U.S. patent application number 17/499950 was filed with the patent office on 2022-05-26 for optimal ultrasound-based organ segmentation.
The applicant listed for this patent is KONINKLIJKE PHILIPS N.V.. Invention is credited to SHYAM BHARAT, DIRK BINNEKAMP, JEAN-LUC ROBERT, AMIR MOHAMMAD TAHMASEBI MARAGHOOSH.
Application Number | 20220160333 17/499950 |
Document ID | / |
Family ID | |
Filed Date | 2022-05-26 |
United States Patent
Application |
20220160333 |
Kind Code |
A1 |
BHARAT; SHYAM ; et
al. |
May 26, 2022 |
OPTIMAL ULTRASOUND-BASED ORGAN SEGMENTATION
Abstract
A segmentation selection system includes a transducer configured
to transmit and receive imaging energy for imaging a subject. A
signal processor is configured to process imaging data received to
generate processed image data. A segmentation module is configured
to generate a plurality of segmentations of the subject based on
features or combinations of features of the imaging data and/or the
processed image data. A selection mechanism is configured to select
one of the plurality of segmentations that best meets a criterion
for performing a task. A graphical user interface permits a user to
select features or combinations of features of imaging data or
processed image data to generate the plurality of segmentations and
to select a segmentation that best meets criterion for performing a
task.
Inventors: |
BHARAT; SHYAM; (ARLINGTON,
MA) ; TAHMASEBI MARAGHOOSH; AMIR MOHAMMAD;
(ARLINGTON, MA) ; ROBERT; JEAN-LUC; (CAMBRIDGE,
MA) ; BINNEKAMP; DIRK; (WEERSELO, NL) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
KONINKLIJKE PHILIPS N.V. |
Eindhoven |
|
NL |
|
|
Appl. No.: |
17/499950 |
Filed: |
October 13, 2021 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
15560055 |
Sep 20, 2017 |
11166700 |
|
|
PCT/IB2016/051454 |
Mar 15, 2016 |
|
|
|
17499950 |
|
|
|
|
62169577 |
Jun 2, 2015 |
|
|
|
62138075 |
Mar 25, 2015 |
|
|
|
International
Class: |
A61B 8/08 20060101
A61B008/08; A61B 8/00 20060101 A61B008/00; G06T 7/10 20060101
G06T007/10; G06V 10/147 20060101 G06V010/147; A61B 8/14 20060101
A61B008/14 |
Claims
1. A segmentation selection system, comprising: an ultrasound
transducer configured to transmit and receive ultrasound energy for
imaging a subject; a B-mode processor configured to process imaging
data received to generate processed image data; a segmentation
module configured to generate a plurality of segmentations of the
subject based on one or more combinations of input data and
segmentation metrics wherein the segmentation metrics are derived
from imaging data that has not been compressed for image
processing; and a graphical user interface that permits a user to
select features or combinations of features of imaging data and/or
processed image data to generate the plurality of segmentations and
to select a segmentation that best meets criterion for performing a
task.
2. The system as recited in claim 1, wherein the input data
includes at least one of raw radiofrequency data, envelope
detection data or B-mode display data
3. The system as recited in claim 1, wherein the segmentation
metrics include at least one of a statistical model signal to noise
ratio data, contrast data, texture data or edge detection data.
4. The system as recited in claim 1, further comprising an image
processor configured to automatically select a segmentation that
best meets the criterion for performing a task based upon
programmed criteria.
5. The system as recited in claim 1, further comprising a display
for displaying images of the plurality of segmentations, wherein
the images are displayed one of concurrently or sequentially on a
B-mode image.
6. A method for segmentation selection, comprising: receiving
imaging energy for imaging a subject; processing image data
received to generate processed image data; generating a plurality
of segmentations of the subject based on features or combinations
of features of raw imaging data and/or processed imaging data
wherein the raw imaging data and/or processed imaging data includes
at least one of raw radiofrequency data and envelope detection
data; and selecting at least one of the plurality of segmentations
that best meets a segmentation criterion.
7. The method as recited in claim 6, wherein selecting includes
automatically selecting a segmentation based upon programmed
criteria.
Description
CROSS-REFERENCE TO PRIOR APPLICATIONS
[0001] This application is a divisional application of U.S. patent
application Ser. No. 15/560,055, filed on Sep. 20, 2017, which in
turn is the U.S. National Phase application under 35 U.S.C. .sctn.
371 of International Application No. PCT/IB2016/051454, filed on
Mar. 15, 2016, which claims the benefit of U.S. Provisional Patent
Application No. 62/138,075, filed on Mar. 25, 2015, and U.S.
Provisional Patent Application No. 62/169,577, filed on Jun. 2,
2015. These applications are hereby incorporated by reference
herein.
BACKGROUND
Technical Field
[0002] This disclosure relates to medical instruments and more
particularly to systems, methods and user interfaces for image
component segmentation for imaging applications.
Description of the Related Art
[0003] Ultrasound (US) is a low-cost, easy-to-use imaging modality
that is widely used for intra-procedural real-time guidance and
treatment monitoring. Ultrasound image segmentation is mainly
driven by the clinical need for extracting organ boundaries in
B-mode images as a step towards dimension measurement (e.g., tumor
size and extent). The appearance of geometric boundaries of organs
in US images mainly depends on the acoustic impedance between
tissue layers. Despite its low cost and ease-of-use, US B-mode
imaging is not necessarily the most suitable for anatomical
imaging. B-mode US images, compared to other imaging modalities,
such as, magnetic resonance (MR) and computed tomography (CT),
suffer from poor signal-to-noise ratio and background speckle
noise. Existing ultrasound-based segmentation methods utilize pixel
(or voxel, in 3D) information from the B-mode images as input to a
metric calculator (where `metric` refers to the quantity
calculated, namely, signal-to-noise ratio (SNR), contrast, texture
etc.).
[0004] US B-mode images are also afflicted with the problem of poor
contrast between an organ and its immediately surrounding tissue
(e.g., prostate-bladder, prostate-rectal wall) due to isoechoic
pixel values on the B-mode images. This limits the robustness of
many existing automated segmentation methods. Manual segmentation
remains the only possible alternative, and the achievable accuracy
with this method is heavily dependent on the skill-level of the
clinician. The inter-operator variability in manual US segmentation
is on the order of 5 mm, with the Dice coefficient (a measure of
similarity) being 20-30% lower than for MRI segmentations.
SUMMARY
[0005] In accordance with the present principles, a segmentation
selection system includes a transducer configured to transmit and
receive imaging energy for imaging a subject. A signal processor is
configured to process imaging data received to generate processed
image data. A segmentation module is configured to generate a
plurality of segmentations of the subject based on features or
combinations of features of the imaging data and/or the processed
image data. A selection mechanism is configured to select one of
the plurality of segmentations that best meets a criterion for
performing a task.
[0006] Another segmentation selection system includes an ultrasound
transducer configured to transmit and receive ultrasound energy for
imaging a subject. A B-mode processor is configured to process
imaging data received to generate processed image data. A
segmentation module is configured to generate a plurality of
segmentations of the subject based on one or more combinations of
input data and segmentation metrics. A graphical user interface
permits a user to select features or combinations of features of
imaging data and/or processed image data to generate the plurality
of segmentations and to select a segmentation that best meets
criterion for performing a task.
[0007] A method for segmentation selection includes receiving
imaging energy for imaging a subject; image processing data
received to generate processed image data; generating a plurality
of segmentations of the subject based on features or combinations
of features of raw imaging data and/or processed imaging data; and
selecting at least one of the plurality of segmentations that best
meets a segmentation criterion.
[0008] These and other objects, features and advantages of the
present disclosure will become apparent from the following detailed
description of illustrative embodiments thereof, which is to be
read in connection with the accompanying drawings.
BRIEF DESCRIPTION OF DRAWINGS
[0009] This disclosure will present in detail the following
description of preferred embodiments with reference to the
following figures wherein:
[0010] FIG. 1 is a block/flow diagram showing an imaging system
with a segmentation selection module in accordance with one
embodiment;
[0011] FIG. 2 is a block/flow diagram showing data or signal inputs
to a segmentation selection module for generating a plurality of
segmentations in accordance with one embodiment;
[0012] FIG. 3 is a diagram showing an illustrative graphical user
interface for providing selection criteria and selecting a
segmentation in accordance with one embodiment; and
[0013] FIG. 4 is a flow diagram showing a segmentation selection
method in accordance with illustrative embodiments.
DETAILED DESCRIPTION OF EMBODIMENTS
[0014] In accordance with the present principles, systems and
methods are provided to estimate an optimal segmentation from
multiple segmentations of a single imaging modality (e.g.,
ultrasound (US)). The multiple segmentations may employ, e.g., raw
beam-summed US radiofrequency (RF) data and/or other data matrices
along the pipeline of B-mode (brightness mode) image formation, in
conjunction with various segmentation metrics. Both manual and
automatic methods may be selectable by a user through a user
interface to choose the optimal segmentation.
[0015] Ultrasound B-mode images are created by detecting an
envelope of raw RF data, followed by logarithmic compression and
scan conversion. The logarithmic data compression permits
concurrent visualization of a wide range of echoes. However, this
also suppresses subtle variations in image contrast that may be
vital to achieving an accurate segmentation. In accordance with the
present principles, a computation of segmentation metrics is
provided on different US data forms (prior to the generation of
conventional B-mode data). The segmentation will be derived from
one or more metrics (texture, contrast, etc.). Based on the data
matrix and segmentation metric used, multiple segmentation outputs
will be presented to the user, with the possibility of manually or
automatically choosing the optimal segmentation. The chosen
segmentation will be superimposed on the B-mode image visualized on
the screen, without adding any complexity from the user's
perspective. Overall, the unprocessed RF data is potentially rich
in information about the organ under investigation and the
boundaries between tissues. Involving several different data
streams in the segmentation process, complementary to the B-mode
image, can lead to a more accurate segmentation that is less
sensitive to speckle noise. Hence, the computation of multi-channel
segmentation metrics on pre-log compressed data is provided. The
present principles increase the accuracy and robustness of
segmentation algorithms, thereby improving and streamlining
clinical workflow.
[0016] It should be understood that the present invention will be
described in terms of medical instruments for US imaging; however,
the teachings of the present invention are much broader and are
applicable to any imaging modality where multiple segmentation
options can be provided for that modality. In some embodiments, the
present principles are employed in tracking or analyzing complex
biological or mechanical systems. In particular, the present
principles are applicable to internal tracking procedures for
biological systems, and may include procedures in all areas of the
body such as the lungs, gastro-intestinal tract, excretory organs,
blood vessels, etc. The elements depicted in the FIGS. may be
implemented in various combinations of hardware and software and
provide functions which may be combined in a single element or
multiple elements.
[0017] The functions of the various elements shown in the FIGS. can
be provided through the use of dedicated hardware as well as
hardware capable of executing software in association with
appropriate software. When provided by a processor, the functions
can be provided by a single dedicated processor, by a single shared
processor, or by a plurality of individual processors, some of
which can be shared. Moreover, explicit use of the term "processor"
or "controller" should not be construed to refer exclusively to
hardware capable of executing software, and can implicitly include,
without limitation, digital signal processor ("DSP") hardware,
read-only memory ("ROM") for storing software, random access memory
("RAM"), non-volatile storage, etc.
[0018] Moreover, all statements herein reciting principles,
aspects, and embodiments of the invention, as well as specific
examples thereof, are intended to encompass both structural and
functional equivalents thereof. Additionally, it is intended that
such equivalents include both currently known equivalents as well
as equivalents developed in the future (i.e., any elements
developed that perform the same function, regardless of structure).
Thus, for example, it will be appreciated by those skilled in the
art that the block diagrams presented herein represent conceptual
views of illustrative system components and/or circuitry embodying
the principles of the invention. Similarly, it will be appreciated
that any flow charts, flow diagrams and the like represent various
processes which may be substantially represented in computer
readable storage media and so executed by a computer or processor,
whether or not such computer or processor is explicitly shown.
[0019] Furthermore, embodiments of the present invention can take
the form of a computer program product accessible from a
computer-usable or computer-readable storage medium providing
program code for use by or in connection with a computer or any
instruction execution system. For the purposes of this description,
a computer-usable or computer readable storage medium can be any
apparatus that may include, store, communicate, propagate, or
transport the program for use by or in connection with the
instruction execution system, apparatus, or device. The medium can
be an electronic, magnetic, optical, electromagnetic, infrared, or
semiconductor system (or apparatus or device) or a propagation
medium. Examples of a computer-readable medium include a
semiconductor or solid state memory, magnetic tape, a removable
computer diskette, a random access memory (RAM), a read-only memory
(ROM), a rigid magnetic disk and an optical disk. Current examples
of optical disks include compact disk--read only memory (CD-ROM),
compact disk--read/write (CD-R/W), Blu-Ray.TM. and DVD.
[0020] Reference in the specification to "one embodiment" or "an
embodiment" of the present principles, as well as other variations
thereof, means that a particular feature, structure,
characteristic, and so forth described in connection with the
embodiment is included in at least one embodiment of the present
principles. Thus, the appearances of the phrase "in one embodiment"
or "in an embodiment", as well any other variations, appearing in
various places throughout the specification are not necessarily all
referring to the same embodiment.
[0021] It is to be appreciated that the use of any of the following
"/", "and/or", and "at least one of", for example, in the cases of
"A/B", "A and/or B" and "at least one of A and B", is intended to
encompass the selection of the first listed option (A) only, or the
selection of the second listed option (B) only, or the selection of
both options (A and B). As a further example, in the cases of "A,
B, and/or C" and "at least one of A, B, and C", such phrasing is
intended to encompass the selection of the first listed option (A)
only, or the selection of the second listed option (B) only, or the
selection of the third listed option (C) only, or the selection of
the first and the second listed options (A and B) only, or the
selection of the first and third listed options (A and C) only, or
the selection of the second and third listed options (B and C)
only, or the selection of all three options (A and B and C). This
may be extended, as readily apparent by one of ordinary skill in
this and related arts, for as many items listed.
[0022] Referring now to the drawings in which like numerals
represent the same or similar elements and initially to FIG. 1, an
ultrasound imaging system 10 constructed in accordance with the
present principles is shown in block diagram form. The ultrasound
system 10 includes a transducer device or probe 12 having a
transducer array 14 for transmitting ultrasonic waves and receiving
echo information. The transducer array may be configured as, e.g.,
linear arrays or phased arrays, and can include piezoelectric
elements or capacitive micromachined ultrasonic transducers (CMUT)
elements. The transducer array 14, for example, can include a two
dimensional array of transducer elements capable of scanning in
both elevation and azimuth dimensions for 2D and/or 3D imaging.
[0023] The transducer array 14 is coupled to a microbeamformer 16
in the probe 12, which controls transmission and reception of
signals by the transducer elements in the array. The
microbeamformer 16 may be integrated with the flexible transducer
device 12 and is coupled to a transmit/receive (T/R) switch 18,
which switches between transmission and reception and protects a
main beamformer 22 from high energy transmit signals. In some
embodiments, the T/R switch 18 and other elements in the system can
be included in the transducer probe rather than in a separate
ultrasound system base. The transmission of ultrasonic beams from
the transducer array 14 under control of the microbeamformer 16 is
directed by a transmit controller 20 coupled to the T/R switch 18
and the beamformer 22, which may receive input from the user's
operation of a user interface or control panel 24.
[0024] One function controlled by the transmit controller 20 is the
direction in which beams are steered. Beams may be steered straight
ahead from (orthogonal to) the transducer array 14, or at different
angles for a wider field of view. The partially beamformed signals
produced by the microbeamformer 16 are coupled to a main beamformer
22 where partially beamformed signals from individual patches of
transducer elements are combined into a fully beamformed
signal.
[0025] The beamformed signals are coupled to a signal processor 26.
The signal processor 26 can process the received echo signals in
various ways, such as bandpass filtering, decimation, I and Q
component separation, and harmonic signal separation. The signal
processor 26 may also perform additional signal enhancement such as
speckle reduction, signal compounding, and noise elimination. The
processed signals are coupled to a B mode processor 28 or other
mode processor, e.g., an M-mode processor 29), which can employ
amplitude detection for the imaging of structures in the body. The
signals produced by the mode processors 28, 29 are coupled to a
scan converter 30 and a multiplanar reformatter 32. The scan
converter 30 arranges the echo signals in the spatial relationship
from which they were received in a desired image format. For
instance, the scan converter 30 may arrange the echo signal into a
two dimensional (2D) sector-shaped format, or a pyramidal three
dimensional (3D) image. The multiplanar reformatter 32 can convert
echoes which are received from points in a common plane in a
volumetric region of the body into an ultrasonic image of that
plane.
[0026] A volume renderer 34 converts the echo signals of a 3D data
set into a projected 3D image as viewed from a given reference
point. The 2D or 3D images are coupled from the scan converter 30,
multiplanar reformatter 32, and volume renderer 34 to an image
processor 36 for further enhancement, buffering and temporary
storage for display on an image display 38. A graphics processor 40
can generate graphic overlays for display with the ultrasound
images. These graphic overlays or parameter blocks can contain,
e.g., standard identifying information such as patient name, date
and time of the image, imaging parameters, frame indices and the
like. For these purposes, the graphics processor 40 receives input
from the user interface 24, such as a typed patient name. The user
interface 24 can also be coupled to the multiplanar reformatter 32
for selection and control of a display of multiple multiplanar
reformatted (MPR) images.
[0027] In accordance with the present principles, ultrasound data
is acquired and stored in memory 42. The memory 42 is depicted as
being centrally placed; however, the memory 42 may store data and
interact at any position in the signal path. Corrections may be
employed as feedback for correcting a beam steering signal (Beam
Steer) in accordance with the positions of the elements in the
array 14.
[0028] Display 38 is included for viewing internal images of a
subject (patient) or volume. Display 38 may also permit a user to
interact with the system 10 and its components and functions, or
any other element within the system 10. This is further facilitated
by the interface 24, which may include a keyboard, mouse, a
joystick, a haptic device, or any other peripheral or control to
permit user feedback from and interaction with the system 10.
[0029] Ultrasound B-mode images are output from the B-mode
processor 28 and are created by detecting an envelope of raw RF
data, followed by logarithmic compression and scan conversion by
the scan converter 30. The logarithmic data compression by the
B-mode processor 28 permits concurrent visualization of a wide
range of echoes.
[0030] Ultrasound M-mode images are output from an M-mode processor
29. M-mode images may be processed (e.g., compressed) and sent for
scan conversion by the scan converter 30. In M-mode (motion mode)
ultrasound, pulses are emitted in quick succession when A-mode or
B-mode images are taken to record successive images. As the organ
boundaries produce reflections relative to the probe 12, this can
be used to determine the velocity of specific organ structures.
[0031] A computation of segmentation metrics is provided on
different US data forms (prior to the generation of conventional
B-mode data or M-mode data) and received by a segmentation module
50. The segmentation module 50 derives a plurality of segmentation
images generated from one or more metrics (e.g., texture, contrast,
etc.) measured from the US data. Based on the data matrix and
segmentation metric employed, multiple segmentation outputs will be
generated using the image processor 36 (or graphics processor 40)
for display on the display 38. The segmentation outputs are
presented to the user on a graphical user interface (GUI) 52
generated by the image processor 36 (or the graphics processor
40).
[0032] The segmentation outputs are presented to the user with the
possibility of manually or automatically choosing an optimal
segmentation. The chosen segmentation may be superimposed on the
B-mode image visualized on the display 38 by the image processor 36
(or the graphics processor 40). The unprocessed RF data provided
prior to B-mode processing 28 (or even before signal processing 26)
is potentially rich in information about an organ or region under
investigation and boundaries between tissues. Introducing several
different data types into the segmentation process by the
segmentation module 50 can be complementary to the B-mode image
from the B-mode processor 28 and can lead to a more accurate
segmentation that is less sensitive to speckle noise or other image
degradation phenomena. Hence, the computation of multi-channel
segmentation metrics on pre-log compressed data can increase the
accuracy and robustness of segmentation algorithms, thereby
improving and streamlining clinical workflow.
[0033] The image processor 36 is configured to generate a graphical
user interface (GUI) or other selection mechanism 52 for user
selection of an optimal segmentation. The segmentation module 50
provides different segmentation outputs resulting from multiple
combinations of input data and segmentation metrics to determine an
optimal segmentation for an application at hand. Optimality can
change based on the organ being segmented and the task at hand.
Input data refers to the use of different forms of US data in
segmentation algorithms and include but are not limited to the
following potential possibilities, e.g., raw beam-summed RF data,
prior to envelope detection and logarithmic compression,
envelope-detected data, formed from raw beam-summed RF data, B-mode
images, formed without logarithmic compression, B-mode images,
formed without logarithmic compression and without application of
other filters, conventional B-mode images, M-mode images, etc.
Segmentation metrics refer to a quantity used to characterize
tissue regions, e.g., Signal-to-Noise Ratio (SNR), contrast,
texture, model-driven algorithms, etc. In some cases, optimality
can be a subjective measure that is decided by the user.
[0034] The different segmentations obtained will then be presented
to the user on a display 38 through, e.g., GUI 52, for manual
selection of the optimal segmentation. Automatic selection of the
optimal segmentation is also contemplated as will be described. The
image processor 36 (or graphics processor 40) provides
visualizations of the segmentations. The different segmentation
outputs are superimposed on the original B-mode image, to permit
the user to choose the best segmentation. In one embodiment, the
image processor 36 automatically cycles through different
segmentation results periodically. In another embodiment, all
outputs may concurrently be visualized in a color-coded or other
visual format. If automatic selection of the optimal segmentation
is performed, the selected optimal segmentation will be
superimposed on the B-mode image shown on the display 38.
[0035] Referring to FIG. 2, a block/flow diagram illustratively
shows two sets of inputs to the segmentation module 50 where input
data 102 and segmentation metrics 104 are depicted. Inputs to the
segmentation module 50 show a pipeline of US B-mode image formation
where uncompressed data is employed at various stages of the
pipeline. A specific combination input data 102 and segmentation
metrics 104 will provide the `optimal` segmentation for a
particular organ site, US system capabilities, etc. The input data
102 may include raw RF data 106 (from anywhere in the signal path
prior to the B-mode processor 28, FIG. 1), envelope detection data
108 (from signal envelope (carrier waves)) and from a B-mode
display 114 (B-mode images or image data). M-mode images or data
115 (or other image modes) may also be employed as input data. The
B-mode or M-mode images may be derived from a scan conversion 112
with or without logarithmic compression 110. The segmentation
metrics 104 may include statistical models 116 or the US imaged
volume, SNR 118, contrast 120, texture 122, edge detection 124 or
any other image characteristics.
[0036] For the input data 102, the logarithmic compression of the
data used is avoided to employ the entire range of the information
captured for calculation of the segmentation metric 104. Multiple
segmentations may be provisioned from which the `optimal` or `best`
segmentation can be selected by the segmentation module 50. This
selection can be made automatically or manually.
[0037] For an automatic selection of the optimal segmentation,
appropriate criteria may be defined. Examples of potential criteria
that may be used include, e.g., Whether the segmented volume is at
least `x` cm.sup.3?; Does the segmented volume include certain
pre-annotated anatomical landmarks?; Does the segmented volume
differ from the mean population-based segmentation model by greater
than `x` %?; Does the segmented shape differ from the mean
population-based shape model by greater than `x` %?; etc. The
metrics that are not utilized to generate the segmentation may be
utilized to rate the quality of the segmentation.
[0038] For a manual selection of the optimal segmentation, the
different segmentations can be presented to the user by
superimposing them on the B-mode image(s), e.g., all segmentations
may be superimposed, each with a different color/line style, a
single segmentation can be superimposed at any given time, with the
user having the ability to cycle through all the available
segmentations through mouse clicks and/or keyboard shortcuts.
[0039] Referring to FIG. 3, an illustrative GUI 52 is shown in
accordance with one exemplary embodiment. GUI 52 includes an image
panel 202 where an image may include superimposed candidate
segmentations 204, 206. Each candidate segmentation 204, 206 may be
shown superimposed on a B-mode image 205. The candidate
segmentations 204, 206 may all be shown concurrently or
sequentially (or any combination thereof). The user may select
input factors such as an organ to be segmented in check boxes 208,
210, 212, 214 and 216. The user may also select data on which
segmentation is to be performed by selecting check boxes 218, 220,
222. The user may also select data on which segmentation metric is
used to perform the segmentation by selecting check boxes 224, 226,
228. A final segmentation may be elected by presses button 230.
[0040] If the segmentations are performed in an automated manner,
the segmentations resulting from each combination of input factors
are shown to the user for selection of the `optimal` segmentation.
If the segmentations are to be manually performed by the user, the
user is sequentially shown multiple images (e.g., the beam-summed
RF image, the envelope detected image, B-mode images with and
without filtering, etc.). The user performs a manual segmentation
on the displayed image and has the choice of accepting or rejecting
it. The user may select the automatic or manual modes of operation.
After the optimal segmentation is selected by the user, it is
finalized by clicking the `Finalize segmentation` button 230.
[0041] It should be understood that the GUI 52 depicted in FIG. 3
is for illustrative purposes. The interface may be developed or
extended as needed to include more functionality. The type and
positioning of features on the GUI 52 may be changed or reorganized
as needed or desired. Additional buttons or controls may be
employed or some buttons or controls may be removed.
[0042] The present principles provide segmentation techniques,
which find application in a variety of areas. Examples of accurate
segmentation leading to accurate image registration find utility in
areas such as, e.g., adaptive treatment planning for radiation
therapy, intra-procedural therapy monitoring (e.g., brachytherapy,
RF ablation), real-time biopsy guidance, etc. It should be
understood that while described in terms of US imaging, the
segmentation selection aspects in accordance with the present
principles may be employed for other imaging modalities instead of
US. For example, the segmentation selection may be employed for
MRI, CT or other imaging modalities.
[0043] The present embodiments may be employed to provide enhanced
organ segmentation capabilities that complement tracking
technologies (e.g., EM tracking, ultrasound tracking) for
interventional devices. Also, the segmentation methods and user
interface can be integrated in existing commercial systems, without
the need to provide user access to raw data. Automated segmentation
capability in accordance with the present principles may be
employed with improved accuracy on any of the clinically-available
imaging systems, and in particular US imaging systems.
[0044] Referring to FIG. 4, a method for segmentation selection is
shown in accordance with illustrative embodiments. In block 302,
imaging energy is received for imaging a subject. The imaging
energy may include ultrasound, although other imaging modalities
and energy types may be employed. In block 304, the received data
is image processed to generate processed image data. The processed
image data may include logarithmic compression, or other filtering
or compression. The processing may include scan converting the data
and/or B-mode processing. Other forms of processing are also
contemplated.
[0045] In block 306, a plurality of segmentations are generated for
the subject based on features or combinations of features of raw
imaging data and/or processed imaging data. The raw imaging data
may include raw radiofrequency data, envelope detection data,
signal to noise ratio data, contrast data, texture data, edge
detection data, etc. The processed imaging data may include a
statistical model comparison, compressed data, converted data,
B-mode processed display data, etc.
[0046] Segmentations may be generated based on raw data and
processed data in different combinations to generate segmentations
that differ from one another. Generating the plurality of
segmentations may include generating the plurality of segmentations
based on one or more combinations of input data and segmentation
metrics. For example, one segmentation may be generated using a
particular segmentation metric and a particular type of input data.
The input data may include raw RF data, envelope detection data,
B-mode display data, etc. The segmentation metric information may
include statistical model comparison data, signal to noise data,
contrast data, texture data, edge detection data, etc. Other
segmentations may combine input data with segmentation metrics or
combined aspects of the input data with aspects of the segmentation
metrics. As an example, raw RF data may be combined with contrast
and texture data to generate a segmentation. Other combinations are
contemplated.
[0047] In block 308, at least one of the plurality of segmentations
is selected that best meets a segmentation criterion. The selection
criteria may include use desired aspects or automatic criteria. The
features or combinations of features of the raw imaging data and/or
the processed imaging data may be employed to generate the
plurality of segmentations. The segmentation may be manually
selected, which best meets user defined criteria or automatically
selected based on programmed criteria (for example, contrast or
pixel thresholds, a best fit image with a statistical model shape,
etc.).
[0048] In block 310, a graphical user interface is generated and
displays the plurality of segmentations. The segmentations are
preferably displayed on a B-mode image (background), wherein the
segmentation images are displayed concurrently or sequentially in
accordance with a user preference.
[0049] In block 312, the selected segmentation is employed to
perform an operative procedure or other task.
[0050] In interpreting the appended claims, it should be understood
that: [0051] a) the word "comprising" does not exclude the presence
of other elements or acts than those listed in a given claim;
[0052] b) the word "a" or "an" preceding an element does not
exclude the presence of a plurality of such elements; [0053] c) any
reference signs in the claims do not limit their scope; [0054] d)
several "means" may be represented by the same item or hardware or
software implemented structure or function; and [0055] e) no
specific sequence of acts is intended to be required unless
specifically indicated.
[0056] Having described preferred embodiments for optimal
ultrasound-based organ segmentation (which are intended to be
illustrative and not limiting), it is noted that modifications and
variations can be made by persons skilled in the art in light of
the above teachings. It is therefore to be understood that changes
may be made in the particular embodiments of the disclosure
disclosed which are within the scope of the embodiments disclosed
herein as outlined by the appended claims. Having thus described
the details and particularity required by the patent laws, what is
claimed and desired protected by Letters Patent is set forth in the
appended claims.
* * * * *