Method and system for system parameter self-tuning in ultrasound imaging

Abstract

The methods and systems process ultrasound images based on user parameters determined by user selected images having different parameters. In one aspect, a method includes generating at least a first set of test images comprising at least one pair of ultrasound images having first parameter values different from each other, and transmitting the first set of test images to a user interface accessible by a user. The method may also include receiving one or more user selected images comprising at least one ultrasound image. The method may also include feeding the one or more user selected images into one or more data analysis methods to generate one or more ultrasound processing parameters comprising at least one of 1) the first parameter having second parameter values, or 2) one or more second parameters. The method may also include receive and process one or more ultrasound signals based on the one or more ultrasound processing parameters for display.

Description

METHOD AND SYSTEM FOR SYSTEM PARAMETER SELF-TUNING IN ULTRASOUND IMAGINGCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S Provisional Application Serial No. 63 / 624,694, filed January 24, 2024, entitled “METHOD AND SYSTEM FOR SYSTEM PARAMETER SELF-TUNING IN ULTRASOUND IMAGING,” which is incorporated by reference herein in its entirety.BACKGROUND

[0002] The present disclosure relates to medical image processing technology7. Specifically, the present disclosure relates to parameter tunning for ultrasound imaging.SUMMARY

[0003] Various embodiments of systems, methods, and devices within the scope of the appended claims each have several aspects, no single one of which is solely responsible for the desirable attributes described herein. Without limiting the scope of the appended claims, some prominent features are described herein.

[0004] The present disclosure describes methods for self-tuning ultrasound system parameters (‘parameters”) of an ultrasound system based on a plurality of individual user preferences. The parameters can include, but are not limited to. one or more of system front-end acquisition control parameters, algorithm type parameters, parameters for image formation processing, and / or user control parameters that are accessible via a user interface. Some of these parameters may be linked to programmable preset parameters of the ultrasound system. The present disclosure describes a parameter tuning module that may be run on ultrasound systems, or remotely to the ultrasound systems. A parameter tuning module may acquire user feedback on a series of image comparisons (e.g., an iterative selection of an image A vs an image B) that are generated using different values of selected ultrasound system parameters. The parameter tuning module may determine optimized parameter values based on the user preferences and / or another “goodness” or “best” criteria. Embodiments of the parameter tuning module may be intuitive for a user to use, and each run where one or more parameter are determined can be completed in a short time (e.g., 1-5 minutes) such that that the parameter tuning model will not be burdensome forusers with limited technical ultrasound knowledge. The parameter tuning module can be applied to a specific imaging mode (e.g., B-mode, Doppler, elastography, and / or the like) or clinical protocol. The resulting parameter values can be linked to a user ID or to the clinical site. In some implementations for B-mode, the parameter tuning module may include self-tuning of auto gain and dynamic range parameters, and image smoothing and edge enhancement parameters.

[0005] In some embodiments, the systems, methods, and devices described herein relate to an ultrasound imaging system. In some embodiments, such ultrasound imaging systems may include a probe configured to generate one or more ultrasound signals, one or more displays configured to display ultrasound images, a memory storing instructions corresponding to one or more data analysis methods associated with ultrasound image processing, and one or more ultrasound image processor in communication with the probe and the one or more displays. In some embodiments, the ultrasound image processor includes one or more computer hardware processors, at least one of the one or more computer hardware processors configured to generate at least a first set of test images. The first set of test images may include at least one pair of ultrasound images having first parameter values that are different from each other. The ultrasound imaging system can be configured to transmit the first set of test images to a user interface accessible by a user, receive, via the user interface, one or more user selected images, the one or more user selected images including at least one ultrasound image of the first set of test images, feed the one or more user selected images into the one or more data analysis methods to generate one or more ultrasound processing parameters, the one or more ultrasound processing parameters including at least one of 1) the first parameter having second parameter values different from the first parameter values, or 2) one or more second parameters different from the first parameter; receive the one or more ultrasound signals from the probe, process the one or more ultrasound signals based at least in part on the one or more ultrasound processing parameters to generate one or more processed ultrasound images, and display the one or more processed ultrasound images via one or more displays.

[0006] In some embodiments, the at least one of the one or more computer hardware processors is further configured to generate a second set of test images including at least one additional ultrasound image having a first parameter value of the first parameter different than the first set of test images, receive, from the user interface, one or more additional user selected images, the one or more additional user selected images including at least one ultrasound image of the second set of test images, and feed the one or moreadditional user selected images into the one or more data analysis methods to generate the one or more ultrasound processing parameters. In some aspects, the techniques described herein relate to an ultrasound imaging system, wherein the second set of test images are configured to be generated based on the one or more user selected images and the one or more data analysis methods. In some embodiments, the first set of test images includes a second pair of ultrasound images having second parameter values of a second parameter that are different than each other. In some embodiments, the one or more user selected images include at least one of the second pair of ultrasound images. In some embodiments, to process the one or more ultrasound signals, at least one of the one or more computer hardware processors is configured to perform B-mode processing. In some embodiments, to process the one or more ultrasound signals, at least one of the one or more computer hardware processors is configured to perform color flow processing. In some embodiments, to process the one or more ultrasound signals, at least one of the one or more computer hardware processors is configured to perform spectral doppler processing. In some embodiments, at least one pair of ultrasound images include images of a same physiological feature.

[0007] In some embodiments, the systems, methods, and devices described herein relate to a method of processing ultrasound images. In some embodiments, the method includes generating, by one or more processors, at least a first set of test images, the first set of test images including at least one pair of ultrasound images having first parameter values of a first parameter that are different from each other, transmitting, by at least one of the one or more processors, the first set of test images to a user interface accessible by a user, receiving, from the user interface, one or more user selected images, the one or more user selected images including at least one ultrasound image of the first set of test images, feeding the one or more user selected images into one or more data analysis methods to generate one or more ultrasound processing parameters, the one or more ultrasound processing parameters including at least one of 1) the first parameter having second parameter values different from the first parameter values, or 2) one or more second parameters different from the first parameter; receiving one or more ultrasound signals from a probe, processing the one or more ultrasound signals based on the one or more ultrasound processing parameters to generate one or more processed ultrasound images, and displaying the one or more processed ultrasound images via one or more displays.

[0008] In some embodiments, the method, further includes generating a second set of test images including at least one additional ultrasound image having a first parametervalue of the first parameter different than the first set of test images, receiving, from the user interface, one or more additional user selected images, the one or more additional user selected images including at least one ultrasound image of the second set of test images, and feeding the one or more additional user selected images into the one or more data analysis methods to generate the one or more ultrasound processing parameters. In some embodiments, the second set of test images are generated based on the one or more user selected images and the one or more data analysis methods. In some embodiments, the first set of test images includes a second pair of ultrasound images having second parameter values of a second parameter that are different than each other. In some embodiments, the one or more user selected images include at least one of the second pair of ultrasound images. In some embodiments, processing the one or more ultrasound signals includes performing B-mode processing. In some embodiments, processing the one or more ultrasound signals includes performing color flow processing. In some embodiments, processing the one or more ultrasound signals includes performing spectral doppler processing. In some embodiments, at least one pair of ultrasound images include images of a same physiological feature.

[0009] In some embodiments, the systems, methods, and devices described herein relate to a non- transitory computer readable recording medium storing instructions, when executed by one or more processors, cause the processors to perform the methods described herein.

[0010] In some embodiments, the systems, methods, and devices described herein relate to a user terminal including one or more processors configured to perform the methods described herein.BRIEF DESCRIPTION OF DRAWINGS

[0011] These and other features, aspects, and advantages of the present application are described with reference to drawings of certain embodiments, which are intended to illustrate, but not limit, the present disclosure. It is to be understood that the attached drawings are for the purpose of illustrating concepts disclosed in the present application and may not be to scale.

[0012] Figure 1A illustrates a block diagram of an example medical imaging system.

[0013] Figure IB illustrates a block diagram of an example ultrasound imaging system.

[0014] Figure 2 illustrates a block diagram of an exemplary' processing environment for processing of ultrasound signals into display data by an ultrasound image processing system.

[0015] Figure 3A illustrates an example implementation of a testing application.

[0016] Figure 3B and 3C illustrate examples of image pairs that can be utilized by a testing application.

[0017] Figure 4A is an example table illustrating example system parameters that may be adjusted by a parameter tuning module, according to various embodiments.

[0018] Figures 4B-4F are example tables illustrating non-limiting example values for parameters such as the parameters provided in the table of Figure 4A.

[0019] Figure 5 illustrates an example of B-mode processing, according to various embodiments.

[0020] Figure 6 illustrates an example amplitude histogram plot of a tissue regions for a typical intra-cardiac image, according to various embodiments.

[0021] Figure 7 illustrates an example flow diagram of a process for tuning system parameters based on user preferences and displaying ultrasound images based on the tuned system parameters.

[0022] Figure 8 illustrates an example flow diagram of a process for tuning a single parameters based on user preferences.

[0023] Figure 9 illustrates an example flow diagram of a process for tuning a two or more system parameters simultaneously based on user preferences.DETAILED DESCRIPTION

[0024] An ultrasound imaging system can include many system parameters that can be optimized (or “tuned”) to produce high image quality for different clinical applications, anatomical views, and / or body habitus. A user interface (“UI”) of an ultrasound imaging system may provide control over some basic parameters such as gain and velocity scale. However, such parameters represent only a small subset of all the adjustable parameters in an ultrasound imaging system. For example, front-end transmit / receive acquisition controls, image formation algorithms, and backend video display processing may all have numerous adjustable parameters.

[0025] Over the past decades, advancements have been made to design more intuitive UI for ultrasound imaging systems, and to provide some user customizationcapabilities via programmable presets. However, programable presets may be disadvantageous in several ways. For example, programable presets generally control only a relatively small subset of the system parameters. As another example, programable presets generally pertain more to high-level and fairly standard system parameters, such as transmit frequency, display gain, dynamic range and gray map. To generate the best quality ultrasound image, a lower level and wider range of system parameters need to be optimized. As such, programmable presets lack the adaptability and range of parameter adjustment to generate high quality ultrasound images for a range of clinical applications.

[0026] Beyond presets, many commercial systems have implemented various automated image optimization algorithms especially for image display presentation including gain and dynamic range settings. These data-adaptive algorithms can often be triggered by, for example, a simple button push. One example is “Auto tissue optimization and continuous tissue optimization, GE LOGIQ series technology, 2021” (included as an Appendix herein). However, the performance consistency of such methods still generally depends on how well the underlying algorithm parameters are tuned to the specific clinical applications and / or user preferences. From a mathematical optimization perspective, algorithms often optimize parameter values based on a goodness metric. In ultrasound image optimization, how to align the goodness metric with the user’s preferences can be a continued challenge.

[0027] In current ultrasound imaging system development processes, ultrasound manufacturers typically perform system parameter optimization by conducting systematic imaging and image analysis on tissue-mimicking phantoms, normal human subjects, and clinical studies. For some applications, such as intracardiac echo, pre-clinical studies, animal models are used instead of human subjects.

[0028] Image quality can be subject to a user’s personal preferences for different clinical applications, settings, and sub-specialties. In interventional cardiology, for example, subspecialties can include electrophysiology, where ultrasound imaging is used to guide the placement of ablation catheters; and structural heart procedures, where ultrasound imaging provides guidance for the repairs of heart defects. The parameter tuning in each of the preceding example subspecialties may differ. As a result, parameter tuning can be a never-ending process. For instance, clinical feedback from one software release may be addressed by another round of parameter tuning for the next software release, and so on.

[0029] The current system parameter tuning process faces a few challenges. For example, because manufacturers can only perform image optimization scanning on a limited number of normal human subjects (or pre-clinical studies) and clinical studies, for each product or software release, the parameter values chosen based on those studies often need further adjustments after the product is deployed widely in clinical settings. This is especially true for intracardiac devices where the pre-clinical studies were mainly performed on animal models. Such animal models may mimic the human anatomy to a large extent but may result in parameter tunings that are not optimized to humans undergoing treatment procedures. As another example, many system or algorithm-level imaging parameters are generally not exposed to the users. As such, any desired adjustments to the parameter values may require a new product software release cycle.

[0030] As such, there is a need for improved system parameter tuning based on an individual user’s (e.g., a doctor’s or a technician's) or clinical site’s preferences for specific ty pes of images and patient cases. Further, there is a need for an easier and faster adaptive tuning process that does not require an image parameter tuning engineer from the manufacturer to be present in the clinical site. This is especially important for an interventional ultrasound system which is generally deployed in clinical settings that have limited floor space, and where ionizing radiation is used during the clinical procedures.

[0031] Generally, conventional medical imaging systems may include a medical imaging device, an imaging processing system, and a user interface / display. In an ultrasound system, the medical image device may comprise an ultrasound imaging device that includes various modules and sensors for transmitting and receiving ultrasound signals. For example, an ultrasound imaging device can include frond-end transmi t / receive (“T / R”), a transducer array, a transmit modules, receivers, filters, analog-to-digital converters (“ADCs”), and / or the like. In some medical imaging systems, when the front-end T / R switches are set to a transmit mode, the transmit module, which can include transmit delay signals and electrical pulsers, excite the array of transducers to emit acoustic energy' into the region of interest. The returning echoes can be received by setting the T / R switches to a receive mode. In some instances, such as in an intracardiac device, for example, a wire bundle that connects the transducer elements to the front-end electronics may be inserted inside a catheter. In some medical imaging systems, the data processing path for received signals begins with the radio frequency receivers, filters, and ADCs for different elements of the transducer array. The digitized data is then transferred to an image processing system, such as an ultrasound image processing system.

[0032] An ultrasound image processing system may be implemented in software on a central processing unit-graphics processing unit (CPU-GPU) module. The ultrasound image processing system may include image formation processing components, such as beamformers, filters, demodulators, and / or the like, color flow processing components, B-mode processing components, spectral doppler processing components, scan conversion, video processing components, and / or the like. An ultrasound image processing system may perform image formation processing using beamformers, filters, and demodulators, to output two digital in-phase (“I”) and quadrature (“Q”) data streams that cany' the information content of received radio-frequency (“RF”) echo signals. These data streams are derived from the echoes along the paths of the transmit beams. The filters may be programmed with coefficients to pass a band of frequencies around at the center frequency of the transmit waveform. The beamformed I and Q data may be fed into different image processors or different image processing modules of the same image processor, such as a color flow processor or module, a B-mode processor or module, and a spectral Doppler processor or module, depending on an activated imaging mode. In some instances, such as for 2D scan modes, the resultant image data from the different image processors may be scan converted from an acoustic line (range-angle) to an X-Y pixel grid for image enhancement and display processing. In some instances, such as in a spectral Doppler processing mode, the resultant image data from the spectral Doppler processing may undergo video processing to produce a Doppler frequency (or corresponding velocity) versus time graphic. The different imaging modes of data can be combined into a composite layout in a Ul / display, such as a physical display unit such as a on system tablet, or an external display screen.

[0033] In an ultrasound system the UI may provide a way for a user to interact with the ultrasound system. Specifically, the UI may enable the user to adjust a set of standard imaging controls such as gain, dynamic range (DR), gray map, image enhancement filters for B-mode processing and Color flow processing, and the velocity scale for Color flow processing and Spectral Doppler processing. The UI may utilize a display touch screen and / or soft-key controls for receiving user selections or adjustments. The user selections or adjustments may be input to a master compute controller implemented on the CPU of the ultrasound system. The compute controller may coordinate all the digital data processing and real time data flow within the software-based image processor.

[0034] In some instances, the ultrasound operations for each imaging processing mode may include many ultrasound system parameters (also referred to herein as “system parameters”). Such system parameters may include, but are not limited to, one or more of front-end transmit / receive acquisition control parameters software algorithmlevel parameters, and user- adjustable parameters that are linked to various labels in the UI such as velocity scale, gain, wall settings, etc. Any combination of these system parameters may be part of programmable presets for the system. These system parameters may be optimized to produce a desired image quality for different clinical applications, anatomical views and / or body habitus. As previously described, in certain ultrasound systems, the system parameters are normally optimized by the manufacturer, and some or all of the system parameters may not be directly accessible by the user. This may be true even though the system parameters are linked to UI controls and / or programmable presets. For example, when the user selects a “high frequency setting” on the UI, a large number of system parameter, such as transmit center frequency parameters, receive time-gain-curve parameters, digital processing parameters, and image post-processing parameters, may be varied in concert with one another. As such, these ultrasound systems do not provide the option to precisely define many of the system parameters, as they are often specific to the ultrasound system architecture, imaging controls, and image processing algorithms of the ultrasound platform.

[0035] For purposes of illustration, an example of B-mode processing is shown in Figure 5. As illustrated in Figure 5, B-mode processing may include envelope detection (denoted by “EnvQ”), gain processing, and logarithmic data compression for each beamsummed line (denoted by “I / Q”). A B-mode processor or a B-mode processing module may perform the above operations. Envelope detection is equivalent to taking the magnitude of the beam-summed I and Q data samples (i.e., taking the sqrt(I2+Q2)). Due to the nature of tissue scattering, the signal that corresponds to one B-line will generally show random fluctuations as illustrated in Figure 5. The signal dynamic range (dB), defined by the smallest to the largest echo envelope value, is typically around 70 dB. However, the larger reflectors can be closer to 100 dB higher than the noise floor. To visualize any subtle changes at the soft tissue level, it may be necessary to display the envelope data on a dB scale via a logarithmic data compression. The addition of a gain adjustment may include a system component, a user component, or a combination of both, to adjust the amplitude of the envelope data such that its maximum value is close to a preset value (maximum numericvalue). In some instances, the amplitude of the envelope data may eventually map to a highest display brightness level.

[0036] In some instances, to control the lowest level of signal to be displayed, a dynamic range (“DR”) setting may be used to limit the floor of logarithmic data compression. The DR setting, which can be expressed in dB, may be controllable by ultrasound system UI controls. For B-mode data, a maximum numeric value of 96 dB is generally sufficient, in which case the minimum output or floor of the log output is 96 dB minus the DR. Conceptually, the output floor should be set close to the noise floor of the system, such that only useful signal content is compressed and displayed. From a user perspective, the gain and DR control may determine respectively the brightness and contrast of the final B-mode image.

[0037] Described herein is a medical imaging system that can obtain userpreferred values or estimates for one or more system parameters, rather than using manufacturer pre-tuned parameter values or programable presets for the system parameters. As such, the medical imaging system according to various embodiments addresses the challenges of the cunent system parameter tuning process, by enabling one or more system parameters to be optimized based on the user’s (or clinical site’s) preferences without the need for an expert in image optimization.

[0038] The medical imaging system can include a parameter tuning module that can run, or interact with, a testing application (e.g., a software program) that interacts with a user through, for example, the UI and display screen. The testing application may include an intuitive, short-duration, test session to determine user preferences for one or more system parameters. Based on the user responses or inputs, the testing application may further perform the system parameter optimization to meet the user preferences. Depending on the number and type of system parameters, the testing application may use different strategies or sequencing schemes between the user response query steps and the analysis steps.

[0039] In various non-limiting implementations, the testing application may not require in-depth technical knowledge or skills in regard to ultrasound imaging. The testing application may present a user with two or more versions (or renditions) of the same anatomical view generated using different system parameter values, and prompt the user to select a preferred version. In some implementations, the two or more versions (or renditions) may be generated images by the testing application, communicated to the testing application, stored by the testing application, and / or originate from another source. Invarious non-limiting implementations, the testing application may receive a user selection through a user input, such as a touchscreen input, a soft button input, a voice command input, and / or other user input techniques. In some examples, a user may select between an image “A” vs image “B” using a finger touch. “Images”, as described herein may include still images, short video loops, or a combination of both, that are pre-recorded and stored on a device as part of the testing application. In various non-limiting implementations, the testing application may be run in a short duration. For example, the testing application may take only about 15 mins or about 30 mins to run through a series of anatomical image examples.

[0040] In some embodiments, system parameters can be optimized at the end of a run of the testing application by performing analysis on each of the user responses. For example, in some embodiments the testing application, or other module, may average the system parameter values of each response. As another example, the user responses may be input into a parameter tuning module. The parameter tuning module may utilize one or more algorithms or combinations of algorithms to output one or more system parameters based on the user responses.

[0041] In some embodiments, parameter optimization may be performed concurrently as a run of the testing application. For instance, intermediate values can be used to generate additional test images. In such cases, a user may be prompted to confirm the additional test images or select one or more additional images. As such, the testing application may generate additional tests in an iterative manner. In various implementations, multiple approaches can be used to generate the sequence of test images in a run of the testing application.

[0042] In some implementations, a parameter tuning module, testing application or other system of an image processing system may optimize one system parameter at a time. For example, the testing application may be repeated using different anatomical images to check for consistency. In these examples, the testing application may check for consistency in user selection and generate additional examples to be presented to the user when user selection is inconsistent.

[0043] In a nonlimiting example, the testing application may optimize one system parameter at a time. In this example, at a step 1) the testing application generates two or more renditions of the same anatomical image by varying the value of the system parameter. All other system parameters can be set to a default value. At a step 2). after receiving one or more selections of the renditions, the testing application may perform finetuning by generating renditions of the anatomical image with a narrower range of values for the system parameter and repeat the test using the next anatomical image in a series of N anatomical image examples. At a step 3) the testing application checks for consistency in user selection and performs further fine tuning as needed. In this example, the testing application may repeat the optimization steps for each system parameter. In some instances, at step 1) the testing application may set system parameters that have already been optimized to an optimized value rather than a default value.

[0044] In some implementations, a parameter tuning module, testing application or other system of an image processing system may optimize two (or more) parameters together as a set. For some image processing algorithms, two or more interdependent system parameters are adjusted together rather than independently. For instance, the visual perception of the effect of changing one system parameter may be affected by the value of one or more of the other system parameter(s). An example of such interdependent system parameters may include system parameters associated with an image enhancement algorithm in B-mode image processing that performs tissue speckle smoothing, edge, and border sharpening.

[0045] By way of explanation, the system parameters to be optimized can be denoted as XI, X2, X3, and so on, and the analysis of the user responses may become a multi-parameter optimization problem. One approach is to treat the parameters as a vector quantity X=[X1. X2, X3,... ], and to take the average of all the outcome vectors of the testing application that correspond to user responses to a series of N image examples. This approach may be effective when the system parameters to be optimized are limited, or when the image quality metrics are very subjective. In this approach the goal can be characterized as selecting approximate system parameter values that meet the user’s preference, or selecting one processing algorithm over another algorithm.

[0046] If there are three or more system parameters to be optimized together as a set, or if greater precision is required, a more formal approach, such as multi-objective optimization or another suitable optimization, may be desirable. In these approaches, the solution process may be iterative with the decision maker continuously interacting with the testing application while searching for a preferred solution. In other words, the decision maker may need to input preferences at several iterations in order to obtain one or more optimal solutions that are of interest to the decision maker and to determine which solutions are attainable.

[0047] In a nonlimiting example, the testing application may optimize two or more parameters at a time. In this example, at step 1) the testing application generates two or more renditions of the same anatomical image by using different combinations of system parameters XI, X2, and so on within reasonable bounds. At step 2) after receiving one or more selections of the renditions, step 1) is repeated using the next image in a series of N anatomical image examples. At step 3) after three or more test cases, the testing application, or a parameter tuning module, may perform statistical or other analysis to determine optimal values of XI, X2, and so on. In some embodiments, the analysis may be performed using one or more algorithms or combination of algorithms. In this example, steps 1-3 may be repeated to confirm the optimal values. In some embodiments, fine tuning may be performed using narrower ranges of values of XI, X2. and so on for the next set of three or more anatomical images in the series of N anatomical images.

[0048] The approaches to system parameter tuning described above are provided as mere examples, and the present disclosure is not limited thereto. Many variations and hybrid approaches may be utilized without departing from the present disclosure. Once the user-preferred system parameter values are determined and stored in the system software, such as in a lookup table (‘"LUT”), they can be applied to process image data to achieve the image appearances preferred by the user.

[0049] In some embodiments, the testing application is run as part of the new system initialization, which will prepare the system for clinical use with the user-preferred settings. For clinical sites that include more than one user for the new device, the testing application may be run for every' new user ID.

[0050] In another embodiment, the testing application is used to optimize based on the clinical procedure rather than user ID. Using structural heart procedures as an example, some image processing parameter values may be more optimal for guiding the repair of atrial septal defects than for other procedures.

[0051] This disclosure may be applied to any system parameter of an ultrasound system or other medical imaging systems. In some instances, the transmit system parameter optimization process may be constrained by the physics of ultrasound imaging. For example, parameter optimization may be constrained by the trade-off between resolution and penetration. Parameter optimization may be used to optimize algorithm-level and other system parameters that affect the more subjective measures of image quality and are more dependent on user preferences, rather than the physics of ultrasound imaging.

[0052] Referring again to Figure 5, for B-mode anatomical data, a typical signal DR is 70 dB. If the maximum numerical amplitude is chosen to correspond to. for example, 96 dB, the log output floor may be, for example, 26 dB. Often a preset value of this DR in a UI is determined during the image parameter tuning process that represents the average or typical DR for a given clinical application. However, in clinical practice, the optimal DR varies from patient to patient, and from region to region within the same anatomical structure of a patient. This means that the DR itself is likely more dependent on the anatomical view than the user’s preference.

[0053] An objective method of setting the DR (for the log compression function) may be to determine the actual range of amplitude values that represent the majority of the tissue region in a given image. As an example, an amplitude histogram of the tissue regions for a typical intra-cardiac image is plotted in the Figure 6. The horizontal axis represents the envelope amplitude expressed in dB, with a maximum value of 96 dB. A suitable DR setting should encompass the majority of the tissue region. The majority can be defined by a minimum and a maximum bound of the histogram based on predefined minimum and maximum percentage threshold values. For example, a reasonable minimum threshold can be about 5%. Samples below this minimum threshold may be considered noise-dominant and can be excluded. On the high end of the histogram, a reasonable maximum threshold may be about 95%. The optimal DR, which is illustrated as 65 dB in the histogram example of Figure 6, is given by the difference between the two bounds in dB.

[0054] In some embodiments, the minimum and maximum threshold values illustrated in the histogram example of Figure 6 are tunable system parameters. Together they provide a means of determining an objective DR setting in an auto optimize algorithm. Based on the B-mode gain and the selected DR setting, the image may be compressed via a log function and the floor of the output is set by the DR. The resultant image is usually presented on an 8-bit display with 256 gray levels. The B-mode envelope values above the maximum histogram threshold can saturate the display dynamic range and ultimately appear bright white on the image display. For this reason, the maximum threshold for the histogram analysis can be alternately expressed as a system parameter:Degree of Saturation (DoS) = 100% - maximum threshold for the amplitude histogram

[0055] In interventional cardiology, where a primary purpose is to visualize a catheter and other treatment devices, some users may prefer a more “binary'’ look (tissueis bright or even saturated) by increasing the DoS parameter value from, for example, 5% to 15%, while other users prefer to see a more traditional ultrasound image with up to 255 shades of gray for an 8-bit display. It should be emphasized that the user’s preference pertains to the degree of saturation, and not the absolute amplitude histogram spread, which varies with the anatomic content of the raw image prior to log compression. In some embodiments, the test application may produce an optimal DoS, noise threshold, and a B- mode gain parameter. These three parameters can serve as complementary functions for auto image optimization. An example of a test application producing an optimal DoS is described in more detail with respect to Figure 3B.

[0056] In some instances, B-mode XY-pixel image enhancement and display processing can include an image enhancement algorithm that involves multiple tunable parameters to suit user preferences. Over the past few decades many adaptive image enhancement algorithms have been proposed that can smooth out the characteristic granular speckle texture while enhancing borders of B-mode images. As an example, one such algorithm, used for speckle reduction fdtering. uses two parameters (degree of smoothness, and number of iterations).

[0057] State-of-the-art commercial scanners may employ sophisticated algorithms for speckle reduction fdtering. These fdters may be adaptive to local tissue features and structures, and can provide two or more types of image enhancement. Each type may involve two or more fdter parameters to be optimized by the manufacturer. For example, one parameter may be used to control the level of low-pass fdtering or image smoothing, another to control the level of edge enhancement. The latter function can potentially be further divided between fine edges and more prominent borders with corresponding tuning parameter(s). In this disclosure, an app can present the user with images obtained using different combinations of values for two or more algorithm parameters. Because the perceived edge enhancement effects generally depend on the background tissue smoothness, these parameters may require optimization together as a set rather than independent optimization.

[0058] Figure 1A illustrates a block diagram of an example medical imaging system 100. The medical imaging system 100 may include a medical imaging device 104, and image processing system 102, and a display 106. The medical imaging device 104 may include sensors, analog processing components, digital processing components, and / or the like to transmit, receive, and process medical imaging signals. The image processing system 102 may include various hardware and software components for further processingthe medical imaging signals. For example, the image processing system 102 may include one or more processing modules that process the medical imaging signals based on one or more system parameters. The display 106 may include various hardware and software components for displaying processed medical images. For example, the display 106 may include a monitor, screen, or the like that can display medical images to a user. The display 106 may also include a UI that can receive user input.

[0059] In some implementations, the medical imaging system 100 may include one or more user devices 120 connected to the image processing system 102 via a network 110. The user devices 120 may be used to provide user input into the image processing system 102. The user devices 120 can include smart devices, cellular phones, personal computers, application specific computing devices, or any other suitable devices for providing user input to the image processing system 102. In some implementations, the user input is received via the UI on the display 106 rather than the user devices 120. In some implementations, the user devices 120 may include software, such a testing application, designed to interact with one or more system parameters and / or with one or more programs on the image processing system 102. The network 110 can include any type of communication network. For example, the network 110 can include one or more of a wide area network (WAN), a local area network (LAN), a cellular network, an ad hoc network, a satellite network, a wired network, a wireless network, and so forth. In some embodiments, the netw ork 110 can include the Internet.

[0060] Figure 1 B illustrates a block diagram of an example ultrasound imaging system 180. In some embodiments, ultrasound imaging system 180 is an example implementation of the medical imaging system 100. The ultrasound imaging system 180 can include an ultrasound imaging device 130. an ultrasound image processing system 140, and a display 106.

[0061] The ultrasound imaging device 130 can transmit ultrasound signals from a source signal and received reflected ultrasound signals. In some embodiments, the ultrasound imaging device 130 is an example implementation of a medical imaging device 104. The ultrasound imaging device 130 can include a transducer array 131, T / R switches 132, a transmit module 133, and an analog processing unit 134. The transducer array 131 can include an ultrasound sensor, such as a piezoelectric cry stal or other similar device. The transducer array 131 can generate ultrasound waves based on a transmission signal and generate a reception signal from reflected ultrasound waves. The T / R switches 132 can control when the ultrasound imaging device 130 is in a transmission mode or a receptionmode. The transmit module 133 can include components used to generate an ultrasound transmission signal to be sent to the transducer array 131, such as transmit delay signals, beamformers, electrical pulsers, and the like. The analog processing unit 134 can include receivers, filters, ADCs, and the like. The analog processing unit 134 can take received input data and convert it into an analog transmission signal to be used by the transmit module 133. The analog processing unit 134 can also receive reception signals from the transducer array 131 and convert it to digital ultrasound signals to be used by the ultrasound image processing system 140.

[0062] When the T / R switches 132 are set to a transmit mode, the transducer array 131 generates ultrasound signals based on a transmission signal from the transmit module 133. When the T / R switches 132 are set as a receive mode, the transducer array 131 sends a reception signal from reflected ultrasound waves to the analog processing unit 134 to be converted into a digital ultrasound signal. The ultrasound imaging device 130 may include multiple data lines to cany7signals to and from individual elements of the transducer array 131. For example, if the transducer array 131 is an 8x1 array, the ultrasound imaging device 130 may include eight discrete data paths, each configured to carry signals to and from an element of the 8x1 array. Other configurations of the transducer array 131 are possible that may require a different number of data lines.

[0063] The ultrasound image processing system 140 can process received digital ultrasound signals to produce display data of ultrasound images that can be displayed on the display 106. In some embodiments, the ultrasound image processing system 140 is a specific implementation of an image processing system 102. The ultrasound image processing system 140 can include various hardware and software components to process the received digital ultrasound signals. For example, the ultrasound image processing system 140 can include various CPU / GPU resources and software programs. In the illustrated example, the ultrasound image processing system 140 includes an image processing module 150 and a master compute controller 160.

[0064] The image processing module 150 can include software components for processing digital ultrasound signals, such as a beamformer, filter, and demodulator module 151, a color flow processing module 152, a B-mode processing module 153, a spectral doppler processing module 154, a scan conversion module 155, and a video processing module 156. Each software component of the image processing module 150 may have one or more system parameters used to process the digital ultrasound images. At least one of the modules 151-156 can be implemented as a separate processor. At least one of themodules 151-156 can be implemented as software, hardware, firmware, or a combination thereof.

[0065] The beamformer, filter, and demodulator module 151 may receive the digital ultrasound signal and convert it into one or more digital in-phase (“I”) data streams and one or more quadrature (“Q"’) data streams that earn- information associated with received RF signals (e g., a digitized reflected signal from the ultrasound imaging device 130). The filter of the beamformer, filter, and demodulator module 151 may be programmed w ith coefficients to pass a band of frequencies around at the center frequency of a transmitted waveform. The I and Q data streams may be fed into the color flow processing module 152, the B-mode processing module 153, and / or the spectral doppler processing module 154. The color flow processing module 152 may perform color flow processing based on one or more system parameters. The B-mode processing module 153 may perform B-mode processing based on one or more system parameters. The spectral doppler processing module 154 may perform spectral doppler processing based on one or more system parameters.

[0066] The scan conversion module 155 may receive acoustic-line (e.g., rightangle) information from the color flow' processing module 152 and / or the B-mode processing module 153 and convert it to X-Y pixel grid information. The video processing module 156 may perform image enhancement and display processing on the X-Y pixel grid information. The spectral doppler processing module 154 may output information to the video processing module 156 which, in-tum, can produce a Doppler frequency (or corresponding velocity) verse time graphic. In some embodiments, the data from the different modules may be combined at the video processing module 156 into a composite layout to be transmitted to the display 106.

[0067] The master compute controller 160 may be implemented on a CPU or a processor of the ultrasound image processing system 140 and may coordinate the digital data processing performed by the image processing module 150. The master compute controller 160 may include a parameter tuning module 162 that may control the values of the one or more system parameters utilized by the image processing module 150. The module 162 may include software and hardware resources for determining optimized system parameter values. For example, the parameter tuning module 162 may include one or more algorithms combinations of algorithms. In some embodiments, all or a portion of the parameter tuning module 162 may be performed outside of the master compute controller 160. For example, the parameter tuning module 162 may be performed on adevice such as the display 106 or the user device 120 of Figure 1A and communicated to the master compute controller 160. In some embodiments, the parameter tuning module 162 may optimize one or more system parameters based on input received from an interface such as the display 106 or the user device 120 of Figure 1A and / or from another user interface, such as user interface user interface 173. Examples of uses of the parameter tuning module 162 are described in more detail with respect to Figure 2.

[0068] The display 106 can receive one or more processed ultrasound images from the image processing module 150 and display them to a user. It should be understood the processed ultrasound images received from the image processing module 150 may include still images, video images, or a combination of both. The display 106 can include a display screen 171. an audio output 172, and a user interface 173. The display screen 171 may be any suitable display for displaying digital images to a user. For example, the display screen 171 may be a tablet, a computer monitor, a television, a projector, or any other display. The audio output 172 may be any speaker and may output audio signals received from the image processing module 150, such as audio signals received from the spectral doppler processing module 154. The user interface 173 may receive input from a user of the ultrasound imaging system 180. The user interface 173 can include, for example, a touch screen, soft key inputs, mouse and keyboard inputs, voice activated inputs, or any other computer input method. In some embodiments, the user interface 173 is integrated with the display 106. In other embodiments, all or a portion of the user interface 173 may be received from an external device and communicated to the display 106 and / or ultrasound image processing system 140. For example, all or a portion of the user interface 173 may be implemented on a user device 120 as depicted in Figure 1A.

[0069] Figure 2 illustrates a block diagram 200 of an exemplary processing environment for the processing of ultrasound signals 206 into display data 208 by an ultrasound image processing system 140. In the illustrated example, ultrasound signals 206, which can include the digital ultrasound signals discussed in Figure IB, are processed by the image processing module 150 into display data 208. The display day 208 may be one or more ultrasound images that can be displayed to a user (e.g., by a display 106). As discussed above, the image processing module 150 may process the ultrasound signals 206 by running the ultrasound signals 206 through various software components that utilize one or more ultrasound processing parameters (also referred to herein as “system parameters’'). The value of these system parameters may be set based on ultrasound processing parameter information 204 received from the master compute controller 160.

[0070] The ultrasound processing parameter information 204 may be generated by a parameter tuning module 162 in communication with a testing application on a user interface 173 of a display (or on a user device 120). As described above, a testing application 210 may display a series of images of anatomical images to a user with one or more system parameters varied. The testing application 210 may receive a user selection of some of the anatomical images. For example, a series of image pairs may be displayed to a user and the testing application 210 may receive a series of user selections of one image from each pair. However, other arrangements of image displays and selections may be used. For example, more than two images may be displayed to a user at once and a user may select more than one image at a time.

[0071] The testing application 210 may compile the series of user selections into user selected images 202 and output the user selected images 202 to the parameter tuning module 162. The parameter tuning module 162 may perform analysis on the user selected images 202 to produce the ultrasound processing parameter information 204. As previously described, the parameter tuning module 162 may include one or more algorithms or combinations of algorithms to produce the ultrasound processing parameter information 204. In one example, the parameter tuning module 1 2 may average parameter values in the user selected images 202 to produce the ultrasound processing parameter information 204. In another example, the parameter tuning module 162 includes one or more other data analysis methods or algorithms configured to produce one or more system parameters based on input ultrasound images. The ultrasound processing parameter information 204 may be saved and tied to a user identification, a clinical site identification, a clinical procedure identification, or other configuration. The parameter tuning module 162 may update the system parameters with the stored ultrasound processing parameter information 204. As such, the testing application 210 need not be run every time to alter the system parameters. The ultrasound processing parameter information 204 may be updated by a user. For example, by running the testing application 210.

[0072] As described above, the testing application 210 and / or the parameter tuning module 162 may be run as part when anew ultrasound image processing system 140 initialization, which can prepare the ultrasound image processing system 140 for clinical use with the user-preferred settings.

[0073] In some embodiments, the ultrasound imaging system 180 is an edge device that is connected to a cloud service. For the testing application 210 , image examples and parameter optimization algorithms may be updated periodically. In anotherembodiment, the testing application 210 may be run periodically, such as monthly or once every half a year. As the ultrasound imaging system 180 operates in clinical applications, over time, raw clinical data can be stored on the ultrasound imaging system 180, or uploaded to a local or remote cloud. A user may periodically re-run the app to further optimize the system parameters based on the increasingly larger and richer data sets.

[0074] In another embodiment, the ultrasound imaging system 180 may be connected to a data archiving system of a clinical site, the ultrasound imaging system 180 can utilize prior image data from the image archive that may hold images obtained by ultrasound scanners from other manufacturers. In some instances, an option can be provided to the user to allow the new device to fetch image examples from the image archive. In addition to the preset information of the prior images, additional image analysis of their prior images may be done to infer information about how the algorithm parameters can be adjusted to best match user or clinical site preferences.

[0075] Figure 3A illustrates an example implementation of the testing application 210. In the illustrated example, the same anatomical image is presented to the user twice with at least one system parameter varied. For example, parameter XI may have a first value in image A and parameter XI may have a second value in image B. As previously described, image A and image B may have more than one system parameter varied. For example, image A and image B may have multiple interdependent system parameters varies between them. A user may be prompted to select either image A or image B. After selecting image A or image B the user may be shown a new' set of images. In some implementations, the new' set of images may be based at least in part on the user selection of either image A or image B. For example, the new set of images may show' have the same system parameters with less variance.

[0076] In Figure 3B a short sequence of A vs B comparison images are shown to illustrate how Strategy 1 can be used to optimize the DoS parameter as the first system parameter, while assuming the default values for the two other parameters. In this simplified but exemplary’ test, three pairs of images are presented to the user sequentially from the top to the bottom. In the first pair, anatomical view 1 is rendered using a typical DoS value 5% (Left) and an aggressive value of 30% (Right). The hypothetical user’s pick is the image on the right (R) side, which indicates to the app that this user prefers the saturated look. In the next frame, keeping the same anatomical example, a new rendition using a DoS value of 20% is created and compared to the value of 30% that was chosen by the user in frame 1. The user chooses “L” which indicates 20% is closer to the preferredchoice. In the third frame, a new anatomical example is used, and the two renderings correspond to DoS of 20% vs a new value of 25%. The user still prefers 20% even though the anatomy has changed. This exemplary test concludes and the optimal saturation threshold is 20% for this particular user.

[0077] It is important to keep in mind that the absolute histogram spread between the two anatomical examples is different, such that if the histogram minimum noise and DoS values are kept the same at typical factory values of 5%, an auto optimize algorithm would have recommended DR settings that correspond to the respective histogram spreads of the two anatomical examples. This would have missed the user’s preference which is to allow more saturation and a more binary look. By running the app in this invention, even just going through the two anatomical examples, it is shown that the user consistently selects a more saturated look obtained using a higher DoS value, and after three iterations, the optimal value converges to 20%.

[0078] It should be noted that Figure 3B illustrates a simplified and shortened image test sequence for illustration purposes. In practice, a more intermediate values for fine tuning the DoS. and additional anatomical examples may be used to achieve more reliable optimization. Further, the new' anatomical examples may be interspersed with the back-and-forth fine tuning steps in different orders. Furthermore, the degree of saturation percentages shown in Figure 3B are merely examples and other percentages can be used.

[0079] In Figure 3C. a short sequence of A (images on the left) vs B (images on the right) is shown to illustrate how multiple, interrelated system parameters may be tuned at during a single run of the testing application 210. In the illustrated example, the left side of the top two rows are generated by using the default values for an edge enhancement and a smoothing filter parameter. The right side shows the effect of changing the level of smoothing. After the second comparison test, a user may select the right side of the middle row, which represents a preferred balance of edge enhancement and smoothing. As discussed above, the level of smoothing and edge enhance filter illustrated in Figure 3C may not be tied to a single a system parameter or a single algorithm. Rather each may refer to the type of algorithm (e.g. filter type 1 vs filter type 2). This may be particularly useful for image enhancement as there are a large number of filter classes (e.g., linear, nonlinear, adaptive, iterative, and so on) and specific algorithms in each class that can produce very different characteristic texture patterns to suit different user preferences. In the bottom row of Figure 3C, the left side represents another combination of system parameter values using the same filters as for the middle row; but using an anatomicalexample 4. The right side of the bottom row, however, represents an entirely different algorithm that utilizes a different border enhancement fdter which produces a different tissue texture appearance. In this illustrated example, the user may select the new algorithm on the right side.

[0080] As is illustrated in the example of Figure 3C, it should be clear that many different combinations of anatomical views generated by two or more filter parameter values, and / or by different image enhancement filter types or algonthms, can be sequenced in different ways in the testing application to better meet user preferences for image enhancement.

[0081] The examples illustrated in Figures 3B and 3C both pertain to system parameters used in B-mode processing. It should be noted that similar processes can be used for other system parameter types, such as system parameters used in Color flow processing or Spectral Doppler imaging processing. While Figures 3B and 3C show ultrasound images, the present disclosure can be used to other medical imaging technology, for example, MRI or CT. Moreover, while Figures 3B and 3C show an image of a particular internal organ of a person, the present disclosure can be used to process a medical image of other internal organs.

[0082] The testing application 210 is not limited to use of a touch screen or button clicks for a user to select their preferred images, such as the example implementation of the testing application 210 illustrated in Figure 3 A. In some embodiments, the testing application 210 supports voice commands for a user to verbally express user preferences. In another embodiment, the testing application 210 may utilize speech-to-text technologies that are rapidly advancing on portable devices such as a tablet. The use of speech-to-text may include several advantages, such as capturing additional information in written text regarding subjective factors for why a user may prefer one image over another. For example, using the example illustrated in Figure 3C, a user may use voice inputs to indicate that they prefer the right image over the left image. The user may also speak a few additional words such as “I like the right image more because I like its paint-brush-like tissue texture,” and these spoken words can be captured and translated into written text by the testing application 210 and / or the parameter tuning module 162. This additional textual information may provide additional data in the system parameter optimization process. In some implementations, a chat-bot may be enabled to provide conversational style voice feedback to the user to acknowledge and / or to clarify their inputs.

[0083] In some embodiments, the testing application 210 and / or the parameter tuning module 162 may implement a Large Language Model (“LLM”). The LLM model can be pre-trained to interpret requests such as '‘I like a paint-brush-like texture’’ into algorithmic decisions, such as “Apply image enhancement algorithm type XYZ.” Such a LLM can be implemented on the user device 120, the ultrasound imaging system 180, or on a remote cloud service depending on the compute resource requirements.

[0084] Figure 4A is an example table 400 illustrating example system parameters that may be adjusted by the parameter tuning module 162, according to various embodiments. Table 400 is provided merely as an example, and the present disclosure is not limited thereto. For example, one or more other parameters in addition to or instead of the parameters of Table 400 can be used. The parameters listed in table 400 may or may not be used and / or adjusted by the parameter tuning module 162, or any other system described herein. In some instances, the system parameters listed in a data field of table 400 may include multiple system parameters, algorithms, and / or the like. For example, as previously described, “degree of smoothness,’’ as listed as a system parameter of B-mode processing may include multiple inter-dependent system parameters and / or types of algorithms. Further, other system parameters, not shown in table 400 may be used and / or adjusted by the parameter tuning module 162 without departing from this disclosure.

[0085] Additionally, while table 400 includes system parameters categorized in “Color Flow Processing,” “B-mode Processing.” Spectral Doppler Processing.” Scan Conversion,” and “Video Processing,” these categorizations are provided for illustration only. As such, more categories may be used and / or system parameters listed under one category may be used in another category. Further, system parameters may be interdependent with another system parameter in the same category and / or a different category or may be independent.

[0086] In the context of the present disclosure, ultrasound system parameters can be divided into two broad categories. The first category consists of parameters that are usually adjusted by the user based on objective reasons or needs. For example, in color flow processing, the user may adjust the “velocity scale” that sets the maximum velocity limit which is proportional to the pulse repetition frequency. The velocity scale may be set higher if the user (or system) wishes to measure higher blood flow velocities without ambiguities. The second category consists of parameters that are chosen by the user based at least partly on subjective or personal preferences.

[0087] The present disclosure generally describes methods for self-tuning the second category of system parameters, which can be standard controls available on a user interface, as well as lower level engineering parameters that may be non-standard and often proprietary to the manufacturer’s machine.

[0088] Figures 4B-4F provide example values for the parameters provided in table 400. The example parameter values provided in Figures 4B-4F are provided by way of example only. The values in Figures 4B-4F are not intended to limit the scope of the present disclosure, but only to provide non-limiting examples of types of values that may used in conjunction with the parameters of table 400 in Figure 4 A.

[0089] Referring now to tables 410 and 420 in Figures 4B and 4C, in 2D or 3D spatial imaging including color flow and B-mode, it is well known to those skilled in the art that the choice of transmit frequency can represent a tradeoff between spatial resolution and penetration. For example, higher transmit frequencies may enhance the lateral resolution but at the expense of penetration due to the frequency-dependent attenuation effects of biological tissue. Once the transmit frequency is set appropriately, (e.g., 6 MHz for a given transducer and clinical application), any additional fine tuning, e.g. 5.5-6.5 MHz, becomes a user preference. In the specific example of ICE, typical transmit frequency ranges are as shown in the respective tables 410, 420.

[0090] For color flow (table 410) and B-mode (table 420) processing, one specific non-limiting example of the present technology described in detail (Figures 5-6) includes auto dynamic range (DR) and auto gain. As non-limiting examples, the DR may vary from 30 to 100 dB, and gain may vary7from 0-30 dB in a ty pical B-mode processor. The additional parameters such as the targeted degree of saturation (0-100%) or target average brightness level can be linked to the manufacturer’s non-standard and proprietary algorithm. Another specific non-limiting example of the present technology' already described in detail (Figure 3) is image enhancement filter algorithms which often refer to the manufacturers’ non-standard and proprietary recipes that may include, for example, weighting factors (0 to 100%) between a edge-enhanced image component, and a smoothed or speckle-reduced image component.

[0091] For spectral Doppler processing (table 430), a power spectrum is generally computed via a Fast Fourier Transform (FFT) but the exact FFT analysis parameters may vary. As an example, the FFT analysis parameter is the FFT window' length which can vary from 64 to 256. While there are objective rationales for choosing a larger or smaller FFT size depending on the pulse repetition frequency, the exact size around areasonable value represents a tradeoff between temporal vs velocity resolution, which can be a user preference. On most ultrasound machines, the design rules are typically fixed by the manufacturer, but the present technology would enable the user to customize according to their personal preference within reasonable bounds. Another parameter is the spectral display time scale, which refers to the length of the horizontal axis in seconds, for the spectral waveform display. Ultrasound machines often provide a choice of 3 settings, short, medium and long, such as 2. 4 and 8 seconds respectively. But some users may prefer to have the ability to select from, for example, 1, 5 and 10 seconds. The example of image enhancement parameters for spectral Doppler is similar to that for B-mode and color flow, although the image enhancement filtering may be simply a smoothing filter that reduces the Doppler spectrum speckle texture. That is. the user-selectable filter parameter may represent the size of the smoothing filter.

[0092] For scan conversion (table 440), there generally are relatively few tuning parameters, as its function is to convert image data sampled in polar coordinates into Cartesian or XY coordinates for a pixel display. Still, the XY image width and height in terms of pixel count at the output of scan conversion can be subjective choices. The present technology would enable the user to indicate their personal preferences of a slightly larger or smaller total XY image display size.

[0093] Similarly, for the video processing block in the ultrasound signal processing chain (table 450), the duplex or triplex mode image layout parameters may be selected by the user using the present technology. Tn duplex mode (e.g., both B-mode and Spectral Doppler mode are displayed), the relative sizes of the background B-mode image and the spectral Doppler waveform display are subject to user preferences. In triplex mode (e.g., B-mode, color flow, and spectral Doppler mode are all displayed), the relative sizes of the color flow (on the background B-mode anatomical image) and the spectral Doppler display, are also subject to user preferences. Additionally, the gray map or color map choices can affect the perceived contrast of the B-mode and / or color flow image display, and they are subject to user perceptions and preferences that could be inferred from the self-tuning app in accordance with the present technology.

[0094] Figure 7 illustrates an example flow diagram of process 700 for tuning system parameters based on user preferences and displaying ultrasound images based on the tuned system parameters. Process 700 may be performed by the ultrasound imaging system 180, such as by the parameter tuning module 162. a user device 120, or other system. For convenience, process 700 will be described from the perspective of ultrasound imagingsystem 180. Process 700 may be implemented in a different order than illustrated without departing from this disclosure. Further, additional processes may be added to process 700 or processes shown in process 700 may be removed without departing from this disclosure.

[0095] At block 702, the ultrasound imaging system 180 generates one or more test images, such as the images illustrated in Figures 3B and 3C. In some implementations, the ultrasound generating system 180 may generate the one or more test images by selecting one or more images pre-stored images on the ultrasound imaging system 180 and / or the user device 120, using one or more generative artificial intelligence, and / or otherwise generate the one or more test images. In some implementations, the one or more test images are generated based on a setting of the ultrasound imaging system 180, such as the type of imaging procedure (e.g., intracardiac echocardiography (“ICE”)), a user and / or facility identification, a previous iteration of process 700, and / or the like. In some implementations, the one or more test images are generated by varying the system parameters within a defined range of values. As will be described in more detail with respect to Figures 8 and 9, in various implementations, the one or more test images may each have one parameter varied or have multiple parameters varied.

[0096] At block 704, the test images are transmitted to a user interface, such as a user interface on a user device 120. For example, the user images can be transmitted to a user device and displayed in a manner as illustrated in Figure 3A. In some embodiments, the test images are sent to the user device two images at a time. The test images may be a series of anatomical images with one or more system parameter varied, such as the images described in Figures 3B and 3C.

[0097] At block 706, the ultrasound imaging system 180 receives user selected images. The user selected images may be transmitted to the ultrasound imaging system 180 from a user device 120. The user selected images may represent user selections from each pair of images generated at block 702. In some embodiments, blocks 702, 704, and 706 may be implemented iteratively. For example, at block 702 a first pair of images may be generated, at block 704, the first pair of images may be sent to the user interface, and at block 706. a selection of one image from the first pair may be received. In some embodiments, a subsequent pair of images is generated at block 702 based on the selected image at block 706.

[0098] At block 708, the ultrasound imaging system 180 feeds the user selected images into one or more data analysis methods to generate ultrasound processing parameters, such as the system parameters described herein. The data analysis methodsmay be deterministic or statistical. For example, for system parameters with discrete values or choices (e.g., a set of selectable color maps for a display of color flow images), the data analysis may determine the color map option most frequently chosen by the user over a number of image pair comparisons. For system parameters with a continuous range of values, a statistical data analysis may be performed to determine an optimal value that may best fit the user preference based on the selected images. For example, the ultrasound imaging system 180 can perform statistical estimation (e.g., compute averages), perform regression analysis (e.g., a best line fit), perform multi-objective optimization, and / or other statistical method based on the user selected parameter values. For example, as previously described, statistical analysis may be performed on one or more system parameters in the selected images. In general, statistical data analysis may invoke machine learning or deep learning methods, although in some cases a large number of image examples may be required and could make the total run time impractical. In some embodiments, other information may be used by the parameter tuning module 162 to determine the ultrasound processing parameters. For example, a speech interpretation technology such as LLM may be used to analyze speech data received from the user interface.

[0099] In some embodiments, block 708 may be implemented iteratively with blocks 702, 704, and 706. For example, subsequent images generated at block 702 may be based on output of an iterative data analysis algorithm, and / or may be generated by other processing by the ultrasound imaging system 180 of user selected images.

[0100] At block 710, the ultrasound imaging system 180 receives one or more ultrasound signals from a probe. For example, the ultrasound imaging system 180 may receive digital ultrasound signals from an ultrasound imaging device 130 as described in Figure IB. At block 712, the ultrasound imaging system 180 processes the ultrasound signals based on the ultrasound processing parameters to generate one or more ultrasound images. For example, the ultrasound processing parameters may be used by the beamformer, filter, and demodulator module 151, the color flow processing module 152, the B-mode processing module 153, the spectral doppler processing module 154, the scan conversion module 155, the video processing module 156, and / or by another module that processes ultrasound images. At block 714, the ultrasound images are displayed. For example, the ultrasound images may be transmitted to a display 106 to be displayed to a user.

[0101] Figure 8 illustrates an example flow diagram of process 800 for tuning a single parameter based on user preferences. Process 800 may be performed by theultrasound imaging system 180, such as by the parameter tuning module 162, a user device 120, such as by the testing application 210, or other system. For convenience, process 800 will be described from the perspective of testing application 210. Process 800 may be implemented in a different order than illustrated without departing from this disclosure. Further, additional processes may be added to process 800 or processes shown in process 800 may be removed without departing from this disclosure.

[0102] At block 802, the testing application 210 generates images with the system parameter varied. For example, the testing application 210 may generate tw o images with degree of saturation varied, as illustrated in the first row7of Figure 3B. In some implementations, the testing application 210 may generate the images by selecting one or more images pre-stored images on the ultrasound imaging system 180 and / or the user device 120, using one or more generative artificial intelligence, and / or otherwise generate the images. In some implementations, the images are generated based on a setting of the ultrasound imaging system 180, such as the type of imaging procedure (e.g., intracardiac echocardiography (“ICE”)), a user and / or facility identification, and / or the like. In some implementations, the images are generated by varying the system parameters within a defined range of values. The testing application testing application 210 may display the images on a user interface of a user device, such as user device 120. At block 804, the testing application 210 received a selection of an image. The user selection may be associated with an image the user prefers. For example, the testing application 210, may receive a selection of the right image of the first row of Figure 3B, signifying the user prefers a 30% degree of saturation over a 5% degree of saturation.

[0103] At block 806, the testing application 210 may generate additional images with a narrower range of parameter values. The additional images can be based on the selection received at 804. For example, after receiving the selection at block 804, the testing application 210 may generate the second row7of images illustrated in Figure 3B with a narrower range of degree of saturation. At block 808, the testing application 210 receives a selection of an image generated at block 806 in a similar manner as described in block 804. For example, the testing application 210 may receive a selection of the left image of the second row7of Figure 3B, signifying the user prefers, for example, a 20% degree of saturation over, for example, a 30% degree of saturation.

[0104] At decision block 810, the testing application 210 determines whether more tuning is needed. The determination at block 810 may be based on a number of iterations. For example, the testing application 210 may require a minimum of threeiterations of images. The determination may be based on a determination that parameter tuning is incomplete, for example, the testing application 210 may determine that two or more user selections of images produced an inconsistent result. In response to the determination at block 810 being affirmative, the testing application 210 proceeds to block 806 and generates additional images. For example, if the testing application 210 determines two or more user selections of images produced an inconsistent result, the testing application 210 may generate images to resolve the inconsistency (e.g., generating the same parameter values on a different anatomical example). In response to the determination at block 810 being negative, the testing application 210 proceeds to block 812.

[0105] At block 812. the testing application 210 transmits tuning data for the parameter. F or example, the testing application 210 may transmit the value of the parameter in the final selected images, an average of the parameter value across a portion of the selected images, an average of the parameter value across the final generated images, or transmit other information from gathered during process 800. In some embodiments, all the selected images from process 800 are transmitted to a processing module to determine an ideal value of the parameter. Process 800 may be performed iteratively and / or concurrently to tune one or more additional system parameters.

[0106] Figure 9 illustrates an example flow diagram of process 900 for tuning a two or more system parameters simultaneously based on user preferences. Process 900 may be performed by the ultrasound imaging system 180. such as by the parameter tuning module 162, a user device 120, such as by the testing application 210, or other system. For convenience, process 900 will be described from the perspective of testing application 210. Process 900 may be implemented in a different order than illustrated without departing from this disclosure. Further, additional processes may be added to process 900 or processes shown in process 900 may be removed without departing from this disclosure.

[0107] At block 902, the testing application 210 generates a first set of images with at least a first parameter and second parameter varied. In some implementations, the testing application 210 may generate the first set of images by selecting one or more images pre-stored images on the ultrasound imaging system 180 and / or the user device 120, using one or more generative artificial intelligence, and / or otherwise generate the images. In some implementations, the images are generated based on a setting of the ultrasound imaging system 180. such as the type of imaging procedure (e.g., intracardiac echocardiography ("ICE")), a user and / or facility identification, and / or the like. In some implementations, the first set of images are generated by varying the first and secondsystem parameters within a defined range of values. The first and second parameters may be inter-dependent. For instance, the first and second parameters may relate to level of smoothing, as illustrated in Figure 3C. As previously described, level of smoothing may involve multiple inter-dependent parameters. For example, the testing application 210 may display the top row of images illustrated in Figure 3C with a default image on the left and a high level of smoothing on the right (indicating the first and second parameter are set to values other than default.

[0108] At block 904, the testing application 210 received a selection of an image. The user selection may be associated with an image the user prefers. For example, the testing application 210, may receive a selection of the left image of the first row of Figure 3C, signifying the user prefers the default level of smoothing.

[0109] At block 906, the testing application 210 generates additional sets of images with the first and second parameters varied. The additional sets of images may be based on the selection received at block 904. For example, the testing application 210 may generate the second row of images illustrated in Figure 3C with the default values on the left and a mid-level of smoothing on the right. In some instances, the sets of images generated at block 906 may vary additional parameters. For example, the testing application 210 may generated the third row of images illustrated in Figure 3C with the level of smoothing and edge enhance filters varied.

[0110] At block 908, the testing application 210 receives a selection of an image generated at block 906 in a similar manner as described in block 904. For example, the testing application 210 may receive a selection of the right image of the second row of Figure 3C, signifying the user prefers a mid level of smoothing over the default values.[OHl] At decision block 910, the testing application 210 determines whether more tuning is needed. The determination at block 910 may be based on a number of iterations. For example, the testing application 210 may require a minimum of three iterations of images, to properly tune the first and second parameters. The determination may be based on a determination that parameter tuning is incomplete, for example, the testing application 210 may determine that two or more user selections of images produced an inconsistent result. In response to the determination at block 910 being affirmative, the testing application 210 proceeds to block 906 and generates additional images. For example, if the testing application 210 determines two or more user selections of images produced an inconsistent result, the testing application 210 may generate images to resolve the inconsistency (e.g., generating the same parameter values on a different anatomicalexample). In response to the determination at block 910 being negative, the testing application 210 proceeds to block 912.

[0112] At block 912, the testing application 210 transmits tuning data for the parameter. For example, the testing application 210 may transmit the value of the first and second parameters in the final selected images, an average of the first and second parameters values across a portion of the selected images, an average of the first and second parameters values across the final generated images, or transmit other information from gathered during process 900. In some embodiments, all the selected images from process 900 are transmitted to a processing module to determine an ideal value of the first and second parameters. Process 900 may be performed iteratively and / or concurrently to tune one or more additional system parameters.

[0113] The foregoing description details certain embodiments of the systems, devices, and methods disclosed herein. It will be appreciated, however, that no matter how detailed the foregoing appears in text, the systems, devices, and methods can be practiced in many ways. As is also stated above, it should be noted that the use of particular terminology when describing certain features or aspects of the invention should not be taken to imply that the terminology is being re-defined herein to be restricted to including any specific characteristics of the features or aspects of the technology' with which that terminology is associated.

[0114] It will be appreciated by those skilled in the art that various modifications and changes may be made without departing from the scope of the described technology'. Such modifications and changes are intended to fall within the scope of the embodiments. It will also be appreciated by those of skill in the art that parts included in one embodiment are interchangeable with other embodiments; one or more parts from a depicted embodiment can be included with other depicted embodiments in any combination. For example, any of the various components described herein and / or depicted in the figures may be combined, interchanged or excluded from other embodiments.

[0115] As is also stated above, it should be noted that the use of particular terminology when describing certain features or aspects should not be taken to imply that the terminology is being re-defined herein to be restricted to including any specific characteristics of the features or aspects of the technology7with which that terminology7is associated. Conditional language such as, among others, “can,” “could,” “might” or “may,” unless specifically stated otherwise, are otherwise understood within the context as used in general to convey that certain embodiments include, while other embodiments do notinclude, certain features, elements and / or steps. Thus, such conditional language is not generally intended to imply that features, elements and / or steps are in any way required for one or more embodiments or that one or more embodiments necessarily include logic for deciding, with or without user input or prompting, whether these features, elements and / or steps are included or are to be performed in any particular embodiment.

[0116] Headings are included herein for reference and to aid in locating various sections. These headings are not intended to limit the scope of the concepts described with respect thereto. Such concepts may have applicability throughout the entire specification.

[0117] As used herein in the specification and claims, including as used in the examples and unless otherwise expressly specified, all numbers may be read as if prefaced by the word "‘about” or “approximately.” even if the term does not expressly appear. The phrase “about” or “approximately” may be used when describing magnitude and / or position to indicate that the value and / or position described is within a reasonable expected range of values and / or positions. For example, a numeric value may have a value that is + / - 0.1% of the stated value (or range of values), + / - 1% of the stated value (or range of values). + / - 2% of the stated value (or range of values), + / - 5% of the stated value (or range of values). + / - 10% of the stated value (or range of values), etc. Any numerical values given herein should also be understood to include about or approximately that value, unless the context indicates otherwise.

[0118] For example, if the value “10” is disclosed, then “about 10” is also disclosed. Any numerical range recited herein is intended to include all sub-ranges subsumed therein. It is also understood that when a value is disclosed that “less than or equal to” the value, “greater than or equal to the value” and possible ranges between values are also disclosed, as appropriately understood by the skilled artisan. For example, if the value “X” is disclosed the “less than or equal to X” as well as “greater than or equal to X” (e.g., where X is a numerical value) is also disclosed. It is also understood that the throughout the application, data is provided in a number of different formats, and that this data, may represent endpoints or starting points, and ranges for any combination of the data points. For example, if a particular data point “10” and a particular data point “15” may be disclosed, it is understood that greater than, greater than or equal to, less than, less than or equal to, and equal to 10 and 15 may be considered disclosed as well as between 10 and 15. It is also understood that each unit between two particular units may be also disclosed. For example, if 10 and 15 may be disclosed, then 11, 12, 13, and 14 may be also disclosed.

[0119] Disjunctive language such as the phrase "at least one of X, Y. or Z,” unless specifically stated otherwise, is otherwise understood with the context as used in general to present that an item, term, etc., may be either X, Y, or Z, or any combination thereof (e.g., X, Y, and / or Z). Thus, such disjunctive language is not generally intended to, and should not, imply that certain embodiments require at least one of X, at least one of Y, or at least one of Z to each be present.

[0120] The phrase ‘"based on7’ does not mean “based only on,” unless expressly specified otherwise. In other words, the phrase “based on” describes both “based only on” and “based at least on.” Unless otherwise explicitly stated, articles such as “a” or “an” should generally be interpreted to include one or more described items. Accordingly, phrases such as “a device configured to” are intended to include one or more recited devices.

[0121] It will be appreciated by those skilled in the art that various modifications and changes may be made without departing from the scope of the described technology. Such modifications and changes are intended to fall within the scope of the embodiments. It will also be appreciated by those of skill in the art that parts included in one embodiment are interchangeable with other embodiments; one or more parts from a depicted embodiment can be included with other depicted embodiments in any combination. For example, any of the various components described herein and / or depicted in the figures may be combined, interchanged or excluded from other embodiments.

[0122] The above description also discloses methods and materials of the present application. The devices described herein may be susceptible to modifications in the methods and materials, as well as alterations in the fabrication methods and equipment. Such modifications will become apparent to those skilled in the art from a consideration of this disclosure or practice of the invention disclosed herein. Consequently, it is not intended that this invention be limited to the specific embodiments disclosed herein, but that it cover all modifications and alternatives coming within the true scope and spirit of the invention as embodied in the attached claims. Applicant reserves the right to submit claims directed to combinations and sub-combinations of the disclosed inventions that are believed to be novel and non-obvious. Inventions embodied in other combinations and sub-combinations of features, functions, elements and / or properties may be claimed through amendment of those claims or presentation of new claims in the present application or in a related application. Such amended or new claims, whether they are directed to the same invention or a different invention and whether they are different, broader, narrower or equal in scopeto the original claims, are to be considered within the subject matter of the inventions described herein.

Claims

WHAT IS CLAIMED IS:

1. An ultrasound imaging system comprising: a probe configured to generate one or more ultrasound signals; one or more displays configured to display ultrasound images; a memory' storing instructions corresponding to one or more data analysis methods associated with ultrasound image processing; and an ultrasound image processor in communication with the probe and the one or more displays, the ultrasound image processor comprising one or more computer hardware processors: at least one of the one or more computer hardware processors configured to: generate at least a first set of test images, the first set of test images comprising at least one pair of ultrasound images having first parameter values of a first parameter that are different from each other; transmit the first set of test images to a user interface accessible by a user; receive, via the user interface, one or more user selected images, the one or more user selected images comprising at least one ultrasound image of the first set of test images; feed the one or more user selected images into the one or more data analysis methods to generate one or more ultrasound processing parameters, the one or more ultrasound processing parameters comprising at least one of 1 ) the first parameter having second parameter values different from the first parameter values, or 2) one or more second parameters different from the first parameter; receive the one or more ultrasound signals from the probe; process the one or more ultrasound signals based at least in part on the one or more ultrasound processing parameters to generate one or more processed ultrasound images; and display the one or more processed ultrasound images via one or more displays.

2. The ultrasound imaging system of Claim 1, wherein at least one of the one or more computer hardware processors is further configured to:generate a second set of test images comprising at least one additional ultrasound image having a first parameter value of the first parameter different than the first set of test images; receive, from the user interface, one or more additional user selected images, the one or more additional user selected images comprising at least one ultrasound image of the second set of test images; and feed the one or more additional user selected images into the one or more data analysis methods to generate the one or more ultrasound processing parameters.

3. The ultrasound imaging system of Claim 2, wherein the second set of test images are configured to be generated based on the one or more user selected images and the one or more data analysis methods.

4. The ultrasound imaging system of any one of Claims 1-3, wherein the first set of test images comprises a second pair of ultrasound images having second parameter values of a second parameter that are different than each other, and wherein the one or more user selected images comprise at least one of the second pair of ultrasound images.

5. The ultrasound imaging system of Claim 1, wherein to process the one or more ultrasound signals, at least one of the one or more computer hardware processors is configured to perform B-mode processing.

6. The ultrasound imaging system of any one of Claims 1-5 wherein to process the one or more ultrasound signals, at least one of the one or more computer hardware processors is configured to perform color flow processing.

7. The ultrasound imaging system of any one of Claims 1 -6, wherein to process the one or more ultrasound signals, at least one of the one or more computer hardware processors is configured to perform spectral doppler processing.

8. The ultrasound imaging system of any one of Claims 1-7, wherein at least one pair of ultrasound images comprise images of a same physiological feature.

9. A method of processing ultrasound images, the method comprising:generating, by one or more processors, at least a first set of test images, the first set of test images comprising at least one pair of ultrasound images having first parameter values of a first parameter that are different from each other; transmitting, by at least one of the one or more processors, the first set of test images to a user interface accessible by a user; receiving, from the user interface, one or more user selected images, the one or more user selected images comprising at least one ultrasound image of the first set of test images; feeding the one or more user selected images into one or more data analysis methods to generate one or more ultrasound processing parameters, the one or more ultrasound processing parameters comprising at least one of 1) the first parameter having second parameter values different from the first parameter values, or 2) one or more second parameters different from the first parameter; receiving one or more ultrasound signals from a probe; processing the one or more ultrasound signals based on the one or more ultrasound processing parameters to generate one or more processed ultrasound images; and displaying the one or more processed ultrasound images via one or more displays.

10. The method of Claim 9, further comprising: generating a second set of test images comprising at least one additional ultrasound image having a first parameter value of the first parameter different than the first set of test images; receiving, from the user interface, one or more additional user selected images, the one or more additional user selected images comprising at least one ultrasound image of the second set of test images; and feeding the one or more additional user selected images into the one or more data analysis methods to generate the one or more ultrasound processing parameters.

11. The method of Claim 10, wherein the second set of test images are generated based on the one or more user selected images and the one or more data analysis methods.

12. The method of any of Claims 9-11, wherein the first set of test images comprises a second pair of ultrasound images having second parameter values of a second parameter that are different than each other; and wherein the one or more user selected images comprise at least one of the second pair of ultrasound images..

13. The method of any one of Claims 9-12. wherein processing the one or more ultrasound signals comprises performing B-mode processing.

14. The method of any one of Claims 9-13, wherein processing the one or more ultrasound signals comprises performing color flow processing.

15. The method of any one of Claims 9-14, wherein processing the one or more ultrasound signals comprises performing spectral doppler processing.

16. The method of any one of Claims 9-1 , wherein at least one pair of ultrasound images comprise images of a same physiological feature.

17. A non-transitory computer readable recording medium storing instructions, when executed by one or more processors, cause the processors to perform the method of any one of Claims 9-16.

18. A user terminal comprising one or more processors configured to perform the method of any one of Claims 9-16.

Images