Contrast-enhanced ultrasound image representation

Abstract

A framework for characterizing contrast-enhanced ultrasound (CEUS) images. The framework derives a brightness mode (B-mode) and a contrast image sequence from a contrast-enhanced ultrasound (CEUS) image sequence. The B-mode and contrast images in the B-mode and contrast image sequence are co-registered spatially and subsampled temporally. A region of interest can be selected in frames of the B-mode and contrast image sequence. The region of interest is then characterized based on the B-mode and contrast image sequence.

Description

Technical Field

[0001] This disclosure generally relates to medical imaging, and more particularly to a frame for characterizing contrast-enhanced ultrasound images. Background Technology

[0002] Ultrasound imaging is a non-invasive medical test that helps physicians diagnose and treat medical conditions. Ultrasound imaging involves using a small probe called a transducer, which is placed directly on the skin to transmit and receive high-frequency sound waves traveling through the body. These sound waves are used to create medical diagnostic images of the body's internal structures.

[0003] Contrast-enhanced abdominal ultrasound (CEUS) combines abdominal ultrasound with a special type of intravenously injected contrast agent to improve visualization of blood vessels and organs. The contrast agent may include inflated microbubbles for better visualization of organs and blood vessels within the abdomen and pelvis. Abdominal CEUS can assess the liver, spleen, kidneys, pancreas, intestines, and / or bladder for various diseases and conditions.

[0004] CEUS can significantly improve the diagnostic accuracy of ultrasound in detecting and characterizing focal liver lesions. In a liver CEUS scan, the aim is to determine the type of focal liver lesion by injecting a contrast agent into the patient and observing the perfusion (wash-in) and wash-out properties. Perfusion generally refers to the enhancement of tissue or mass after contrast injection and is seen in benign or non-cancerous tissue. Wash-out generally indicates that tissue or mass loses contrast more quickly compared to normal liver tissue and is seen in cancerous or malignant tissue. Because perfusion occurs relatively quickly (e.g., within 20–30 seconds), video can be saved to observe the perfusion process. Wash-out images can be acquired discretely, such as spaced 1 minute apart, up to approximately 5 minutes apart. However, physicians may find it difficult to track any lesions in the video and / or individual images.

[0005] In some applications, data frames acquired over time are synthesized. The resulting images can provide useful diagnostic information, such as showing smaller blood vessels or perfusion channels. However, due to operator movement (e.g., sweeping motion) or the patient's internal movements (e.g., heavy breathing), the combination of information from different frames can result in blurry images or inaccurate information. Summary of the Invention

[0006] This paper describes a framework for characterizing contrast-enhanced ultrasound (CEUS) images. This framework derives a brightness pattern (B-mode) and a sequence of contrast images from a sequence of CEUS images. The B-mode and contrast images in the B-mode and contrast image sequences are spatially co-registered and temporally subsampled. Regions of interest can be selected within frames of the B-mode and contrast image sequences. The regions of interest are then characterized based on the B-mode and contrast image sequences. Attached Figure Description

[0007] A more complete understanding of the present disclosure and its many accompanying aspects will be readily available as the present disclosure and its many accompanying aspects become better understood when considered in conjunction with the accompanying drawings and by referring to the following detailed description.

[0008] Figure 1 This is a block diagram illustrating an exemplary imaging system; Figure 2 An exemplary method of characterization is shown; and Figure 3 An exemplary timeline diagram is shown. Detailed Implementation

[0009] In the following description, numerous specific details, such as examples of particular components, devices, methods, etc., are set forth to provide a thorough understanding of implementations of this framework. However, it will be apparent to those skilled in the art that these specific details need not be used in practicing implementations of this framework. In other instances, well-known materials or methods have not been described in detail to avoid unnecessarily obscuring implementations of this framework. While this framework is susceptible to various modifications and alternatives, specific embodiments thereof are shown by way of example in the accompanying drawings and will be described in detail herein. However, it should be understood that the invention is not intended to be limited to the specific forms disclosed, but rather, it is intended to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the invention. Furthermore, for ease of understanding, certain method steps are depicted as separate steps; however, these separately depicted steps should not be construed as necessarily depending on the order in which they are performed.

[0010] Unless otherwise stated, as will be apparent from the following discussion, terms such as “segmentation,” “generation,” “registration,” “determination,” “alignment,” “positioning,” “processing,” “computation,” “selection,” “estimation,” “detection,” “tracking,” or the like can refer to the actions and processes of a computer system or similar electronic computing device that manipulate and transform data represented as physical (e.g., electronic) quantities within the registers and memories of the computer system into other data similarly represented as physical quantities within the computer system's memory or registers or other such information storage, transmission, or display devices. Embodiments of the methods described herein can be implemented using computer software. If written in a programming language conforming to recognized standards, sequences of instructions for implementing these methods are designed to be compilable for execution on various hardware platforms and to interface with various operating systems. Furthermore, implementations of this framework are not described with reference to any particular programming language. It will be understood that various programming languages ​​can be used.

[0011] One aspect of this framework provides tools for generating summaries of acquired data and allowing physicians to review the data in a single static image timeline, rather than watching videos and observing individual images. Another aspect of this framework derives a spatially co-registered sequence of selected frames from an ultrasound image sequence containing B-mode and contrast images. Both B-mode and contrast images are used to ensure reliable lesion tracking throughout the contrast lifecycle and to compensate for motion (e.g., operator and / or internal motion) to facilitate accurate visualization and measurement. Accurate tracking is not possible using only B-mode images because they are affected by contrast perfusion and are therefore unreliable when used during tracking. These and other exemplary advantages and features will be described in more detail below.

[0012] Figure 1This is a block diagram illustrating an exemplary medical imaging system 100. In some embodiments, system 100 includes a computer system 101 coupled to a medical imaging device 114. Computer system 101 includes a processor device 104 coupled to one or more non-transitory computer-readable media 105 (e.g., computer storage devices or memory devices), input / output devices 108 (e.g., monitors, mice, touchpads, or keyboards), and a communication module 110. Processor device 104 may include, for example, a central processing unit (CPU), a graphics processing unit (GPU), a field-programmable gate array (FPGA), or a combination thereof. Computer system 101 may also include support circuitry such as caches, power supplies or batteries, clock circuitry, and communication buses (not shown). Various other peripheral devices, such as client devices (e.g., workstations) and additional data storage and printing devices, may also be connected to computer system 101.

[0013] This technology can be implemented in various forms of hardware, software, firmware, dedicated processors, or combinations thereof, as part of the following executed via an operating system: microinstruction code, or part of an application or software product, or a combination thereof. In some embodiments, the technology described herein is implemented as computer-readable program code tangibly embodied in one or more non-transitory computer-readable media 105. In particular, this technology can be implemented by a processing engine 107. The non-transitory computer-readable media 105 may include random access memory (RAM), read-only memory (ROM), floppy disk, flash memory, and other types of memory, or combinations thereof. The computer-readable program code is executed by processor device 104 to process data acquired, for example, by medical imaging device 114. The computer-readable program code is not intended to be limited to any particular programming language and its implementation. It will be appreciated that the teachings of the disclosure contained herein can be implemented using various programming languages ​​and their encodings. The same or different computer-readable media 105 may be used to store databases, including but not limited to image datasets, knowledge bases, individual subject data, medical records, subject diagnostic reports (or documents), or combinations thereof.

[0014] The communication module 110 enables the computer system 101 to communicate with external systems and / or networks. In some embodiments, the communication module 110 includes a high-speed digital interface, such as Thunderbolt. TMUniversal Serial Bus (USB), multi-gigabit Ethernet, fiber optic, waveguide technology, or wireless interfaces are all acceptable. Other types of interfaces are also useful. In some implementations, the communication module 110 includes a wireless transceiver that transmits signals using common communication protocols such as GSM, Wi-Fi, Bluetooth, Zigbee, LoRa, and TCP / IP.

[0015] Medical imaging device 114 is a radiological imaging device that acquires medical image data revealing internal structures hidden beneath the skin and bones of a subject. In some embodiments, medical imaging device 114 is a contrast-enhanced ultrasound (CEUS) system. The CEUS system acquires and displays B-mode images of tissue along with an overlaid contrast image or a contrast image displayed side-by-side with the B-mode image. The CEUS system may include a transmit beamformer, a transducer, and a receive beamformer. Additional, different, or fewer components may be provided. For example, a separate memory may be provided to buffer or store data frames.

[0016] It should also be understood that, because some of the system components and method steps depicted in the accompanying drawings can be implemented in software, the actual connections between system components (or process steps) can vary depending on how this framework is programmed. Given the teachings provided herein, those skilled in the art will be able to anticipate these and similar implementations or configurations of this framework.

[0017] Figure 2 An exemplary method 200 for characterization is shown. It should be understood that the steps of method 200 may be performed in the order shown or a different order. Additional, different, or fewer steps may also be provided. Furthermore, it is possible to utilize... Figure 1 System 100, different systems or combinations of the above to achieve the method 200.

[0018] At 202, processing engine 107 receives a sequence of contrast-enhanced ultrasound (CEUS) images of at least one structure of interest. The CEUS image sequence may be acquired by medical imaging device 114 (e.g., in real time) or previously generated and retrieved from, for example, a non-transitory computer-readable medium 105. The CEUS image sequence may be substantially continuous or periodic (e.g., acquired once or multiple times per heartbeat cycle). The CEUS sequence may be a video comprising data frames representing the region or structure of interest scanned at different times following administration of a contrast agent.

[0019] The structure of interest can be any anatomical structure identified for further research (e.g., the liver). Structures of interest include contrast agents or areas that may contain already administered contrast agents. The contrast agent can respond to ultrasound energy. Some or all frames in the sequence include information from the contrast agent, as well as responses from the tissue or fluid.

[0020] In some implementations, CEUS image sequences are acquired in two modes. Both B-mode images and contrast images are recorded simultaneously at, for example, the same frame rate and duration. B-mode images, also known as brightness mode images, are two-dimensional ultrasound images generated by converting the amplitude of the echoes into varying shades of gray. Contrast images are two-dimensional color images reflecting the spatial distribution of the contrast agent (e.g., microbubbles). Contrast images are formed by detecting and separating the nonlinear components of both linear and nonlinear signals from the contrast agent in the tissue.

[0021] At position 204, processing engine 107 derives a B-mode and contrast image sequence from the CEUS image sequence. The B-mode and contrast image sequence comprises a series of B-mode images and a corresponding series of contrast images extracted from the CEUS image sequence. Each frame in the B-mode series corresponds to a frame in the contrast image series and is spatially co-registered with frames in that contrast image series. In concurrent mode, B-mode and contrast image pairs can be acquired concurrently. In alternating mode, B-mode and contrast image pairs can be acquired alternately. Generally, B-mode images are of better quality in alternating mode, but the contrast and B-mode images may be temporally misaligned. Concurrent mode provides temporally aligned B-mode and contrast images.

[0022] In the resulting image sequence, the Mode B images and contrast images are spatially co-registered and temporally subsampled. In some implementations, deriving the image sequence involves subsampling frames from a CEUS image sequence that is temporally uniformly spaced. In other words, the time interval between the acquisition times of consecutive frames in each series is the same.

[0023] Alternatively, the resulting image sequence comprises subsampling frames from a CEUS image sequence that is non-uniformly spaced in time. In such an implementation, specific frames are selected from the CEUS image sequence to form a B-mode image sequence and a contrast image sequence. Frames included in each series can be selected based on at least one characteristic of the image data. Such characteristics include, but are not limited to, motion displacement and similarity between frames. For example, frames associated with substantial motion and / or frames very similar to previously selected frames are not selected for inclusion. The selected frames are “optimal” for quantification of tissue perfusion. “Optimal” frames have no substantial motion (e.g., out-of-plane motion) and have sufficient signal intensity to indicate adequate contrast agent penetration and adequate temporal dynamics sampling. Out-of-plane motion cannot be compensated for, while in-plane motion can be compensated for using registration techniques.

[0024] Frames within each series can be spatially aligned to compensate for in-plane motion between frames. In-plane motion may be caused by transducer movement, patient movement, and / or tissue movement within the region of interest. To compensate for motion, relative translations and / or rotations along one or more dimensions are determined. Data from one frame is correlated with different regions in another data frame to identify the best or sufficient match. Correlation, cross-correlation, minimum sum of absolute differences, and / or another similarity metric can be used. Rigid or non-rigid motion models can be provided, such as distortions in addition to translations and rotations in non-rigid motion models. The entire data frame or data window can be used to determine the best match and corresponding motion.

[0025] In some implementations, the extracted B-mode and contrast images are spatially co-registered to compensate for (or correct) motion (e.g., in-plane motion). Unlike conventional methods, this framework combines B-mode images with contrast images for motion correction. B-mode and / or contrast images can be used for motion correction based on timing of contrast dosing. For example, B-mode images acquired only during the early perfusion phase can be used for motion correction because they provide a better representation of lesion boundaries, while contrast images acquired only during the late clearance phase can be used for motion correction because they provide better information. Early perfusion generally refers to perfusion that is clearly detectable earlier than a predetermined time (e.g., 60 seconds) after contrast dosing. Late clearance generally refers to clearance that is clearly detectable at or after a predetermined time (e.g., 60 seconds) after contrast dosing. Alternatively, if the liver lesion has high contrast and the B-mode image is not affected by contrast filling, both B-mode and contrast images can be used for motion correction.

[0026] At point 206, processing engine 107 selects at least one region of interest (ROI) in the first frame of the B-mode image series and / or contrast image series. The operator can select the ROI in the first frame via, for example, a user interface. Alternatively, the ROI can be detected semi-automatically or automatically in the first frame using, for example, machine learning techniques. The ROI may correspond to a lesion or other abnormality in the structure of interest. More than one ROI can be selected (e.g., multiple lesions).

[0027] At point 208, processing engine 107 characterizes the region of interest based on the derived image sequence. The output of such characterization can be, for example, qualitative parameters of the region of interest, quantitative parameters of the region of interest, a timeline plot, or a combination of these. The characterization output can be presented (or displayed) in a user interface, for example, on a user device, to facilitate diagnosis. The characterization output can also be stored, for example, in an electronic health record, for access by, for example, patients and healthcare providers.

[0028] To generate qualitative parameters, the resulting image sequences can be input into the diagnostic function module. B-mode image sequences can be used to evaluate, for example, the size, volume, shape, echogenicity, and arrangement of the region of interest, while contrast image sequences can be used to evaluate, for example, hemodynamics and angiogenesis patterns, such as arterial phase contrast perfusion characteristics (e.g., filling pattern, dynamics, timing) and hilar phase clearance characteristics (e.g., degree of clearance, timing). Additionally, characteristics may also include lesion type (e.g., benign or malignant). The diagnostic function module can implement deep learning techniques or any other suitable techniques to, for example, extract contrast characteristics or predict lesion type (e.g., malignant, benign), lesion class (e.g., hepatocellular carcinoma or HCC, focal nodular hyperplasia or FNH, hemangioma), or Liver Imaging Reporting and Data System (LI-RADS) category.

[0029] In some implementations, processing engine 107 performs quantification of regions of interest based on the resulting image sequence to generate quantitative parameters. Quantification can be performed by extracting a calculated time-intensity ratio curve between lesion and liver intensity or using deep learning-based techniques. As previously mentioned, such a curve can provide information about contrast perfusion and clearance characteristics. Processing engine 107 can select the “best” frame for quantification from each of these series, for example, using similarity measurements to reference frames (e.g., the first frame) or deep learning-based techniques used in object tracking.

[0030] In some implementations, processing engine 107 generates a timeline plot based on the resulting image sequence that tracks the region of interest. A timeline plot is a visual representation of a series of B-mode images adjacent to (e.g., above or below) a series of contrasting images that progress over time. The timeline plot can provide a single static representation of the characteristics of the region of interest, such as arterial phase (AP) perfusion patterns (e.g., centrifugal, centripetal, rim-like, spherical), clearance timing, etc.

[0031] Figure 3 An exemplary timeline 302 is shown. Timeline 302 includes a series of Mode B images 304a-b displayed above a series of contrast images 306a-b. Although a static horizontal timeline 302 is shown, it should be understood that other configurations of the timeline (e.g., a vertical configuration) are also possible. Timeline 302 includes arterial phase (AP) perfusion images (304a, 306a). The AP perfusion images (304a, 306a) are finely subsampled and preselected. Additionally, portal vein phase clearance images (304b, 306b) may also be included. The portal vein phase clearance images (304b, 306b) may be sparsely sampled. Acquisition time 308 may be displayed above timeline 302.

[0032] Additional timeline plots can be generated to represent additional regions of interest or multiple contrast agent administrations to a single region of interest. For example, there may be multiple lesions in a person's liver, or multiple contrast agent injections may be administered for visualization of a single lesion. Multiple timeline plots can be shown to the user to provide additional information about the contrast process. Multiple timeline plots of multiple lesions can be shown concurrently to facilitate visual comparison of lesions.

[0033] The following is a list of non-limiting illustrative embodiments disclosed herein: Illustrative Example 1. An image characterization method includes: receiving a sequence of contrast-enhanced ultrasound (CEUS) images; deriving a brightness pattern (B-mode) and a contrast image sequence from the CEUS image sequence, wherein the B-mode image and the contrast image in the B-mode and contrast image sequences are spatially co-registered and temporally subsampled; selecting at least one region of interest in the B-mode and contrast image sequences; and characterizing at least one region of interest based on the B-mode and contrast image sequences.

[0034] Illustrative Example 2. The method according to Illustrative Example 1, wherein deriving the B mode and the contrast image sequence includes subsampling frames that are uniformly spaced in time.

[0035] Illustrative Example 3. The method according to any one of Illustrative Examples 1-2, wherein deriving the B mode and the contrast image sequence includes subsampling the frames, including subsampling the frames that are non-uniformly spaced in time.

[0036] Illustrative Example 4. According to the method of Illustrative Example 3, frames are selected from the CEUS image sequence based on motion displacement, similarity between frames, or a combination of the above.

[0037] Illustrative Example 5. The method according to any one of Illustrative Examples 1-4, wherein obtaining the B-mode and contrast image sequences includes spatially co-registering the B-mode and contrast images for motion correction.

[0038] Illustrative Example 6. The method according to Illustrative Example 5, wherein B-mode images acquired only in the early perfusion phase and contrast images acquired only in the late clearance phase are used for motion correction.

[0039] Illustrative Example 7. According to the method described in Illustrative Example 5, both the B-mode image and the contrast image are used for motion correction.

[0040] Illustrative Example 8. The method according to any one of Illustrative Examples 1-7, wherein characterizing at least one region of interest based on B-mode and contrasting image sequences includes generating one or more qualitative parameters.

[0041] Illustrative Example 9. The method according to any one of Illustrative Examples 1-8, wherein characterizing at least one region of interest based on B-mode and contrast image sequences includes generating one or more quantitative parameters.

[0042] Illustrative Example 10. The method according to any one of Illustrative Examples 1-9, wherein characterizing at least one region of interest based on B-mode and a contrasting image sequence includes generating a timeline plot.

[0043] Illustrative Example 11. According to the method of Illustrative Example 10, generating a timeline plot includes generating a static horizontal timeline plot.

[0044] Illustrative Example 12. The method according to any one of Illustrative Examples 1-11, wherein characterizing at least one region of interest based on B-mode and contrast image sequences includes generating multiple timeline plots for administering multiple contrast agents to the region of interest.

[0045] Illustrative Example 13. An image processing system includes: a non-transitory memory device for storing computer-readable program code; and a processor device in communication with the non-transitory memory device, the processor device operating with the computer-readable program code to perform steps including: deriving a brightness pattern (B-mode) and a contrast image sequence from a sequence of contrast-enhanced ultrasound (CEUS) images, wherein the B-mode image and the contrast image in the B-mode and contrast image sequences are spatially co-registered and temporally subsampled; selecting at least one region of interest in the B-mode and contrast image sequences; and characterizing the at least one region of interest based on the B-mode and contrast image sequences.

[0046] Illustrative Example 14. The image processing system according to Illustrative Example 13, wherein the processor device operates together with computer-readable program code to derive a B-mode and contrast image sequence by subsampling frames that are uniformly spaced in time.

[0047] Illustrative Example 15. An image processing system according to any one of Illustrative Examples 13-14, wherein the processor device operates together with the computer-readable program code to derive a B-mode and contrast image sequence by subsampling frames that are non-uniformly spaced in time.

[0048] Illustrative Example 16. An image processing system according to any one of illustrative Examples 13-15, wherein the processor device operates together with the computer-readable program code to select at least one region of interest by selecting one or more lesions.

[0049] Illustrative Example 17. An image processing system according to any one of Illustrative Examples 13-16, wherein the processor device operates together with the computer-readable program code to characterize at least one region of interest based on B-mode and a contrasting image sequence by generating one or more qualitative parameters, quantitative parameters, timeline plots, or combinations thereof.

[0050] Illustrative Example 18. An image processing system according to any one of illustrative Examples 13-17, wherein the processor device operates together with the computer-readable program code to characterize at least one region of interest based on a sequence of B-mode and contrasting images by generating a static horizontal timeline plot.

[0051] Illustrative Example 19. An image processing system according to any one of Illustrative Examples 13-18, wherein the processor device operates together with the computer-readable program code to generate multiple timeline plots based on the B-mode and the contrast image sequence by administering multiple contrast agents to a region of interest.

[0052] Illustrative Example 20. One or more non-transitory computer-readable media embodying machine-executable instructions to perform operations including: deriving a brightness pattern (B-mode) and a contrast image sequence from a sequence of contrast-enhanced ultrasound (CEUS) images, wherein the B-mode image and the contrast image in the B-mode and contrast image sequences are spatially co-registered and temporally subsampled; selecting at least one region of interest in the B-mode and contrast image sequences; and characterizing the at least one region of interest based on the B-mode and contrast image sequences.

[0053] While this framework has been described in detail with reference to exemplary embodiments, those skilled in the art will appreciate that various modifications and substitutions can be made therein without departing from the spirit and scope of the invention as set forth in the appended claims. For example, within the scope of this disclosure and the appended claims, elements and / or features of different exemplary embodiments may be combined with and / or substituted for each other.

Claims

1. An image representation method, comprising: Receive contrast-enhanced ultrasound (CEUS) image sequences; A luminance mode (B mode) and a contrast image sequence are derived from the CEUS image sequence, wherein the B mode image and the contrast image in the B mode and contrast image sequence are spatially co-registered and temporally subsampled; Select at least one region of interest from the B mode and the contrast image sequence; as well as The at least one region of interest is characterized based on the B-mode and the contrast image sequence.

2. The method of claim 1, wherein deriving the B mode and the contrast image sequence comprises subsampling frames that are uniformly spaced in time.

3. The method of claim 1, wherein deriving the B mode and the contrast image sequence comprises subsampling the frames, including subsampling frames that are non-uniformly spaced in time.

4. The method of claim 3, wherein the frame is selected from the CEUS image sequence based on motion displacement, similarity between frames, or a combination of the above.

5. The method of claim 1, wherein obtaining the B-mode and contrast image sequence includes spatially co-registering the B-mode and contrast images for motion correction.

6. The method of claim 5, wherein the B-mode images acquired only in the early perfusion phase and the contrast images acquired only in the late clearance phase are used for motion correction.

7. The method of claim 5, wherein both the B-mode image and the contrast image are used for motion correction.

8. The method of claim 1, wherein characterizing the at least one region of interest based on the B mode and the contrast image sequence includes generating one or more qualitative parameters.

9. The method of claim 1, wherein characterizing the at least one region of interest based on the B mode and the contrast image sequence includes generating one or more quantitative parameters.

10. The method of claim 1, wherein characterizing the at least one region of interest based on the B mode and the contrasting image sequence includes generating a timeline plot.

11. The method of claim 10, wherein generating the timeline plot includes generating a static horizontal timeline plot.

12. The method of claim 1, wherein characterizing the at least one region of interest based on the B mode and the contrast image sequence comprises generating multiple timeline plots for administering multiple contrast agents to the region of interest.

13. An image processing system, comprising: Non-transitory memory devices used for storing computer-readable program code; as well as A processor device communicating with the non-transitory memory device, the processor device operating together with the computer-readable program code to perform steps, the steps including A brightness pattern (B-mode) and a contrast image sequence are derived from a contrast-enhanced ultrasound (CEUS) image sequence, wherein the B-mode and contrast images in the B-mode and contrast image sequences are spatially co-registered and temporally subsampled. Select at least one region of interest from the B-mode and the contrast image sequence, and The at least one region of interest is characterized based on the B-mode and the contrast image sequence.

14. The image processing system of claim 13, wherein the processor device operates together with the computer-readable program code to derive the B-mode and contrast image sequence by subsampling frames that are uniformly spaced in time.

15. The image processing system of claim 13, wherein the processor device operates together with the computer-readable program code to derive the B-mode and contrast image sequence by subsampling frames that are non-uniformly spaced in time.

16. The image processing system of claim 13, wherein the processor device operates together with the computer-readable program code to select the at least one region of interest by selecting one or more lesions.

17. The image processing system of claim 13, wherein the processor device operates together with the computer-readable program code to characterize the at least one region of interest based on the B-mode and the contrast image sequence by generating one or more qualitative parameters, quantitative parameters, timeline plots, or combinations thereof.

18. The image processing system of claim 13, wherein the processor device operates together with the computer-readable program code to characterize the at least one region of interest based on the B-mode and contrasting image sequence by generating a static horizontal timeline plot.

19. The image processing system of claim 13, wherein the processor device operates together with the computer-readable program code to generate multiple timeline plots based on the B-mode and the contrast image sequence by administering multiple contrast agents to the region of interest.

20. One or more non-transitory computer-readable media embodying machine-executable instructions for performing operations, said operations including: A brightness pattern (B-mode) and a contrast image sequence are derived from a contrast-enhanced ultrasound (CEUS) image sequence, wherein the B-mode image and the contrast image in the B-mode and contrast image sequences are spatially co-registered and temporally subsampled; Select at least one region of interest from the B mode and the contrast image sequence; as well as The at least one region of interest is characterized based on the B-mode and the contrast image sequence.

Images