Method and system for ultrasound imaging

Abstract

A method of ultrasound imaging based on an analog signal from an ultrasound transducer system is provided. The ultrasound transducer system includes: a piezoelectric ultrasound transducer; a resonance matching network connected to the piezoelectric ultrasound transducer and configured to receive an electrical signal from the piezoelectric ultrasound transducer and provide a passive voltage gain; and a low-noise amplifier connected to the resonance matching network and configured to receive and amplify the electrical signal from the resonance matching network. The method includes: converting the analog signal received from the ultrasound transducer system to a digital signal; attenuating ringing artifact in the digital signal based on a ringing artifact attenuation machine learning model to produce a ringing artifact attenuated digital signal; constructing an ultrasound image by beamforming imaging based on the ringing artifact attenuated digital signal; and enhancing a resolution of the ultrasound image based on a resolution optimization function to produce an enhanced ultrasound image, the resolution optimization function being configured based on a point spread function (PSF) of a transducer degradation model of the piezoelectric ultrasound transducer. There is also provided a corresponding ultrasound imaging system and a corresponding ultrasound transducer and imaging system.

Description

METHOD AND SYSTEM FOR ULTRASOUND IMAGINGCROSS-REFERENCE TO RELATED APPLICATION

[0001] This application claims the benefit of priority of Singapore Patent Application No. 10202401786U filed on 19 June 2024, the content of which being hereby incorporated by reference in its entirety for all purposes.TECHNICAL FIELD

[0002] The present invention generally relates to a method of ultrasound imaging, and a system thereof, as well as an ultrasound transducer and imaging system.BACKGROUND

[0003] Piezoelectric transducers have been widely used in areas such as non-destructive evaluation (NDE), structural health monitoring (SHM), biomedical imaging, and energy harvesting, and so on. Compared to other cross-sectional imaging modalities (e.g., computed tomography (CT), microwave, magnetic resonance imaging (MRI), and so on), ultrasound imaging technique provides simpler implementation, lower cost and non-invasive capability. Hence, ultrasound detection provides huge potential for convenient home health monitoring (HHM), fingerprint recognition and implementation of wearable devices. To ultrasonic sensor systems, a powerful piezoelectric ultrasound transducer with high sensitivity and wideband response is of great demand, especially for biomedical imaging with high resolution and superior contrast.

[0004] FIG. 1A depicts a schematic diagram of a first-order low-pass filter with a passive amplification matching configuration which may be applied to provide voltage amplification and noise figure (NF) reduction when the inductor resonates with the capacitor at operating frequency ωos. For example, this matching technique appeared in RF systems (e.g., disclosed in Ru et al., “A tunable 300-800MHz RF-sampling receiver achieving 60dB harmonic rejection and 0.8dB minimum NF in 65nm CMOS”, 2009 IEEE Radio Frequency Integrated Circuits Symposium, pages 21-25, 2009 (hereinafter referred to as the Ru reference)) and Nuclear Magnetic Resonance (NMR) systems (e.g., disclosed in Sun et al., “CMOS RF Biosensor Utilizing Nuclear Magnetic Resonance” in IEEE Journal of Solid-State Circuits, vol. 44, no. 5, pages 1629-1643, May 2009 (hereinafter referred to as the Sun reference)) to maximally deliver the output voltage rather than the output power by predictable passive gain, and the minimumNF can be further obtained. To maximize the output voltage, the input impedance of the low- noise amplifier (LNA) connected after the matching network needs to be high impedance rather than 50 Q. Hence, the electrical matching (EM) technique also makes the design of ultrasound receivers more convenient as the amplifier chips working at the ultrasound frequency generally possess high input impedance. Tn the Ru reference, they added an inductor to resonate with the parasitic capacitor of LNA while in the Sun reference, they added a capacitor for the internal inductor of NMR coil. However, their first-order matching networks contain only two components and one of them is fixed, resulting in fixed voltage gain and NF. In addition, the bandwidth for matching is very narrow in this electrical matching technique since the first-order matching network only generates a sharp voltage peak at the operating frequency, leading to ringing after the main signal pulse.

[0005] When applying the electrical matching technique to ultrasound imaging, the ringing will limit the resolution of constructed ultrasound images. In this regard, as shown in FIG. IB, a broadband electrical matching technique is proposed in Yang ei al.,Broadband Resonant Noise Matching Technique for Piezoelectric Ultrasound Transducers,” in IEEE Sensors Journal, vol. 20, no. 8, pages 4290-4299, 15 April 2020 (hereinafter referred to as the Yang reference) to broaden the passband of the matching network and reduce the ringing artifact in the electrical signal. In addition, a de-Q resister is applied in the matching network to make the gain in passband flatter. Consequently, the ringing artifact in the electrical signal is mitigated compared with the narrow-band matching described above. However, the ringing artifact is not completely or sufficiently removed. Furthermore, due to the passive enhancement (i.e., the above-mentioned de-Q resister), there may be an increase in the ringing artifact in the electrical signal.

[0006] A need therefore exists to provide a method of ultrasound imaging, as well as an ultrasound imaging system thereof or an ultrasound transducer and imaging system thereof, that seek to overcome, or at least ameliorate, one or more deficiencies in conventional ultrasound imaging methods, and more particularly, that improves or enhances a quality or resolution of the ultrasound image constructed, including attenuation of ringing artifact caused by a matching network implemented in an ultrasound transducer system. It is against this background that the present invention has been developed.SUMMARY

[0007] According to a first aspect of the present invention, there is provided a method of ultrasound imaging based on an analog signal from an ultrasound transducer system. The ultrasound transducer system comprises: a piezoelectric ultrasound transducer; a resonance matching network connected to the piezoelectric ultrasound transducer and configured to receive an electrical signal from the piezoelectric ultrasound transducer and provide a passive voltage gain; and a low-noise amplifier connected to the resonance matching network and configured to receive and amplify the electrical signal from the resonance matching network. The method comprises: converting the analog signal received from the ultrasound transducer system to a digital signal; attenuating ringing artifact in the digital signal based on a ringing artifact attenuation machine learning model to produce a ringing artifact attenuated digital signal; constructing an ultrasound image by beamforming imaging based on the ringing artifact attenuated digital signal; and enhancing a resolution of the ultrasound image based on a resolution optimization function to produce an enhanced ultrasound image, the resolution optimization function being configured based on a point spread function (PSF) of a transducer degradation model of the piezoelectric ultrasound transducer.

[0008] According to a second aspect of the present invention, there is provided an ultrasound imaging system for ultrasound imaging based on an analog signal from an ultrasound transducer system. The ultrasound transducer system comprises: a piezoelectric ultrasound transducer; a resonance matching network connected to the piezoelectric ultrasound transducer and configured to receive an electrical signal from the piezoelectric ultrasound transducer and provide a passive voltage gain; and a low-noise amplifier connected to the resonance matching network and configured to receive and amplify the electrical signal from the resonance matching network. The ultrasound imaging system comprises: at least one memory; and at least one processor communicatively coupled to the at least one memory and configured to: convert the analog signal received from the ultrasound transducer system to a digital signal;attenuate ringing artifact in the digital signal based on a ringing artifact attenuation machine learning model to produce a ringing artifact attenuated digital signal; construct an ultrasound image by beamforming imaging based on the ringing artifact attenuated digital signal; and enhance a resolution of the ultrasound image based on a resolution optimization function to produce an enhanced ultrasound image, the resolution optimization function being configured based on a PSF of a transducer degradation model of the piezoelectric ultrasound transducer.

[0009] According to a third aspect of the present invention, there is provided a computer program product, embodied in one or more non-transitory computer-readable storage mediums, comprising instructions executable by at least one processor to perform the method of ultrasound imaging according to the above-mentioned first aspect of the present invention.

[0010] According to a fourth aspect of the present invention, there is provided an ultrasound transducer and imaging system comprising: the ultrasound imaging system for ultrasound imaging according to the above-mentioned second aspect of the present invention; and an ultrasound transducer system communicatively coupled to the ultrasound imaging system and comprises: a piezoelectric ultrasound transducer configured to produce an electrical signal based on an ultrasound signal received; a resonance matching network connected to the piezoelectric ultrasound transducer and configured to receive the electrical signal from the piezoelectric ultrasound transducer and provide a passive voltage gain; and a low-noise amplifier connected to the resonance matching network and configured to receive and amplify the electrical signal from the resonance matching network and output the amplified electrical signal as an analog signal for the ultrasound imaging system for ultrasound imaging.BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Embodiments of the present invention will be better understood and readily apparent to one of ordinary skill in the art from the following written description, by way of example only, and in conjunction with the drawings, in which:FIG. 1A depicts a schematic diagram of a first-order low-pass filter with a passive amplification matching configuration which may be applied to provide voltage amplification and noise figure (NF) reduction when the inductor resonates with the capacitor at operating frequency;FIG. IB depicts a broadband electronic matching configuration to broaden the passband of a matching network;FIG. 2 depicts a schematic diagram of a method of ultrasound imaging based on an analog signal from an ultrasound transducer system, according to various embodiments of the present invention;FIG. 3 depicts a schematic block diagram of an ultrasound imaging system for ultrasound imaging based on an analog signal from an ultrasound transducer system, according to various embodiments of the present invention;FIG. 4 depicts a schematic drawing of an ultrasound transducer and imaging system, according to various embodiments of the present invention;FIG. 5 depicts a schematic block diagram of an ultrasound transducer and imaging system, according to various example embodiments of the present invention;FIG. 6 depicts a schematic diagram of an example resonance matching network according to various example embodiments of the present invention configured to provide a passive voltage gain;FIG. 7 depicts a schematic diagram (equivalent noise circuit) of an example ultrasound transducer system (or an example ultrasound receiver) comprising a piezoelectric ultrasound transducer; and a resonance matching network and a low-noise amplifier (LNA), according to various example embodiments of the present invention;FIG. 8 depicts a schematic diagram (thermal noise distribution) of the LNA, according to various example embodiments of the present invention;FIGs. 9A to 9C show the simulation results of the equivalent input-referred noise of the LNA (i.c., noise variations) for different sizes of the input and load transistors;FIG. 10 depicts an example deep neural network for a ringing artifact attenuation machine learning model, according to various example embodiments of the present invention;FIG. 1 1 depicts a schematic block diagram illustrating the method of obtaining the ringing artifact attenuation training database for training the ringing artifact attenuation machine learning model described above, according to various example embodiments of the present invention;FIG. 12A illustrates a plane circular transducer and its field pattern, according to various example embodiments of the present invention;FIG. 12B illustrates a spherical focused transducer and its field pattern at a focal plane, according to various example embodiments of the present invention;FIG. 13 depicts a schematic flow diagram of an image enhancing method, according to various example embodiments of the present invention;FIG. 14 shows the sensitivity improvement illustrated by a transient output for an acoustic signal after applying an electrical matching network according to various example embodiments of the present invention; andFIG. 15 shows enhancement of in vivo results from ultrasonic imaging, according to various example embodiments of the present invention.DETAILED DESCRIPTION room Various embodiments of the present invention relate to a method of ultrasound imaging, and a system thereof, as well as an ultrasound transducer and imaging system.

[0013] As described in the background, although conventional matching techniques or matching networks may be implemented in ultrasound transducer systems to provide passive voltage amplification or gain and reduce noise figure (NF), such matching techniques or matching networks introduce ringing artifact in the electrical signal, deteriorating or limiting the quality or resolution of the ultrasound image constructed. In this regard, various embodiments of the present invention seek to provide a method of ultrasound imaging, as well as an ultrasound imaging system thereof and an ultrasound transducer and imaging system thereof, that seek to overcome, or at least ameliorate, one or more deficiencies in conventional ultrasound imaging methods, and more particularly, that improves or enhances a quality or resolution of the ultrasound image constructed, including attenuation of ringing artifacts caused by a matching network implemented in an ultrasound transducer system.

[0014] FIG. 2 depicts a schematic diagram of a method 200 of ultrasound imaging based on an analog signal from an ultrasound transducer system. The ultrasound transducer system comprises a piezoelectric ultrasound transducer; a resonance matching network connected to the piezoelectric ultrasound transducer and configured to receive an electrical signal from the piezoelectric ultrasound transducer and provide a passive voltage gain and a low-noise amplifier connected to the resonance matching network and configured to receive and amplify the electrical signal from the resonance matching network. The method 200 comprises: converting (at 206) the analog signal received from the ultrasound transducer system to a digital signal; attenuating (at 208) ringing artifact in the digital signal based on a ringing artifact attenuation machine learning model to produce a ringing artifact attenuated digital signal; constructing (at 210) an ultrasound image by beamforming imaging based on the ringing artifact attenuateddigital signal; and enhancing (at 212) a resolution of the ultrasound image based on a resolution optimization function to produce an enhanced ultrasound image. The resolution optimization function is configured based on a point spread function (PSF) of a transducer degradation model of the piezoelectric ultrasound transducer.

[0015] In various embodiments, the ringing artifact attenuation machine learning model is trained based on a ringing artifact attenuation training database comprising: digital signals converted from analog signals from a first training ultrasound transducer system comprising a piezoelectric ultrasound transducer, a resonance matching network and a low-noise amplifier respectively corresponding to the piezoelectric ultrasound transducer, the resonance matching network and the low-noise amplifier of the ultrasound transducer system; and digital signals converted from analog signals from a second training ultrasound transducer system comprising a piezoelectric ultrasound transducer and a low-noise amplifier respectively corresponding to the piezoelectric ultrasound transducer and the low-noise amplifier of the ultrasound transducer system, without a resonance matching network corresponding to the resonance matching network of the ultrasound transducer system.

[0016] In various embodiments, the above-mentioned enhancing (at 212) the resolution of the ultrasound image further comprises further enhancing the enhanced ultrasound image based on a target biological structure machine learning model to produce a further enhanced ultrasound image. The target biological structure machine learning model is trained based on target biological structure images.

[0017] In various embodiments, the above-mentioned enhancing (at 212) the resolution of the ultrasound image based on the resolution optimization function to produce the enhanced ultrasound image and the above-mentioned further enhancing the enhanced ultrasound image based on a target biological structure machine learning model to produce the further enhanced ultrasound image are performed in each iteration of a plurality of iterations. In this regard, the further enhanced ultrasound image produced by an iteration of the plurality of iterations is input to a next iteration of the plurality of iterations as an ultrasound image for enhancing the resolution thereof.

[0018] In various embodiments, in the ultrasound transducer system, the resonance matching network is configured to increase the impedance of the piezoelectric ultrasound transducer to provide the passive voltage gain for the low-noise amplifier (LNA), which contributes to the ringing artifact in the digital signal.

[0019] In various embodiments, the resonance matching network comprises a second order LC circuit comprising two inductors and two capacitors arranged in a ladder configuration, and the inductances of the two inductors are optimized to maximize the passive voltage gain of the resonance matching network and minimize a noise figure (NF) of the low-noise amplifier. Furthermore, a first-stage resonance frequency is configured to be lower than a second-stage resonance frequency of the second order LC circuit.

[0020] FIG. 3 depicts a schematic block diagram of an ultrasound imaging system 300 for ultrasound imaging based on an analog signal from an ultrasound transducer system, according to various embodiments of the present invention, corresponding to the above-mentioned method 200 of ultrasound imaging as described hereinbefore with reference to FIG. 2 according to various embodiments of the present invention. Accordingly, the ultrasound transducer system comprises a piezoelectric ultrasound transducer; a resonance matching network connected to the piezoelectric ultrasound transducer and configured to receive an electrical signal from the piezoelectric ultrasound transducer and provide a passive voltage gain; and a low-noise amplifier connected to the resonance matching network and configured to receive and amplify the electrical signal from the resonance matching network. The ultrasound imaging system 300 comprises: at least one memory 302; and at least one processor 304 communicatively coupled to the at least one memory 302 and configured to: convert the analog signal received from the ultrasound transducer system to a digital signal; attenuate ringing artifact in the digital signal based on a ringing artifact attenuation machine learning model to produce a ringing artifact attenuated digital signal; construct an ultrasound image by beamforming imaging based on the ringing ar tifact attenuated digital signal; and enhance a resolution of the ultrasound image based on a resolution optimization function to produce an enhanced ultrasound image. The resolution optimization function is configured based on a point spread function (PSF) of a transducer degradation model of the piezoelectric ultrasound transducer.

[0021] It will be appreciated by a person skilled in the art that the at least one processor 304 may be configured to perform various functions or operations through set(s) of instructions (e.g., software modules) executable by the at least one processor 304 to perform various functions or operations. Accordingly, as shown in FIG. 3, the ultrasound imaging system 300 may comprise: an analog-to-digital converting module (or an analog-to-digital converting circuit or an analog-to-digital converter (ADC)) 306 configured to convert the analog signal received from the ultrasound transducer system to a digital signal; a ringing artifact attenuation module (or a ringing artifact attenuation circuit) 308 configured to attenuate ringing artifact inthe digital signal based on a ringing artifact attenuation machine learning model to produce a ringing artifact attenuated digital signal; an ultrasound image constructing module (an ultrasound image constructing circuit) 310 configured to construct an ultrasound image by beamforming imaging based on the ringing artifact attenuated digital signal; and an image resolution enhancing module (or an image resolution enhancing circuit) 312 configured to enhance a resolution of the ultrasound image based on a resolution optimization function to produce an enhanced ultrasound image. The resolution optimization function is configured based on a PSF of a transducer degradation model of the piezoelectric ultrasound transducer.

[0022] It will be appreciated by a person skilled in the art that the above-mentioned modules are not necessarily separate modules, and two or more modules may be realized by or implemented as one functional module (e.g., a circuit or a software program) as desired or as appropriate without deviating from the scope of the present invention. For example, two or more of the analog-to-digital converting module 306, the ringing artifact attenuating module 308, the ultrasound image constructing module 310 and the image resolution enhancing module 312 may be realized (e.g., compiled together) as one executable software program, which for example may be stored in the at least one memory 302 and executable by the at least one processor 304 to perform the corresponding functions or operations as described herein according to various embodiments of the present invention.

[0023] In various embodiments, the ultrasound imaging system 300 corresponds to the method 200 of ultrasound imaging as described hereinbefore with reference to FIG. 2, therefore, various operations, functions or steps configured to be performed by the least one processor 304 may correspond to various operations, functions or steps of the method 200 of ultrasound imaging described hereinbefore according to various embodiments, and thus need not be repeated with respect to the ultrasound imaging system 300 for clarity and conciseness. In other words, various embodiments described herein in context of methods (e.g., the method 200 of ultrasound imaging) are analogously valid for the corresponding systems or devices (e.g., the ultrasound imaging system 300), and vice versa. For example, in various embodiments, the at least one memory 302 may have stored therein the analog-to-digital converting module 306, the ringing artifact attenuating module 308, the ultrasound image constructing module 310 and / or the image resolution enhancing module 312, which respectively correspond to various operations, functions or steps of the method 200 of ultrasound imaging as described hereinbefore according to various embodiments, which are executable by the at least one processor 304 to perform the corresponding operations, functions or steps as described herein.

[0024] A computing system, a controller, a microcontroller or any other system providing a processing capability may be provided according to various embodiments in the present invention. Such a system may be taken to include one or more processors and one or more computer-readable storage mediums. For example, the ultrasound imaging system 300 described hereinbefore includes at least one processor 304 and at least one computer- readable storage medium (or memory) 302 which are for example used in various processing carried out therein as described herein. A memory or computer-readable storage medium used in various embodiments may be a volatile memory, for example a DRAM (Dynamic Random Access Memory) or a non-volatile memory, for example a PROM (Programmable Read Only Memory), an EPROM (Erasable PROM), EEPROM (Electrically Erasable PROM), or a flash memory, c.g., a floating gate memory, a charge trapping memory, an MRAM (Magnetoresistive Random Access Memory) or a PCRAM (Phase Change Random Access Memory).

[0025] In various embodiments, a “circuit” may be understood as any kind of a logic implementing entity, which may be special purpose circuitry or a processor executing software stored in a memory, firmware, or any combination thereof. Thus, in an embodiment, a “circuit” may be a hard-wired logic circuit or a programmable logic circuit such as a programmable processor, e.g., a microprocessor (e.g., a Complex Instruction Set Computer (CISC) processor or a Reduced Instruction Set Computer (RISC) processor). A “circuit” may also be a processor executing software, e.g., any kind of computer program, e.g., a computer program using a virtual machine code, e.g., lava. Any other kind of implementation of various functions or operations may also be understood as a “circuit” in accordance with various other embodiments. Similarly, a “module” may be a portion of a system according to various embodiments in the present invention and may encompass a “circuit” as above, or may be understood to be any kind of a logic -implementing entity therefrom.

[0026] Some portions of the present disclosure may be explicitly or implicitly presented in terms of algorithms and functional or symbolic representations of operations on data within a computer memory. These algorithmic descriptions and functional or symbolic representations are the means used by those skilled in the data processing arts to convey most effectively the substance of their work to others skilled in the art. An algorithm may be, and generally, conceived to be a self-consistent sequence of steps leading to a desired result.

[0027] The present specification also discloses a system (e.g., which may also be embodied as one or more devices or apparatuses), such as the ultrasound imaging system 300, for performing various operations, functions or steps of various methods described herein. Such asystem may be specially constructed for the required purposes or may comprise a general purpose computer system selectively activated or reconfigured by a computer program stored in the computer system. In general, various algorithms that may be presented herein are not limited to being implemented or executed by any particular computer system. Alternatively, the construction of more specialized computer system to perform various operations, functions or steps of various methods described herein may be provided as desired or as appropriate without going beyond the scope of the present invention.

[0028] In addition, the present specification also at least implicitly discloses computer program(s) or softwarc / functional modulc(s), in that it would be apparent to a person skilled in the art that various operations, functions or steps of various methods described herein may be put into effect by computer code. The computer program(s) is not intended to be limited to any particular programming language and implementation thereof, and it will be appreciated by a person skilled in the art that a variety of programming languages and coding thereof may be used to implement the computer program(s). Moreover, the computer program(s) is not intended to be limited to any particular control flow as there are a variety of programming languages which can use different control flows. It will be appreciated by a person skilled in the art that a computer program may be stored on any computer-readable storage medium (non- transitory computer-readable storage medium), such as but not limited to, a magnetic disk, an optical disk or a memory chip. For example, a computer program stored on a computer-readable storage medium may be loaded and executed on a computer system to implement various operations, functions or steps of various methods described herein according to various embodiments of the present invention.

[0029] Accordingly, in various embodiments, there is provided a computer program product, embodied in one or more computer-readable storage mediums (non-transitory computer-readable storage medium), comprising instructions (e.g., the analog-to-digital converting module 306, the ringing artifact attenuating module 308, the ultrasound image constructing module 310 and / or the image resolution enhancing module 312) executable by one or more computer processors to perform a method 200 of ultrasound imaging as described hereinbefore with reference to FIG. 2 according to various embodiments of the present invention. Accordingly, various computer programs or software modules described herein may be stored in a computer program product receivable by a system therein, such as the ultrasound imaging system 300 as shown in FIG. 3, for execution by at least one processor 304 of theultrasound imaging system 300 to perform various operations, functions or steps of various methods described herein according to various embodiments of the present invention. roo3o] It will be appreciated by a person skilled in the art that various modules described herein (e.g., the analog-to-digital converting module 306, the ringing artifact attenuating module 308, the ultrasound image constructing module 310 and / or the image resolution enhancing module 312) may be software module(s) realized by computer program(s) or set(s) of instructions executable by a computer processor to perform various functions or operations. Various modules described herein (e.g., the analog-to-digital converting module 306, the ringing artifact attenuating module 308, the ultrasound image constructing module 310 and / or the image resolution enhancing module 312), together with the at least one processor 304 and the at least one memory 302, may also be implemented as hardware modulc(s) being functional hardware unit(s) designed to perform various functions or operations. More particularly, in the hardware sense, a module is a functional hardware unit designed for use with other components or modules. For example, a module may be implemented using discrete electronic components, or it can form a portion of an entire electronic circuit such as an Application Specific Integrated Circuit (ASIC) or a Field Programmable Gate Array (FPGA). Numerous other possibilities exist. It will also be appreciated by a person skilled in the art that a combination of hardware and software modules may be implemented. Furthermore, various operations, functions or steps of various methods described herein may be performed in parallel rather than sequentially as desired or as appropriate (e.g., as long as it does not render the method(s) inoperable or unsatisfactory for its intended purpose).

[0031] FIG. 4 depicts a schematic drawing of an ultrasound transducer and imaging system 400, according to various embodiments of the present invention. The ultrasound transducer and imaging system 400 comprises: the ultrasound imaging system 300 as described herein according to various embodiments of the present invention; and an ultrasound transducer system 350 communicatively coupled to the ultrasound imaging system 300 and comprises: a piezoelectric ultrasound transducer 360 configured to produce an electrical signal based on an ultrasound signal received; a resonance matching network 370 (or a resonance matching circuit) connected to the piezoelectric ultrasound transducer 360 and configured to receive the electrical signal from the piezoelectric ultrasound transducer 360 and provide a passive voltage gain; and a low-noise amplifier (LNA) 380 connected to the resonance matching network 370 and configured to receive and amplify the electrical signal from the resonance matching network370 and output the amplified electrical signal as an analog signal for the ultrasound imaging system 300 for ultrasound imaging.

[0032] In various embodiments, the resonance matching network 370 is configured to increase the impedance of the piezoelectric ultrasound transducer 360 to provide the passive voltage gain for the LNA 380, which contributes to the ringing artifact in the digital signal obtained by the ultrasound imaging system 300 based on the analog signal received from the ultrasound transducer system 300.

[0033] In various embodiments, the resonance matching network 370 comprises a second order LC circuit comprising two inductors and two capacitors arranged in a ladder configuration, and the inductances of the two inductors are optimized to maximize the passive voltage gain of the resonance matching network 370 and minimize a noise figure (NF) of the LNA 380. Furthermore, a first-stage resonance frequency is configured to be lower than a second- stage resonance frequency of the second order LC circuit.

[0034] In various embodiments, the LNA 380 has a common-source configuration for increasing its input impedance.

[0035] In various embodiments, the LNA 380 comprises an input transistor and a cascode transistor coupled to the input transistor, and the input transistor is a P-MOS input pair fabricated in an N-Well region.

[0036] In various embodiments, the low-noise amplifier comprises a common-mode feedback circuit for stabilizing a common-mode output voltage.

[0037] It will be appreciated by a person skilled in the art that the terminology used herein is for the puipose of describing various embodiments only and is not intended to be limiting of the present invention. As used herein, the singular forms “a”, “an” and “the” arc intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “comprises” and / or “comprising,” when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0038] Any reference to an element or a feature herein using a designation such as “first”, “second” and so forth does not limit the quantity or order of such elements or features, unless stated or the context requires otherwise. For example, such designations may be used herein as a convenient way of distinguishing between two or more elements or instances of an element. Thus, a reference to first and second elements does not necessarily mean that only two elementscan be employed, or that the first element must precede the second element, unless stated or the context requires otherwise. In addition, a phrase referring to “at least one of’ a list of items refers to any single item therein or any combination of two or more items therein.

[0039] In order that the present invention may be readily understood and put into practical effect, various example embodiments of the present invention will be described hereinafter by way of examples only and not limitations. It will be appreciated by a person skilled in the art that the present invention may, however, be embodied in various different forms or configurations and should not be construed as limited to the example embodiments set forth hereinafter. Rather, these example embodiments arc provided so that this disclosure will be thorough and complete, and will fully convey the scope of the present invention to those skilled in the art.

[0040] In various example embodiments, an enhanced ultrasound transducer and imaging system is provided with integrated passive resonance matching and artificial intelligence (Al) image processing. In various example embodiments, the electrical matching (EM) network and the LNA in the ultrasound transducer system are co-designed to achieve the maximum passive gain and minimum noise factor (NF). As discussed in the background, however, the electrical matching network introduces ringing artifact in the electrical signal, which deteriorates or limits the quality or resolution of the ultrasound image if not attenuated or removed. To address this technical problem, according to various example embodiments, an Al-based ringing artifact attenuating method or algorithm is provided or implemented in the ultrasound imaging system to attenuate (or suppress) or remove the adverse ringing artifact in the electrical signal due to the electrical matching network, thus advantageously enhancing the quality or resolution of the ultrasound image. In addition, an Al-based image resolution enhancing method or algorithm is further provided or implemented in the ultrasound imaging system, that is integrated with an existing beamforming imaging method or algorithm, for post image processing to enhance the quality or resolution of the ultrasound image constructed by the existing beamforming imaging method and obtain the ultrasound image with high quality or resolution.

[0041] Accordingly, the ultrasound transducer and imaging system according to various example embodiments of the present invention does not only achieve optimal or minimum NF through co-design of the electrical matching network and the LNA, but is further able to effectively attenuate (or suppress) or completely remove the ringing artifact in the main ultrasound signal output from the ultrasound transducer system using the Al-based ringing artifact attenuating method or algorithm. In addition, the ultrasound image constructed isenhanced using the Al-based image resolution enhancing method or algorithm. As a result, super-high resolution ultrasound images can be achieved. For example, various example embodiments found that the ultrasound transducer and imaging system (or the corresponding ultrasound imaging method) is able to improve the overall ultrasound imaging resolution close to physical limit, which has not achieved elsewhere.

[0042] FIG. 5 depicts a schematic block diagram of an ultrasound transducer and imaging system 500 according to various example embodiments of the present invention. The ultrasound transducer and imaging system 500 comprises an ultrasound imaging system 502 and an ultrasound transducer system 550 (which may also be referred to as an ultrasound receiver) communicatively coupled to the ultrasound imaging system 502. The ultrasound transducer system 550 comprises a piezoelectric ultrasound transducer 560 configured to produce an electrical signal based on an ultrasound signal received; a resonance matching network 570 (or a resonance matching circuit) connected to the piezoelectric ultrasound transducer 560 and configured to receive the electrical signal from the piezoelectric ultrasound transducer 560 and provide a passive voltage gain; and an analog front-end 380, comprising a LNA, connected to the resonance matching network 570 and configured to receive and amplify the electrical signal from the resonance matching network 570 and output the amplified electrical signal as an analog signal for the ultrasound imaging system 502 for ultrasound imaging. In various example embodiments, the analog front-end 380 may further comprise a filter and a variable-gain amplifier (VGA). The ultrasound imaging system 502 may comprise an analog-to-digital converter (ADC) module 506, a ringing artifact attenuating module 508 (e.g., may also be referred to as a pulse reshaping module), an ultrasound image constructing module 510 (e.g., may also be referred to as a bcamforming imaging module) and an image resolution enhancing module 512 (e.g., may also be referred to as a post image enhancement module).

[0043] The resonance matching network 570 is connected to the ultrasound transducer 560 and is co-optimized with the LNA. The ringing artifact attenuating module 508 is configured to perform pulse reshaping to attenuate ringing artifact in the digital signal based on a ringing artifact attenuation machine learning model (or an Al model) to produce a ringing artifact attenuated digital signal. The ultrasound image constructing module 510 is configured to construct an ultrasound image by beamforming imaging based on the ringing artifact attenuated digital signal. The image resolution enhancing module 512 is configured to enhance a resolution of the ultrasound image based on a resolution optimization function (Al-based) to produce an enhanced ultrasound image for image resolution enhancement. Accordingly, the ultrasoundtransducer 560 converts the ultrasound signal received to electrical signal, which then goes through the resonance matching network 570 and is amplified (passive voltage gain), thereby enhancing the SNR (signal-to-noise ratio) of the electrical signal. The AFE 580, which includes the LNA, provides voltage gain and filtering before feeding the amplified electrical signal as an analog signal to the ultrasound imaging system 502 for ultrasound imaging. In this manner, the amplitude of the analog signal is suitable to be processed by the ADC 506 and the out-of- band noise is reduced.

[0044] FIG. 6 depicts a schematic diagram of an example resonance matching network according to various example embodiments of the present invention configured to provide a passive voltage gain. In this regard, the passive voltage gain (or voltage amplification) is achieved through impedance transformation, and thus may also be referred to as an impedance transformer. As shown in FIG. 6, the resonance matching network comprises an LC circuit comprising an inductor Lmand a capacitor Cm. Resistor Rd and capacitor Co respectively represent the equivalent resistance and capacitance of the transducer’s source impedance at working frequency, and inductor Lmand capacitor Cmconstitute the resonance matching network. The series input admittance of the transducer and the inductor Lmmay be expressed as:(Equation 1 )

[0045] When a shunt capacitor is employed to resonate out the imaginary part, the resistance can be raised to:(Equation 2)

[0046] Since the output power Poutis constant if Lmand Cmare ideal, the voltage is amplified accordingly, which can be well explained by the Thevenin's theorem. Accordingly, when the input resistance is transformed from Rsito Rinm, the open terminal voltage is increased from Vs

[0047] The co-dcsign of the resonance matching network 570 with the LNA will now be described according to various example embodiments of the present invention.

[0048] FIG. 7 depicts a schematic diagram (equivalent noise circuit) of an example ultrasound transducer system (which may also be referred to as an ultrasound receiver)comprising a piezoelectric ultrasound transducer; and a resonance matching network and a LNA, according to various example embodiments of the present invention. In various example embodiments, the resonance matching network is a second order electrical matching network, which is designed for each channel. As shown, the resonance matching network comprises a second order LC circuit comprising two inductors Lmj, Lm2 and two capacitors Cmj, Cm2 and arranged in a ladder configuration. The electrical impedance of the transducer may be extracted through testing its S parameters. Then the corresponding Butterworth- Van Dyke (BVD) model may be built through tuning the element values based on the extracted impedance of the transducer. In this regard, the electrical matching network is configured to enhance or optimize the passive gain and minimize the NF of the LNA, which includes optimizing the inductances of the two inductors Lmi, Lm2. Therefore, through the resonance matching network, the impedance of the transducer is improved or increased and the passive gain is increased or optimized.

[0049] The effect of front-end (including the LNA) with the electrical matching network will now be described below according to various example embodiments of the present invention.

[0050] Referring to the equivalent noise circuit of the ultrasound transducer system shown in FIG. 7, the and epresent the transducer noise density and input-referrednoise density of the LNA, respectively. GEM and A represent the gain of the electrical matching network and the LNA, respectively. The NF of the LNA may thus be determined as:(Equation 3)

[0051] Evidently, with the increase of GEM, the input-referred noise of the LNA is2 ( V2(f} equivalently reduced. Note that if the GEM is large to make f Vn in(f) df »n'L?Adf, the NF would be near 0 dB ideally. Consequently, the NF is less affected by the noise of LNA. In this sense, in various example embodiments, the electrical matching network is configured to be high enough to relax the noise requirement of the LNA. Through this way, the enhancement of the output electrical signal (with enhanced passive gain) and the reduction (or minimization) of the NF of the LNA are obtained concurrently. In various example embodiments, according to Equation (2), to improve the gain of the electrical matching network, the inductance of the Lmis increased and the capacitance of the capacitor Cmis correspondingly reduced in order to keepthe resonant frequency at the same level. In this regard, the conventional matching technique transfers the source impedance to its conjugate value while according to various example embodiments, transfers the impedance is increased to a high impedance.

[0052] Various example embodiments identified and address a critical trade-off relating to the electrical matching network: while high gain electrical matching networks improve sensitivity, they also result in a sharper frequency response, which can lead to undesirable ringing artifacts in the time-domain signal. To address this technical problem, as will be described later below, various example embodiments introduce an Al-based ringing artifact attenuation model (c.g., corresponding to the ringing artifact attenuating module 508 to be described later below) and a resolution enhancement module (e.g., corresponding to the image resolution enhancing module 512 to be described later below) specifically to compensate for or remedy this signal degradation introduced by the steep electrical matching network filtering profile. In various example embodiments, the electrical matching network adopts a second- order resonance matching structure and is optimized for capacitive transducers using an LC ladder network, rather than heuristic or empirical tuning. Furthermore, in various example embodiments, the first-stage resonance frequency is specifically configured to be lower than the second-stage resonance frequency, which has been found to improve both gain bandwidth and flatness. This configuration provides enhanced matching and gain performance over a broader frequency range.

[0053] Another contribution of the electrical matching network is that it improves the inputSNR of the LNA. For example, le and / V represent the in-band noise powerdensity and out-of-band noise power density of the transducer, respectively. In addition, let GFMI and GEM„ represent the in-band gain and out-of-band attenuation respectively. As insignificant noise comes from the matching network, the input SNR with electrical matching may be given as:(Equation 4)

[0054] On the other hand, the input SNR without matching may be given as:(Equation 5)

[0055] Therefore, the SNR improvement may be easily obtained as:

[0056] The signal is amplified by a factor of GEMI- Simultaneously, as the matching network has out-of-band attenuation due to the filtering effect, the out-of-band noise is reduced dramatically. Hence, the input SNR with the electrical matching network according to various example embodiments is improved as explained in Equation (6).

[0057] In various example embodiments, to obtain a high input impedance in the LNA, the LNA is configured to have a common-source architecture. FIG. 8 depicts a schematic diagram (thermal noise distribution) of the LNA according to various example embodiments of the present invention. As the input of this amplifier is a capacitor which is small in 65nm process, the impedance is high in the working frequency band of the transducer. The P-MOS transistor, which is fabricated in N-Well, is adopted to build the input pair to reduce noise from the substrate further, whereby the N-Well can isolate the noise from the substrate. Moreover, as the center frequency of the acoustic signals could be low to hundreds of kHz, the flicker-noise may be dominant. Thus, in various example embodiments, P-MOS transistor is more attractive as it has lower 1 / f noise comer frequency than N-MOS transistor. To obtain a stable common-mode output voltage, a common-mode feed-back circuit is applied as shown in FIG. 8.

[0058] The thermal noise of the LNA is shown in FIG. 8, and the equivalent input-referred noise V£qof the LNA may be determined as:(Equation 7) where gm, roand V„ represent the transconductance, channel resistance and input noise of the transistors, respectively.

[0059] Evidently, the noise from the cascode transistor M4 is reduced by the intrinsic gain of M2 Thus, it is negligible.

[0060] The flicker-noise is calculated as:(Equation 8) where Vf and KF represent the flicker noise density and the noise coefficient respectively. K= [inCOxW / L, where is the electronic mobility, Cox is the parasitic capacitance of a transistor. L represent the length of the transistors.

[0061] Accordingly, in various example embodiments, the LNA is advantageously codesigned with a capacitive transducer in a 65nm process, where the small gate capacitance naturally contributes to a high impedance over the target frequency band. In various example embodiments, to further reduce substrate noise (which is particularly critical at low frequencies), a P-MOS input pair is utilized and fabricated within an N-Well, allowing effective isolation from substrate coupling. This structural design, combined with the inherently lower 1 / f noise of P-MOS devices compared to N-MOS, addresses the flicker noise challenge systematically at both device and layout levels. Additionally, in various example embodiments, a tailored common-mode feedback (CMFB) circuit is employed to maintain output stability under varying transducer loading and process conditions. These clcmcnts / componcnts and configuration advantageously results in a LNA design optimized not only for high input impedance, but also for low-noise, low-frequency capacitive sensing applications, which is distinct from generic RF LNA architectures.

[0062] FIGs. 9A to 9C show the simulation results of the equivalent input-referred noise Vtgqof the LNA (i.e., noise variations) for different sizes of the input and load transistors, namely, for L6=280 nm (FIG. 9A), L6=L2 urn (FIG. 9B), L6=2.8 urn (FIG. 9C). FIG. 9A shows the noise density when the L6=280 nm. As can be seen, the flicker noise decreases with the increase with L2 while the W / L remains the same, where L2 and Le denote the length of transistor M2 and M6, respectively. The thermal noise at high frequency is 3.3 nNI\lHz. The reducing trend is more evident in FIG. 9B and FIG. 9C. With the increase of Ls, the flicker noise is also reduced, and the thermal noise are reduced to 1 .77 nV / v / Tfz and 1 .76 nNlyTHz respectively. This trend is consistent with the conclusion of Equations (7) and (8). Accordingly, for example, the length of the transistor may be designed as 1.2pm which contributes less noise with smaller input capacitance.

[0063] With the development of the CMOS process, SoC is developing fast in various fields. The digital circuits generate considerable effects on the analog circuits on the same die. Thus, the substrate -noise is also considered:(Equation 10)

[0064] According to various example embodiments, to reduce the substrate noise, three methods are used: (1) layout isolation by deep-well between the digital and analog circuits, (2) increase the distance of digital and analog circuits, (3) use separated power supply and ground.

[0065] In various example embodiments, after the design of the LNA is finished, the input capacitance CPLNA is also confirmed. Thereafter, the parasitic capacitance in PCB is also considered to make the EM network achieve the highest passive gain. The routing of the LNA’ s input in PCB is extracted and behaves as the loading of the EM network together with the CPLNA- In this regard, the input capacitance CPI,NA is the parasitic of the input transistors, when the size and the DC point of the transistor are confirmed, the CPLNA is also confirmed. The parasitic capacitance of the routings in PCB works like the CPLNA, it will also affect the gain of the matching network as explained based on Equation (2) above.

[0066] The ringing artifact attenuating module 508 (e.g., Al-based ringing artifact attenuating algorithm) configured to attenuate or suppress the ringing artifact in the digital signal (pulse reshaping) and the image resolution enhancing module 512 (e.g., Al-based image resolution enhancing algorithm) configured to enhance a resolution of the ultrasound image constructed (post image resolution enhancement) will now be described in further detail according to various example embodiments of the present invention. In various example embodiments, the Al-based ringing artifact attenuating algorithm and the Al-based image resolution enhancing algorithm are configured to be compatible with existing ultrasound image reconstruction algorithms (e.g., DAS (delay and sum) and SAFT (synthetic aperture focusing technique) image reconstruction algorithms), such as capable plug and play. The ringing artifact attenuating module 508 is configured to attenuate (or suppress) or eliminate the side effect (the ringing artifact) introduced by the resonance matching enhancement module and the image resolution enhancing module 512 is configured to comprehensively enhance the ultrasoundimage constructed by the ultrasound image reconstruction algorithm to obtain high spatial resolution with more biological structure or tissue details.

[0067] The ringing artifact attenuating module 508 will now be described in further detail according to various example embodiments of the present invention. Firstly, before the image reconstruction algorithm, the input signal can be considered to be generated with transducer’s response function of h(t) as well as the side effect e(t) introduced by resonance matching enhancement module 508. Therefore, the output signal y(t) of a signal channel of the transducer array from the ADC, obtained with the resonance matching network, can be represented as: y(t) = s(t) * h(t) * e(t)(Equation 11) where s(t) is the latent clean signal. As explained hereinbefore, due to the introduction of resonance matching enhancement module 570, ringing artifact appears in the A-scan ID signal, which is attenuated or removed according to various example embodiments of the present invention before sending to the image reconstruction module 510. Since the ringing artifact attenuating algorithm is configured to be compatible with existing ultrasound image reconstruction algorithms, no modification of existing image reconstruction algorithm is required. The output signal y0(t) from the ADC, obtained without the resonance matching network, may be expressed as: y0(t) = s(t) * h(t)(Equation 12)

[0068] According to various example embodiments, a large amount of experimental paired training data is acquired with and without the resonance matching enhancement module. In this regard, a ringing artifact attenuation machine learning model (or Al model) is trained to provide a data driven method for mapping an output signal obtained with the resonance matching network (and thus having the ringing artifact) to an output signal with the ringing artifact attenuated or removed. For example, an example dedicated deep neural network illustrated in FIG. 10 is trained. In various example embodiments, to obtain the ringing artifact attenuation training database for training the ringing artifact attenuation machine learning model, two different experiments may be performed to obtain paired training data. In a first experiment, digital signals are obtained which are converted from analog signals from a first training ultrasound transducer system comprising a piezoelectric ultrasound transducer, a resonance matching network and a low-noise amplifier respectively corresponding to (e.g., the same as)the piezoelectric ultrasound transducer 560, the resonance matching network 570 and the low- noise amplifier of the ultrasound transducer system 550. In a second experiment, digital signals converted are obtained which are converted from analog signals from a second training ultrasound transducer system comprising a piezoelectric ultrasound transducer and a low-noise amplifier respectively corresponding to (e.g., the same as) the piezoelectric ultrasound transducer 560 and the low-noise amplifier of the ultrasound transducer system 550, without a resonance matching network corresponding to the resonance matching network 570 of the ultrasound transducer system 550. From the digital signals collected from the two different experiments, corresponding paired data for each channel can thus be extracted for training. FIG. 11 depicts a schematic block diagram illustrating the method of obtaining the ringing artifact attenuation training database for training the ringing artifact attenuation machine learning model described above.

[0069] Accordingly, the ringing artifact attenuating module 508 provides an Al-based algorithm that effectively attenuates or eliminates ringing artifacts before image reconstruction. In various example embodiments, the ringing artifact attenuating module 508 is designed to be plug-and-play, meaning that it is configured to be compatible with existing ultrasound image reconstruction algorithms such as DAS (Delay-And-Sum) and SAFT (Synthetic Aperture Focusing Technique), thus requiring no modification of these traditional algorithms.

[0070] In various example embodiments, after optimization and gradient decent based on a loss function, such as Mean Squared Error (MSE) loss function, for training the ringing artifact attenuation machine learning model, the trained ringing artifact attenuation machine learning model is further applied to map the signal to its original form that is fit for image reconstruction as input signal.

[0071] The image resolution enhancing module 512 will now be described in further detail according to various example embodiments of the present invention.

[0072] Various example embodiments note that, after the ultrasound image is reconstructed by the ultrasound image constructing module 510 based on the ringing artifact attenuated digital signal, the ultrasound image reconstructed may still suffer from degraded lateral resolution due to the essential constraint posed by the beamforming imaging theory and gaussian noise with standard deviation of cr. In this regard, the lateral resolution is determined by the beamwidth in the focal region. Therefore, the beamforming theory' is first explored using the Rayleigh- Sommerfeld diffraction formula as follows:(Equation 13)

[0073] By weighting function into the Rayleigh-Sommerfeld diffraction formula, the field pattern of acoustic beam E (f^, co) may be derived as:(Equation 14) where ( ) is the first order Bessel function of the first kind.

[0074] For illustration purpose, FIG. 12A illustrates a plane circular transducer and its field pattern, and FIG. 12B illustrates a spherical focused transducer and its field pattern at a focal plane.

[0075] The PSF kernel K (which may function as a degradation kernel) may be determined as the integral of transducer’s field pattern towards the transducer frequency spectrum 7'(cn) across the whole bandwidth of the ultrasound transducer: co)dco(Equation 15)

[0076] By transforming the PSF kernel K into Toeplitz form, the degradation kernel matrix may be obtained as K. In various example embodiments, to enhance the constructed ultrasound images, an optimization problem may be formulated as follows:(Equation 16) where the <t> (x) is prior knowledge of the target imaging biological structure, which may be domain specific and can be learned externally by a clean imaging dataset, a denotes the standard variance of noise presented in the input image, and y denotes the input image (e.g., degraded input low quality image represents the data fidelity term, for aligning therestored image with both physical model and input image structures.

[0077] In various example embodiments, the relative importance of prior term (<£(x)) and data fidelity term can be adjusted or balanced by a weight parameter A. In variousexample embodiments, by introducing an ADMM (alternating direction method of multipliers) technique, an additional penalty term can be incorporated by parameter g. and the optimization problem may be split into three sub-problems as follows:(Equation 18) cfc+l ~ck ~xk+l +zk+l-(Equation 19) In the ADMM framework, a constraint term is introduced, where c = x - z. ||x — zk— ck|| corresponds to a penalty term, g denotes a weighting parameter, <r denotes the standard variance of noise, and ckdenotes an auxiliary parameter used for introducing constraint term (c.g., may be set close to 0). In Equation (18), || (xk+1— ck) — z\\2corresponds to a penalty term introduced through the ADMM framework, and denotes a weight parameter, which determines the denoising level, <?(z) refers to the prior information extracted from all the training database, which may depends on how much data are used for training. In various example embodiments, the Equation 18 is considered as a denoising problem and a machine learning model (c.g., FFDNct) is used for denoising.

[0078] FIG. 13 depicts a schematic flow diagram of an image enhancing method based on the above-mentioned three sub-problems, according to various example embodiments of the present invention. As shown in FIG. 13, a constructed ultrasound image (e.g., a low-quality input PA (photoacoustic) image) is first enhanced by a model-based resolution optimization function (corresponding to Equation 17) developed based on a transducer degradation model, and then the preliminarily enhanced image xtis sent to the trained machine learning model (e.g., FFDNet) (corresponding to Equation 18), where the noise generated with the model-based resolution optimization function is reduced or removed. For example, the FFDNet is trained by a large amount of vasculature PA images, the structure and shape of target biological structures (e.g., vasculature) are learnt by the neural network model. As illustrated in FIG. 13, this resolution enhancement based on the model-based resolution optimization function and the FFDNet may be performed for a number of iterations (e.g., 8 iterations). For each iteration, both model-based resolution optimization function and the FFDNet are performed. The four parameters K, a, g and X shown in FIG. 13 may be predefined, and during each of the iteration, the predefined values are fed into the modules for processing as shown in FIG. 13. Accordingly, the image resolution enhancing module 512 advantageously integrates model-based algorithm (i.e., the model-based resolution optimization function) and learning-based algorithm (i.e., themachine learning model (e.g., FFDNet), which has been found to significantly enhance interpret-ability and flexibility.

[0079] Therefore, in various example embodiments, a resolution of the ultrasound image is enhanced based on a resolution optimization function to produce an enhanced ultrasound image. In this regard, the resolution optimization function is configured based on a PSF of a transducer degradation model of the transducer. In various example embodiments, the enhanced ultrasound image is further enhanced based on a target biological structure machine learning model to produce a further enhanced ultrasound image. In this regard, the target biological structure machine learning model is trained based on target biological structure images. In various example embodiments, the above-mentioned enhancing the resolution of the ultrasound image based on the resolution optimization function and the above-mentioned further enhancing the enhanced ultrasound image based on a target biological structure machine learning model are performed in each iteration of a plurality of iterations, whereby the further enhanced ultrasound image produced by an iteration of the plurality of iterations is input to a next iteration of the plurality of iterations as an ultrasound image for enhancing the resolution thereof.

[0080] Various experimental results will now be discussed.

[0081] FIG. 14 shows the sensitivity improvement (>10.4 dB) illustrated by a transient output. As shown in FIG. 14, the effective acoustic signal is enhanced after applying the electrical matching network according to various example embodiments of the present invention, and also the output SNR is highly improved which means the NF is reduced dramatically. The output SNR is improved by 22.5 dB.

[0082] FIG. 15 shows the enhancement of in vivo results from ultrasonic imaging according to various example embodiments of the present invention, including (a) mouse leg vasculature image; enhancement result of (a) with FDU-Net; (c) enhancement result of (a) with total variation; (d) enhancement result of (a) with the image resolution enhancing module 512; (e) mouse ear vasculature image; (f) Enhancement result of (e) with FDU-Net; (g) enhancement result of (e) with total variation; (h) Enhancement result of (e) with the image resolution enhancing module 512. In FIG. 15, the scale bar is 1 mm. Accordingly, the enhancement results to in vivo ultrasound images by employing the image resolution enhancing module 512 can be seen in the FIG. 15.

[0083] It will be appreciated by a person skilled in the art that the method of ultrasound image, or the corresponding ultrasound transducer and imaging system, can be applied to a variety of practical applications. For example, for all the sensors based on piezoelectricmaterials, the method or system can be applied to improve the sensitivity, such as sensing sensors, imaging sensors, probes, and so on, covering ultrasonic applications in all aspects of industrial and medical fields. Especially, for the imaging sensors, the image resolution can be highly enhanced as the ringing of the ultrasound pulse can be deeply removed. Further the image enhancement algorithm according to various example embodiments can greatly improve the resolution of ultrasound image systems.

[0084] While embodiments of the invention have been particularly shown and described with reference to specific embodiments, it should be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the scope of the invention as defined by the appended claims. The scope of the invention is thus indicated by the appended claims and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced.

Claims

CLAIMS1. A method of ultrasound imaging based on an analog signal from an ultrasound transducer system, the ultrasound transducer system comprising a piezoelectric ultrasound transducer; a resonance matching network connected to the piezoelectric ultrasound transducer and configured to receive an electrical signal from the piezoelectric ultrasound transducer and provide a passive voltage gain; and a low-noise amplifier connected to the resonance matching network and configured to receive and amplify the electrical signal from the resonance matching network, the method comprising: converting the analog signal received from the ultrasound transducer system to a digital signal; attenuating ringing artifact in the digital signal based on a ringing artifact attenuation machine learning model to produce a ringing artifact attenuated digital signal; constructing an ultrasound image by beamforming imaging based on the ringing artifact attenuated digital signal; and enhancing a resolution of the ultrasound image based on a resolution optimization function to produce an enhanced ultrasound image, the resolution optimization function being configured based on a point spread function (PSF) of a transducer degradation model of the piezoelectric ultrasound transducer.

2. The method according to claim 1, wherein the ringing artifact attenuation machine learning model is trained based on a ringing artifact attenuation training database comprising: digital signals converted from analog signals from a first training ultrasound transducer system comprising a piezoelectric ultrasound transducer, a resonance matching network and a low-noise amplifier respectively corresponding to the piezoelectric ultrasound transducer, the resonance matching network and the low-noise amplifier of the ultrasound transducer system; and digital signals converted from analog signals from a second training ultrasound transducer system comprising a piezoelectric ultrasound transducer and a low-noise amplifier respectively corresponding to the piezoelectric ultrasound transducer and the low-noise amplifier of the ultrasound transducer system, without a resonance matching network corresponding to the resonance matching network of the ultrasound transducer system.

3. The method according to claim 1 or 2, wherein said enhancing the resolution of the ultrasound image further comprises further enhancing the enhanced ultrasound image based on a target biological structure machine learning model to produce a further enhanced ultrasound image, the target biological structure machine learning model is trained based on target biological structure images.

4. The method according to claim 3, wherein said enhancing the resolution of the ultrasound image based on the resolution optimization function to produce the enhanced ultrasound image and said further enhancing the enhanced ultrasound image based on a target biological structure machine learning model to produce the further enhanced ultrasound image are performed in each iteration of a plurality of iterations, wherein the further enhanced ultrasound image produced by an iteration of the plurality of iterations is input to a next iteration of the plurality of iterations as an ultrasound image for enhancing the resolution thereof.

5. The method according to any one of claims 1 to 4, wherein in the ultrasound transducer system, the resonance matching network is configured to increase the impedance of the piezoelectric ultrasound transducer to provide the passive voltage gain for the low -noise amplifier, which contributes to the ringing artifact in the digital signal.

6. The method according to claim 5, wherein the resonance matching network comprises a second order LC circuit comprising two inductors and two capacitors arranged in a ladder configuration, the inductances of the two inductors are optimized to maximize the passive voltage gain of the resonance matching network and minimize a noise figure of the low-noise amplifier, and a first-stage resonance frequency is configured to be lower than a second-stage resonance frequency of the second order LC circuit.

7. An ultrasound imaging system for ultrasound imaging based on an analog signal from an ultrasound transducer system, the ultrasound transducer system comprising a piezoelectric ultrasound transducer; a resonance matching network connected to the piezoelectric ultrasound transducer and configured to receive an electrical signal from the piezoelectric ultrasound transducer and provide a passive voltage gain; and a low-noise amplifier connected to theresonance matching network and configured to receive and amplify the electrical signal from the resonance matching network, the ultrasound imaging system comprising: at least one memory; and at least one processor communicatively coupled to the at least one memory and configured to: convert the analog signal received from the ultrasound transducer system to a digital signal; attenuate ringing artifact in the digital signal based on a ringing artifact attenuation machine learning model to produce a ringing artifact attenuated digital signal; construct an ultrasound image by beamforming imaging based on the ringing artifact attenuated digital signal; and enhance a resolution of the ultrasound image based on a resolution optimization function to produce an enhanced ultrasound image, the resolution optimization function being configured based on a point spread function (PSF) of a transducer degradation model of the piezoelectric ultrasound transducer.

8. The ultrasound imaging system according to claim 7, wherein the ringing artifact attenuation machine learning model is trained based on a ringing artifact attenuation training database comprising: digital signals converted from analog signals from a first training ultrasound transducer system comprising a piezoelectric ultrasound transducer, a resonance matching network and a low-noise amplifier respectively corresponding to the piezoelectric ultrasound transducer, the resonance matching network and the low-noisc amplifier of the ultrasound transducer system; and digital signals converted from analog signals from a second training ultrasound transducer system comprising a piezoelectric ultrasound transducer and a low-noise amplifier respectively corresponding to the piezoelectric ultrasound transducer and the low-noise amplifier of the ultrasound transducer system, without a resonance matching network corresponding to the resonance matching network of the ultrasound transducer system.

9. The ultrasound imaging system according to claim 7 or 8, wherein said enhance the resolution of the ultrasound image further comprises further enhancing the enhanced ultrasound image based on a target biological structure machine learning model to produce a furtherenhanced ultrasound image, the target biological structure machine learning model is trained based on target biological structure images.

10. The ultrasound imaging system according to claim 9, wherein said enhance the resolution of the ultrasound image based on the resolution optimization function to produce the enhanced ultrasound image and said further enhancing the enhanced ultrasound image based on a target biological structure machine learning model to produce the further enhanced ultrasound image are performed in each iteration of a plurality of iterations, wherein the further enhanced ultrasound image produced by an iteration of the plurality' of iterations is input to a next iteration of the plurality of iterations as an ultrasound image for enhancing the resolution thereof.

11. The ultrasound imaging system according to any one of claims 7 to 10, wherein in the ultrasound transducer system, the resonance matching network is configured to increase the impedance of the piezoelectric ultrasound transducer to provide the passive voltage gain for the low-noise amplifier, which contributes to the ringing artifact in the digital signal.

12. The ultrasound imaging system according to claim 11, wherein the resonance matching network comprises a second order LC circuit comprising two inductors and two capacitors arranged in a ladder configuration, the inductances of the two inductors are optimized to maximize the passive voltage gain of the resonance matching network and minimize a noise figure of the low-noise amplifier, and a first-stage resonance frequency is configured to be lower than a second-stage resonance frequency of the second order LC circuit.

13. An ultrasound transducer and imaging system comprising: the ultrasound imaging system for ultrasound imaging according to any one of claims 7 to 10; and an ultrasound transducer system communicatively coupled to the ultrasound imaging system and comprises: a piezoelectric ultrasound transducer configured to produce an electrical signal based on an ultrasound signal received;a resonance matching network connected to the piezoelectric ultrasound transducer and configured to receive the electrical signal from the piezoelectric ultrasound transducer and provide a passive voltage gain; and a low-noise amplifier connected to the resonance matching network and configured to receive and amplify the electrical signal from the resonance matching network and output the amplified electrical signal as an analog signal for the ultrasound imaging system for ultrasound imaging.

14. The ultrasound transducer and imaging system according claim 13, wherein the resonance matching network is configured to increase the impedance of the piezoelectric ultrasound transducer to provide the passive voltage gain for the low-noise amplifier, which contributes to the ringing artifact in the digital signal obtained by the ultrasound imaging system based on the analog signal received from the ultrasound transducer system.

15. The ultrasound transducer and imaging system according to claim 14, wherein the resonance matching network comprises a second order LC circuit comprising two inductors and two capacitors arranged in a ladder configuration, the inductances of the two inductors are optimized to maximize the passive voltage gain of the resonance matching network and minimize a noise figure of the low-noise amplifier, and a first-stage resonance frequency is configured to be lower than a second-stage resonance frequency of the second order LC circuit.

16. The ultrasound transducer and imaging system according to any one of claims 13 to 15, wherein the low-noise amplifier has a common- source configuration for increasing its input impedance.

17. The ultrasound transducer and imaging system according to any one of claims 13 to 16, wherein the low-noise amplifier comprises an input transistor and a cascode transistor coupled to the input transistor, and the input transistor is a P-MOS input pair fabricated in an N-Well region.

18. The ultrasound transducer and imaging system according to any one of claims 13 to 17, wherein the low-noise amplifier comprises a common-mode feed-back circuit for stabilizing a common-mode output voltage.

19. A computer program product, embodied in one or more non-transitory computer- readable storage mediums, comprising instructions executable by at least one processor to perform the method of ultrasound imaging according to any one of claims 1 to 6.

Images