Method and system for data transmission reduction in ultrasound imaging - Patents.com
By identifying a region of interest and applying tailored transmit and receive schemes, the method addresses power and data transmission challenges in ultrasound imaging, achieving ultrafast imaging with reduced power consumption and optimized data processing.
Patent Information
- Application Number
- JP2022580512
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2020-06-29
- Filing Date
- 2021-06-18
- Publication Date
- 2024-11-14
- Estimated Expiration
- 2041-06-18
AI Technical Summary
Conventional ultrasound imaging techniques face challenges in achieving high temporal and spatial resolution while managing power consumption and data transmission loads, particularly in handheld and wireless systems, due to the need for high acoustic energy and complex hardware to process large amounts of radio frequency data.
The method involves selecting a region of interest (ROI) within the imaging area and applying distinct transmit and receive schemes for this region, reducing the number of ultrasound exposures and data transmission by focusing on the ROI with higher temporal resolution, while using simplified hardware and optimized data processing.
This approach reduces power consumption and data transmission rates, enabling ultrafast ultrasound imaging with improved spatial and temporal resolution, particularly suitable for echocardiography and other applications requiring high frame rates.
Smart Images

Figure 0007583839000004 
Figure 0007583839000005 
Figure 0007583839000006
Abstract
Description
[Technical field]
[0001] The present invention relates to a method and a corresponding system for data transmission reduction in an ultrasound (US) imaging system. [Background technology]
[0002] In conventional ultrasound imaging, a 2D image is acquired by transmitting a transmit pulse in 200 different directions in a plane covering an angle of 90° and performing about 200 pulse-echo measurements. Thus, for each ultrasound transmission event, one line is acquired (single line acquisition, SLA). With this approach, a 2D image with an imaging depth of 130 mm can be acquired at a frame rate of about 30 Hz. In cardiac ultrasound imaging, such a frame rate is sufficient to visualize the overall motion of the heart. However, some cardiac phases are known to be of very short duration and associated with very fast motions and deformations of the myocardium and valves. Accurate observation of these cardiac phases requires a higher frame rate.
[0003] Therefore, ultrafast ultrasound imaging techniques have been developed, such as parallel beamforming (i.e., multiple line acquisition, MLA). In this approach, multiple image lines are simultaneously reconstructed for each transmit pulse. To accommodate the increased number of parallel receive beams, the transmit beam must be broadened (e.g., by reducing the transmit aperture). However, since a smaller aperture transmits less energy, the maximum number of parallel receive beams remains limited (e.g., 2, 4, or 8). As an alternative, plane wave or dispersive wave imaging has been proposed. This technique allows a high degree of parallel receive beamforming, but the spatial resolution of the image is significantly reduced due to the very wide transmit beams. To compensate for this effect, different transmit beams are typically combined, resulting in an overall gain in frame rate.
[0004] In general, ultrafast imaging requires a relatively large number of transmissions (>10 for 2D imaging, >100 for 3D imaging) to produce an image of acceptable quality. Typically, ultrafast imaging techniques are operated at frame rates greater than 1000 Hz. Thus, the number of ultrasound exposures per second can be greater than 10 for 2D and >100 for 3D imaging. 4 Larger, 10 for 3D 5 To achieve sufficient penetration depth and to perform harmonic imaging at depth, the surface pressure used needs to be high. For example, it can be as high as 200 mm (assuming a high voltage driver power efficiency of 30% and a pulse generator on-off duty cycle of 0.2%). 2 Approximately 1.5 W / mm to generate a surface pressure of 1.5 MPa using a transducer aperture of 2 Therefore, a significant amount of acoustic energy is required to insonify the field of view. When operating at high frame rates, acoustic power needs to be generated many times.
[0005] These requirements result in higher power consumption in ultrasound (US) probes. Especially for small handheld and / or wearable (wireless) imaging systems, it is important to keep probe costs and power consumption down. A typical USB handheld probe limits its power consumption to 2-3W so that power can be provided over the USB3 interface. Without active cooling, the 2-3W limit is further required to comply with medical regulations related to (short-term) skin heating. Wireless patch probes for monitoring applications also limit power consumption to 100mW to preserve battery life and prevent long-term skin heating.
[0006] Furthermore, ultrafast ultrasound techniques result in an increased data transmission load, which standard techniques do not yet support. Especially for matrix transducers consisting of more than 1000 transducer elements, the envisaged radio frequency (RF) data volumes are very high. To achieve high spatial resolution, the raw data of many transducer elements (channels) needs to be collected and processed independently. This results in high consumption and sometimes complex hardware, e.g. multi-Gbps serial data links.
[0007] It is therefore desirable to limit the number of sonications and the amount of RF data that is generated, processed and transmitted, which further reduces probe costs since simpler hardware components can be used, and since the main power consumption is due to sonications, it is further important to collect all relevant echo information to make further sonications unnecessary.
[0008] In the prior art, region of interest based image compression has been proposed (e.g., Heng-Ming Tails, Men Long, Wei He, and Hao Yarg, “AN EFFICIENT REGION OF INTEREST CODING FOR MEDICAL IMAGE COMPRESSION”, Proceedings of the Second Joint EMEISEIMES Conference Houston, TX, USA *October 2+26, 2W2, and “Review on Region of Interest Coding Techniques for Medical Image Compression”, International Journal of Computer Applications (0975-8887) Volume 134-No. 10, January 2016). In this approach, a region of an image (region of interest, ROI) is coded with higher quality (possibly losslessly) than non-ROI regions (which are typically compressed using lossy methods). This is usually achieved by allocating more bits to the ROI compared to the non-ROI regions.
[0009] US Patent Application Publication No. 2016 / 0262720(A1) is concerned with ultrasound imaging and proposes scanning the entire volume and sub-volumes separately, where the sub-volumes are scanned using different settings for beamforming parameters, thus allowing higher image quality for the sub-volumes while providing context from the entire volume. Thus, the volume and sub-volume slices are interleaved.
[0010] No. 5,873,830(A) discloses an ultrasound system and method in which a region of interest is selected. A first set of imaging parameters is applied inside the region of interest to improve spatial resolution, and a different set of parameters is applied outside the region of interest. A composite image is then constructed that includes the region of interest and the region outside the region of interest.
[0011] However, even with such advanced methods, the amount of data that needs to be processed and transmitted remains an impediment to applicability and the goal of minimizing associated costs. Summary of the Invention [Problem to be solved by the invention]
[0012] It is therefore an object of the present invention to provide a method and associated system for ultrasound imaging that provides high quality ultrasound images with sufficient temporal and spatial resolution and sufficiently wide field of view coverage while reducing the power consumption of the ultrasound probe and the amount of data that needs to be transmitted. [Means for solving the problem]
[0013] This object is realized or exceeded by a method for ultrasound imaging as claimed in claim 1 and an ultrasound imaging system as claimed in claim 15. Any feature, advantage or alternative embodiment described in this specification in relation to the claimed method may also be applied to the system and vice versa.
[0014] According to the present invention, there is provided a method for ultrasound imaging using an ultrasound system comprising an ultrasound probe including a transducer array and an analog-to-digital converter (ADC) array, the ultrasound probe being configured to insonify an imaging region, to receive echo signals, to digitize the received echo signals using the ADC array, and to transmit the digitized radio frequency (RF) data via a bandwidth limited channel to a data processing unit, the data processing unit being adapted to beamform the RF data, the method comprising the following steps: - acquiring at least one scouting image of an imaging region, the scouting image including an anatomical structure of interest; - selecting at least one region of interest (ROI) within a scouting image, the at least one ROI including a portion of an anatomical structure; - configuring a transmit (TX) and receive (RX) scheme for ultrasound imaging of an imaging region, wherein at least one ROI is insonified using different TX and RX settings than an imaging region outside the at least one ROI; - acquiring a plurality of ultrasound images according to a TX scheme and a RX scheme; A method is provided comprising:
[0015] The ultrasound probe is configured to transmit to the data processing unit only received RF data corresponding to a depth range of at least one ROI for at least some of the specified receive beams reflected from the ROI, and is configured to transmit received RF data corresponding to the entire depth of the imaging region for at least some of the receive beams reflected by the imaging region outside the at least one ROI.
[0016] The steps described above are generally performed in the order in which they are presented, except that the TX and RX schemes may be arranged in any order.
[0017] An important application of the present invention is echocardiography. Considering the heart as a collection of high moving objects, the valves of the heart move faster than the atrial / ventricular walls. Therefore, in this and other applications, not all parts of the moving anatomical structures require high-speed or ultra-high-speed imaging techniques. Therefore, the present invention provides a method in which one or more ultrasound images, called scouting images, of an anatomical structure of interest, for example the whole heart, are first acquired. For this ultrasound image, a region of interest is selected, which preferably includes a part of the anatomical structure, such as a part of a particular interest or a part that moves faster than the rest of the anatomical structure. The ROI usually covers a part of the imaging area of the scouting image, for example 5% to 40% thereof. Preferably, the ROI includes a part of the anatomical structure that needs to be imaged using different scan settings, in particular different transmit and / or receive settings (also called transmit and receive parameters), in order to obtain images with higher temporal resolution and / or better quality from the ROI compared to the rest of the imaging area. The TX and RX schemes include settings such as the general imaging / measurement mode (e.g., plane wave, dispersive wave, Doppler imaging), the number, type, and width of the TX and RX beams, frame rate / time resolution, transmit power, pulse repetition rate, focusing scheme, etc.
[0018] The ROI is selected manually or automatically. There may be more than one ROI, for example, two to five ROIs. Based on the location of one or more ROIs in the imaging region relative to the ultrasound probe and the desired image quality or time resolution for the ROI, then a transmit and receive scheme for ultrasound imaging of the imaging region is set, preferably automatically, according to a predefined user specification, which results in scanning the ROI using different TX and RX settings, in particular using higher time resolution, contrast, power, focusing scheme, or generally using a different type of transmit beam, compared to the imaging region outside the ROI. For example, based on the spatiotemporal location of the ROI, the transmit beams illuminating the ROI are scheduled at smaller time intervals, and thus RF data is acquired from the ROI at a faster rate. For other parts of the anatomical structure, the prescribed transmit beams are scheduled at longer time intervals to match their slower speed of movement.
[0019] Furthermore, the receiving scheme comprises configuring the ultrasound probe to transmit only received RF data for at least some of the defined receiving beams reflected from the ROI, i.e., corresponding to the depth range of the ROI for the receiving beams used to reconstruct the image of the ROI, to the data processing unit. The depth direction corresponds to the distance from the ultrasound probe. In contrast, for non-ROI regions, at least some, preferably all, receiving beams are digitized and transmitted to the data processing unit (DPU). In other words, the receiving scheme can be adjusted to keep samples only in the depth range corresponding to the ROI of the defined transmitting beams. Thus, the data transmission rate to the DPU is significantly reduced. After being digitized, the RF data is shifted in time and interleaved as necessary to reconstruct a video stream of an ultrasound image.
[0020] Therefore, the present invention describes a method to reduce the number of transmit (TX) events, preferably resulting in a faster frame rate, reducing the amount of RF data that needs to be transmitted from the probe to the data processing unit, and thus reducing the amount of further beamforming calculations that need to be performed. This is achieved by an adaptive TX-RX scheme that achieves improvements in data compression and power consumption due to 1) an optimized number of transmit or TX / RX settings, and 2) limited depth range reconstruction, resulting in a reduced RF data transmission load.
[0021] To achieve these advantages, the present invention is implemented using a digital ultrasound probe, i.e. an ultrasound probe with in-probe digitization of the received echo signals (RF signals). The digitized RF data is transmitted to a data processing unit. In a next step, the digitized RF data is processed and mixed together into a continuous video stream showing the ROI region with better image quality, e.g. with higher time resolution, compared to the non-ROI region, i.e. the imaging region outside the ROI. In one embodiment, the data processing unit is adapted to beamform the RF data into a plurality of ultrasound images. In a next step, the plurality of ultrasound images are mixed together into a video stream. In a preferred embodiment, the data processing unit is not part of the ultrasound probe, but is located in the main ultrasound system and is connected to the probe via a digital interface. Such a digital interface comprises a data connection or channel with a limited bandwidth, i.e. a maximum data rate, e.g. a serial data link with a defined maximum number of Gbps. In other words, the ultrasound probe and the data processing unit (DPU) are preferably accommodated in separate housings connected by a data cable or even by a wireless data connection. Thus, the method of the present invention can be realized by a simplified digital front end, e.g. a digital ultrasound probe.
[0022] The scouting image may be acquired using any imaging scheme, preferably in B-mode, with lower resolution or lower frame rate to limit the data rate involved, but the scouting image of the imaging region (including the anatomical structures) may also be one or more of the multiple ultrasound images acquired by implementing the method of the present invention, in which the position of the ROI is likewise detected and adjusted if necessary.
[0023] In one embodiment, several scouting images are acquired so that fast moving regions or parts of the anatomical structure can be automatically detected and selected as regions of interest (ROI), e.g., by computer vision or artificial intelligence (AI) algorithms. Such algorithms are specifically adapted to a particular clinical application. Alternatively, the ROI is manually selected, e.g., by a user displaying the scouting images on a screen and setting one or more markers on the parts of interest. The user further delineates the ROI by drawing a graphical object, e.g., a circle or a square, on the ROI or by using a cursor to draw the boundary of the ROI.
[0024] Then, given a ROI, different transmit and receive settings are set for the ROI and the imaging area outside the ROI (non-ROI area).
[0025] In other words, the transmission and reception schemes according to the method embodiments of the present invention include two types of insonification: Type A for the entire imaging region and Type B for the ROI.
[0026] Type A: In a first series of ultrasound exposures, the target anatomical structure (e.g., the heart) is insonified, for example, using the same plane wave transmission. RF data from all transducers (e.g., 128 channels) is used for the entire imaging area. Digitization and beamforming are performed at medium resolution to limit the amount of data collected. By comparing and interpreting the data of the sequential ultrasound exposures or scouting images, the system (especially the ultrasound system controller) can detect fast moving areas, such as heart valves.
[0027] Type B: In the second series of insonification, for example, a focused transmit beam is transmitted and data from a limited number of RF signal channels is collected. The data is digitized and transmitted with a relatively high resolution. Data acquisition and transmission is only performed over a limited period of time corresponding to a limited depth range to access a specific region of interest. For example, acquisition begins 50 μs after the insonification event and lasts for 20 μs (in real time).
[0028] It takes 1000 μs to transmit data for sonication type A and 100 μs to transmit data for sonication type B. The controller is aware of the in-probe data storage capacity (if any) and the available transmission rate between the probe and the DPU, and adapts the timing for new sonication requests accordingly. Plane wave sonications (e.g. for type A) and targeted sonications (type B) are generated in a regular time interleaved manner. However, the interleaving may be irregular / dynamic and / or triggered by temporal signal characteristics. Temporal signal characteristics are particularly the speed of motion detected within a particular region of the imaging field. Artificial intelligence may be used to recognize these temporal signal characteristics to support intelligent image formation or data detection.
[0029] The amount of data reduction is shown in FIG. 4. In the cube shown in FIG. 4, the fast time or depth axis is shown horizontally, the vertical axis shows the transmit events, and the third axis is the receive element axis. For multi-angle plane wave or dispersive wave imaging, the entire cube shows the entire RF data array acquired for a 2D image of the imaging area. The smaller cubes correspond to the RF data transmitted for at least some of the ROI scans. It is clear that the smaller cubes are made smaller in depth and on the transmit axis because the transmit scheme often requires fewer transmit events to image the ROI compared to the entire imaging area, since the ROI typically does not cover the full azimuth.
[0030] In some embodiments, the cube representing the RF data transmitted for the ROI is also made smaller in the receive element axis. In other words, only a subset of the receive elements are active when receiving RF data from a transmit beam that illuminates the ROI, and / or only data from a limited number of signal channels is collected and digitized. In other words, only RF data from a portion of the transducer elements is transmitted to the DPU for at least some of the TX / RX events that illuminate the ROI. Thus, the data transmission load can be further reduced.
[0031] By transmitting only RF data corresponding to a particular depth range (d, also called fast time) and only when transmitting to a particular transmit angle, the amount of data transmitted can be reduced according to the following equation:
number
[0032] Thus, the data transmission rate can be significantly reduced. For example, in an ultrasound probe using 128 transducer elements, using an 8 GHz sound frequency, sampled with 12-bit resolution and 4 samples per wave, the resulting data rate is 8M x 128 x 4 x 12 bits / sec = 49 Gbit / sec. By reducing the amount of data transmitted for those scans related to the ROI, the data rate over the cable of the interface between the probe and the DPU can be reduced, for example, to less than 20 Gbit / sec, more preferably less than 10 Gbit / sec.
[0033] After digitization and transmission to the DPU, the digitized RF data is beamformed into multiple ultrasound images. Since the data from the ROI region contains less data, the beamforming calculations are also computationally less expensive. Beamforming can be applied to the RF data from the ROI by setting the data corresponding to the depth range outside the ROI to zero, thus receiving an image of the imaging region with the non-ROI region black. This type of image is easily blended with the beamformed ultrasound image of the entire imaging region. Alternatively, only the ROI region can be beamformed, resulting in a smaller image. This results in fewer beamforming calculations, but makes the video blending step more complex, especially with respect to blending at the correct location in the overall image.
[0034] The result of the method is a plurality of ultrasound images, particularly a temporal sequence, showing the entire imaging area, but showing the ROI with better quality, in different imaging or measurement modes, and / or with higher temporal resolution. The higher temporal resolution for the non-ROI area is achieved by upsampling, as will be explained in more detail later, and therefore does not contain more information. However, the ROI area has been sampled with a higher temporal resolution in some embodiments, and therefore fast moving parts of the anatomical structure can be analyzed. Thus, ultrafast ultrasound imaging can be achieved with a lower required data transmission rate, resulting in less power consumption, without sacrificing the frame rate.
[0035] According to an embodiment, the TX scheme is configured such that at least one ROI is insonified using a higher time resolution and / or using a different type of transmit beam and / or using a different pulse repetition rate compared to imaging areas outside the at least one ROI. In a preferred embodiment, the transmit parameters are configured such that the ROI is insonified more frequently than non-ROI areas. According to the invention, frame rates of about 60-1000 Hz, preferably 120 Hz-500 Hz, can be realized in the ROI, which corresponds to very fast imaging, but without higher power consumption and data transmission rates, as would be required if the entire imaging area were imaged using such high frame rates. Additionally or alternatively, depending on the position of the ROI relative to the ultrasound probe, the transmit power and / or the transmit focusing scheme are adapted. For example, focused TX is useful for performing harmonic imaging in the ROI. The pulse repetition rate is the number of ultrasound pulses emitted per second. The pulse repetition rate is limited by the speed of sound in a medium, e.g., human tissue or anatomical structures, and the distance the pulse must travel on the return trip from the probe to the measurement area to avoid superposition of the returning echo signals. Therefore, it may be considered to increase the pulse repetition rate when insonifying an ROI, e.g., if the ROI region is relatively close to the ultrasound probe, unless the transducer receives a signal from a different insonification while listening to echoes from the ROI.
[0036] According to an embodiment, the ROI is insonified by an alternative measurement mode, for example by Doppler mode, i.e. the TX and RX scheme comprises measuring the Doppler shift in the ROI. This is performed by selecting a Doppler gate and a specific region where the Doppler shift is measured, i.e. the ROI for which a different transmit / receive scheme is used. In a dual mode, by interlacing the whole frame with the Doppler measurement, one can track the motion, in particular the blood flow, in the ROI and adjust the position of the Doppler gate accordingly. Thus, the data reduction method of the present invention can further be used for dynamically imaging an anatomical structure, in particular the heart or parts thereof, by Doppler mode, allowing for example the evaluation of valve function or other blood vessels. Other alternative measurement modes include strain imaging, for example measuring the deformation rate of parts of an anatomical structure, such as the myocardium, or ultrasound elasticity imaging.
[0037] Furthermore, a receive scheme is configured according to the transmit scheme, i.e., configured to receive beams from the area illuminated by the transmit beams.
[0038] According to a preferred embodiment, the ROI acquisition is interleaved with the global acquisition of the entire imaging area. In particular, the TX scheme comprises interleaving at least one ROI acquisition cycle in which only at least one ROI is insonified between two global acquisition cycles in which the entire imaging area of the ultrasound probe is insonified, and according to the RX scheme, in the ROI acquisition cycle, the ultrasound probe is configured to transmit only the received RF signals corresponding to the depth range of at least one ROI for at least some of the defined receive beams reflected by the at least one ROI, and in the global acquisition cycle, the ultrasound probe is configured to transmit the received RF signals corresponding to the entire imaging area of the ultrasound probe. For example, 1 to 8, preferably 2 to 4 ROI acquisition cycles are interleaved between two global acquisition cycles.
[0039] According to a preferred embodiment, the ultrasound probe comprises an in-probe memory, which is in particular integrated with the ADC array. The method further comprises the steps of selecting digitized RF data corresponding to a depth range of at least one ROI for at least some of the defined receive beams reflected by the at least one ROI for transmission to a data processing unit, and buffering at least some of the selected digitized RF data in the in-probe memory to delay transmission of the RF data from the ultrasound probe when a predefined maximum data rate is exceeded. Thus, in some embodiments, a constant output data rate is achieved. The ADC array preferably corresponds to the transducer array. A digital ultrasound probe may easily generate more data than can be transmitted from the probe to the system (DPU), for example if the ADC samples at a frequency of 12-40 MHz resulting in a required data transmission rate of up to 50 Gbps, whereas the data link speed defined by a channel with limited bandwidth may be only 3-6 Gbps. Thus, the data connection between the probe and the DPU is a bottleneck in many systems. It is therefore important to use the maximum available data rate of the interface as much as possible. Therefore, this embodiment of implementing the invention involves selecting relevant RF data and discarding other digitized RF data, especially RF data corresponding to a depth range outside the ROI region for a beam reflected from the ROI in an ROI acquisition cycle. This allows the relevant RF data to be flushed out while the ultrasound probe is already performing the next acquisition. When insonifying an ROI with higher time resolution, the interface simply must be able to transmit all digitized (full imaging region and ROI) RF data over a cycle with one or more ROI acquisition cycles and one full acquisition cycle. However, the interface does not necessarily need to be able to transmit RF data at the native ADC data rate.The in-probe memory is preferably integrated with or located in the ADC array, as this is often more energy efficient.
[0040] Therefore, it is possible to first digitize all received RF data and then select the relevant data corresponding to the depth range of the ROI. Alternatively, it is possible to control the ADC array so that only the received signals corresponding to the depth range of the ROI for a given transmit beam illuminating the ROI (or a given receive beam reflected from the ROI) are digitized. Thus, the ADC array is switched on only in the required depth range. In this embodiment, since only the necessary RF data is digitized in the first place, the in-probe memory may or may not be used, but it is still beneficial to achieve a constant data rate.
[0041] According to an embodiment, before the digitized RF data is transmitted to the data processing unit, the digitized RF data is processed, in particular compressed. For example, a data compression method capable of compressing a dense RF array / tensor / buffer is used, such as wavelet compression and / or discrete wavelet transform (DWT). An example of such a compression method is the MPEG data compression protocol.
[0042] The digitized RF data from the ROI and non-ROI regions may be coded / processed in different ways. In particular, the digitized RF data reflected by at least one ROI may be compressed with a different bit depth, in particular using a larger bit depth, and / or using a different sub-sampling pattern, e.g. sampling the RF data at a different sampling rate, and / or using a different wavelet compression, compared to the digitized RF data reflected by the imaging region outside the at least one ROI. In particular, in order to reduce the data rate over a channel with limited bandwidth, the RF data from outside the ROI is compressed to a greater extent than the data from the ROI, while at the same time maintaining a sufficiently high quality of the RF data from within one or more ROIs. It is also conceivable to use different coding algorithms for the RF data from the ROI compared to the RF data from outside the ROI.
[0043] According to an embodiment, the transmit and receive scheme uses a wide transmit beam imaging mode, e.g. dispersive wave, plane wave, or angle-specified plane wave imaging mode, with multiple transmit events to insonify the imaging region. For example, plane wave imaging can be realized by transmitting 10-50 transmit beams, e.g. 1°-5° transmit beams, in different directions from which multiple receive beams are reconstructed. Only a portion of these transmit events insonify at least one ROI. Preferably, the transmit events insonifying at least one ROI are performed with a higher time resolution. For example, in angle-specified plane wave imaging mode, different transmit beams are directed in different directions. Depending on the size of the ROI, only a subset of the transmit beams are needed to image the ROI. In this example, the ROI acquisition cycle includes only a subset of the transmit events, whereas the full acquisition cycle includes transmit events in all directions covering the entire azimuth angle, e.g. 90°. According to an embodiment, in the TX scheme, in particular in the ROI acquisition cycle, the transmit power and / or focusing scheme of the transmit beams insonifying the ROI are adapted to the ROI, in particular to the position of the ROI relative to the ultrasound probe. For example, the transmit beam can be focused to have the best focus at the depth range of the ROI. Furthermore, the transmit power can be adjusted to the distance of the ROI from the ultrasound probe, e.g., the longer the distance, the more power is needed.
[0044] According to an embodiment, two or more regions of interest (ROIs) are selected in the scouting image, each of the two or more ROIs including a part of the anatomical structure, and in a transmit (TX) and receive (RX) scheme, the two or more ROIs are insonified using TX and RX settings different from an imaging area of the ultrasound probe outside the two or more ROIs. According to an embodiment, at least two different ROIs are insonified using TX and RX settings different from each other. In other words, different TX and RX settings are applied for the insonification of each ROI from a selection of two or more different ROIs with different grades or aspects of interest, e.g. very high level-medium level-low level. This makes it possible to distinguish between several different parts of the anatomical structure, each of which has different properties, e.g. different movement speeds, or has different importance for the planned medical analysis and therefore requires different measurement settings, e.g. different time resolutions, to image them properly.
[0045] The digitized RF data are mixed together into a sequence of images (video stream) showing the ROI region, preferably using a higher temporal resolution than the non-ROI region. According to an embodiment, this further comprises a step of temporally upsampling the RF data acquired from the imaging region outside the ROI to the temporal resolution of the RF data acquired from the ROI. In case of one or more ROI acquisition cycles interleaved with the overall acquisition cycle, first the images are beamformed to reconstruct the ROI image and the overall ultrasound image. Then the overall image is temporally upsampled to a frame rate the same as the ROI frame rate. In a next step, the temporally upsampled overall ultrasound image is mixed together with the ROI image to generate a sequence of overall ultrasound images at the frame rate of the ROI images. In one embodiment, a simple linear temporal upsampling / interpolation method is used, where missing data between two overall ultrasound images is obtained by duplicating data from the available overall ultrasound images or by applying linear interpolation from the neighbors. To mix the images together, a simple mask-based mixing method is applied. Preferably, the mixing is performed using floating point, i.e. using soft mask boundaries. For example, a linear or radial gradient is applied to the boundary to blend the images. In it, a mask corresponding to the ROI region is defined in the imaging domain, and the ROI image is copied into this mask in the upsampled global ultrasound image. According to this embodiment, blending is performed in the image domain in a Cartesian image plane, or in a polar plane, e.g., r-theta. When the global image and the ROI image are expressed in polar coordinates, blending is preferably performed after beamforming and before scan conversion when the global image and the ROI image are blended together.
[0046] According to an alternative embodiment, the mixing step is performed in the RF domain rather than in the beamformed image.
[0047] According to an embodiment, the upsampling interpolation is more complex in that it considers the motion between at least one ROI and a non-ROI region. This helps to compensate for object motion, e.g. motion of anatomical structures and / or parts of anatomical structures, and / or patient motion between multiple acquisition cycles. Such an algorithm is known as Philips natural motion, which is used e.g. in video streaming technology to make motion appear smoother. Alternatively, AI methods, e.g. based on neural networks, are used. For example, the methods described in https: / / heartbeat.fritz.ai / research-guide-for-video-frame-interpolation-with-deep-learning-519ab2eb3dda or in SuperSloMo: High Quality Estimation of Multiple Intermediate Frames for Video Interpolation, Huaizu Jiang, Dequing Sun, Varun Jampani, Ming-Hsuan Yang, Erik Learned-Miller, Jan Kautz, https: / / arxiv.org / abs / 1712.00080v2 are used.
[0048] When performing an ROI acquisition cycle, according to an embodiment, the transmit events insonifying the ROI must start with the same time offset as in the full acquisition cycle. In other words, if the entire imaging area requires multiple transmit events and if only a portion of the transmit events insonify the ROI in the ROI acquisition cycle, similar (spatially equivalent) transmit events, i.e. transmit events in the same direction, start with the same time offset in the ROI acquisition cycle than in the full acquisition cycle. This is beneficial because it makes the blending relatively simple, since the ROI and non-ROI insonifications are generated in a regular time-interleaved manner. Moreover, it prevents violent oscillations in the motion that could otherwise occur due to differences between the frame rate of the generated video sequence and the frame rate of the actual measurement, especially when the offset between the measured frames is irregularly varied. Thus, a constant time offset, i.e. a uniform interval between corresponding transmit events, provides smooth motion.
[0049] According to an embodiment, at least one ROI and the imaging area outside the ROI are insonified by overlapping transmit beams, with more overlapping transmit beams in at least one ROI compared to non-ROI areas. In other words, large overlapping transmit beams are used, such that for example only 3-8 transmit events are needed to insonify the entire imaging area. This results in much higher time resolution in the zone where all transmit beams overlap compared to the edges. By adjusting the transmit pattern to provide maximum overlap near the ROI zone, improved (time) sampling of the ROI is realized.
[0050] According to an embodiment, the at least one ROI is automatically detected and tracked by a computer vision or AI algorithm on the scouting ultrasound image, but in particular on the multiple ultrasound images acquired using the method itself, and the position of the at least one ROI is automatically adjusted accordingly.
[0051] The invention further relates to an ultrasound imaging system adapted to perform the method according to the invention. The ultrasound system is adapted for ultrafast imaging. However, the invention is not limited to use for ultrafast ultrasound systems, but may also be applied to average speed and / or low cost systems. The invention is generally suitable for improving the imaging quality and / or frame rate in a ROI in an imaging region. The system comprises an ultrasound probe including a transducer array and an ADC array and configured to insonify the imaging region, to receive echo signals, to digitize the received echo signals, and to transmit the digitized RF data to a DPU, the DPU being adapted to beamform and optionally process the RF data. The ultrasound probe of the invention is preferably a high-end probe, for example the probe includes a suitable integrated circuit for digitization. The invention makes it possible to realize ultrafast imaging with a low power consumption of up to 2-3 W. The transducer array is a 1D or 2D (matrix) array, for example a linear array or a phased array including 32-128 transducer elements (channels). The ultrasound probe is equipped for in-probe digitization, i.e. the transducer elements are connected to an array of ADCs. Optionally, an in-probe memory is available for temporarily storing digitized RF data from the ADCs. According to an embodiment, the array of ADCs is connected to an encoder, which is adapted to stream the RF data through a data interface, e.g. a serial data link and / or a limited bandwidth channel, to a decoder connected to a DPU, which is part of the ultrasound probe, but is preferably external to the ultrasound probe, and is part of the main control hardware of the ultrasound system. Such control hardware further comprises a graphical user interface (GUI) that allows a user to view multiple ultrasound images and possibly select an ROI if required.
[0052] The DPU is adapted to beamform the digitized RF data. A mix of ROI and non-ROI acquisition is performed before or after beamforming. The ADC, in-probe memory, and encoder are implemented by one or more integrated circuits. Furthermore, the decoder and the DPU are implemented by integrated circuits.
[0053] Useful embodiments of the present invention will now be described with reference to the accompanying drawings, in which similar elements or features are indicated by the same reference numerals in the figures, and in which the different embodiments shown are expressly permitted to be combined unless otherwise stated. [Brief description of the drawings]
[0054] [Figure 1] FIG. 1 is a schematic diagram of a total ultrasound (ultrasound) image including a marked region of interest. [Diagram 2] FIG. 2 is a schematic diagram of an angle-specified plane wave imaging mode including a region of interest according to an embodiment of the present invention. [Diagram 3] FIG. 2 is a schematic diagram of an alternative angle-specified plane wave imaging mode including a region of interest according to an embodiment of the present invention. [Figure 4] FIG. 2 illustrates an example of a receiving scheme according to an embodiment of the present invention. [Diagram 5] FIG. 2 is a flow diagram illustrating a method according to an embodiment of the present invention. [Figure 6] 1 is a schematic diagram of a system according to an embodiment of the present invention. [Figure 7] FIG. 4 is a further flow diagram of a method according to an embodiment of the present invention. [Figure 8] FIG. 4 is a further flow diagram of a method according to an embodiment of the present invention. [Figure 9] FIG. 3 illustrates the concept of interleaving ROI acquisition with overall acquisition according to an embodiment of the present invention based on data acquired using an imaging mode similar to that shown in FIG. 2. [Figure 10]FIG. 10 illustrates the concept of upsampling the entire ultrasound image to the same frame rate as the ROI frame rate based on data similar to that shown in FIG. 9. [Figure 11] FIG. 4 illustrates the concept of interleaving ROI acquisition with overall acquisition according to an embodiment of the present invention based on data acquired using the imaging modes shown in FIG. 3. [Figure 12] FIG. 2 illustrates another schematic representation of an ultrasound system in accordance with an embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0055] FIG. 1 shows a schematic representation of a total ultrasound (ultrasound) image 18a corresponding to an imaging region of an ultrasound probe. The ultrasound image 18a shows a four-chamber view of a human heart 30. Within this total ultrasound image 18a, a region of interest (ROI) 16 is marked. Some parts of the heart, such as the valves, move faster than other parts, such as the ventricle 27 or the atrial walls. To resolve the fast movement of the valves 28, it is necessary to apply a sufficiently fast imaging technique, for example ultrafast ultrasound imaging. On the other hand, other parts of the anatomical structure do not require such a fast imaging technique. Therefore, after selecting the ROI 16, which in this example includes the fast moving valves 28, the regions inside the ROI 16 are imaged with a higher frame rate than the regions outside the ROI 16. For the other parts of the organ 30, the corresponding transmissions of the ultrasound probe 10 are scheduled at larger intervals to match their (slower) movement speeds. As a result, there are slow and fast time acquisitions. Such an adaptive TX-RX scheme helps to reduce the radio frequency (RF) data transmission load compared to a fixed TX-RX scheme, while still allowing data of the fast moving valve 28 to be acquired using a sufficiently high frame rate.
[0056] 2 shows a schematic diagram of the angle-specified plane wave imaging mode according to the present invention, in which different transmit events, i.e., different transmit beams 22, numbered TX1, TX2, TX3, ..., TX7, are directed in different directions, each transmit beam subtending an azimuth angle of, for example, 5° to 15°. In this embodiment, the transmit beams 22 propagate such that successive transmit beams 22 are close to each other, and each sector of a given area corresponding to the entire imaging region 18 is covered by one transmit beam. In this example, only transmit beams TX4, TX5, and TX6 cover the ROI 16. Thus, only three of the seven transmit beams 22 are needed to image the ROI 16.
[0057] The transmit and receive parameters for the insonification of the ROI are different from those for the entire imaging region. For example, the transmit power and / or the transmit focusing scheme are adapted depending on the location of the ROI 16. Focused TX helps to perform harmonic imaging in the ROI 16. Furthermore, the transmit beams 22 covering the ROI 16 (TX4-TX6) are transmitted using a higher frequency than the transmit beams 22 for the rest of the imaging region 18.
[0058] FIG. 3 shows a schematic diagram of an alternative angle-specified plane wave TX scheme according to an embodiment of the present invention. In this embodiment, there are three partially overlapping transmit events (TX1, TX2, and TX3). This results in a much higher time resolution in the zone where all transmit beams 22 overlap compared to the edge, which is covered by only one transmit beam 22 (TX1 and TX3, respectively). Improved sampling of the ROI 16 is achieved by adjusting the transmit pattern to provide the most overlap near the ROI 16 zone. In this example, the central zone containing the ROI 16 is imaged three times more frequently, i.e. by all three transmit beams 22 (TX1-TX3), compared to the two zones at the edge.
[0059] 4 shows an example of a receive scheme according to an embodiment of the invention. The receive scheme can be adjusted to keep the sample only in a depth range corresponding to the ROI 16 of a given transmit beam 22. In this figure, the entire imaging area 18 is represented by a large cube, the horizontal axis is the depth axis (also called the fast time axis), the vertical axis Tx corresponds to transmit events, and the third axis Rx corresponds to receive elements. The ROI 16 is represented by a smaller cube within the larger cube. To reduce the RF data 14 transferred to the analysis system 40 during an ROI acquisition cycle, a particular depth range, in this case d start From d stop Only RF data 14 corresponding to a range up to tx is transferred. Furthermore, RF data 14 is transferred only when transmitting at a specific transmit angle that covers the ROI 16. In FIG. 4, this range of transmit angles is denoted by tx start From TX stop According to this scheme, the amount of RF data 14 can be reduced by:
number
[0060] FIG. 5 shows a diagram illustrating a method according to an embodiment of the present invention. As a first step, a scouting ultrasound image of a patient's anatomical structure, e.g., the heart, is acquired (15). To limit the data rate involved, the scouting ultrasound image 15 has a lower resolution or supports a lower frame rate than the images acquired later for the actual data analysis. As a next step, an ROI is selected (116) within the scouting ultrasound image 15. The ROI 16 in particular covers fast moving parts of the anatomical structure, such as the valves 28 of the heart. The selection of the ROI 16 is performed, for example, manually by a user or the ROI 16 is automatically detected by AI.
[0061] Next, a transmission scheme and a reception scheme are adapted in the ultrasound imaging system 114 based on the position of the ROI 16 relative to the ultrasound probe 10 of the system. The transmission scheme consists, for example, in insonifying the ROI using a higher time resolution and / or using a different type of transmission beam 22 compared to the imaging area 18 outside the ROI 16. On the other hand, the reception scheme is according to the embodiment shown in FIG. 4, i.e. only the received RF data 14 corresponding to the depth range of the ROI 16 is processed in a predefined time interval corresponding to the transmission scheme. It is also conceivable to apply a different bit depth in the ROI 16 than in the area outside the ROI 16, for example 8-bit precision, for example 12-bit precision, or to measure the Doppler shift within the ROI 16 but not outside the ROI 16. The next step is to acquire RF whole frame data (118) and acquire ROI frame data (120), according to which the ROI 16 is insonified using a higher time resolution than the imaging area 18 outside the ROI 16, for example. This is achieved by interleaving one or more ROI acquisition cycles, where only the ROI 16 is insonified between two whole acquisition cycles in which the entire imaging region 18 is insonified. Thus, for each acquired RF whole frame data set 118, at least one other data set including only the ROI frame is further acquired (120). According to the receiving scheme, in the ROI acquisition cycle, the ultrasound probe 10 is configured to transmit only the received RF signals corresponding to the depth range of the ROI 16 for the defined receive beam 24 reflected from the ROI 16, and in the whole acquisition cycle, the ultrasound probe 10 is configured to transmit the received RF signals corresponding to the entire imaging region 18.
[0062] The next steps are beamforming the whole frame data (119) and beamforming the ROI frame data (121). The ROI image 16a and the whole ultrasound image 18a are then blended together into a sequence of images to generate a time sequence of the imaging area, i.e., a video. Thus, a series of ultrasound images 26 is generated. Optionally, it is conceivable to use one or more of the blended images as further scouting ultrasound images 15 to adjust the ROI 16 to accommodate relative motion between the ultrasound probe 10 and the ROI 16. Thus, the steps of the method are repeated in a cyclical manner to generate multiple images, with the position of the ROI 16 being continually adjusted.
[0063] 6 shows a schematic diagram of a system according to an embodiment of the present invention. The system is configured to acquire a scouting ultrasound image 15 of an anatomical structure and to automatically or manually select a ROI 16 that includes a portion of the anatomical structure in the scouting ultrasound image 15. The system is further configured to set a transmit and receive scheme based on the acquired scouting ultrasound image 15 and the selected ROI 16. The system comprises a receive element 34 including a transducer array 8 adapted to emit a transmit beam 22 and to detect a receive beam 24 as shown in FIG. 6.
[0064] Furthermore, one analog-to-digital converter (ADC) or preferably an ADC array 32 is provided for each transducer element. The ADC is adapted to digitize the received echo signals and to transfer them as RF data 14 to the encoder 42. The encoder 42 is, for example, an application specific integrated circuit (ASIC) with an in-probe memory 50 adapted to temporarily store relevant data, for example RF data corresponding to a depth range of the ROI according to the receiving scheme described herein. The encoder 42 is configured to transfer selected RF data 14 via the interface 36 to the analysis system 40. According to the receiving scheme set by the system, the encoder 42 is adapted to transfer only the received RF data 14 corresponding to a depth range of the ROI 16 for at least some of the defined receiving beams 24 reflected from the ROI 16, in particular the receiving beams 24 corresponding to the transmitting beams 22 of the ROI acquisition cycle. The encoder 42 is further configured to transmit received RF data 14 corresponding to the entire depth of the imaging region 18 for at least some of the receive beams 24 reflected by the FOV outside the ROI 16, particularly those receive beams 24 corresponding to the transmit beams 22 of the entire acquisition cycle.
[0065] Since the interface 36 has a limited capacity for the transmission of data, the encoder 42 is preferably adapted to use the in-probe memory 50 as a buffer to control and equalize the data stream transferred through the interface 36. For example, if the ROI 16 is imaged using a higher frequency than areas outside the ROI 16, the amount of data transferred varies with time and is always larger when the entire imaging area 18 is imaged compared to when only the ROI 16 is imaged. To balance this, the data is buffered by clipping the data, i.e., when a large amount of RF data 14 is generated, storing some of the generated RF data 14 in the in-probe memory 50, and when a relatively small overall amount of RF data 14 is generated, transferring the stored data together with the relatively small amount of currently generated RF data 14. Thus, a uniform stream of data is transferred without any data peaks. Preferably, all relevant data can be transferred within one entire cycle, including both at least one ROI acquisition cycle and one entire acquisition cycle, in a relatively constant data stream, in particular such that the data rate does not rise above a predefined threshold. Beneficially, the encoder 42 further comprises means for compressing the RF data 14, for example wavelet compression, to achieve a further minimization of the transferred data.
[0066] The system further comprises an analysis system 40 including a decoder 44 configured to receive the transmitted RF data 14, possibly decompress the transmitted RF data 14, and provide the transmitted RF data 14 to a data processing unit 12 including a beamformer. By using the data processing unit 12, the analysis system 40 is adapted to process the RF data 14 and to generate an ultrasound image 26.
[0067] 7 and 8 show another schematic representation of a method according to an embodiment of the present invention. FIG. 7 shows the first stage in which an ROI 16 is selected and a corresponding TX and RX scheme is set, in which a TX beamforming pattern 38a is first transmitted to the transducer array 8. The transducer array 8 emits a transmit beam 22 and detects a corresponding receive beam 24, which is digitized and processed via the RX beamformer 38b. Then, a corresponding B-mode image 52 is used to select the ROI 16. Then, according to the selected ROI 16, a corresponding TX and RX scheme 46, 48 is selected (148) and redirected to the transducer array 8. As shown in FIG. 8, the TX scheme 46 is then used for TX beamforming 38a via the transducer array 8, and the incoming signal received by the transducer array 8 is digitized and processed via the RX beamformer 38b according to the set RX scheme 48. This RX scheme 48 produces one or more ROI images 16a and a full ultrasound image 18a of the imaging region 18. These images are then blended in a final step 122 to produce a series of ultrasound images 26.
[0068] FIG. 9 illustrates the concept of interleaving ROI acquisition with overall acquisition according to an embodiment of the present invention. The concept is to acquire images of the ROI 16 at a higher frequency, i.e., acquire more ROI images 16a per unit time, while still keeping the entire imaging area 18 imaged, albeit by using a lower frequency, i.e., by acquiring fewer overall ultrasound images 18a per unit time. In this embodiment, there are three ROI images 16a per overall ultrasound image 18a. The time axis 145 indicates the order in which the images are acquired. Thus, after each overall ultrasound image 18a, three ROI images 16a are acquired consecutively before the next overall ultrasound image 18a is acquired. Preferably, the time offset between corresponding TX and RX events in each of the acquired images is constant. This makes it easier to blend the images together and to achieve a smooth motion in the corresponding resulting video. In general, by interweaving one overall ultrasound image 18a with n-1 ROI images 16a, and by reducing the aforementioned reduction factor R frame Based on this, the overall reduction in data rate is calculated by the reduction factor R tot This is explained by:
number
[0069] Therefore, the total reduction in data rate in the example shown in FIG. tot =(3R frame +1) / 4. Depending on the size of the particular ROI, the total data reduction factor R can range from 2 to 5. tot can be realized as a result.
[0070] FIG. 10 is based on the data shown in FIG. 9, where the whole ultrasound image 18a is upsampled to the same frame rate as the ROI frame rate. The hatched portion of the bar corresponds to the measured RF data 14, whereas the open portion of the bar corresponds to the upsampled data. The upsampling of the whole ultrasound image 18a is realized by a simple linear temporal interpolation method. Thus, an interpolated whole image 18b based on the previous and subsequent acquired whole ultrasound images 18a is generated at a time step corresponding to the time step of the acquired ROI image 16a. Thus, since the ROI image 16a exists, the same number of images of the whole imaging area are generated, including the interpolated whole image 18b and the original measured whole ultrasound image 18a. The upsampled whole frame images are then mixed with the ROI image 16a to generate a whole video sequence of the imaging area 18. The resulting video is of higher quality in the ROI 16 compared to the areas outside the ROI 16, i.e. it is more accurate in time. During image interpolation, it is conceivable that motion between the ROI 16 and regions outside the ROI 16 is taken into account. This is supported by using neural networks. Furthermore, the blending can be performed in the image domain on a pixel basis as described above, i.e. by adding or averaging the intensities of the two images, or the blending can be performed in the RF domain before the actual image is generated, by using a Gaussian intensity profile, in particular a weighted Gaussian intensity profile. Blending in the image domain can be performed in polar coordinates, which are natural coordinates resulting from the geometry of the ultrasound image acquisition system, or the blending can be performed in Cartesian coordinates calculated from the polar coordinates during scan conversion.
[0071] The images shown in Fig. 9 and Fig. 10 are acquired, for example, by dispersive waves or by an angle-specified planar imaging mode similar to that shown in Fig. 2. When acquiring images by the embodiment shown in Fig. 3 where different transmit events overlap, the interleaving scheme shown in Fig. 11 is applied. Similarly, the hatched areas of the bars represent the actual measured RF data 14, while the empty parts of the bars are brought about by image interpolation. In the central area close to the time axis 154 corresponding to the ROI 16, the image data brought about by all three different transmit beams 22 (TX1, TX2, and TX3) overlap, whereas in the upper (TX3) and lower (TX1) parts of the figures corresponding to the areas outside the ROI 16, only one transmit beam (respectively TX1 and TX3) contributes to the actual image data. As in the embodiment shown in Fig. 9 and Fig. 10, the remaining empty spaces are filled by interpolation, so that the image is upsampled to show the whole area at each time step. Similarly, the interpolated overall images 18b are then mixed together to generate a video sequence in which the ROI 16 is more accurately represented in time.
[0072] FIG. 12 shows a schematic diagram of an ultrasound system 200 according to an embodiment of the present invention and configured to implement the method of the present invention. The ultrasound system 200 is configured to operate in particular as shown diagrammatically in FIG. 5. The ultrasound system 200 includes a conventional ultrasound hardware unit 202 with a CPU 204, a GPU 206, and a digital storage medium 208, such as a hard disk or a solid-state disk. A computer program is loaded into the hardware unit from a CD-ROM 210 or via the Internet 212. The hardware unit 202 is connected to a user interface 214 with a keyboard 216 and optionally a touchpad 218. The touchpad 218 also serves as a display device for displaying imaging parameters. The hardware unit 202 is connected to an ultrasound probe 10 including a transducer array 8 and optionally an in-probe memory 50 (not shown). The ultrasound probe 10 is configured to insonify an imaging region, to receive echo signals, to digitize the received echo signals, and to transmit the digitized RF data to a data processing unit 12, which is part of the CPU 204 and / or the GPU 206. The acquired ultrasound image 26 is displayed on a screen 226, which may be any commercially available display unit, such as a screen, a television set, a flat screen, a projector, etc.
[0073] The above description is intended to be merely illustrative of the system and should not be construed to limit the appended claims to any particular embodiment or group of embodiments. Thus, while the system has been described in particular detail with reference to exemplary embodiments, it should also be understood that many variations and alternative embodiments may be devised by those skilled in the art without departing from the broader and intended spirit and scope of the invention as set forth in the following claims. Accordingly, the specification and drawings are to be considered in an illustrative manner and are not intended to limit the scope of the appended claims. [Explanation of symbols]
[0074] 8 Transducer Array 10 Ultrasound probe 12 Data Processing Unit 14 RF Data 15 Scouting ultrasound images 16 ROI 16a ROI image 18 Imaging area 18a Whole body ultrasound image 18b Interpolated whole image 22 Transmit Beam 24 receiving beam 26 Ultrasound Image Series 27 Ventricle 28 Valves 30 Heart 32 ADC Array 34 Receiving elements 34a Receiving element axis 35a Sending Event Axis 36 Interface 38 Beam forming section 38a TX beam forming section 38b RX beam forming section 40 Analysis Systems 42 Encoder 44 Decoder 46 TX Scheme 48 RX Scheme 50 Probe memory 52 B-mode 114 Adapting the TX / RX scheme 116 Selecting an ROI 118 Get the entire RF frame 119 Whole Frame Beamforming Get 120 ROI frames 121 Beamforming ROI Frame 122 Video / Image Mix 148 Select TX and RX schemes 154 Timeline 200 Ultrasound System 202 Ultrasound Hardware Unit 204 CPU 206 GPU 208 Digital Storage Media 210 CD-ROM 212 Internet 214 User Interface 216 Keyboard 218 Touchpad 226 screens
Claims
1. 1. A method for ultrasonic ultrasound imaging using an ultrasound system including an ultrasound probe including a transducer array and an ADC array, the ultrasound probe insonifying an imaging region, receiving echo signals, digitizing the echo signals received by the ADC array, and transmitting the digitized RF data over a bandwidth limited channel to a data processing unit, the data processing unit beamforming the RF data, the method comprising: acquiring at least one scouting image of the imaging region, the scouting image including an anatomical structure of interest; selecting at least one region of interest within the scouting image, the at least one region of interest including a portion of the anatomical structure; configuring a transmit scheme and a receive scheme for ultrasound imaging of the imaging region, wherein the at least one region of interest is insonified using transmit and receive settings that are different from the imaging region outside the at least one region of interest; acquiring a plurality of ultrasound images according to the transmit scheme and the receive scheme; having the ultrasonic probe transmits to the data processing unit only the received RF data corresponding to a depth range of the at least one region of interest for at least some of the defined receive beams reflected by the at least one region of interest, and transmits the received RF data corresponding to the entire depth of the imaging region for at least some of the receive beams reflected by the imaging region outside the at least one region of interest. method.
2. the transmit and receive schemes comprising insonifying the at least one region of interest using a higher time resolution and / or using a different type of transmit beam and / or using a different pulse repetition rate compared to the imaging region outside the at least one region of interest. The method of claim 1.
3. wherein the transmit power and / or focusing scheme of a transmit beam insonifying the at least one region of interest is adapted to the at least one region of interest. The method according to claim 1 or claim 2.
4. the transmit and receive schemes include interleaving at least one region of interest acquisition cycle in which only the at least one region of interest is insonified between two full acquisition cycles in which an entire imaging region of the ultrasound probe is insonified; According to the receiving scheme, in the region of interest acquisition cycle, the ultrasound probe: transmit only the received RF data corresponding to a depth range of the at least one region of interest for at least some of the defined receive beams reflected by the at least one region of interest, and in the entire acquisition cycle, the ultrasound probe transmits the received RF data corresponding to a depth range of the entire imaging region of the ultrasound probe. The method according to any one of claims 1 to 3.
5. The ultrasound probe includes an in-probe memory, and the method further comprises: selecting digitized RF data corresponding to the depth range of the at least one region of interest for at least some of the defined receive beams reflected by the at least one region of interest for transmission to the data processing unit; buffering at least some of the selected digitized RF data in an in-probe memory to delay transmission of the RF data from the ultrasound probe when a predetermined maximum data rate is exceeded; The method according to any one of claims 1 to 4, further comprising:
6. before the digitized RF data is transmitted to the data processing unit, the digitized RF data is compressed via a discrete wavelet transform; 6. The method according to any one of claims 1 to 5.
7. the digitized RF data reflected by the at least one region of interest is compressed using a different bit depth and / or using a different sub-sampling pattern and / or using a different wavelet compression compared to the digitized RF data reflected by the imaging area outside the at least one region of interest. The method according to claim 6.
8. the transmit and receive schemes use a wide transmit beam imaging mode including a plurality of transmit events to insonify the imaging region, only a portion of the plurality of transmit events insonifying the at least one region of interest, and the transmit events insonifying the at least one region of interest are performed with a higher temporal resolution.
8. The method according to any one of claims 1 to 7.
9. the transmit scheme uses a plurality of transmit events to insonify the imaging region, a portion of the plurality of transmit events insonifying the at least one region of interest, and spatially equivalent transmit events begin at the same time offset in the at least one region of interest acquisition cycle and in the overall acquisition cycle.
10. The method of claim 4 or any one of claims 5 to 8 which directly or indirectly derives from claim 4.
10. two or more regions of interest are selected within the scouting image, each of the two or more regions of interest including a portion of the anatomical structure; wherein the two or more regions of interest are insonified using transmit and receive settings that differ from the imaging region outside the two or more regions of interest; at least two different regions of interest are insonified using mutually different transmit and receive settings; 10. The method according to any one of claims 1 to 9.
11. reconstructing a region of interest image from the region of interest acquisition cycles and a full ultrasound image from the full acquisition cycle; upsampling the entire ultrasound image in time to achieve a frame rate for the entire imaging field of view that is the same as the frame rate for the region of interest image; blending the upsampled overall ultrasound image together with the region of interest image; The method of any one of claims 3 to 10, further comprising:
12. The method of claim 11, wherein the upsampling is performed using interpolation, the interpolation taking into account motion between the at least one region of interest and the remaining imaging region by using a trained neural network. The method of claim 11.
13. in the transmit scheme, the at least one region of interest and the imaging region outside the at least one region of interest are insonified by overlapping transmit beams, with more of the transmit beams overlapping in the at least one region of interest than in the imaging region outside the at least one region of interest; 13. The method according to any one of claims 1 to 12.
14. the at least one region of interest is automatically detected and tracked in the at least one scouting image and / or in the plurality of ultrasound images by computer vision and / or artificial intelligence algorithms.
14. The method according to any one of claims 1 to 13.
15. 15. An ultrasound imaging system for implementing the method of any one of claims 1 to 14, comprising an ultrasound probe including a transducer array and an ADC array, the ultrasound probe insonifying an imaging region, receiving echo signals, digitizing the received echo signals, and transmitting the digitized RF data to a data processing unit via an interface comprising a bandwidth limited channel, the data processing unit processing the RF data for beamforming. Ultrasound imaging system.
Citation Information
Patent Citations
Doppler ultrasonograph and its method
JP1998033535A
Ultrasonic diagnostic equipment and ultrasonic image generation method
JP2013090827A
Ultrasonic diagnostic equipment and control program for ultrasonic diagnostic device
JP2013141545A
Ultrasonic measuring apparatus, ultrasonic image apparatus and ultrasonic measuring method
JP2015071028A
Ultrasonic diagnostic apparatus
JP2018102771A
