image frames in a sequence.

[0040] In some examples, Figure 2 The various components shown can be combined. For example, image processor 236 and graphics processor 240 can be implemented as a single processor. In another example, scan converter 230 and multi-plane reformer 232 can be implemented as a single processor. In some examples, Figure 2The various components shown can be implemented as separate components. For example, signal processor 226 can be implemented as a separate signal processor for each imaging mode (e.g., B-mode, Doppler). In some examples, Figure 2 One or more of the various processors shown can be implemented by a general-purpose processor and / or microprocessor configured to perform a specified task. In some examples, one or more of the various processors can be implemented as dedicated circuitry. In some examples, one or more of the various processors (e.g., image processor 236) can be implemented using one or more graphics processing units (GPUs).

[0041] Figure 3 This is a block diagram illustrating an example processor 300 according to the principles of this disclosure. Processor 300 can be used to implement one or more processors described herein, such as... Figure 2 The image processor 236 is shown. The processor 300 can be any suitable processor type, including but not limited to microprocessors, microcontrollers, digital signal processors (DSPs), field-programmable arrays (FPGAs) (where the FPGA has been programmed to form a processor), graphics processing units (GPUs), application-specific integrated circuits (ASICs) (where the ASIC has been designed to form a processor), or combinations thereof.

[0042] Processor 300 may include one or more cores 302. Core 302 may include one or more arithmetic logic units (ALUs) 304. In some examples, in addition to or in place of ALU 304, core 302 may include a floating-point logic unit (FPLU) 306 and / or a digital signal processing unit (DSPU) 308.

[0043] Processor 300 may include one or more registers 312 communicatively coupled to core 302. Registers 312 may be implemented using dedicated logic gates (e.g., flip-flops) and / or any memory technology. In some examples, registers 312 may be implemented using static memory. Registers may provide data, instructions, and addresses to core 302.

[0044] In some examples, processor 300 may include one or more levels of cache memory 310 communicatively coupled to core 302. Cache memory 310 may provide computer-readable instructions to core 302 for execution. Cache memory 310 may provide data for processing by core 302. In some examples, computer-readable instructions may already be provided to cache memory 310 by local memory (e.g., local memory attached to external bus 316). Cache memory 310 may be implemented using any suitable cache memory type, such as metal-oxide-semiconductor (MOS) memory, such as static random access memory (SRAM), dynamic random access memory (DRAM), and / or any other suitable memory technology.

[0045] Processor 300 may include controller 314, which can control other processors and / or components included in the system (e.g., Figure 1 The control panel 252 and scan converter 230 shown are used for input to the processor 300 and / or from the processor 300 to other processors and / or components included in the system (e.g., Figure 1 The outputs of the display 238 and volume plotter 234 shown are illustrated. The controller 314 can control the data paths in the ALU 304, FPLU 306, and / or DSPU 308. The controller 314 can be implemented as one or more state machines, data paths, and / or dedicated control logic. The gates of the controller 314 can be implemented as stand-alone gates, FPGAs, ASICs, or any other suitable technology.

[0046] Register 312 and cache 310 can communicate with controller 314 and core 302 via internal connections 320A, 320B, 320C and 320D. The internal connections can be implemented as a bus, multiplexer, cross switch and / or any other suitable connection technology.

[0047] The inputs and outputs of the processor 300 may be provided via a bus 316, which may include one or more wires. The bus 316 may be communicatively coupled to one or more components of the processor 300, such as the controller 314, cache 310, and / or register 312. The bus 316 may be coupled to one or more components of the system, such as the aforementioned display 238 and control panel 252.

[0048] Bus 316 may be coupled to one or more external memories. The external memory may include read-only memory (ROM) 332. ROM 332 may be a mask ROM, electrically programmable read-only memory (EPROM), or any other suitable technology. The external memory may include random access memory (RAM) 333. RAM 333 may be static RAM, battery-backed static RAM, dynamic RAM (DRAM), or any other suitable technology. The external memory may include electrically erasable programmable read-only memory (EEPROM) 335. The external memory may include flash memory 334. The external memory may include a magnetic storage device, such as a disk 336. In some examples, the external memory may be included in the system, for example... Figure 2 The ultrasound imaging system 200 shown includes, for example, local memory 242.

[0049] Figure 4 This is an example of an image processor (such as...) according to this disclosure. Figure 1 The flowchart 400 shows the method executed by the image processor 236 shown.

[0050] In some examples, multi-frame playback (e.g., image sequences) of conventional side-by-side contrast and tissue images can be used as input to the signal processor, as indicated by box 402. Image formats can be DICOM, AVI, WMV, JPEG, and / or other formats. In some examples, image-domain-based processing can be implemented as an offline processing feature. In some applications, images may be logarithmically compressed with a limited dynamic range; therefore, the image-domain implementation of adaptive CAI may have limited performance. In some examples, adaptive CAI can also be implemented in the IQ domain (where the input is IQ data) or the RF domain (where the input is RF data) instead of multi-frame playback as shown in 4. In some applications, IQ data and / or RF data can provide better performance (e.g., better clutter suppression). In examples using IQ and / or RF data, the image processor can receive data from a signal processor and / or beamformer, for example... Figure 1 The signal processor 226 and / or beamformer 222 are shown.

[0051] At box 404, the image processor may perform image formatting. In some examples, multi-frame playback is processed to separate tissue and contrast images so that they can be processed independently, as indicated by boxes 406 and 408. The tissue and contrast images may be appropriately formatted for subsequent processing boxes. For example, a red-green-blue (RGB) image may be converted to a grayscale image (or indexed image) with a desired dynamic range (e.g., normalized from 0 to 1). In some examples, the image processor may receive tissue and contrast images from separate imaging streams that do not require separation. In examples where enhanced CAI is performed on RF data and / or IQ data instead of multi-frame playback, image formatting may include separating the signal generated by the contrast agent from the signal generated by the tissue.

[0052] At box 410, motion estimation can be performed on the tissue image. Frame-to-frame displacement for each image pixel can be estimated using motion estimation methods (e.g., speckle tracking and / or optical flow). Rigid motions (e.g., translation and rotation) and non-rigid motions (e.g., deformation) can be estimated. Spatial and / or temporal smoothing methods can be applied to the estimated displacement of the tissue image.

[0053] At box 412, motion compensation is performed at least in part based on the motion estimation performed on the tissue image at box 410. In some examples, the tissue image may not be used for motion estimation, and the motion estimation techniques discussed in reference box 410 above may be performed directly on the contrast image. However, in these examples, motion compensation may not be as robust.

[0054] Optionally, at box 414, clutter suppression filtering can be performed on the contrast image. This can reduce effects such as stationary echoes (especially in the near field), reverberation, etc. The clutter suppression filter can be implemented as a high-pass filter based on finite impulse response (FIR), infinite impulse response (IIR) (with a sufficient number of coefficient delay pairs (e.g., taps)), polynomial least squares curve fitting filter, and / or a high-pass filter based on singular value decomposition (SVD). The filter parameters can be optimized to suppress most of the residual clutter while preserving most of the contrast signal. In some examples, adaptive algorithms (such as the "Tissue Suppression" feature in VM6.0 distributed by Philips) can be used to perform tissue clutter removal before accumulation, where the nonlinear difference between tissue and microbubbles is used to mask and then suppress pixels containing tissue on a per-frame basis. In some examples, box 414 can be omitted.

[0055] After motion compensation (and optional clutter suppression), an adaptive PSF refinement technique can be performed at block 416. Figure 5 An adaptively refined functional block diagram according to an example of this disclosure is shown. Figure 5As shown, the adaptive thinning technique can receive motion-compensated tissue images (e.g., the image generated at box 410), motion-compensated contrast images (e.g., the image generated at box 412), and timing data from contrast agent perfusion timer 506. In some examples, the timing data can be generated at least in part based on user input received via a user interface (e.g., user interface 252) indicating when the contrast agent should be injected.

[0056] At box 508, the images from boxes 502 and 504 can be segmented using appropriate image segmentation techniques. In some examples, contrast images can be used alone to define different spatial regions with different contrast agent concentrations. In these examples, tissue images may not be segmented. Furthermore, temporal (slow-time) filtering can be performed on the contrast images to segment different spatial regions of flow velocity (i.e., vessel size) based on slow-time frequencies. Different levels of aggression can be assigned to different regions based on segmentation, as described below.

[0057] At box 510, the segmented image can be analyzed to generate spatially adaptive aggression parameters. For example, the aggression parameters can be based at least in part on blood vessel size. Figure 6A This is an illustration of a carotid artery 602 with plaque 604, which can be imaged using CEUS. In this example, low aggression (e.g., low-value aggression parameter) can be used for the carotid lumen 606 to preserve computational power, and high aggression can be used for plaque 604 to enable visualization of the internal microvascular system of plaque 604 at high spatial resolution.

[0058] In another example of spatially adjusting the aggression parameter, aggression can be based at least in part on contrast agent concentration. For example, large vessels can be perfused much earlier than smaller vessels. Therefore, in the early stages, the concentration of microbubbles will be much higher in large vessels. Signal intensity thresholds can be set to visualize microvessels with specific concentrations of microbubbles, such that large vessels are masked or their intensity is reduced to enhance smaller vessels. In other words, regions with high signal intensity can be assigned high aggression parameters, and regions with low signal intensity can be assigned low aggression parameters. In other examples, instead of a single threshold, the function can define how the aggression parameter varies with signal intensity.

[0059] Return to Figure 5 Box 512, contrast agent perfusion timing data can be used to generate time-adaptive aggressive parameters. For example, as... Figure 6BAs shown in graph 600, the aggression parameter can be gradually increased with contrast agent perfusion time to avoid discontinuous and low-sensitivity imaging in the early perfusion phase. In another example, the aggression parameter can be set to a minimum for a first number of frames (or a first time interval), before the contrast agent reaches the imaging area, and can be ramped up to a higher value after the first number of frames (or after the first time interval) for better resolution of the microbubbles. In some examples, dynamic range / luminance can also be adjusted based on aggression to ensure that the color map is effective as the aggression parameter changes over time. In some examples, segmented images can also be analyzed to generate a temporally adaptive aggression parameter.

[0060] Although the examples described in this paper use both spatially and temporally adaptive aggressive parameters, in other examples, aggressive parameters may be adaptive in only one of the spatial or temporal domains.

[0061] Return to Figure 5 In box 514, spatially and temporally adaptive aggressive parameters are used to perform an adaptive thinning technique on the contrast image. In box 516, the thinned contrast image is provided as output. Any suitable thinning technique can be used in conjunction with one or more adaptive aggressive parameters to provide the adaptive thinning technique. For example, techniques using morphological operations or techniques using spatial smoothing or low-pass filtering can be used. For illustrative purposes, several examples of suitable thinning techniques are provided herein, but the principles of this disclosure are not limited to the examples provided.

[0062] In the first example refinement technique, morphological operations using image erosion can be employed. Image erosion can be performed on each frame of the contrast playback to erode the boundaries of regions and leave a reduced area of ​​contrast agent signal. In this example, aggression can be the size of the structured element (e.g., an aggression parameter defines the size of the structured element). The structured element can be the shape used to apply the function to the image frame. The shape of the structured element can be a simple square or rectangle in some examples, but more complex shapes in others. In this technique, a high aggression parameter (e.g., an aggression parameter with a high value) corresponds to a large-sized structured element. The larger the structured element, the more the boundaries are eroded, leaving a smaller area of ​​residual contrast agent signal. A low aggression parameter (e.g., an aggression parameter with a low value) corresponds to a smaller-sized structured element, and the less the boundaries are eroded, leaving a larger area of ​​residual contrast agent signal. The size of the structured element can be adjusted spatially and / or temporally.

[0063] In some examples, image erosion can be achieved by applying a grayscale erosion algorithm to each frame of an image based on constructed structured elements. In this technique, the erosion of an image pixel is the minimum value of that pixel within its neighborhood, defined by the structured elements. The output of the image erosion operation is the output of box 514.

[0064] In some examples, the output of the image erosion operation can be referred to as a mask. The mask can be normalized using values ​​between 0 and 1. Powers can be applied to the mask. The final output of box 514 can be the product of the original input and the normalized mask with an exponent, as shown in the equation below:

[0065] Mask = Erosion (Input) Equation (1)

[0066] Output = Input × (Normalized Mask) 指数 Equation (2)

[0067] In addition to the structured elements, the exponent (index) can be used to control the aggression parameter. Aggression increases as the exponent increases (e.g., greater than 1).

[0068] In the second example of the thinning technique, morphological operations using image dilation can be employed. In this example, structural elements, such as those described with reference to the first example, are also used. Image dilation is provided by the following formula:

[0069]

[0070] The output is the output (e.g., thinned) image, and the input is the input image (e.g., the contrast image from box 502). The output image is equal to the input image divided by the input image after the dilation operation. In this example, aggression can again be the size of the structured element in the image dilation. In the case of high aggression (large structured elements), more boundaries are magnified, leaving a smaller area of ​​remaining contrast agent signal. In the case of low aggression (smaller structured elements), fewer boundaries are magnified, leaving a larger area of ​​remaining contrast agent signal. The size of the structured element can be adjusted spatially and / or temporally. The dilation technique can be applied to each frame.

[0071] In some examples, image dilation can be a grayscale dilation algorithm. In this algorithm, the dilation of an image pixel is the maximum value of the image pixel in its neighborhood, which is defined by structured elements. After the dilation operation, the PSF expands to a larger size according to an aggressiveness parameter (the size of the structured elements). The output of this step (each frame of the image) is referred to as dilation(input) in Equation 3. The input image can then be scaled based on the dilated output. According to Equation 3, the input image is scaled using the output of the dilation step. Specifically, each pixel output (x,y) of the output image is the product of the image pixel input (x,y) and the corresponding scaling factor 1 / [dilation(input)(x,y)]. Due to the scaling factor, the output PSF shrinks to a smaller size according to the aggressiveness parameter. Optionally, a normalization step can be applied to the output to remove outliers (e.g., infinite elements) and normalize the output dynamic range to some limits.

[0072] exist Figure 7 Examples of image dilation-based thinning with different aggression parameters (e.g., structured elements of different sizes) according to this disclosure are shown. Images 700, 702, and 704 are CEUS images of human liver lesions 701 during the early arterial phase of contrast agent perfusion. Image 700 is obtained by performing image dilation using a low-value aggression parameter. The adjusted (e.g., regulated, shaped) PSF has a slightly smaller size than the original PSF (not shown). Image 702 is obtained by performing image dilation using a median aggression parameter. The adjusted PSF has a smaller size than the original PSF and the PSF of image 700. Image 704 is obtained by performing image dilation with a high-value aggression parameter. The adjusted PSF has a much smaller size than the original PSF and the PSF of images 700 and 702.

[0073] Similar to the first example describing image erosion, the output of Equation 3 can be referred to as a mask, which can be normalized and boosted to an exponent that can be used to control the aggressiveness as described in Equation 2. Then, the final output of box 514 can be the original image multiplied by the normalized mask with the exponent, as described in Equation 2.

[0074] In the third example, spatial smoothing thinning techniques can be used. Spatial smoothing can be described by the following equation:

[0075]

[0076] The output is the output (e.g., thinned) image, and the input is the input image (e.g., the angiographic image from box 502). The output image is equal to the input image divided by the input image after the spatial smoothing operation. In this example, aggression can be the size of the smoothing kernel used for spatial smoothing. A high aggression parameter corresponds to a large smoothing kernel size (e.g., 8×8), and more boundaries are smoothed, leaving a smaller area of ​​residual angiographic signal. A low aggression parameter corresponds to a smaller smoothing kernel size (e.g., 3×3), and fewer boundaries are smoothed, leaving a larger area of ​​residual angiographic signal. Generating the smoothing kernel can be similar to generating the structured elements described above. In some examples, the shape of the smoothing kernel can be a simple square or rectangle. The size of the smoothing kernel can be adjusted temporally and / or spatially.

[0077] The output of the spatial smoothing operation (the smoothing(input) in Equation 4) can be the result of a 2D spatial convolution between the input image and the smoothing kernel. In some examples, if the filter coefficients of the smoothing kernel are 1 (e.g., all elements have a value of 1), spatial smoothing can be reduced to a 2D moving average. If the filter coefficients of the smoothing kernel are based on some distribution (e.g., a 2D Gaussian distribution), spatial smoothing can be a weighted moving average. After the smoothing operation, the PSF is expanded to a larger size according to aggression. The output of this step (each frame of the image) is referred to as smoothing(input) in Equation 4. The input image can then be scaled based on the smoothed output. According to Equation 4, the input image is scaled using the output of the smoothing step. Specifically, each pixel output (x,y) of the output image is the product of the image pixel input (x,y) and the corresponding scaling factor 1 / [smoothing(input)(x,y)]. Due to the scaling factor, the output PSF shrinks to a smaller size according to the aggression parameter. Optionally, a normalization step can be applied to the output to remove outliers (e.g., infinite elements) and normalize the output dynamic range to some limits.

[0078] Similar to the examples describing morphological operations, the output of Equation 4 can be referred to as a mask, which can be normalized and boosted to an exponent that can be used to control the aggression described in Equation 2. Then, the final output of box 514 can be the original image multiplied by the normalized mask with the exponent, as described in Equation 2.

[0079] In the fourth example, a low-pass filter (LPF) refinement technique can be used. The LPF technique can be described by the following equation:

[0080]

[0081] The output is the output (e.g., thinned) image, and the input is the input image (e.g., the contrast image from box 502). The output image is equal to the input image divided by the input image after the spatial smoothing operation. In this example, aggression can be the spatial cutoff frequency of the spatial LPF. A high aggression parameter corresponds to a lower cutoff frequency, and more boundaries are filtered, leaving a smaller residual contrast signal region. A low aggression parameter corresponds to a higher cutoff frequency, and fewer boundaries are filtered, leaving a larger residual contrast signal region. The cutoff frequency can be adjusted spatially and / or temporally.

[0082] A 2D spatial Fast Fourier Transform (FFT) is performed on the input image. An LPF is then applied to the output of the FFT. This step removes and / or suppresses high spatial frequency components of the input image (e.g., spatial frequency components above the cutoff frequency). An inverse FFT is then performed on the filtered image, which brings the image back from the frequency domain to the image (e.g., spatial) domain. Because the high-frequency components have been removed and / or suppressed, the PSF expands to a larger size according to aggression. The output of this step is referred to as LPF(input) in Equation 5. The input image can then be scaled based on the LPF output. According to Equation 5, the input image is scaled using the output of the LPF step. Specifically, each pixel output (x,y) of the output image is the product of the image pixel input (x,y) and the corresponding scaling factor 1 / [LPF(input)(x,y)]. Due to the scaling factor, the output PSF shrinks to a smaller size according to the aggression parameter. Optionally, a normalization step can be applied to the output to remove outliers (e.g., infinite elements) and normalize the output dynamic range to certain limits.

[0083] Similar to the examples describing morphological operations, the output of Equation 5 can be referred to as a mask, which can be normalized and boosted to an exponent that can be used to control the aggression described in Equation 2. Then, the final output of box 514 can be the original image multiplied by the normalized mask with the exponent, as described in Equation 2.

[0084] The examples provided in this article are for illustrative purposes only, and other adaptive thinning techniques can be used. For example, multi-resolution pyramid decomposition, blob detection, and / or tube detection image processing techniques can be used. With all thinning techniques, the aggressiveness parameters used can vary spatially and / or temporally.

[0085] Return to Figure 4 At box 418, the refined image output from box 416 can be (e.g., Figure 5The time accumulation (e.g., combining multiple sequential image frames) is performed on the images in box 516. In some examples, time accumulation can be performed on all images in a multi-frame playback to provide an infinite time window. In other examples, time accumulation can be performed on a moving window in the time domain to provide a finite time window. The time accumulation window (e.g., an average integration window) can be set to not exceed a certain time interval (e.g., 1 second, 5 seconds, 10 seconds, 30 seconds). Time accumulation can provide a final adaptive CAI image sequence, which can be referred to as high-resolution image playback, as indicated by box 420.

[0086] Any suitable time accumulation method can be used. Two examples are provided for illustrative purposes, but the principles of this disclosure are not limited to the examples provided. In some examples, peak preservation or maximum intensity projection methods can be used for time accumulation. In this method, the final CAI image frame only shows the maximum intensity among all previous input frames at each image pixel. For illustration, an example MATLAB algorithm is provided below:

[0087]

[0088] The input and output playback have the same dimensions [Nz, Nx, Nt], where Nz is the axial dimension, Nx is the horizontal dimension, and Nt is the time dimension.

[0089] In some examples, averaging using exponential correction or average intensity projection can be used. The final CAI image frame shows the time-averaged intensity (using exponential correction) at each image pixel across all previous input frames. For illustration, an example MATLAB algorithm is provided below:

[0090]

[0091] expCoef is an exponential coefficient used to correct the dynamic range of the final sequence of adaptive CAI images (e.g., output playback), which is typically set to 0.5, but can be adjusted based on the nature of the ultrasound imaging system (e.g., imaging system 100) and / or the type of examination.

[0092] For reference Figure 2 As noted, in some examples, the user can provide input via a user interface (e.g., user interface 252) to select the adaptive thinning technique used and / or have additional control over the thinning technique used to generate the final adaptive CAI image sequence. That is, in some examples, the user can have control over the reference... Figure 4 and 5 Control of the described method.

[0093] In some examples, users can manually control and / or rewrite the aggressive parameters of the adaptive PSF refinement / skeletonization steps. For example, after processing is complete (e.g., after execution...). Figure 4-5 After the box shown, the user can manipulate the user control to view the results with different aggressive parameters (e.g., final CAI image sequence / high-resolution playback).

[0094] In some examples, at the adaptive PSF thinning / skeletonization box 416, multiple levels of adjusted aggression parameters can be applied, and the results can be stored (e.g., in local memory 242) for use in the same downstream processing (e.g., box 418). Users can view results with different levels of aggression without reprocessing the dataset. In other examples, instead of computing different versions of the output playback at multiple levels of aggression at box 416, a blending algorithm can be performed between a regular CAI image (e.g., the image at box 502) and a highly thinned (e.g., high aggression parameter) adaptive CAI image at each pixel and frame. Users can adjust the blending ratio between the two results to achieve the optimal blended image. In some examples, the original image can be combined with its own thinned version based on one or more weights. In some examples, the weights can vary over time. For example, the same thinning operation can be applied to all image frames of a sequence (e.g., playback), and the image can contribute to the time accumulation process by summing the weights of the original image (e.g., percentages) with the weights of the thinned image, where the weights can be high (e.g., 90-100%) for the first set of frames and gradually decrease for subsequent frames (e.g., down to 10-0%).

[0095] Figure 8 A conventional CAI image and an adaptive CAI image according to an example of this disclosure are shown. Images 800 and 802 are CEUS images of a human thyroid gland. Image 800 is generated using conventional CAI techniques (e.g., combining two or more sequential image frames in playback). Image 802 is generated using an adaptive thinning technique according to the principles of this disclosure. Image 802 provides better visualization of both larger and smaller perfusion areas. For example, in region 804, image 800 suffers from saturation and blurring of the perfusion area. In image 802, saturation and blurring are reduced, providing a clearer view of the perfusion area 804. In another example, region 806 is blurred in image 800, making it difficult to see individual blood vessels. However, in image 802, the microvascular system in region 806 can be seen more clearly.

[0096] Figure 9Conventional CAI images and adaptive CAI images according to examples of this disclosure are shown. Images 900 and 902 are CEUS images of a human liver including lesion 901. Image 900 is generated using conventional CAI techniques (e.g., combining two or more sequential image frames in playback). Image 902 is generated using an adaptive thinning technique according to the principles of this disclosure. Image 902 provides better visualization of both larger and smaller perfusion areas. For example, the microvascular system within lesion 900 can be more clearly discerned in image 902 compared to image 900. In another example, large vessels 904 and 906 suffer significant blurring and saturation in image 900. In image 902, saturation is reduced, and the shapes of vessels 904 and 906 are more clearly defined.

[0097] This paper describes systems and methods for performing adaptive CAI techniques. Adaptive CAI techniques can adjust (e.g., regulate, change) the aggression parameters of PSF thinning / skeletonization techniques. These aggression parameters can be adjusted spatially and / or temporally. Adjusting the aggression parameters allows adaptive CAI techniques to provide improved visualization. In some applications, the adaptive thinning techniques disclosed herein can allow for improved visualization of both high-intensity and low-intensity signals (e.g., high-perfusion and low-perfusion areas, large vessels and small vessels) within a CAI image frame.

[0098] In various embodiments that implement components, systems, and / or methods using programmable devices such as computer-based systems or programmable logic, it should be appreciated that the aforementioned systems and methods can be implemented using various known or later-developed programming languages ​​such as "C", "C++", "FORTRAN", "Pascal", "VHDL", etc. Accordingly, various storage media, such as magnetic computer disks, optical disks, electronic storage devices, etc., can be prepared, which can contain information that can boot devices such as computers to implement the aforementioned systems and / or methods. Once a suitable device accesses the information and programs contained on the storage medium, the storage medium can provide the information and programs to the device, thereby enabling the device to perform the functions of the systems and / or methods described herein. For example, if a computer disk containing appropriate materials (such as source files, object files, executable files, etc.) is provided to a computer, the computer can receive this information, appropriately configure itself, and perform the functions of the various systems and methods outlined in the above figures and flowcharts to implement various functions. That is, a computer can receive various portions of information relating to different elements of the aforementioned systems and / or methods from the disk, implement individual systems and / or methods, and coordinate the functions of the individual systems and / or methods described above.

[0099] In view of this disclosure, it should be noted that the various methods and apparatuses described herein can be implemented in hardware, software, and / or firmware. Furthermore, the various methods and parameters are included by way of example only and not in any limiting sense. In view of this disclosure, those skilled in the art can implement the teachings to determine their own techniques and the apparatus required to implement these techniques, while remaining within the scope of the invention. The functionality of one or more of the processors described herein can be incorporated into a smaller number or a single processing unit (e.g., a CPU) and can be implemented using application-specific integrated circuits (ASICs) or general-purpose processing circuitry programmed in response to executable instructions that perform the functions described herein.

[0100] Although this system may have been described with particular reference to ultrasound imaging systems, it is also contemplated that this system can be extended to other medical imaging systems in which one or more images are acquired systematically. Therefore, this system can be used to acquire and / or record image information relating to, but not limited to, the kidneys, testes, breasts, ovaries, uterus, thyroid gland, liver, lungs, musculoskeletal system, spleen, heart, arteries, and vascular system, as well as other imaging applications related to ultrasound-guided interventions. Furthermore, this system may include one or more procedures that can be used with conventional imaging systems, such that they can provide the features and advantages of this system. Certain additional advantages and features of this disclosure will be apparent to those skilled in the art upon studying this disclosure, or may be experienced by those employing the novel systems and methods of this disclosure. Another advantage of this system and method is that conventional medical imaging systems can be easily upgraded to incorporate the features and advantages of this system, device, and method.

[0101] Of course, it should be recognized that any of the paradigms, examples, or processes described herein may be combined with one or more other paradigms, examples, and / or processes, or may be separated and / or performed within a separate device or device portion according to this system, device, and method.

[0102] Finally, the foregoing discussion is intended merely to illustrate the system and method and should not be construed as limiting the claims to any particular example or set of examples. Therefore, although the system has been described in particular detail with reference to exemplary examples, it should be appreciated that many modifications and alternative examples can be devised by those skilled in the art without departing from the broader and contemplated spirit and scope of the system and method as set forth in the following claims. Thus, the specification and drawings are to be viewed in an illustrative manner and not intended to limit the scope of the claims.

Claims

1. An ultrasound imaging system (200), comprising: An ultrasonic probe (212) is used to receive ultrasonic signals for multiple transmit / receive events; as well as At least one processor (250) communicating with the ultrasound probe, the at least one processor being configured to: An adaptive thinning technique (416) is performed on the ultrasound signals for the plurality of transmit / receive events, wherein the adaptive thinning technique is at least partially based on an aggression parameter adjusted in the time domain, wherein the aggression parameter of the adaptive thinning technique is configured to control the spatial resolution of the adaptive contrast accumulation image; and The ultrasonic signals for the plurality of transmit / receive events are accumulated over time using the adaptive thinning technique (418) to generate the adaptive contrast accumulation image.

2. The ultrasound imaging system according to claim 1, wherein, The aggressiveness parameter is also adjusted in the spatial domain.

3. The ultrasound imaging system according to claim 1 further includes a beamformer, wherein, The ultrasonic signal corresponds to radio frequency data provided by the beamformer to the at least one processor.

4. The ultrasound imaging system according to claim 1 further includes a signal processor, wherein, The ultrasound signal corresponds to IQ data provided by the signal processor to the at least one processor.

5. The ultrasound imaging system according to claim 1 further includes a user interface, wherein, The user interface is configured to receive user input indicating the type of technique to be used in the adaptive refinement technique.

6. The ultrasound imaging system according to claim 5, wherein, The technology type is morphological manipulation technology, spatial smoothing technology, or low-pass filtering technology.

7. The ultrasound imaging system according to claim 1 further includes a user interface, wherein, The user interface is configured to receive user input indicating how the aggressive parameters should be adjusted.

8. The ultrasound imaging system according to claim 7, wherein, The method includes applying different values ​​of the aggression parameter at different locations in the image corresponding to the ultrasound signal of at least one of the plurality of transmission / reception events.

9. The ultrasound imaging system according to claim 7, wherein, The method includes applying different values ​​of the invasiveness parameter, at least in part, based on the time the contrast agent is injected into the subject, the ultrasound signal being received from the subject.

10. An ultrasound imaging method (400), comprising: Receive (406) multiple contrast-enhanced ultrasound images; An adaptive thinning technique (416) is performed on an individual contrast-enhanced ultrasound image among the multiple contrast-enhanced ultrasound images, wherein the adaptive thinning technique is at least partially based on an aggression parameter adjusted in the time domain, wherein the aggression parameter of the adaptive thinning technique is configured to control the spatial resolution of the adaptive contrast accumulation image; and At least two ultrasound images from the individual ultrasound images in the plurality of contrast-enhanced ultrasound images are accumulated over time (418) to provide the adaptive contrast-accumulated image.

11. The method according to claim 10, wherein, The aggressiveness parameter is also adjusted in the spatial domain.

12. The method of claim 10, further comprising: Receive multiple tissue ultrasound images; Motion estimation is performed on individual tissue ultrasound images from the multiple tissue ultrasound images; and Before performing the adaptive thinning technique, motion compensation is performed on individual contrast-enhanced ultrasound images among the multiple contrast-enhanced ultrasound images, at least in part based on the motion estimation.

13. The method of claim 11, further comprising segmenting at least one of the plurality of contrast-enhanced ultrasound images to define how to adjust the aggression parameter in the spatial domain.

14. The method of claim 10, further comprising: Receive timing data from the contrast agent perfusion timer; and The aggression parameter is adjusted in the time domain based on the timing data.

15. The method according to claim 10, wherein, The adaptive thinning technique is an image erosion technique, wherein the aggressiveness parameter defines the size of the structured elements of the erosion technique.

16. The method of claim 10, wherein, The adaptive thinning technique is an image dilation technique, wherein the aggressiveness parameter defines the size of the structured elements of the dilation technique.

17. The method according to claim 10, wherein, The adaptive thinning technique is a spatial smoothing technique, wherein the aggressive parameter defines the size of the smoothing kernel.

18. The method according to claim 10, wherein, The adaptive refinement technique is a low-pass filtering technique, wherein the aggressive parameter defines the cutoff frequency of the low-pass filter.

19. The method according to claim 10, wherein, Accumulation over time includes generating either the maximum intensity projection or the average intensity projection.

20. A non-transient computer-readable medium encoded with executable instructions, said executable instructions causing an ultrasound imaging system, when executed, to: Receive (406) multiple contrast-enhanced ultrasound images; An adaptive thinning technique (416) is performed on individual contrast-enhanced ultrasound images from the multiple contrast-enhanced ultrasound images, wherein, The adaptive thinning technique is based, at least in part, on an aggressive parameter adjusted in the time domain, wherein the aggressive parameter of the adaptive thinning technique is configured to control the spatial resolution of the adaptive contrast-accumulated image; and At least two ultrasound images from the individual ultrasound images in the plurality of contrast-enhanced ultrasound images are accumulated over time (418) to provide the adaptive contrast-accumulated image.

Images