Systems and methods for conducting non-visual diagnostics using artificial intelligence with multiple aperture ultrasound imaging

Abstract

A diagnostic ultrasound imaging method is provided, that can include transmitting an omni-directional unfocused ultrasound waveform into a target region with a transmit aperture of a multiple aperture ultrasound imaging probe; receiving ultrasound echoes from the target region with a plurality of receive apertures of the multiple aperture ultrasound imaging probe; with a processor of the multiple aperture ultrasound imaging probe, extracting echo data from the received ultrasound echoes for a tissue density sampling region within the target region, the echo data comprising at least one of a frequency, a voltage level, and an acoustic impedance for pixels or voxels in the tissue density sampling region; with the processor, identifying one or more patterns in the echo data; with the processor, correlating the one or more patterns to previously collected ultrasound echo data to identify a probability that pixels or voxels in the tissue density sampling region comprise a specific tissue type.

Description

SYSTEMS AND METHODS FOR CONDUCTING NON- VISUAL DIAGNOSTICS USING ARTIFICIAL INTELLIGENCE WITH MULTIPLE APERTURE ULTRASOUND IMAGINGPRIORITY CLAIM

[0001] This patent application claims priority to U.S. provisional patent application no. 63 / 582,683, titled “SYSTEMS AND METHODS FOR CONDUCTING VISUAL AND NON- VISUAL DIAGNOSTICS USING ARTIFICIAL INTELLIGENCE WITH MULTIPLE APERTURE ULTRASOUND IMAGING,” and filed on September 14, 2023, which is herein incorporated by reference in its entirety.INCORPORATION BY REFERENCE

[0001] Unless otherwise specified herein, all patents, publications and patent applications mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication or patent application was specifically and individually indicated to be incorporated by reference.RELATED PATENTS AND APPLICATIONS

[0002] This application is also related to the following US Patent Applications: U.S. Patent No. 8,007,439 titled "Method and Apparatus to Produce Ultrasonic Images Using Multiple Apertures;" U.S. Patent No. 9,247,926 titled "Multiple Aperture Medical Ultrasound Transducers;" U.S. Patent No. 9,146,313 titled "Point Source Transmission and Speed-of- Sound Correction Using Multi-Aperture Ultrasound Imaging;" U.S. Patent No. 9,668,714 titled "Systems and Methods for Improving Ultrasound Image Quality by Applying Weighting Factors;" U.S. Patent No. 9,883,848 titled "Ultrasound Imaging Using Apparent Point-Source Transmit Transducer;" U.S. Patent No. 10,401,493 titled "Network-Based Ultrasound Imaging System;" and U.S. Patent Publication No. 2017 / 0209121 titled "Ultrasound Imaging with Sparse Arrays;" International Patent Application No.PCT / US2020 / 05662 titled “Systems and Methods for Tissue Characterization Using Multiple Aperture Ultrasound;” International Patent Application No. PCT / US2021 / 57725 titled “Systems and Methods for Improving Ultrasound Image Quality;” International Patent Application No. PCT / US 2023 / 062069 titled “Synthetic Aperture Unfocused Transmission for Multiple Aperture Ultrasound Systems;” and International Patent Application No. PCT / US2013 / 027120 titled “Determining Material Stiffness Using Multiple ApertureUltrasound.” The entire contents of each of these patents and patent applications is incorporated herein by reference. These related applications may be referred to herein as “Applicant’s prior patents and applications.”FIELD

[0003] This invention generally relates to medical diagnostics and specifically to diagnostic data collected through the use of ultrasound.BACKGROUND

[0004] Medical imaging today uses multiple modalities and often requires significant expertise of both the technician and or the physician to capture image data and then to conduct a diagnosis. That same level of expertise is required to conduct interventional procedures requiring the placement of instruments and therapies under image guidance.

[0005] Separate but related to diagnostic ultrasound imaging, is the fact that understanding a physiological condition or establishing a diagnosis can sometimes be concluded through quantitative analysis of the tissue being examined without visualization of that tissue. Providing qualitative and quantitative assessment of the patient’s condition has become accepted for everything determining bladder volume for catheterization to determining gastric volume prior to aestheticizing high risk patients. Perhaps most well- known is the non-imaging evaluation of bone density using ultrasound or x-ray. In these types of cases, a diagnostic data point is acceptable as a result of a medical ultrasound examination even when an image is not produced. For example, ultrasound bone densitometers produce t-score and z-score predictors of osteoporosis.

[0006] Much of the reason that ultrasound imaging equipment is not used to deliver more qualitative or quantitative data is that a trained technologist must aim the equipment at the appropriate tissue to be evaluated, which usually requires imaging expertise on the part of the user. Additionally, most conventional ultrasound imaging systems cannot image beyond traditional barriers to ultrasound wave transmission such as bone or gas. These natural barriers are caused by speed-of-sound changes as well as signal attenuation due to large mismatch of the acoustic impedance between tissues and bones. For instance, when imaging the liver from behind a rib, the transmit energy is largely attenuated by the rib bone, and only the portions of the transducer which extend beyond the rib over to the intercostal space are able to acquire decipherable image.

[0007] Significant improvements have been made in the field of ultrasound imaging with the creation of Ping Based Multiple Aperture Imaging (PMA), examples of which are shownand described in Applicant's prior patents and applications. PMA is sometimes referred to in prior works as Computed Echo Tomography (CET). Multiple aperture imaging methods and systems (PMA and CET) allow for ultrasound signals to be both transmitted and received via separate apertures, either electronically or physically. With this method of data capturing, it makes collecting data from around traditional barriers possible. The resultant data sets include multiple tissue types being collected simultaneously. A data set could also include solids, gases, liquids and devices. For instance, in cardiology a data set may consist of the complete thoracic cavity including heart, rib bones, an artificial heart valve (if present), moving blood and the upper lung are all possible with PMA systems. Physiological data can in many cases be simultaneously collected from a multiple aperture probe and system of multiple tissue types. Viewing and analyzing the total data set is possible in total, by subsections or solely at the data level without image conversion. Complete data sets of all tissue types greatly expanding the ability of qualitative and quantitative diagnostic assessments to be made.

[0008] This work is concerned with the collection and utilization of the data made possible by a PMA system. Further, it addresses utilizing that raw data for qualitative and quantitative analysis, and the use of algorithms to both train and assess that data.SUMMARY OF THE DISCLOSURE

[0009] A diagnostic ultrasound imaging method is provided, comprising the steps of: transmitting an omni-directional unfocused ultrasound waveform into a target region with a transmit aperture of a multiple aperture ultrasound imaging probe; receiving ultrasound echoes from the target region with a plurality of receive apertures of the multiple aperture ultrasound imaging probe; with a processor of the multiple aperture ultrasound imaging probe, extracting echo data from the received ultrasound echoes for a tissue density sampling region within the target region, the echo data comprising at least one of a frequency, a voltage level, and an acoustic impedance for pixels or voxels in the tissue density sampling region; with the processor, identifying one or more patterns in the echo data; with the processor, correlating the one or more patterns to previously collected ultrasound echo data to identify a probability that pixels or voxels in the tissue density sampling region comprise a specific tissue type; and presenting the tissue type to the user on a display.

[0010] In some aspects, the method comprises, prior to identifying the one or more patterns in the echo data, inputting the echo data into a trained machine learning model,wherein the trained machine learning model is configured to identify the one or more patterns in the echo data.

[0011] In some aspects, the trained machine learning model is configured to perform the correlating step.

[0012] In one aspect, correlating the one or more patterns further comprises comparing the one or more patterns to previously collected ultrasound echo data stored in a database.

[0013] A diagnostic ultrasound imaging method is provided, comprising the steps of transmitting an omni-directional unfocused ultrasound waveform into a target region with a transmit aperture of a multiple aperture ultrasound imaging probe; receiving ultrasound echoes from the target region with a plurality of receive apertures of the multiple aperture ultrasound imaging probe; with a processor of the multiple aperture ultrasound imaging probe, extracting echo data from the received ultrasound echoes for a tissue density sampling region within the target region, the echo data comprising at least one of a frequency, a voltage level, and an acoustic impedance for pixels or voxels in the tissue density sampling region; with the processor, inputting the echo data into a trained machine learning model configured to identify one or more patterns in the echo data and correlate the pattern to a known tissue type; with the processor, receiving an output from the trained machine learning model identifying a tissue type of pixels or voxels in the tissue density sampling region; and presenting the tissue type to the user on a display.

[0014] In some aspects, the tissue density sampling region is received as a user input.

[0015] A diagnostic ultrasound imaging method is provided, comprising the steps of transmitting an omni-directional unfocused ultrasound waveform into a target region with a transmit aperture of a multiple aperture ultrasound imaging probe; receiving ultrasound echoes from the target region with a plurality of receive apertures of the multiple aperture ultrasound imaging probe; receiving an input from a user indicating a tissue density sampling region within the target region; with a processor of the multiple aperture ultrasound imaging probe, extracting echo data from the received ultrasound echoes comprising at least one of a frequency, a voltage level, and an acoustic impedance for pixels or voxels in the tissue density sampling region; with the processor, inputting the echo data into a trained machine learning model configured to identify one or more patterns in the echo data and correlate the pattern to a known tissue type; with the processor, receiving an output from the trained machine learning model identifying a tissue type of pixels or voxels in the tissue density sampling region; and presenting the tissue type to the user on a display.

[0016] In some aspects, the method comprises, prior to identifying the one or more patterns in the echo data, inputting the echo data into a trained machine learning model, wherein the trained machine learning model is configured to identify the one or more patterns in the echo data.

[0017] In some aspects, the trained machine learning model is configured to perform the correlating step.

[0018] In one aspect, correlating the one or more patterns further comprises comparing the one or more patterns to previously collected ultrasound echo data stored in a database.

[0019] An ultrasound imaging system is provided, comprising: at least one transmitting transducer element configured to transmit an omni-directional unfocused ultrasound waveform into a target region; a plurality of receiving transducer elements configured to receive ultrasound echoes from the target region; a processor configured to extract echo data from the received ultrasound echoes for a tissue density sampling region within the target region, the echo data comprising at least one of a frequency, a voltage level, and an acoustic impedance for pixels or voxels in the tissue density sampling region, the processor being further configured to identify one or more patterns in the echo data and correlate the one or more patterns to previously collected ultrasound echo data to identify a probability that pixels or voxels in the tissue density sampling region comprise a specific tissue type.

[0020] In some aspects, the system includes a display configured to present the tissue type to the user on a display.

[0021] In one aspect, the processor is further configured to, prior to identifying the one or more patterns in the echo data, input the echo data into a trained machine learning model, wherein the trained machine learning model is configured to identify the one or more patterns in the echo data.

[0022] In other aspects, the trained machine learning model is configured to correlate the one or more patterns to previously collected ultrasound echo data to identify a probability that pixels or voxels in the tissue density sampling region comprise a specific tissue type.

[0023] In some aspects, the processor is further configured to compare the one or more patterns to previously collected ultrasound echo data stored in a database.

[0024] An ultrasound system is provided, comprising: a transmit aperture comprising at least one transducer element configured to transmit an omni-directional unfocused ultrasound waveform into a target region; a plurality of receive apertures each comprising at least one transducer element configured to receive ultrasound echoes from the target region; a processor configured to extract echo data from the received ultrasound echoes for a tissuedensity sampling region within the target region, the echo data comprising at least one of a frequency, a voltage level, and an acoustic impedance for pixels or voxels in the tissue density sampling region, the processor being further configured to input the echo data into a trained machine learning model configured to identify one or more patterns in the echo data and correlate the pattern to a known tissue type, the processor also being configured to receive an output from the trained machine learning model identifying a tissue type of pixels or voxels in the tissue density sampling region.

[0025] In some aspects, the system includes a display configured to present the tissue type to the user on a display.

[0026] In one aspect, the system includes a user-input device configured to receive the tissue density sampling region as a user input.BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The novel features of the invention are set forth with particularity in the claims that follow. A better understanding of the features and advantages of the present invention will be obtained by reference to the following detailed description that sets forth illustrative embodiments, in which the principles of the invention are utilized, and the accompanying drawings of which:

[0028] FIG. l is a schematic illustration of a multiple aperture imaging probe with three transducer arrays and several points to be imaged.

[0029] FIG. 2 is a schematic diagram of an embodiment of a concave curvilinear transducer demonstrating how transmit and receive apertures can be widened around a desired view angle to achieve greater resolution of the target area.

[0030] FIG. 2A is a schematic diagram of an embodiment of a concave transducer where the transmit aperture and multiple receive apertures can be electronically controlled to operate in different positions.

[0031] FIG. 2B is a schematic diagram of an embodiment of a concave transducer where elements of the transmit and receive apertures can be utilized in rapid succession to gather more data from multiple transmit and receive apertures.

[0032] FIG. 2C is a schematic diagram of an embodiment of a concave transducer where demonstrating how transmit and receive apertures may be widened around a desired view angle to achieve greater resolution of the target area.

[0033] FIG. 2D is a schematic diagram of an embodiment of a concave curvilinear matrix with curvature in two orthogonal directions, also referred to as a Three-Dimensional (3D) array. Each element in a 3D array is displaced relative to adjacent elements in all of x, y, and z axes. In this illustration, an element or elements of a transmit aperture is designated to insonify the medium. Multiple targets in the medium are illustrated for the purpose of demonstrating how volumetric data may be gathered. Multiple receive apertures are illustrated to demonstrate how simultaneous gathering of data may involve timing and tissue speed of sound variations.

[0034] FIG. 2E illustrates an embodiment of a 3D array. Multiple transmit apertures Ti through TN are indicated for the purpose of demonstrating transmit pulses being received on one or more receive apertures R2 and / or R3. A single target is indicated for the purpose of demonstrating how data may be gathered.

[0035] FIG 3 illustrates a 2D grid of pixels.

[0036] FIG 3 A is an example of pixels used in a low-resolution imaging system to define point A and point B .

[0037] FIG. 3B is an example of pixels used in a high-resolution imaging system to define point A and point B.

[0038] FIG. 3C is an example of pixels used in a high-resolution imaging system where individual pixels on their own would not be visual to the human eye, but a cluster of pixels that can be seen clearly

[0039] FIG. 4 is an example of data collection methods using multiple aperture imaging.

[0040] FIG. 5 is a representation of a raster map plotted with values from frequencies in different samples.

[0041] FIG. 6 is a representation of a raster map plotted with values from values or signal amplitudes at different samples.

[0042] FIG. 7 is a representation of a raster map plotted with values from acoustic impedance at different samples.

[0043] FIG. 8 is an example of a data collection method using multiple aperture imaging.DETAILED DESCRIPTION

[0044] The various embodiments will be described in detail with reference to the accompanying drawings. References made to particular examples and implementations are for illustrative purposes and are not intended to limit the scope of the invention or the claims.

[0045] The present disclosure provides systems and methods for utilizing artificial intelligence to analyze raw data produced by ping-based multiple aperture imaging (“PMA” imaging) systems The resultant data can be used to improve image quality and to conduct quantitative or qualitative analysis of both visual and non-visual diagnostic conditions.

[0046] As described in more detail below and in Applicant’s prior patents and applications, PMA imaging involves transmission of ultrasound “pings,” echoes of which may be received by receive elements located at some distance from the transmitter. Each received echo signal lies along an ellipse defined by the transmitter and receiver positions and the time interval between ping transmission and echo reception. A data element can be created virtually anywhere in a 2D grid or 3D volume. Typically, ellipses crossing at a location in a 2D grid is known as a pixel. Ellipses crossing in a 3D volume are known as a voxel. An image may be formed by combining such pixels or voxels (i.e., data elements) in a way that their intersections become emphasized. Each intersection may be an image point in a two-dimensional or three-dimensional image. Concurrently and separately, groupings of data elements known as data sets may be evaluated for quantitative and or qualitative indications of diagnostic conditions.

[0047] FIG. 1 demonstrates a ping based multiple aperture probe 100 S, with arrays 12, 14, and 16. Subarrays or often individual elements within each array are indicated as points a, b, c, d, e, f, g, h and i. However, sub-arrays can be located across physical gaps between arrays and should not be considered limited to individual elements on an individual array. A ping transmission is represented by the wavefront 13 (dashed wavefront(s)) generated by a transmit aperture at ‘a’ on array 12 and is indicated by wavelets. Point A in the medium or tissue 20 is meant to represent a hard structure (e.g., calcium or hardened plaque from atherosclerosis, or other hard objects such as bone), which would immediately reflect or scatter the transmitted wavefront 13 in multiple directions represented here as reflected wavefront 15 (solid wavefront(s)). The reflected wavefront emanating from point A may provide a relatively bright signal to the receive elements in arrays 12, 14, and 16. The transmitted wavefront 13 may also continue on through the medium or tissue 20 to point B, which is meant to represent an anechoic structure (e.g., a blood vessel or other soft tissue) that would provide a relatively weaker reflected wavefront 17 back to the individual transducer elements on arrays 12, 14, and 16.

[0048] Echoes from points A and B can be received by receive elements in arrays 12, 14, and 16 and used by the probe 100 to create data sets and frames used to form multiple aperture ultrasound images or for analysis by artificial intelligence engines. An electroniccontroller(s) or processor(s) associated with any probe on a PMA enabled system can begin the process of analyzing the data in the region of interest. Data being collected after Analog to Digital conversion is stored into data strings for a first receiver element. This receiver element may be part of an array, or it may be an independent element used as an omnidirectional receiver. It need not be used in conjunction with other elements to collect and compound data in real time. A second receiver element can be used to produce a second string of data coming off of the same ping utilized to provide receiver element data to the first receiver element. Similarly, echo data coming off of the same ping transmission can be used by a plurality of receiver elements (e.g., third, fourth, fifth, etc. receiver elements).

[0049] Transducers may be contiguous in some embodiments or may configured using multiple arrays in other embodiments. No matter the type of array, transducer elements operate independently and can be utilized to form any number of either transmit or receive apertures within the structure. As used in the embodiment in FIG. 1, the terms "transducer array" or "array" generally refers to a collection of transducer elements mounted to a common backing block. Such arrays may have one dimension (ID), two dimensions (2D), 1.5 dimensions (1.5D) or three dimensions (3D). Other dimensioned arrays as understood by those skilled in the art may also be used. Transducer arrays may also be collections of pMUT or cMUT transducer elements. An element of a transducer array may be the smallest discretely functional component of an array. For example, in the case of an array of piezoelectric transducer elements, each element may be a single piezoelectric crystal or a single machined section of a piezoelectric crystal.

[0050] In the embodiment illustrated in FIG. 2 data collected at the receiver aperture Ri is coming from a single transmit source Ti, and therefore will have a singular consistent speckle noise pattern. By itself, this is the same limiting factor of conventional ultrasound today. However, data from the transmission Ti may also be collected at aperture R2 at the same time. The channel data collected has a different aspect angle on the target, and when compounded with the aperture data at Ri will provide higher resolution data collection for the pixels at the target location. In either case, energy is transmitted toward a region of interest which has at least one reflector 170. Receiver aperture Ri and aperture R2 may be electronically designated to collect data for this transmit cycle by the MAUI Electronics 140. The subject of how to combine pixel data for receivers located along different physical pathways at different apertures and sometime on different arrays altogether, is the subject of US Patent 9,146,313 titled "PointSource Transmission and Speed-of-Sound Correction Using Multi-Aperture Ultrasound Imaging.”

[0051] In the embodiment illustrated in FIG. 2A the transmitter Ti is now located in a different physical separated aperture. Uike FIG. 2, this sequence of transmitting from Ti will create a unique speckle pattern for data received by transducer(s) at apertures Ri and R2. However, when the data set created at a first time in FIG. 2 is combined with the data collected at a second time in FIG. 2A, the aggregated data set will benefit from the constructive interference of the speckle noise patterns emanating from the different transmit locations. Data clarity specifically around signal frequency and amplitude benefit greatly over conventional ultrasound systems collecting data from only a single transmit aperture or origin. Similaraly, energy is transmitted toward a region of interest which has at least one reflector 170. Thereafter, receiver aperture Ri and aperture R2 may be electronically designated to collect data for this transmit cycle by the MAUI Electronics 140.

[0052] The embodiment illustrated in FIG. 2B is an extension of the concept illustrated in FIG. 2A. In FIG. 2B, multiple receiver apertures Ri, R2 and R3 collect data from Ti at a first time. Subsequently, all receive apertures receiver apertures Ri, R2 and R3 receive additional data from the transmission at T2 at a second time, and the transmission at T3 at a third time. The natural benefit of this process is to create wider and wider apertures which improve resolution. FIG. 2B further demonstrates another unique feature of some embodiments of multiple aperture arrays called view angle control. As seen in FIG. 2B, the view angle may be defined as the angle a between lines 180 and 190 that may be drawn from Ti to 170 along line 180 and from Ri to 170 along line 190. In some embodiments, the MAUI electronics 140 may be configured to automatically move the transmit Ti and receive Ri apertures along the array or arrays without changing the total distance between the transmit Ti and receive Ri apertures. This process of improving lateral resolution using PMA systems is described in U.S. Patent No. 8,007,439 titled "Method and Apparatus to Produce Ultrasonic Images Using Multiple Apertures.” The curvature of the concave probe also improves trans lateral (or resolution measured diagonally from the center of the transducer) and axial resolution (measured vertically from the center of the transducer. These improvements are in part due to the reduction in angular offset and associated “view angle” by receiving elements, and in part due to being physically closer to the image targets. Further, differenttransmission frequencies can be interlaced into the different transmit and receive apertures, can be utilized to further create higher resolution data sets.

[0053] FIG. 2C illustrates another important capability of some embodiments of multiple aperture arrays, referred to herein as total aperture size control. In this embodiment, the view angle created by TiA and RIA provides an obstruction-free view of the region of interest including reflector 170 which avoids obstruction 150. In this example, the system provides multiple aperture imaging with a total aperture width of AIA. Using total aperture size control, the total aperture width can be varied either inward or outward on the array, while maintaining a fixed view angle center. Thus, in some embodiments, as a total aperture size is adjusted, both transmit and receive apertures may be electronically moved at the same rate outward or inward from the fixed view angle center so as to maintain the original view angle.

[0054] Radian resolution is proportional to X / d, where X is wavelength and d is total aperture width. The wider the total aperture, the higher the resolution; the smaller the total aperture, the lower the resolution. The total aperture width may be varied to get the best possible resolution for a chosen view angle. For example, in the embodiment of FIG. 2C, the total aperture width may be maximized by selecting the total aperture width between TIB and RIB, resulting in a total aperture width of AIB.

[0055] U.S. Patent No. 9,220,478 titled “Concave Ultrasound Transducers and 3D Arrays,” demonstrates many iterations of arrays configurations to produce volumetric data sets and their associated voxels. Illustration of FIG. 2D is presented here to represent a 3D array used with a PMA system to create 3D volumetric data sets. In FIG. 2D, a snapshot of multiple aperture data collection is depicted enroute to building an image of an entire volume 310. As seen in FIG. 2D, a concave 3D transducer array 300 may have a curvature about two orthogonal axes. In some embodiments, a 3D concave curved array 300 may be constructed using machined piezoelectric transducers. Calibration of a 3D array may be needed as each element of the array 300 will have slightly different positions in the x axis 301, y axis 302 and z axis 303.

[0056] As further seen in FIG. 2D, a snapshot of multiple aperture data collection is depicted enroute to building an image of an entire volume 310. Here, an element or elements of a transmit aperture Ti transmit a pulse into the volume that includes scatterers such as 321 and 332. The elements making up receive aperture R2 may be assembled in a variety of shapes. Here, a square of elements makes up the receive apertures R2 and R3. As mentioned above, the speed of sound along the path from the transmit aperture Ti to the reflector 321 or332 is irrelevant to the coherent addition of the received signals as long as a single aperture is used to receive, however, speed of sound corrections can be made to improve image quality when using multiple receive apertures Ri and R2.

[0057] In some embodiments, the size of the receive aperture R2 may be as large as for a conventional phased array (e.g., about 2cm). But unlike a conventional array, the total aperture 340 determining the lateral and transverse resolution of the system is much larger comprising the distance from the transmitter Ti to the group of receiver elements R2, and could be as wide as the entire array 300 or wider if a transmitter was located on another array within the probe (or in a separate probe in electronic communication). The elements located in the receive aperture R2 each collect volumetric data from the Ti transmit pulse.Transmitter and receiver aperture sizes and sequences may be altered throughout the entire array, similar to the sequences described FIG. 2B, but now transmitters and receivers may be located anywhere within the array and in any axis.

[0058] FIG. 2E illustrates another embodiment of a 3D array 300. As seen in FIG. 2E, multiple transmit apertures Ti through TN are indicated for the purpose of demonstrating transmit pulses being received on one or more receive apertures R2 and / or R3. A single target 321 or reflector 321 is indicated for the purpose of demonstrating how data may be gathered.

[0059] Collecting data through varying tissue types naturally requires the accommodation of differing tissue acoustic impedance with differing speeds-of-sound. Pixel and voxel computations can only be accurate when these accommodations are made. U.S. Patent No. 9,146,313 titled "Point Source Transmission and Speed-of-Sound Correction Using MultiAperture Ultrasound Imaging" covers these methods in detail.

[0060] A processor in the PMA system, in the probe itself or in communication with the probe (e.g., wirelessly) may then be configured to conduct calculations using all data set values for all data strings. In some implementations, the data string may be collected for an entire region of interest (i.e., large sample period). In other implementations, the data may be collected for only a specific pixel and voxel (i.e. specific sample period).

[0061] The processor can then initiate the beamforming process to create digital values at a pixel location in the region of interest. In the case of 3D diagnostics and imaging, the same process can be utilized to create data values by voxel. The process can be repeated for multiple ping transmissions and echo data strings collected for multiple receivers.

[0062] As seen in FIG. 3, 300 represents a 2D grid of pixels. The number of pixels that are used in a PMA can be controlled by the user. In imaging, this has the benefit of controlling frame rate. When less pixels are utilized, processing can be faster and imageframe rates higher. However, low numbers of pixels may not produce the high quality images desired. Therefore, higher numbers of pixels utilized in imaging require more processing power and a commensurate decrease in frame rate.

[0063] Pixels sizes selected need not be limited to what can be presented visually. Very large numbers of pixels can be selected even when individual pixels are not discernable with the human eye. Similarly, if presenting a 3D volume of voxels, a large number of voxels may also be beamformed where individual voxels presented by themselves would not be visible to the human eye.

[0064] The number of pixels used to define points “reflectors" A and point B can vary and are a function of desired resolution. FIG. 3 A demonstrated a grid pattern where a pixel is located at the intersection of each X and Y axis. Higher resolution systems are constructed more easily with Ping Based Multiple Aperture (PMA) systems since there is no limit to probe width, shape or frequency. See U.S. Patent No. 9,220,478, titled “Concave Ultrasound Transducers and 3D Arrays” for a demonstration of how PMA system achieve higher resolution in ultrasound. Such higher resolution imaging systems may have pixels located closer to each other than low-resolution system. FIG. 3 A shows that only a few pixels are able to cover points A and only a few more pixels to define B respectively with the low- resolution system. Higher resolution systems obviously increase the clarity with which a reflector is presented as more pixels can be utilized to define both the body and edges of the reflector. FIG. 3B shows a greater number of pixels are used to define those same points with a higher resolution system.

[0065] With even higher processing power, it is possible to define pixels and voxels that individually are in the sub-visual range. That is, at 20” away from a display the human eye can have adequate resolution for 1 arc minute which is about 170 dpi (dots or pixels per inch), which equals around 0.14-0.22mm. Pixels presenting below this value may not be discernable to the human eye. However, all of the data collected that does or does not meet the 170 dpi requirement can still be utilized as beamformed or beamformed data.

[0066] FIG. 3C demonstrations that in very high resolution PMA systems, pixel value can be located below visual range. Note, both points A and B have hundreds of pixels, each of which individually would not be recognizable by the human eye.

[0067] As a result of the beamforming process, two key pieces of data are collected for each pixel or voxel. The first is frequency. Peak frequency is determined as a function of the bandwidth selected in the system settings. Larger receiver bandwidth, selectable by theend user, enables more dynamic range and therefore each pixel can have a more accurately determined frequency value.

[0068] The second key piece of data collected is value or amplitude at that pixel or voxel location. The pixel or voxel value is determined by the strength of the return signal in millivolts (e.g., the amplitude or level of the return signal).

[0069] MAUI Assigned Acoustic Impedance (MAAI) is a value generated from lab testing of all body tissues (e.g. bone, blood, muscles, organs, etc.) for different depths. Note, MAAI is not the natural acoustic impedance of body tissues, but instead can be inferred or calculated by the system based on the return echoes and any number of system parameters. The MAAI values may be stored locally on a PMA system, PMAbase station or in the cloud, and are regularly updated. The variations of acoustic impedances are equivalent in function to a gray scale.

[0070] Pixel or voxel data will be assigned MAAI values by comparison to the lab generated table. The assigned value will be relative to data energy level . The higher the energy level, the higher MAAI value will be. In certain embodiments, data can be classified based on density because the higher MAAI value is the harder tissue the target it.

[0071] The table below is an example of MAAI. The table entries are result of lab characterization of data collected with a PMA system. Each entry can be used as a pointer to more data as the artificial Intelligence application (Al) gets trained over time. The data is dependent on transducer frequency as well as target depth in the body due to ultrasound signal attenuation. Also note the value is unitless, it is relative to acoustic impedance of body soft tissue. The numbers are reduced with depth, but the ratios will stay relatively constant.

[0072] Table 1 :MAUI Assigned AcousticImpedance (MAAI)Method for Creating Tissue Density Maps

[0073] To this point, the method provided was always discussed in conjunction with conducting imaging. Beamforming pixel and voxel values need not proceed to scan conversion for image presentation to the user. For instance, it may be diagnostically desirable to determine if there is fluid inside the skull as a first step in trauma triage. Imaging may not be necessary or even achievable with the personnel on scene; however, quantitative analysis of the data achieved through an unguided or “blind” scan where the probe is held up against the region of interest (e.g., the skull) without using imaging, may provide critical physiological or diagnostic information.

[0074] As seen in FIG. 4, 400 describes a process for achieving tissue density that may be relevant to diagnosis. Using a PMA system with Multiple Aperture Probe, the probe may be placed over the region of interest. The system may provide imaging data at the time per step 401 or may be used without any imaging data being presented per 402. Data may be collected in real-time, can be stored in system memory or even transmitted and stored in the cloud during step 404 reference US Patent Network Based Beamforming.

[0075] Steps 406 and 408 are naturally produced as a function of beamforming in step 405. Frequency and Voltage levels are then grouped independently and respectively in steps 407 and 409. Acoustic Impedance in step 410 must utilize both frequency and voltage data for a pixel or voxel, inputs those values into the known Acoustic Impedance Table, and then extracts the Acoustic Impedance for the pixels or voxels into groups in Steps 411.

[0076] User controls are utilized to define a pixel or voxel region size. That is, the number of pixels to be analyzed together for pattern recognition is selectable by the user. The smallest grouping would be one pixel, and the largest would be equivalent to the resolution of the system (e.g. 1260 x 1260 pixels). 302, 304, 306 represent three pixel regions in FIG. 3. Each of 302, 304 and 306 are using 6 by 3 pixels to define their regions. However, a typicalselection could be on the order of 8 by 8 pixels. In a 3D system, the desired cube might be 8 by 8 by 8 voxels.

[0077] Once a region size is selected by the user or by a computer sampling regions automatically in search of pattern data, a raster map can be created see FIG. 5, 6, 7. These maps are visual representations of the values of acoustic impedance MAAI, frequency and voltage respectively at that point in time for that data set. The data set may have been the function of just a single transmit (ping) and all collected channel data; or it could be the function of multiple transmits and their collected data sets being compounded. Ping-Based Multiple Aperture Imaging (PMA) is less concerned with compounding data from differing tissue types than convention ultrasound systems for reasons discussed in US Patent 9,146,313 titled "Point Source Transmission and Speed-of-Sound Correction Using Multi -Aperture Ultrasound Imaging.” Therefore, data samples from all tissue types can often be collected in spite of barriers that would normally prevent such collection with conventional ultrasound systems.

[0078] In artificial intelligence or machine learning, the algorithms are based on two fundamental building blocks: (1) data and (2) training and pattern recognition. The bigger the pieces (or pixel sets) to coalesce into discernable patterns, the less processing is required. The smaller the pieces (e.g., into the subvisual range) the more processing is required. The selectable size of the baseline pattern building blocks has the natural effect of either improving data patterns that can be easily recognized or even improving imaging in the same fashion. That is. The smallest possible data block utilized will likely improve data detail and in turn image quality. The larger data block sized will likely reduce data detail, but could improve speed of assessment.

[0079] The second part of the process of using Al to “recognize” a pattern is called training. That is, using a library of images, raster maps, or quantitative parameters verified of known tissue type. For instance, the training pattern could be of a type of carcinoma associated with known stages of development. The known carcinoma data set can be used to train the processor of many known patterns. The trained patterns can then be cross correlated against the patterns collected through images, raster maps and other quantitative data to provide probabilities of correlation to specific pathologies (e.g., carcinoma, heart disease, vascular stenosis, and so on). Al will interpret the raster maps and other data searching for patterns that are consistent with known diagnostic conditions from a library of raster maps that have confirmed pathology.

[0080] Tissue density can be reliably inferred by association of the Acoustic Impedance data from the Table in step 410. However, real-time analysis of tissue density is best to be computed by beamforming all data in step 412. Beamforming of frequency, energy and acoustic impedance data can then be weighted via user controls in 412. The resultant tissue density map may then be useful either in the visual range 414 or the sub-visual range 416.

[0081] In some embodiments, imaging from the use of the PMA system is happening concurrently with the data sample beamforming. In Step 418 then, that imaging can be interlaced into the display of the tissue density maps being displayed in step 420.

[0082] In systems where imaging is not being conducted concurrently with data sample beamforming, then both visual and sub-visual tissue analysis can be conducted. In the case of step 416, data regions that are clearly sub-visual can be designated by the user. These subvisual regions, when utilized together may decipher conditions that are diagnostically important to the user as is the case in Step 422. Additionally, Al can be utilized to conduct sub-visual data analysis against data sets with known pathologies.

[0083] Finally, the process is repeated in real-time per Step 426.

[0084] FIG. 8 illustrates another example of a data collection method and system using multiple aperture imaging. In step 801, the system or method may transmit an omnidirectional unfocused ultrasound waveform into a target region with a transmit aperture of a multiple aperture ultrasound imaging probe. In step 802, the system or method may receive ultrasound echoes from the target region with a plurality of receive apertures in the multiple aperture ultrasound imaging probe. Data may be collected in real-time, may be stored in a system memory, or may be transmitted and stored in the cloud during step 801 or step 802.

[0085] Thereafter, the system or method may analyze and store the collected echo data in several ways, individually or simultaneously. For example, in step 803, the system or method may analyze the sample data by pixel or voxel location based on the sample frequency of the received echoes. Thereafter, in step 806, the system or method may group the pixel and voxel data by sample frequency. Alternatively or additionally, in step 804, the system or method may analyze the sample data by pixel or voxel location in accordance with voltage levels or voltage amplitudes of the received echoes. Thereafter, in step 807, the system or method may group the pixel and voxel data by voltage level or amplitude. Alternatively or additionally, in step 805, the system or method may analyze the sample data by pixel or voxel location with acoustic impedance information. The acoustic impedance information may be acoustic impedance values of the tissue that are measured, calculated, or derived, or may be inferred or calculated by the system based on the return echoes and any number of system parameters.Here, the system or method may reliably infer tissue density by association of the Acoustic Impedance information. Thereafter, in step 808, the system or method may group the pixel and voxel data by acoustic impedance.

[0086] In step 809, the system or method may beamform pixel or voxel data using weighed data from data received from each of the aforementioned steps 806, 807, and 808. In step 809, the beamforming of frequency, energy and acoustic impedance data may be weighted automatically or via user controls. Thereafter, in step 810, the system or method may identify one or more patterns in the echo data. In step 811, the system or method may correlate the one or more identified patterns in aforementioned step 810 to previously collected ultrasound echo data to identify a probability that pixels or voxels in the tissue density sampling region comprise a specific tissue type. Finally, in step 812, the system or method may present or output the specific tissue type. Indeed, in step 812, the system or method may present a resultant tissue density map. Such a resultant tissue density map may be either in the visual range or the sub-visual range.

[0087] Although this invention has been disclosed in the context of certain preferred embodiments and examples, it will be understood by those skilled in the art that the present invention extends beyond the specifically disclosed embodiments to other alternative embodiments and / or uses of the invention and obvious modifications and equivalents thereof. Various modifications to the above embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, it is intended that the scope of the present invention herein disclosed should not be limited by the particular disclosed embodiments described above, but should be determined only by a fair reading of the claims that follow.

[0088] In particular, materials and manufacturing techniques may be employed as within the level of those with skill in the relevant art. Furthermore, reference to a singular item includes the possibility that there are plural of the same items present. More specifically, as used herein and in the appended claims, the singular forms "a," "and," "said," and "the" include plural referents unless the context clearly dictates otherwise. As used herein, unless explicitly stated otherwise, the term “or” is inclusive of all presented alternatives, and means essentially the same as the commonly used phrase “and / or.” It is further noted that the claims may be drafted to exclude any optional element. As such, this statement is intended to serve as antecedent basis for use of such exclusive terminology as "solely," "only" and the like in connection with the recitation of claim elements, or use of a "negative" limitation. Unlessdefined otherwise herein, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

Claims

CLAIMS1. A diagnostic ultrasound imaging method, comprising the steps of: transmitting an omni-directional unfocused ultrasound waveform into a target region with a transmit aperture of a multiple aperture ultrasound imaging probe; receiving ultrasound echoes from the target region with a plurality of receive apertures of the multiple aperture ultrasound imaging probe; with a processor of the multiple aperture ultrasound imaging probe, extracting echo data from the received ultrasound echoes for a tissue density sampling region within the target region, the echo data comprising at least one of a frequency, a voltage level, and an acoustic impedance for pixels or voxels in the tissue density sampling region; with the processor, identifying one or more patterns in the echo data; with the processor, correlating the one or more patterns to previously collected ultrasound echo data to identify a probability that pixels or voxels in the tissue density sampling region comprise a specific tissue type; and presenting the tissue type to the user on a display.

2. The method of claim 1, further comprising, prior to identifying the one or more patterns in the echo data, inputting the echo data into a trained machine learning model, wherein the trained machine learning model is configured to identify the one or more patterns in the echo data.

3. The method of claim 2, wherein the trained machine learning model is configured to perform the correlating step.

4. The method of claim 1, wherein correlating the one or more patterns further comprises comparing the one or more patterns to previously collected ultrasound echo data stored in a database.

5. A diagnostic ultrasound imaging method, comprising the steps of: transmitting an omni-directional unfocused ultrasound waveform into a target region with a transmit aperture of a multiple aperture ultrasound imaging probe; receiving ultrasound echoes from the target region with a plurality of receive apertures of the multiple aperture ultrasound imaging probe;with a processor of the multiple aperture ultrasound imaging probe, extracting echo data from the received ultrasound echoes for a tissue density sampling region within the target region, the echo data comprising at least one of a frequency, a voltage level, and an acoustic impedance for pixels or voxels in the tissue density sampling region; with the processor, inputting the echo data into a trained machine learning model configured to identify one or more patterns in the echo data and correlate the pattern to a known tissue type; with the processor, receiving an output from the trained machine learning model identifying a tissue type of pixels or voxels in the tissue density sampling region; and presenting the tissue type to the user on a display.

6. The method of claim 5, wherein the tissue density sampling region is received as a user input.

7. A diagnostic ultrasound imaging method, comprising the steps of: transmitting an omni-directional unfocused ultrasound waveform into a target region with a transmit aperture of a multiple aperture ultrasound imaging probe; receiving ultrasound echoes from the target region with a plurality of receive apertures of the multiple aperture ultrasound imaging probe; receiving an input from a user indicating a tissue density sampling region within the target region; with a processor of the multiple aperture ultrasound imaging probe, extracting echo data from the received ultrasound echoes comprising at least one of a frequency, a voltage level, and an acoustic impedance for pixels or voxels in the tissue density sampling region; with the processor, inputting the echo data into a trained machine learning model configured to identify one or more patterns in the echo data and correlate the pattern to a known tissue type; with the processor, receiving an output from the trained machine learning model identifying a tissue type of pixels or voxels in the tissue density sampling region; and presenting the tissue type to the user on a display.

8. The system of claim 7, further comprising, prior to identifying the one or more patterns in the echo data, inputting the echo data into a trained machine learning model,wherein the trained machine learning model is configured to identify the one or more patterns in the echo data.

9. The method of claim 8, wherein the trained machine learning model is configured to perform the correlating step.

10. The method of claim 7, wherein correlating the one or more patterns further comprises comparing the one or more patterns to previously collected ultrasound echo data stored in a database.

11. An ultrasound imaging system, comprising: at least one transmitting transducer element configured to transmit an omnidirectional unfocused ultrasound waveform into a target region; a plurality of receiving transducer elements configured to receive ultrasound echoes from the target region; a processor configured to extract echo data from the received ultrasound echoes for a tissue density sampling region within the target region, the echo data comprising at least one of a frequency, a voltage level, and an acoustic impedance for pixels or voxels in the tissue density sampling region, the processor being further configured to identify one or more patterns in the echo data and correlate the one or more patterns to previously collected ultrasound echo data to identify a probability that pixels or voxels in the tissue density sampling region comprise a specific tissue type.

12. The system of claim 9, further comprising a display configured to present the tissue type to the user on a display.

13. The system of claim 11, wherein the processor is further configured to, prior to identifying the one or more patterns in the echo data, input the echo data into a trained machine learning model, wherein the trained machine learning model is configured to identify the one or more patterns in the echo data.

14. The system of claim 13, wherein the trained machine learning model is configured to correlate the one or more patterns to previously collected ultrasound echo data to identify aprobability that pixels or voxels in the tissue density sampling region comprise a specific tissue type.

15. The method of claim 11, wherein the processor is further configured to compare the one or more patterns to previously collected ultrasound echo data stored in a database.

16. An ultrasound system, comprising: a transmit aperture comprising at least one transducer element configured to transmit an omni-directional unfocused ultrasound waveform into a target region; a plurality of receive apertures each comprising at least one transducer element configured to receive ultrasound echoes from the target region; a processor configured to extract echo data from the received ultrasound echoes for a tissue density sampling region within the target region, the echo data comprising at least one of a frequency, a voltage level, and an acoustic impedance for pixels or voxels in the tissue density sampling region, the processor being further configured to input the echo data into a trained machine learning model configured to identify one or more patterns in the echo data and correlate the pattern to a known tissue type, the processor also being configured to receive an output from the trained machine learning model identifying a tissue type of pixels or voxels in the tissue density sampling region.

17. The system of claim 16, further comprising a display configured to present the tissue type to the user on a display.

18. The system of claim 16, further comprising a user-input device configured to receive the tissue density sampling region as a user input.

Images