U.S. patent application number 17/463265 was filed with the patent office on 2022-03-17 for modular piezoelectric sensor array with co-integrated electronics and beamforming channels.
This patent application is currently assigned to University of Southern California. The applicant listed for this patent is The Regents of the University of California, University of Southern California. Invention is credited to Thomas Matthew Cummins, Katherine W. Ferrara, Douglas N. Stephens, Robert G. Wodnicki, Qifa Zhou.
Application Number | 20220079559 17/463265 |
Document ID | / |
Family ID | 1000005989048 |
Filed Date | 2022-03-17 |
United States Patent
Application |
20220079559 |
Kind Code |
A1 |
Wodnicki; Robert G. ; et
al. |
March 17, 2022 |
MODULAR PIEZOELECTRIC SENSOR ARRAY WITH CO-INTEGRATED ELECTRONICS
AND BEAMFORMING CHANNELS
Abstract
A modular array includes modular array includes one or more
array modules. Each array module includes one or more transducer
arrays, where each of the one or more transducer arrays includes a
plurality of piezoelectric elements; a conducting interposer
arranged and configured to provide acoustic absorbing backing for
the one or more transducer arrays; and one or more Application
Specific Integrated Circuits (ASICs). The conducting interposer and
the one or more ASICs are in electrical contact with each other at
a first direct electrical interface. Additionally, the conducting
interposer and the one or more transducer arrays are in electrical
contact with each other at a second direct electrical
interface.
Inventors: |
Wodnicki; Robert G.; (Los
Angeles, CA) ; Zhou; Qifa; (Arcadia, CA) ;
Cummins; Thomas Matthew; (Venice, CA) ; Stephens;
Douglas N.; (Davis, CA) ; Ferrara; Katherine W.;
(Davis, CA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
University of Southern California
The Regents of the University of California |
Los Angeles
Oakland |
CA
CA |
US
US |
|
|
Assignee: |
University of Southern
California
Los Angeles
CA
The Regents of the University of California
Oakland
CA
|
Family ID: |
1000005989048 |
Appl. No.: |
17/463265 |
Filed: |
August 31, 2021 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
15999109 |
Aug 17, 2018 |
11134918 |
|
|
PCT/US2017/018537 |
Feb 18, 2017 |
|
|
|
17463265 |
|
|
|
|
62297008 |
Feb 18, 2016 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
A61B 8/4488 20130101;
B06B 1/0622 20130101; H01L 41/1132 20130101; H01L 41/04
20130101 |
International
Class: |
A61B 8/00 20060101
A61B008/00; B06B 1/06 20060101 B06B001/06; H01L 41/04 20060101
H01L041/04; H01L 41/113 20060101 H01L041/113 |
Goverment Interests
STATEMENT AS TO FEDERALLY SPONSORED RESEARCH
[0002] This invention was made with government support under
Contract No. P41-EB002182 awarded by the National Institutes of
Health (NIH). The government has certain rights in the invention.
Claims
1.-32. (canceled)
33. A modular ultrasound system comprising: at least two
piezoelectric sensor modules, each comprising multiple
piezoelectric elements arranged in groups of piezoelectric
elements; and a multi-channel processing unit; wherein a first
element in a first of the groups in a first piezoelectric sensor
module is coupled with a first interconnect bus line through a
first interface unit, and a second element in the first group in
the first piezoelectric sensor module is coupled with a second
interconnect bus line through a second interface unit; wherein a
first element in a first of the groups in a second of the
piezoelectric sensor modules is coupled with the first interconnect
bus line through a third interface unit; wherein the first
interconnect bus line is further coupled to a first channel in the
multi-channel processing unit, and the second interconnect bus line
is further coupled to a second channel in the multi channel
processing unit; and wherein the multi-channel processing unit is
operable to transmit ultrasound pulses to the elements of the
piezoelectric sensor modules in a first operating mode and receive
sensor signals from the elements of the piezoelectric sensor
modules in a second operating mode.
34. The modular ultrasound system of claim 33, wherein
piezoelectric elements of each of the at least two piezoelectric
sensor modules are disposed as a rectangular array of piezoelectric
elements with rows along an azimuthal direction and columns along
an elevation direction, and the groups of piezoelectric elements
are the columns of the rectangular array.
35. The modular ultrasound system of claim 34, wherein the
interconnect bus lines are distributed along an elevation direction
in the rectangular array.
36. The modular ultrasound system of claim 34, wherein the
interconnect bus lines are distributed along an azimuthal direction
in the rectangular array.
37. The modular ultrasound system of claim 34, wherein the
interconnect bus lines are distributed along both azimuthal and
elevation directions, and the modular ultrasound system further
comprises switches arranged and configured to selectively connect
channels in the multi-channel processing unit to horizontal
interconnect bus lines in the first coupling mode, and vertical
interconnect bus lines in the second coupling mode.
38. The modular ultrasound system of claim 33, wherein he interface
units comprise switching circuitry configured to selectively couple
an element in a piezoelectric sensor module of the at least two
piezoelectric sensor modules to another element in the first sensor
module to form a paired grouping.
39. The modular ultrasound system of claim 38, wherein the
switching circuitry comprises a high voltage semiconductor
switch.
40. The modular ultrasound system of claim 38, wherein the
switching circuitry comprises a low voltage semiconductor
switch.
41. The modular ultrasound system of claim 38, wherein the
switching circuitry comprises an electronically-actuated
micromechanical switch.
42. The modular ultrasound system of claim 38, wherein the
switching circuitry comprises a network of three switches which all
share a first terminal, where one of the switches has its second
terminal connected to ground.
43. The modular ultrasound system of claim 38, wherein the elements
of the paired grouping are physically located adjacent to each
other in the piezoelectric sensor module.
44. The modular ultrasound system of claim 43; wherein the elements
of the paired grouping arc part of a same one of the groups of
piezoelectric elements.
45. The modular ultrasound system of claim 43, wherein the elements
of the paired grouping are part of adjacent ones of the groups of
piezoelectric elements.
46. The modular ultrasound system of claim 38, wherein the elements
of the paired grouping are symmetrically situated relative to an
axis of the piezoelectric sensor module.
47. The modular ultrasound system of claim 38, wherein the elements
of the paired grouping are symmetrically situated relative to an
axis of an active aperture of the piezoelectric sensor module.
48. The modular ultrasound system of claim 38, wherein the
switching circuitry are actuated by locally integrated control
circuits.
49. The modular ultrasound system of claim 33, wherein the
interface units comprise electrical buffer circuits.
50. The modular ultrasound system of claim 49, wherein the
electrical buffer circuits are configured to be switched to an off
state in which they draw minimal power.
51. The modular ultrasound system of claim 48, wherein the locally
integrated control circuits are configured to store one or more
switch state bits internally.
52. The modular ultrasound system of claim 51, wherein the locally
integrated control circuits are configured to switch between stored
state bits one or more times during the second operating mode.
53. The modular ultrasound system of claim 38, wherein the
switching circuitry is configured to form the paired grouping of
the piezoelectric sensor module coupled with the first channel, and
a paired grouping of another piezoelectric sensor module of the at
least two piezoelectric sensor modules coupled with the second
channel.
54. The modular ultrasound system of claim 38, wherein the
switching circuitry is configured to form a first and second paired
grouping of the piezoelectric sensor module coupled with a first
channel, and a first and second paired grouping of another
piezoelectric sensor module of the at least two piezoelectric
sensor modules coupled to the second channel.
55. The modular ultrasound system of claim 53 or claim 54, wherein
the elements of the at least two piezoelectric sensor modules are
configured to respond to channels that operate at different
frequencies.
56. The modular ultrasound system of claim 38, wherein the
switching circuitry is configured to implement, in a first mode, a
piezoelectric sensor module with an element pitch greater than half
a transmit wavelength, and, in a second mode, a piezoelectric
sensor module with the element pitch equal to or less than half of
the transmit wavelength.
57. The modular ultrasound system of claim 38, wherein the
switching circuitry is configured to implement, in a first coupling
mode, element grouping for a coarse sampling of the piezoelectric
sensor module with a wide aperture, and, in a second coupling mode,
another element grouping for a fine sampling of the piezoelectric
sensor module with a narrow aperture.
58. The modular ultrasound system of claim 33, wherein the first
channel is configured to operate in a high power transmit mode
while the second channel operates in a low power transmit imaging
mode.
59.-60. (canceled)
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application claims benefit of priority under 35 U.S.C.
.sctn. 119(e)(1) of U.S. Provisional Application No. 62/297,008,
filed on Feb. 18, 2016, which is incorporated by reference
herein.
BACKGROUND
[0003] This specification relates to sensor arrays for imaging
systems, such as for medical and non-destructive evaluation.
[0004] Large area two dimensional (2D) ultrasound arrays for
imaging systems for medical and non-destructive evaluation
(NDE/NDT) require a very large number of interconnections between
the piezoelectric array and the respective buffering and switching
electronics. This large number of interconnects presents a
significant challenge for interconnection of the elements.
[0005] A number of different techniques have been proposed to
address this issue. These include building the transducer arrays on
a high density flexible circuit which is then connected to distal
boards with switching and buffering electronics, laminating the
transducer array directly to the electronic Application Specific
Integrated Circuits (ASICs) with an intervening flex circuit or
anisotropic conductive film (ACF), and building the transducers
directly on top of the ASICs. Electrical attachment methods include
high temperature methods such as bump-bonding solder attach and ACF
bonding, as well as bonding that uses copper pillars or gold stud
bumps. Additional more exotic methods include the use of novel
micro-electro-mechanical system-based ultrasound transducers such
as capacitive micro-machined ultrasonic transducers (cMUTs) and
piezoelectric micro-machined ultrasonic transducers (pMUTs).
[0006] For optimal performance of an ultrasound transducer, it is
preferable to utilize composites of piezo material and epoxy so
that electromechanical coupling efficiency, k.sub.t, can be
increased. Further improvements can be realized by utilizing novel
single crystal PMN-PT and PIN-PMN-PT materials which exhibit higher
kt when compared to traditional PZT piezo materials. Both of these
technologies lead to improved sensitivity and wider bandwidth,
which can be critical to implementation of novel beamforming
algorithms for improved imaging.
[0007] Furthermore, the large number of interconnects presents a
significant challenge for beamforming. A number of different
techniques have been proposed to address this issue, including
sparse arrays, micro-beamformers (SAPs), and Reconfigurable
Arrays.
SUMMARY
[0008] This specification relates to sensor arrays for imaging
systems, such as for medical and non-destructive evaluation.
[0009] With regard to interconnection of the elements, issues with
the previously proposed solutions are the following: systems
utilizing flex circuits are significantly challenged by the
limitation in trace and space widths in existing flex manufacturing
technologies. These systems also incur significant parasitic
capacitance due to the flex circuits between the transducers and
the electronics. Improvements are obtained by laminating the
transducers directly on top of the ASICs with intervening flex or
ACF interconnect. However, these techniques require that the ASIC
be thinned from 500 um to less than 50 um so it does not compromise
the axial resolution of the probe by creating significant ringing
in the transducer response. High temperature attachment methods are
detrimental to the composites and single crystal materials which
can become warped or de-poled. Monolithic methods which build the
transducers directly on top of the ASICs such as cMUTs and pMUTS
compromise the acoustic performance of the transducers themselves
and require dedicated fabrication lines to improve yield.
[0010] The proposed invention creates a system with a co-integrated
high sensitivity and wide bandwidth piezoelectric array and ASICs
while mitigating the negative effects of previous
implementations.
[0011] In one aspect, the disclosed technologies can be implemented
as a modular array including one or more array modules. Each array
module includes one or more transducer arrays, where each of the
one or more transducer arrays includes a plurality of piezoelectric
elements; a conducting interposer arranged and configured to
provide acoustic absorbing backing for the one or more transducer
arrays; and one or more Application Specific Integrated Circuits
(ASICs). The conducting interposer and the one or more ASICs are in
electrical contact with each other at a first direct electrical
interface. Additionally, the conducting interposer and the one or
more transducer arrays are in electrical contact with each other at
a second direct electrical interface.
[0012] Implementations can include one or more of the following
features. In some implementations, a width of the modular array
along an azimuthal direction and a height of the modular array
along an elevation direction can be roughly equal. In some
implementations, a width of the modular array along an azimuthal
direction can be greater than two times a height of the modular
array along an elevation direction. In some implementations, a
height of the conducting interposer is between 5.lamda. and
20.lamda., and .lamda. can be a wavelength of an ultrasound beam
emitted by the modular array. In some implementations, a width of
the modular array along an azimuthal direction can be greater than
two times a height of the modular array along an elevation
direction. In some implementations, pitches of the conducting
interposer along azimuthal and elevation directions can match
respective pitches of a transducer array.
[0013] In some implementations, the plurality of piezoelectric
elements can include a composite of PMN-PT or PIN-PMN-PT
piezoelectric material, and insulating filler material. For
example, the insulating filler material can include a
non-conducting epoxy, and the non-conducting epoxy can include one
or more of a plasticizer, or scattering balloons. In some
implementations, at least one of the one or more transducer arrays
can include multiple acoustic matching layers.
[0014] In some implementations, the conducting interposer can
include an electrically insulating grid frame with holes, and a
conducting material that is acoustically attenuating and fills the
holes of the electrically insulating grid frame. In some cases, a
width along an elevation direction and a width along an azimuthal
direction of the conducting material within a hole can each be at
least 90% of respective pitches of a transducer array of the one or
more transducer arrays. In some cases, the electrically insulating
grid frame can include a non-conducting material that is configured
to suppress transmission of lateral acoustic modes. Here, the
non-conducting material can include one or more of a solid epoxy,
an epoxy with a plasticizer, or an epoxy with scattering balloons.
Further here, the conducting material can have a same acoustic
impedance as the non-conducting filler material. In other cases,
the conducting material can include scattering balloons. In other
cases, the first direct electrical interface can include a silver
loaded epoxy that is plated with a layer of nickel and a layer of
gold. Also, the layer of nickel can be plated with a layer of
palladium.
[0015] In some implementations, the first direct electrical
interface can include a laminated layer of copper that is plated
with a layer of nickel and a layer of gold. In some
implementations, a surface of the conducting interposer adjacent
the first direct electrical interface can include a crossing
pattern of slots that are filled by silver loaded epoxy. In some
implementations, the first direct electrical interface can include
a conductive adhesive and either non-conductive spheres coated with
a conducting metal or solid conductive spheres. In some
implementations, the first direct electrical interface can include
copper pillars, or gold stud bumps.
[0016] In some implementations, the modular array can include
support structures that respectively support the at least two
piezoelectric sensor modules; and a gimbal system mechanically
coupled to the support structures and configured to cause, when
actuated, changes in position and orientation of the at least two
piezoelectric sensor modules relative to each other.
[0017] Another aspect of the disclosure can be implemented as a
method for aligning the transducer arrays of the disclosed modular
array. The method includes disposing a target in front of the one
or more transducer arrays, where a distance from the target to each
piezoelectric element of the one or more transducer arrays is
approximately the same; measuring time of flight information
corresponding to a distance between each respective piezoelectric
element and the target by transmitting and receiving ultrasound
from the respective piezoelectric element; storing the measured
time of flight information at each element in memory; and
calibrating measured signals at each respective piezoelectric
element, while imaging with the modular array, by using the stored
time of flight information.
[0018] Yet another aspect of the disclosure can be implemented as a
method for manufacturing an array module. The method includes
attaching an interposer to a semiconductor substrate of an
Application Specific Integrated Circuit (ASIC) to form a
sub-module; and attaching the sub-module to a transducer array
using a low temperature method to form the array module.
[0019] Yet another aspect of the disclosure can be implemented as
another method for manufacturing an array module. The method
includes forming a block of electrically conducting, acoustically
attenuating material on a surface of a transducer array; machining
slots in the electrically conducting, acoustically attenuating
material; filling the slots with an electrically insulating
material to form a sub-module; coating the sub-module with a metal
and patterning it to create pads; and attaching the sub-module pads
to an Application Specific Integrated Circuit (ASIC) using a low
temperature method to form the array module.
[0020] Yet another aspect of the disclosure can be implemented as
an array module that includes a three dimensionally (3D) patterned
interposer with two or more shelves; one or more transducer arrays
in direct electrical contact with the 3D patterned interposer,
where each of the one or more transducer arrays comprises a
plurality of piezoelectric elements; and application specific
integrated circuit (ASIC) chips assembled to the shelves of, and in
direct electrical contact with, the 3D patterned interposer.
[0021] Implementations can include one or more of the following
features. In some implementations, a surface of the 3D patterned
interposer that is in direct electrical contact with the transducer
arrays can be flat. In some implementations, a surface of the 3D
patterned interposer that is in direct electrical contact with the
one or more transducer arrays can be curved in one dimension. In
some implementations, a surface of the 3D patterned interposer that
is in direct electrical contact with the one or more transducer
arrays can be curved in two dimensions. In some implementations, a
surface of the 3D patterned interposer that is in direct electrical
contact with at least one of the one or more transducer arrays can
be shaped to conform to a curved transducer array.
[0022] In some implementations, the 3D patterned interposer can
include multiple interposers which have been bonded together. In
some implementations, the ASIC chips can be distributed parallel to
an azimuthal direction of the array module. In some
implementations, the ASIC chips can be distributed orthogonal to an
azimuthal direction of the array module.
[0023] In some implementations, the 3D patterned interposer can
include an embedded conducting path which connects a common top
electrode of the one or more transducer arrays to respective
terminals on the ASIC chips. In some implementations, the array
module envelope can be covered by a conducting conformal coating
that is connected to a common top electrode of the one or more
transducer arrays as well as to a terminal on at least one of the
ASIC chips.
[0024] With regard to beamforming, issues with the previously
proposed solutions include compromise on the number of active
elements or the absolute delay length as well as a reduction in the
number of raw data channels available for sophisticated beamforming
algorithms. Newly introduced programmable scanners offer 512-2048
system channels, with a broad range of imaging frequencies,
arbitrary delays, apodization on all channels, large instantaneous
dynamic range (e.g., 14 bits) and programmable transmit waveforms.
There exists a need to integrate these highly versatile ultrasound
processing systems with large 2D ultrasound arrays without
compromising the available data for advanced beamforming
algorithms.
[0025] The proposed invention creates an ultrasound system in which
a large number of beamforming channels are mapped to a large number
of sensor elements to realize a large area ultrasound array system.
The system is composed of multiple modules where each comprises an
ultrasound array directly coupled to respective processing ASICs
and a support structure.
[0026] One way to address the issue of yield for a large array is
to break the array up into smaller (e.g., 16.times.32 piezoelectric
element) modules composed of individual sub-arrays assembled to
their associated interface electronics. Each of the smaller modules
can be screened and selected for yield from a larger pool of
modules to form the final array and, thus, low cost and high yield
methods can be developed for building large area arrays. The
present application describes technologies for integrating one or
more piezoelectric arrays in an array module that, in turn, can be
integrated as part of a modular ultrasound (US) system. Thus, a
system can be implemented with a co-integrated high sensitivity and
wide bandwidth piezoelectric array and ASICs and/or an ultrasound
imaging system in which a large number of beamforming channels are
mapped to a large number of sensor elements to realize a large area
ultrasound array system composed of multiple modules, where each
module comprises an ultrasound array directly coupled to respective
processing ASICs and a support structure.
[0027] As such, in another aspect, the disclosed technologies can
be implemented as a modular ultrasound (US) system including at
least two piezoelectric sensor modules, each including multiple
piezoelectric elements arranged in groups of piezoelectric
elements; and a multi-channel processing unit. A first element in a
first of the groups in a first piezoelectric sensor module is
coupled with a first interconnect bus line through a first
interface unit, and a second element in the first group in the
first piezoelectric sensor module is coupled with a second
interconnect bus line through a second interface unit. A first
element in a first of the groups in a second of the piezoelectric
sensor modules is coupled with the first interconnect bus line
through a third interface unit. The first interconnect bus line is
further coupled to a first channel in the multi-channel processing
unit, and the second interconnect bus line is further coupled to a
second channel in the multi-channel processing unit. Additionally,
the multi-channel processing unit is operable to transmit
ultrasound pulses to the elements of the piezoelectric sensor
modules in a first operating mode and receive sensor signals from
the elements of the piezoelectric sensor modules in a second
operating mode.
[0028] Implementations can include one or more of the following
features. In some implementations, piezoelectric elements of each
of the at least two piezoelectric sensor modules can be disposed as
a rectangular array of piezoelectric elements with rows along an
azimuthal direction and columns along an elevation direction. Here,
the groups of piezoelectric elements are the columns of the
rectangular array.
[0029] In some implementations, the interconnect bus lines are
distributed along an elevation direction in the rectangular array.
In some implementations, the interconnect bus lines can be
distributed along an azimuthal direction in the rectangular array.
In some implementations, the interconnect bus lines can be
distributed along both azimuthal and elevation directions. Here,
the modular ultrasound system can include switches arranged and
configured to selectively connect channels in the multi-channel
processing unit to horizontal interconnect bus lines in the first
coupling mode, and vertical interconnect bus lines in the second
coupling mode.
[0030] In some implementations, the interface units can include
switching circuitry configured to selectively couple an element in
a piezoelectric sensor module of the at least two piezoelectric
sensor modules to another element in the first sensor module to
form a paired grouping. In some cases, the switching circuitry can
include a high voltage semiconductor switch. In some cases, the
switching circuitry can include a low voltage semiconductor switch.
In some cases, the switching circuitry can include an
electronically-actuated micromechanical switch. In some cases, the
switching circuitry can include a network of three switches which
all share a first terminal, where one of the switches has its
second terminal connected to ground. In some cases, the elements of
the paired grouping can be physically located adjacent to each
other in the piezoelectric sensor module. Here, the elements of the
paired grouping can be part of a same one of the groups of
piezoelectric elements. Alternatively, the elements of the paired
grouping can be part of adjacent ones of the groups of
piezoelectric elements. In some cases, the elements of the paired
grouping can be symmetrically situated relative to an axis of the
piezoelectric sensor module. In some cases, the elements of the
paired grouping can be symmetrically situated relative to an axis
of an active aperture of the piezoelectric sensor module.
[0031] In some cases, the switching circuitry are actuated by
locally integrated control circuits. Here, the locally integrated
control circuits can be configured to store one or more switch
state bits internally. Further here, the locally integrated control
circuits can be configured to switch between stored state bits one
or more times during the second operating mode.
[0032] In some cases, the switching circuitry can be configured to
form the paired grouping of the piezoelectric sensor module coupled
with the first channel, and a paired grouping of another
piezoelectric sensor module of the at least two piezoelectric
sensor modules coupled with the second channel. In some cases, the
switching circuitry can be configured to form (i) a first and
second paired grouping of the piezoelectric sensor module coupled
with a first channel, and (ii) a first and second paired grouping
of another piezoelectric sensor module of the at least two
piezoelectric sensor modules coupled to the second channel. In
either of the foregoing two cases, the elements of the at least two
piezoelectric sensor modules can be configured to respond to
channels that operate at different frequencies.
[0033] In some cases, the switching circuitry can be configured to
implement, in a first mode, a piezoelectric sensor module with an
element pitch greater than half a transmit wavelength, and, in a
second mode, a piezoelectric sensor module with the element pitch
equal to or less than half of the transmit wavelength. In some
cases, the switching circuitry can be configured to implement, in a
first coupling mode, element grouping for a coarse sampling of the
piezoelectric sensor module with a wide aperture, and, in a second
coupling mode, another element grouping for a fine sampling of the
piezoelectric sensor module with a narrow aperture.
[0034] In some implementations, the interface units can include
electrical buffer circuits. Here, the electrical buffer circuits
can be configured to be switched to an off state in which they draw
minimal power. In some implementations, the first channel is
configured to operate in a high power transmit mode while the
second channel operates in a low power transmit imaging mode.
[0035] Details of one or more implementations of the disclosed
technologies are set forth in the accompanying drawings and the
description below. Other features, aspects, descriptions and
potential advantages will become apparent from the description, the
drawings and the claims.
BRIEF DESCRIPTION OF THE DRAWINGS
[0036] FIGS. 1A-1B and FIG. 2 show aspects of an example of a
modular piezoelectric sensor array with co-integrated
electronics.
[0037] FIG. 3 shows an example of an interposer used to form direct
electrical connections between a piezoelectric sensor array and its
co-integrated electronics.
[0038] FIGS. 4-7 show examples of processes for attaching an
interposer between a piezoelectric sensor array and its
co-integrated electronics.
[0039] FIG. 8 shows an assembly of a piezoelectric sensor array
co-integrated with electronics via an interposer having multiple
shelves.
[0040] FIG. 9 shows examples of ground connections of piezoelectric
elements of a piezoelectric sensor array.
[0041] FIGS. 10A-10B show aspects of a modular piezoelectric sensor
array with co-integrated electronics coupled with an example of a
gimbal-based alignment system.
[0042] FIGS. 11-12 show aspects of examples of modular ultrasound
systems that use the disclosed modular piezoelectric sensor array
with co-integrated electronics and are to be coupled with an US
imaging system.
[0043] FIGS. 13A-13C show aspects of the disclosed modular
ultrasound systems coupled with an US imaging system in accordance
with an example of a coupling scheme.
[0044] FIGS. 14A-14C show aspects of the disclosed modular
ultrasound systems coupled with an US imaging system in accordance
with another example of a coupling scheme.
[0045] FIGS. 15A-15C show aspects of the disclosed modular
ultrasound systems coupled with an US imaging system in accordance
with yet another example of a coupling scheme.
DETAILED DESCRIPTION
[0046] Detailed examples of one or more implementations are
included below. As will be appreciated, these are merely
illustrative of the various possible implementations. While this
specification contains many implementation details, these should
not be construed as limitations on the scope of the invention or of
what may be claimed, but rather as descriptions of features
specific to particular embodiments of the invention. Certain
features that are described in this specification in the context of
separate embodiments can also be implemented in combination in a
single embodiment. Conversely, various features that are described
in the context of a single embodiment can also be implemented in
multiple embodiments separately or in any suitable subcombination.
Moreover, although features may be described above as acting in
certain combinations and even initially claimed as such, one or
more features from a claimed combination can in some cases be
excised from the combination, and the claimed combination may be
directed to a subcombination or variation of a subcombination.
[0047] FIG. 1A shows an example of a modular transducer system 100
configured to implement the wider intended application of the
proposed invention. The modular transducer system 100 includes a
support structure 110, an array of ASICs 140 disposed on the
support structure, and a large area array 120 of piezoelectric
elements 124 that is directly electrically connected to the array
of ASICs via an interconnecting interposer structure 130 (also
referred to simply as an interposer). The modular transducer system
100 is constructed using multiple smaller transducer modules, for
example like transducer module 102 shown in FIG. 1B. In this
example, the transducer module 102 is composed of a pair of
sub-modules each including a support structure 112A, 112B that
houses an interface ASIC 142A, 142B, an interposer 132A, 132B and a
transducer matrix 122A, 122B. As, in this example, each of the
transducer matrices 122A, 122B has 16.times.16 piezoelectric
elements, a transducer module 102 has 16.times.32 elements arranged
in 32 rows extending along an elevation direction (e.g., along the
y-axis) and 16 columns extending along an azimuthal direction
(e.g., along the x-axis). In this manner, for a piezoelectric
element 124 having a size of 1.6 mm in the elevation direction and
a size of 0.6 mm in the azimuthal direction, the transducer module
102 has a total size of about 51.2 mm in the elevation direction
and 9.6 mm in the azimuthal direction.
[0048] Referring again to FIG. 1A, a desired number of transducer
modules 102 can be stacked along the azimuthal direction to form
the modular transducer system 100. In some implementations, the
array width (i.e., its size along the azimuthal direction) is 2, 5
or 10 times greater than the array height (i.e., its size along the
elevation direction). In other implementations, the array width and
the array height are the same, e.g., within 5%.
[0049] Moreover, the transducer modules 102 of the modular
transducer system 100 can be arranged and oriented relative to each
other to cause a curvature of a surface of the large area array 120
of piezoelectric elements 124 to be zero, negative or positive
corresponding to a piezoelectric element array that is respectively
flat, concave or convex along the azimuthal direction. For the
example of concave piezoelectric element array 120 from FIG. 1A,
the modules are arranged with a radius of curvature that improves
the focusing capability of the complete array for imaging of deep
targets in tissue.
[0050] A more detailed cross-section of the transducer module 102
is shown in FIG. 2. Here, the connections from the ASIC 142 to the
transducer matrix 122 are indicated by pads 148 on the ASIC
substrate 144 which are then connected to the interposer 132 at a
vertical attachment interface 141.
[0051] The ASICs 142 implement high voltage transmit functions as
well as switching for multiplexing and pre-amplification of the
receive signals. The ASICs can also incorporate analog to digital
converters and/or digital or analog micro-beamforming
functionality. In some implementations, adjacent piezoelectric
elements 124 of the transducer matrix 122 can be selectively
coupled together along the elevation direction using switches 145Y.
Alternatively or additionally, adjacent piezoelectric elements 124
of the transducer matrix 122 can be selectively coupled together
along the azimuthal direction using switches 145X.
[0052] The transducer matrix 122 can be composed of PZT material,
PVDF, PMN-PT, PIN-PMN-PT or any other bulk material that is
commonly used to fabricate transducer arrays. In some
implementations, a composite of the piezoelectric material is used
to form the piezoelectric elements 124 of the transducer matrix
122. This can either be a 2-2 composite that is used for linear
arrays, or a 1-3 composite used for 2D arrays. The composite can be
manufactured using a dicing saw and epoxy fill, or by
micro-machining techniques. Additionally, a surface of the
composite that faces a sample to be imaged can include one or more
cast or laminated acoustic matching layers which help to improve
the coupling of acoustic energy from the composite to a surface of
the sample.
[0053] The interposer 132 provides backing for the transducer
module 102, namely it absorbs US energy that propagates from the
transducer matrix 122 backwards away from a surface to be imaged.
Additionally, the interposer 132 transmits beam forming signals
from the ASIC(s) 142 to the piezoelectric elements 124 and/or
detected signals from the piezoelectric elements to the ASIC(s).
FIG. 3 shows an example of such an interposer 132 that includes a
substrate 136 formed from an insulating material with through holes
that go from the top to the bottom of the substrate. The substrate
136, configured in this manner, is also referred to as an
electrically insulating grid frame. The through holes of the
substrate 136 are filled with a conductive material 135 to create
thru-via interconnections from the top to the bottom of the
substrate. These thru-via interconnections are also referred to as
conducting pillars. In some implementations, the conducting
material 135 (e.g., silver loaded epoxy) is also acoustically
damping such that it serves as the backing material for transducer
matrix 122. In some implementations, to further improve the
acoustic absorption of the conducting material 135, it can be
filled with embedded glass or phenolic micro-balloons.
[0054] A pitch p.sub.X along the azimuthal direction (or a pitch
p.sub.Y along the elevation direction) of the through holes of the
substrate 136 corresponds to an azimuthal pitch (or elevation
pitch) of the transducer matrix 132. In some implementations when
the transducer matrix 122 is implemented as a linear array, the
azimuthal pitch or the elevation pitch or both are of order
.lamda., where .lamda. is the wavelength of the US wave
emitted/received by the transducer matrix. In some implementations
when the transducer matrix 122 is implemented as a phased array,
the azimuthal pitch or the elevation pitch or both are of order
.lamda./2. Additionally, the transducer matrix 122 can be operated
using multiple frequencies, and the azimuthal pitch can be
different from the elevation pitch. As such, an example of
transducer matrix 122 can be operated at 1.25 MHz, 2.5 MHz, and 5
MHz. In this example, if the transducer matrix 122 is assumed to be
a linear array in azimuth, then the azimuthal pitch is designed to
be .lamda. at 5 MHz. In this manner, the azimuthal pitch can be
0.25.lamda. at 1.25 MHz (with the elements grouped as described
below in connection with FIGS. 15A-15C). If the transducer matrix
122 is assumed to be a phased array for those frequencies, then a 5
MHz operating mode can use an azimuthal pitch of 0.5 .lamda., and a
1.25 MHz operating mode can use an azimuthal pitch of 0.125
.lamda.. In elevation, a coarser pitch (e.g. 1.5 .lamda., 2
.lamda., 2.5 .lamda.) can be used, especially if the US beam
emitted by the transducer matrix 122 is not being steered in
elevation (e.g., for an application that is not suitable for
volumetric imaging because the US beam is only being focused in
elevation).
[0055] Further, a width w.sub.X along the azimuthal direction (or a
width w.sub.Y along the elevation direction) of each column of
conducting material 135 is a fraction of the corresponding pitch
p.sub.X (or pitch p.sub.Y), for instance w.sub.X (or w.sub.Y)=10%,
30%, 50%, 90%, or 95% of p.sub.X (or p.sub.Y). For large
percentages, the conducting material 135 absorbs most of the
back-emitted US energy, whereas for small percentages the
insulating material 136 absorbs most of the back-emitted US energy.
If the insulating material 136 and the conducting material 135 were
designed to cause similar attenuation and/or have similar acoustic
impedance, then intermediate percentages can also be used.
[0056] Moreover, a height H of each column of conducting material
135 is selected such that a desired degree of attenuation is caused
by the interposer backing. For instance, the height H can be 5, 10
or 20.lamda.. For instance, preferably H.about.10.lamda., depending
on the attenuating properties of the combination of conducting
material 135 and insulating material 136. For instance,
H.about.5.lamda. is possible when the material combination has very
good attenuating properties, but H.about.20.lamda. may be necessary
for a weakly attenuating material.
[0057] The substrate 136 can be fabricated using standard
interposer materials including FR4 material, ceramic, glass, or
silicon. However, in some embodiments, the substrate 136 consists
of a frame fabricated by laser or lithographic micro-machining of a
starting slab of material (e.g. laminated polyimide film, polyether
ether ketone, or acrylic). The substrate 136 creates a frame which
can then be filled with the conductive backing material 135 and
cured. The top and bottom of the substrate 136 can be coated with
patterned gold pads 134 and 138, respectively, which provide an
ohmic connection to the transducers 124 and to the ASICs 142. The
substrate 136 can also be optimally fabricated using
rapid-prototyping fabrication techniques such as stereo-lithography
or microinjection molding. Multiple different materials can be used
to perform such rapid-prototyping fabrication techniques including
(but not limited to) cured epoxy resin, and epoxy resin with
embedded scatterers.
[0058] In some implementations, the interposer substrate frame is
first created, using a 3D printer, as a sacrificial layer that
forms an insulating frame. This insulating frame is then filled
with conducting backing material 135 which is cured. After curing
of the conducting backing material 135, the 3D printed sacrificial
material is removed creating freestanding backing pillars. The
space between the pillars can be filled with an epoxy resin 136 for
structural stability. The epoxy resin 136 can be modified using a
plasticizer and/or embedded glass or phenolic micro-balloons to
reduce propagation of lateral modes.
[0059] The interposer 132 can be further fabricated by casting a
uniform block of electrically conducting, acoustically attenuating
material on the surface of the composite transducer array 122,
dicing or micro-machining slots in the block to create the
conducting backing 135, filling the slots with an electrically
isolating material 136 (e.g. epoxy), and then coating the back of
the interposer 132 with a metal film by sputtering or other
semiconductor fabrication techniques. The metal film can then be
patterned by dicing or using semiconductor lithography to create
the pads 134, 138 for connection to the ASIC 142. This latter
method can provide excellent acoustic connection of the acoustic
backing to the transducer array 122 for optimal performance.
[0060] Interconnection of the interposer 132 to the transducers 124
and to the ASICs 142 can be accomplished using known assembly
techniques which have been developed by the electronics industry.
These include solder attach, gold stud bumps, indium bumping, and
thermo-compression bonding. In addition, metal-coated micro-spheres
can be attached between the ASIC pads 148 and the interposer pads
138. However, in some embodiments, a low temperature conducting
adhesive is used to attach the interposer 132 to the ASICs 142 and
to the transducer matrix 122 above it. An underfill material (e.g.,
epoxy) can be used between the ASIC 142 and the interposer 132 and
between the interposer and the transducer matrix 122 to improve the
reliability of the assembly. In the latter cases, the underfill
material can also ensure an acoustically matched interface between
the transducer matrix 122 and the interposer 132.
[0061] Moreover, the bottom surface of the interposer 132 can be
adapted to improve assembly to the ASIC(s) 142 in the following
ways. In some implementations, a layer of silver loaded epoxy can
be cured on the bottom surface of the interposer 132. Here, in some
cases, the bottom surface of the interposer 132 can have a crossing
pattern of slots that are filled by the silver loaded epoxy. The
cured silver loaded epoxy is then plated with a layer of nickel and
a thin layer of gold. In some cases, the layer of nickel can be
plated with a layer of palladium. In other implementations, a layer
of copper can be laminated on the bottom surface of the interposer
132. The laminated layer of copper is then plated with a layer of
nickel and a thin layer of gold.
[0062] FIG. 4 illustrates one example of a method 400 for assembly
of the transducer matrix 122 to the interposer 132 and the ASIC
142's substrate. At O1, the transducer matrix 122 is received
preferably as a 1-3 composite of PMN-PT or PIN-PMN-PT material that
has an array of conducting electrode metal pads on the bottom. At
O2, the transducer matrix 122 is bonded to the interposer 132 using
one of the techniques described above. In this manner, a transducer
bonding interface 431 is formed between the transducer matrix 122
and the interposer 132. At O3, the interposer and transducer matrix
assembly is attached to the ASICs 142 directly to form an ASIC
bonding interface 441 between the interposer 132 and the substrate
144 of the ASIC. The ASIC's substrate 144 can have an array of pads
(e.g., 148 shown in FIG. 2) which have been bumped with low
temperature conducting adhesive beforehand. At O4 (not shown in
FIG. 4), the completed assembly is then cured in an oven at low
temperature. An alternate method to that shown in FIG. 4 is to
first form the ASIC bonding interface 441 to assemble the
interposer 132 to the ASICs 142, and then form the transducer
bonding interface 431 to assemble the transducer matrix 122 to the
interposer and ASIC assembly. This latter method is advantageous
since a high temperature attachment process can be used to form the
ASIC bonding interface 441 between the interposer and the ASICs
while a low temperature attachment process can be used for forming
the transducer bonding interface 431 thereby preventing damage of
this temperature-sensitive part of the assembly.
[0063] FIG. 5 shows another example of an assembly method 500.
Operations O1 and O2 of the method 500 can be similar to the
corresponding operations of method 400. Except the interposer 532
itself is received with conducting bumps which are created by
sub-dicing the electrically insulating grid frame 136 to expose the
central conducting material 135. At O3, the interposer and
transducer matrix assembly is attached to the ASIC 142 directly to
form an ASIC bonding interface 541 between the interposer 532 and
the substrate 144 of the ASIC. In this case, the ASIC interface
pads will have been prepared ahead of time with an inert
metallization capable of forming an ohmic contact with the
interposer. Non-conducting underfill material is used to secure the
mating components to each other.
[0064] FIG. 6 shows yet another example of an assembly method 600.
Operations O1 and O2 of the method 600 can be similar to the
corresponding operations of method 500. Moreover, at O3, conductive
epoxy is used to form an ASIC bonding interface 641 between the
interposer 532 and the substrate 144 of the ASIC 142 for the
attachment of the ASIC to the interposer and composite assembly. In
some cases, it may be beneficial to sub-dice the front side of the
transducer matrix 122 after bonding to the interposer 532, e.g.,
between operations O1 and O2, such as the case where the bonding
interface 431 consists of a uniform conducting epoxy layer.
[0065] For some fabrication methods, the interposer 132 may be
limited in the height which can be obtained. The height of the
interposer 132 is important for properly attenuating the coupled
acoustic energy from the transducer matrix 122. In this situation,
multiple thinner interposers 132, 132', 132'' can be stacked as
shown in FIG. 7 to realize the correct attenuation distance for the
backing. FIG. 7 shows an example of a method 700 for assembling
transducer matrix 122 to an ASIC 142 through a multi-layer
interposer formed from interposers 132, 132', 132''. Here,
operations O1, O2 and O4 of method 700 can be similar to the
corresponding operations O1, O2 and O3 of method 400. For method
700, at O3, a first inner bonding interface 733 is formed using an
electrically isolating and acoustically transparent material (e.g.,
epoxy) to fill the space between the interposers 132 and 132' that
is in-between their conducting connections 135 in order to insure
good coupling of the multiple layers of the backing stack. At O3',
a second inner bonding interface 733' is formed in a similar manner
to the one used at O3. Conducting backing material 135 for
interposers 132, 132', 132'' may be different, thereby enabling a
graded attenuation profile.
[0066] FIG. 8 shows an assembly 804 that includes a large array 822
of piezoelectric elements (also referred to as a transducer matrix)
that extends along the elevation direction (along the y-axis), a
3D-machined interposer 832, and multiple ASICs 842A', 842A'',
842B', 842B'' with their associated flex circuits 852A', 852A'',
852B', 852B''. The interposer 832 in this case can be manufactured
by laser milling or micro-machining methods such that it contains
multiple shelves (e.g., two shelves) for housing the multiple
ASICs, adjacent shelves separated by respective shoulders 834A,
834B. The importance of this embodiment is that large ASICs cannot
be fabricated with high yield and multiple smaller ASICs provide
much greater chance of yielding a complete working assembly 804. In
addition, attaching multiple ASICs to a flat interposer creates a
challenge for bringing out the ASIC interconnects to an imaging
system. This challenge can be addressed using high voltage
through-Si vias, however the technology is expensive and not mature
enough for building a large inexpensive array. For a small number
of signal channels, this challenge could also be addressed by
laminating a continuous flex circuit between the entire length of
the interposer and the associated ASICs. This later solution may
limit the number of uniquely assigned signal channels which can be
brought into the array due to the coarse pitch available for flex
circuit fabrication. Note that an attach method similar to method
400 can be used to form the transducer bonding interface 431
between the array 822 of piezoelectric elements and the interposer
832, and the ASIC bonding interfaces 441A', 441A'', 441B', 441B''
between the interposer and the respective ASICs 842A', 842A'',
842B', 842B''.
[0067] Moreover, the transducer bonding interface 431 can be shaped
(e.g., either by controlling its thickness along the elevation
direction or by appropriately shaping the interposer 832) such that
the array 822 of piezoelectric elements has a zero, negative or
positive curvature, C, along the elevation direction. In this
manner, the assembly 804 can be used as part of the transducer
module 102 shown in FIG. 1B to impart a desired curvature along the
elevation direction to the transducer module.
[0068] The transducer matrix 822 or 122 further includes a front
side electrode that provides a common ground connection shared by
all elements of the transducer matrix. FIG. 9 illustrates how the
front side electrode 972 providing the common ground connection is
electrically coupled to an ASIC associated with the transducer
matrix. In this example, an assembly 904 includes a transducer
matrix 922 coupled, via an interposer 932, with ASICs 942 disposed
on a support structure 912. Here, the interposer 932 includes
conducting backing 935 filling an insulating frame 936 (e.g.,
formed from epoxy using methods described above). The front side
electrode 972 can be formed, e.g., by sputtering Cr/Au on a top
surface of the piezoelectric elements 924 of the transducer matrix
922. In this manner, the front side electrode 972 provides a common
ground shared by all the piezoelectric elements 924 of the
transducer matrix 922 and, thus, it allows for the piezoelectric
elements to be directly connected to a ground on the ASIC 942. The
connection of the front side electrode 972 with the ground on the
ASIC 942 is formed using a ground plug 974 of the interposer 932.
Note that during operation of the transducer matrix 922, the ground
of the ASIC 942 is tied to the ground of an US imaging system.
[0069] The ground plug 974 shown in FIG. 9 can be formed by first
dicing a respective slot in the fabricated interposer 932 and
transducer matrix 922, and then casting a plug of Ag-filed epoxy.
The latter is located such that it contacts both the front side
electrode 972 and a ground ring 949 of the ASIC 942. In other
implementations, the ground plug 974 can be replaced with another
conducting path that is formed between the front side electrode 972
and the ASIC 942's ground by spraying a conducting conformal
coating over the envelope of the assembly 904. This latter method
would yield a thin ground connection which is advantageous for
reducing the acoustic dead area which surrounds each module.
[0070] FIG. 10A shows an example of a transducer module assembly
1002 that includes a transducer module 102 (as described above in
connection with FIG. 1B), a chassis module 1062 to support the
transducer module, and a flex circuit module 1052 to couple the
transducer module to a US imaging system. In some implementations,
the chassis module 1062 can include a frame 1004 (e.g., made from
anodized aluminum), a gimbal 1006 for adjusting a position of the
transducer module 102 relative to the frame, and bulkhead mounting
hardware 1008 for mounting the transducer module 102 to a desired
US imaging system. FIG. 10B shows a portion of an example of a
transducer system 1000 composed of a tiling of N.times.1 of the
module assemblies 1002, e.g., 1002', 1002'', 1002''', etc. In this
manner, the modular transducer system 100, described above in
connection with FIGS. 1A-1B, can be integrated in the transducer
system 1000 together with a chassis 1060, formed from chassis
modules 1062 described above in connection with FIG. 10A. The
system of gimbals 1006 can be used to adjust, over 6 degrees of
freedom around corresponding pivot points 1005, a position and
orientation of adjacent transducer modules 102 of the modular
transducer system 100 to minimize size of gaps between edges, and
slope difference between surfaces, of their respective transducer
matrices 122. In another embodiment, the gimbal could be an
electro-actively controlled structure (e.g. a MEMs or piezoelectric
device) enabling remote and incremental adjustment for fine tuning
the alignment during use.
[0071] In some implementations, the plurality of transducer arrays
of the transducer system 1000 can be aligned using the following
example of an alignment method. A target is disposed in front of
the plurality of transducer arrays of the transducer system 1000,
such that a distance from the target to each piezoelectric element
of the one or more transducer arrays is the same. Time of flight
information corresponding to a distance between each respective
piezoelectric element of the plurality of transducer arrays and the
target can be measured by transmitting and receiving ultrasound
from the respective piezoelectric element. The measured time of
flight information at each element is stored in memory. Moreover,
signals measured at each respective element are calibrated during
use of the transducer system 1000 for standard imaging (i.e.,
imaging performed outside of the foregoing alignment method) by
using the stored time of flight information.
[0072] The modular transducer system 100 of the transducer system
1000 can be controlled by an US imaging system to form sequences of
US beams, potentially of different apertures, that can be used for
linear array scanning. An architecture of an example of a modular
US system 1100 is illustrated in FIG. 11. Here, the modular US
system 1100 includes an array 1120 of 32.times.128=4096
piezoelectric elements and is comprised of M (e.g., M=8) individual
transducer modules, like the transducer module 102 described above
in connection with FIG. 1B, where each transducer module has a
transducer matrix 1122-k with 32.times.16=512 piezoelectric
elements, where k=1, . . . M. The modular US system 1100 further
includes interconnect bus lines 1154 arranged and configured to
couple the transducer matrices 1122-1, . . . , 1122-M of the
respective transducer modules with an US imaging system 1160. The
US imaging system 1160 is configured to function as a beamforming
system with, e.g., 512 channels, when the modular US system 1100 is
operated in source mode, and as a 512-channel detection system,
when the modular US system is operated in detector mode. The US
imaging system 1100's channels are mapped to the 32.times.128
piezoelectric element large area array 1120 by breaking it into
individual transducer matrices or banks of 16.times.32 uniquely
assigned piezoelectric elements. A 32.times.16 piezoelectric
element active aperture is translated across the 32.times.128
piezoelectric element array 1120 by selectively turning on and off
successive columns of switches in neighboring banks of
piezoelectric elements, {1122-k, 1122-(k+1)}. As described below in
connection with FIGS. 13A-13B, 14A-14B and 15A-15B, different sized
active apertures can be created by trading off the number of
channels corresponding to piezoelectric elements along the
elevation direction with the number of channels corresponding to
piezoelectric elements along the azimuthal direction by using
different switch configurations.
[0073] Switches used to implement the switching configurations
described below may be high voltage electrical switches, low
voltage electrical switches, or micro-electro-mechanical (MEMs)
switches. In some implementations, for optimal reduction in
cross-talk, individual switches can be grouped in a network of
three switches which all share a first terminal, and where one of
the switches has its second terminal connected to ground.
[0074] Operation of the foregoing architecture is as follows: the
US imaging system 1160's channels 1-32 are mapped uniquely to each
element in column #1. For example, in the first bank 1122-1, the
top left-most piezoelectric element is mapped to channel #1, the
one below it to channel #2, etc., e.g., using interconnect bus
lines 1154(m,1,r). Here, the bank index "m" represents any of the M
transducer matrices 1122-k, the column index "1" represents the
first column, and the row index "r" represents any of the 32 rows
of each column. The next column is mapped to channels #33-64, e.g.,
using interconnect bus lines 1154(m,2,r). Here, the column index
"2" represents the second column. In the next bank 1122-2, the same
channels are again mapped uniquely as shown. Note that the
interconnect bus lines 1156 can be disposed either along the
azimuthal direction or the elevation direction. In some
implementations, the interconnect bus lines 1156 can be disposed
along both the azimuthal and elevation directions. In such cases,
switches of the modular ultrasound system 1100 can be arranged and
configured to selectively connect the system channels in the
multi-channel processing unit 1160 to the azimuthally-oriented
interconnect bus lines in a first operating mode, and to the
elevationally-oriented interconnect bus lines in a second operating
mode, for instance.
[0075] Each piezoelectric element 1124 can be selected using a
single mux switch which can either be turned on or off. This switch
is part of an ASIC associated with a respective bank 1122-k of the
modular US system 1100 and is configured to select that
piezoelectric element for a transmit (i.e., source mode)/receive
(i.e., detector mode) connection to the US imaging system 1160 or
to be isolated. For instance, switches can contain locally
integrated control circuits which may further be configured to
switch between stored state bits one or more times during transmit
and receive cycles. An example of a scanning procedure for imaging
is to create a window of piezoelectric elements which translates
linearly from left to right across the face of the array 1120. Such
a window can be created by selecting which piezoelectric elements
are connected to the US imaging system 1160's channels at any
particular time.
[0076] Piezoelectric element #1 (top left-most) in bank 1122-1, and
piezoelectric element #1 in bank 1122-2 are both connected to US
imaging system 1160's channel #1 through their respective mux
switches, e.g., using interconnect bus line combinations
1154(m,1,r)+1156(1,1,r) and 1154(m,1,r)+1156(2,1,r), respectively.
Note that the interconnect bus lines are also referred to simply as
channel lines. Here, the bank index "1" represents the first
transducer matrix 1122-1 and the bank index "2" represents the
second transducer matrix 1122-2. At the start of scanning, the mux
switch in piezoelectric element #1, bank 1122-1 is turned on so
that it can transmit and receive. However, the mux switch in
piezoelectric element #1, bank 1122-2 is turned off. It does not
transmit and does not contribute to receive beamforming.
[0077] At the next stage of scanning, the active window will shift
by one column to the right. This is done by turning the switch in
piezoelectric element #1, bank 1122-1 to the off state, while
simultaneously turning the switch in piezoelectric element #1, bank
1122-2 to the on state. Similarly, all of the piezoelectric
elements in the column below piezoelectric element #1, bank 1122-1
will turn off, and all of the piezoelectric elements in column #1,
bank 1122-2 will turn on. This same procedure continues with every
new shift of the active aperture until it has translated all
completely across the array 1120, e.g., from bank 1122-3 through to
bank 1122-8.
[0078] A second feature of the array architecture, is
interconnection of piezoelectric elements within each bank 1122-k.
In some implementations, interconnection can be provided using
additional mux switches between the piezoelectric elements that
connect neighbors in a given column to each other (e.g.
piezoelectric element #1 connects to piezoelectric element #2 using
a switch 145Y, as shown in FIG. 2) or connects neighbors in
adjacent columns together (e.g. piezoelectric element #1 connects
to piezoelectric element #33 in the second column using a switch
145X, as shown in FIG. 2). In some implementations, interconnection
can also be provided using additional routing at the ASIC level to
connect mirrored piezoelectric elements together. This takes
advantage of the fact that for an array 1120 that is not steered in
the elevation direction, the beamforming delays for piezoelectric
elements that are equidistant relative to the horizontal midline
are identical. In some implementations, interconnection can also be
provided using additional routing at the ASIC level to connect
mirrored piezoelectric elements together that are symmetrically
situated relative to a horizontally (i.e., azimuthally) oriented
midpoint to the active array aperture.
[0079] In each of these cases, the grouping of piezoelectric
elements results in the freeing up of additional beamforming
channels of the US imaging system 1160. These extra beamforming
channels can be used to grow the width of the active aperture along
the azimuthal direction, as described below.
[0080] FIG. 12 illustrates an architecture of another example of a
modular US system 1200. In this example, the modular US system 1200
includes an array 1220 of M.times.(N.sub.R.times.N.sub.C)
piezoelectric elements 1224 and is comprised of M individual
transducer modules, like the transducer module 102 described above
in connection with FIG. 1B, where N.sub.R is the number of rows
(e.g., N.sub.R=8), and N.sub.C is the number of columns of a
transducer matrix 1222. In the example illustrated in FIG. 12, each
transducer module has a transducer matrix 1222-k with
8.times.N.sub.C piezoelectric elements, where k=1 . . . M. Only the
first column of each of the first four transducer matrices 1222-1,
1222-2, 1222-3, 1222-4 are shown in FIG. 12. The modular US system
1200 further includes interconnect bus lines 1154 arranged and
configured to couple the transducer matrices 1222-1, . . . , 1222-M
of the respective transducer modules with an US imaging system
1260. In this example, the US imaging system 1260 has
8.times.N.sub.C channels mapped to the M.times.(8.times.N.sub.C)
piezoelectric element large area array 1220 by breaking it into
individual transducer matrices or banks of 8.times.N.sub.C uniquely
assigned piezoelectric elements. The modular US system 1200 can be
operated in conjunction with the US imaging system 1260 in a manner
similar to the manner of operation of modular US system 1100 in
conjunction with the US imaging system 1160, as described above in
connection with FIG. 11. Operation of modular US system 1200 is
used to illustrate the ability to trade-off channel connections in
the elevation direction and in the azimuthal direction for optimal
use of array 1220's resources. As shown in FIG. 12, each
piezoelectric element in each transducer matrix 1222-k is connected
to two different channel lines 1154(m,1,r)+1158(m,1,r) using two
respective switches 1155(m,1,r). This pair of switches is part of
an interface unit, which may include other components, as described
below. Here, the bank index "m" represents any of the M transducer
matrices 1222-k, the column index "1" represents the first column,
and the row index "r" represents any of the 8 rows of each column.
A multiplicity of channel lines exist within each column, with the
number of lines being equivalent to the number of elements. With
this configuration, it is possible to connect the top piezoelectric
element and the bottom piezoelectric element of a column together
on one channel (e.g., channel 1), and more generally piezoelectric
elements are paired symmetrically relative to a midline parallel to
the azimuthal direction. Alternatively, it is possible to connect
each piezoelectric element of a column to its own separate channel
(e.g., top element to channel 1 and bottom element to channel
8).
[0081] The latter case is described below in connection with FIGS.
13A-13C. In this example, the modular US system 1200 includes eight
transducer modules 1202-1, . . . , 1202-8, each transducer module
having 8.times.16 piezoelectric elements. At ti, the modular US
system 1200 emits along the z-axis a first instance of US beam
1301(t.sub.1)--which has an aperture size d.sub.Y along the
elevation direction and d.sub.X along the azimuthal direction, and
a focal length f --by implementing a first coupling scheme 1300,
over columns 1-16, in the following manner: each piezoelectric
element 1224 of a column is connected to its own channel, e.g., the
first piezoelectric element of column 1 of module 1202-1 being
connected to channel 1 through an appropriate combination of
channel lines 1154(m,1,r)+1156(1,1,r) and a switch 1155(1,1,1) to
receive a first signal S.sub.1; and so on, the eighth piezoelectric
element of column 1 of module 1202-1 being connected to channel 8
through an appropriate combination of channel lines
1154(m,1,r)+1156(1,1,r) and a switch 1155(1,1,8) to receive an
eighth signal S.sub.8; and so on, the eighth piezoelectric element
of column 16 of module 1202-1 being connected to channel 8 through
an appropriate combination of channel lines
1154(m,1,r)+1156(1,16,r) and a switch 1155(1,16,8) to receive a
128.sup.th signal S.sub.128. In this manner, the modular US system
1200 can acquire a first "line" of thickness d.sub.X of a scanned
image of a target that is spaced apart at a distance f. At t.sub.2,
the modular US system 1200 emits along the z-axis a second instance
of US beam 1301(t.sub.2) by implementing the first coupling scheme
1300 over columns 2-17. In this manner, the modular US system 1200
can acquire a second line of thickness d.sub.X of the scanned image
of the target. And so on, at t.sub.112, the modular US system 1200
emits along the z-axis a 112.sup.th instance of US beam
1301(t.sub.112) by implementing the first coupling scheme 1300 over
columns 113-128. In this manner, the modular US system 1200 can
acquire a last (i.e., 112.sup.th) line of thickness d.sub.X of the
scanned image of the target.
[0082] The case where mirrored piezoelectric elements are connected
to the same channel frees up a second channel to be used elsewhere
in the array 1220. This allows the size of the aperture in the
azimuthal direction to be effectively doubled. This case is
described below in connection with FIGS. 14A-14C. In this example,
the same modular US system 1200 used in connection with FIGS.
13A-13C is being used. At t.sub.1, the modular US system 1200 emits
along the z-axis a first instance of US beam 1401(t.sub.1)--which
has an aperture size d'.sub.Y along the elevation direction and
d'.sub.X along the azimuthal direction (that is half d.sub.X
obtained using coupling scheme 1300 which leads to an equivalent
improvement in image resolution), for about the same focal length
f--by implementing a second coupling scheme 1400, over columns
1-32, in the following manner: each of a pair of piezoelectric
elements 1224 mirrored relative a center of a column is connected
to the pair's common channel, e.g., the 1.sup.st and 8.sup.th
piezoelectric elements of column 1 of module 1202-1 being connected
to channel 1 through an appropriate combination of channel lines
1154(m,1,r)+1156(1,1,r) and switch 1155(1,1,1) to each receive a
1.sup.st signal S.sub.1; and so on, the 4.sup.th and 5.sup.th
piezoelectric elements of column 1 of module 1202-1 being connected
to channel 4 through an appropriate combination of channel lines
1154(m,1,r)+1156(1,1,r) and switch 1155(1,1,4) to receive a
4.sup.th signal S.sub.4; and so on, the 4.sup.th and 5.sup.th
piezoelectric elements of column 16 of module 1202-2 being
connected to channel 128 through an appropriate combination of
channel lines 1154(m,1,r)+1156(1,16,r) and switch 1155(2,16,4) to
receive a 128.sup.th signal S.sub.128. In this manner, the modular
US system 1200 can acquire a first "line" of thickness d'.sub.X of
a scanned image of a target that is spaced apart at a distance f.
At t.sub.2, the modular US system 1200 emits along the z-axis a
second instance of US beam 1401(t.sub.2) by implementing the second
coupling scheme 1400 over columns 2-33. In this manner, the modular
US system 1200 can acquire a second line of thickness d'.sub.X of
the scanned image of the target. And so on, at t.sub.96, the
modular US system 1200 emits along the z-axis a 96.sup.th instance
of US beam 1401(t.sub.96) by implementing the second coupling
scheme 1400 over columns 97-128. In this manner, the modular US
system 1200 can acquire a last (i.e., 96.sup.th) line of thickness
d'.sub.X of the scanned image of the target.
[0083] Note that it is possible to focus at a desired focal depth,
f, by adding electronic delays on the different piezoelectric
elements/channels. This can be done both on transmit mode and
receive mode. On transmit mode this can be done at a single focal
depth (or in some cases a small number of depths), and on receive
mode it is done continuously with very fine resolution.
[0084] The coupling scheme 1400, described above in connection with
FIGS. 14A-14C, is based on connecting mirrored piezoelectric
elements to gain larger aperture, and improve image acquisition. A
further grouping is possible between adjacent piezoelectric
elements by integrating switches between adjacent piezoelectric
elements themselves. This grouping can be done either horizontally
(i.e., between piezoelectric elements of adjacent columns, using
switches 145X, as shown in FIG. 2) or vertically (i.e., between
piezoelectric elements of adjacent rows, using switches 1145Y, as
shown in FIG. 15A) or both. The combined groupings of adjacent
piezoelectric elements as well as mirrored piezoelectric elements
allow the aperture to be spread across four modules as illustrated
in FIGS. 15A-15C.
[0085] In this example, the same modular US system 1200 used in
connection with FIGS. 13A-13C is being used. At t.sub.1, the
modular US system 1200 emits along the z-axis a first instance of
US beam 1501(t.sub.1)--which has an aperture size d''.sub.Y along
the elevation direction and d''.sub.X along the azimuthal direction
(that is four times smaller than d.sub.X obtained using coupling
scheme 1300 which leads to an equivalent improvement in image
resolution), for about the same focal length f--by implementing a
third coupling scheme 1500, over columns 1-64, in the following
manner: One of each pair of adjacent ones of a set of four
piezoelectric elements 1224 that is mirrored relative a center of a
column is connected to the pair's common channel and to its
adjacent piezoelectric element, e.g., the 1.sup.st and 8.sup.th
piezoelectric elements of column 1 of module 1202-1 being connected
to channel 1 through an appropriate combination of channel lines
1154(m,1,r)+1156(1,1,r) and switch 1155(1,1,1), the 1.sup.st and
2.sup.nd piezoelectric elements being connected to each other
through switch 1145Y(1,2) and the 7.sup.th and 8.sup.th
piezoelectric elements being connected to each other through switch
1145Y(7,8), such that each receives a 1.sup.st signal S.sub.1; the
3.sup.rd and 6.sup.th piezoelectric elements of column 1 of module
1202-1 being connected to channel 3 through an appropriate
combination of channel lines 1154(m,1,r)+1156(1,1,r) and switch
1155(1,1,3), the 3.sup.rd and 4.sup.th piezoelectric elements being
connected to each other through switch 1145Y(3,4) and the 5.sup.th
and 6.sup.th piezoelectric elements being connected to each other
through switch 1145Y(5,6), such that each receives a 3.sup.rd
signal S.sub.3; and so on, the 2.sup.nd and 7.sup.th piezoelectric
elements of column 16 of module 1202-4 being connected to channel
127 through an appropriate combination of channel lines
1154(m,1,r)+1156(1,1,r) and switch 1155(4,16,2), the 1.sup.st and
2.sup.nd piezoelectric elements being connected to each other
through switch 1145Y(1,2) and the 7.sup.th and 8.sup.th
piezoelectric elements being connected to each other through switch
1145Y(7,8), such that each receives a 127.sup.st signal S.sub.127;
and the 4.sup.th and 5.sup.th piezoelectric elements of column 16
of module 1202-4 being connected to channel 4 through an
appropriate combination of channel lines 1154(m,1,r)+1156(1,1,r)
and switch 1155(4,16,4), the 3.sup.rd and 4.sup.th piezoelectric
elements being connected to each other through switch 1145Y(3,4)
and the 5.sup.th and 6.sup.th piezoelectric elements being
connected to each other through switch 1145Y(5,6), such that each
receives a 128.sup.th signal S.sub.128. In this manner, the modular
US system 1200 can acquire a first "line" of thickness d''.sub.X of
a scanned image of a target that is spaced apart at a distance f.
At t.sub.2, the modular US system 1200 emits along the z-axis a
second instance of US beam 1501(t.sub.2) by implementing the third
coupling scheme 1500 over columns 2-65. In this manner, the modular
US system 1200 can acquire a second line of thickness d''.sub.X of
the scanned image of the target. And so on, at t.sub.64, the
modular US system 1200 emits along the z-axis a 64.sup.th instance
of US beam 1401(t.sub.64) by implementing the third coupling scheme
1500 over columns 65-128. In this manner, the modular US system
1200 can acquire a last (i.e., 64.sup.th) line of thickness
d''.sub.X of the scanned image of the target.
[0086] The third coupling scheme 1500, described above in
connection with FIGS. 15A-15C, is based on combining mirrored and
double groupings to facilitate higher resolution than the ones
obtained based on coupling schemes 1300 and 1400. Such a "high
resolution mode" can be used in conjunction with a high frame rate
"survey" mode in the following manner. The survey mode can be
implemented when adjacent piezoelectric elements 1224 are coupled,
not along the elevation direction (as in coupling schemes 1400 and
1500), but along the azimuthal direction, e.g., using switches
145X(j,k;j+1,k), described in FIG. 2, which directly couple, in
this example the k.sup.th piezoelectric element from column j to
the k.sup.th piezoelectric element from adjacent column j+1. Using
this coupling scheme, the azimuthal pitch will be double the
azimuthal pitch of the coupling schemes 1400, 1500. So in this
case, the array 1220 can be scanned at twice the rate, of course
with half (2.times. worse) resolution. This implements a high frame
rate survey mode which can be useful for quickly obtaining a low
resolution image of a large target region, which can be implemented
prior to a high resolution mode (e.g., based on coupling scheme
1500) for obtaining a zoomed-in, detailed image of a portion of
interest of the target region, where the portion of interest has
been identified in the low resolution image.
[0087] The array 1220 can be programmed with a completely new
configuration (e.g., 1300, 1400, 1500 or other configurations) on
every transmit/receive cycle. In this way the array 1220 can
operate for example as a first window of N.sub.C.times.2
piezoelectric elements on transmit/receive cycle 1, and then
operate as a completely addressed second N.sub.C column.times.8 row
window at the center of the array on transmit/receive cycle 2.
[0088] The advantage of this highly flexible approach is that it
provides near transparent access to the individual 2D piezoelectric
elements 1224 of the array 1220 in order to enable novel
beamforming algorithms which, for example, could be used for
improving image quality in the presence of acoustic aberration or
for deep imaging at higher resolution.
[0089] Within the architectures described above in connection with
FIGS. 11 and 12, the ASICs integrated within each of the modular US
systems 1100 and 1200 can provide buffering of the signals from
each piezoelectric element 1124 or 1224. These buffers are part of
respective interface units, along with switches 1155, and are
disposed between the piezoelectric elements and their respective
switches. It is therefore possible to sum the signals of multiple
piezoelectric elements by connecting them through their switches
1155 to an operational amplifier located outside the modular US
systems 1100 and 1200 with a resistor feedback. The switch on
resistances operates with the amplifier and its feedback resistor
to form an analog summing operation for signals detected by the
piezoelectric elements. This operation can be used to implement the
groupings described above during the receive cycle.
[0090] Multiple frequencies and array pitches of .lamda./4,
.lamda./2 and .lamda. are supported in the architecture described
above in connection with FIGS. 13B-13C, 14B-14C and 15B-15C by
grouping neighboring piezoelectric elements according to their
respective operating frequencies. For example, four piezoelectric
elements at a pitch .lamda./2 can be grouped to form a single large
"piezoelectric element" at a pitch .lamda.. This grouping over the
entire array 1220 can be used to expand the array aperture to cover
a larger area in order to realize finer resolution at the given
operating frequency.
[0091] Thus, particular embodiments of the invention have been
described. Other embodiments are within the scope of the following
claims.
* * * * *