U.S. patent application number 11/505560 was filed with the patent office on 2009-09-10 for automatic identification of orientation in medical diagnostic ultrasound.
This patent application is currently assigned to Siemens Medical Solutions USA, Inc.. Invention is credited to Ismayil Guracar, Bimba Rao, Seshadri Srinivasan.
Application Number | 20090227871 11/505560 |
Document ID | / |
Family ID | 39318863 |
Filed Date | 2009-09-10 |
United States Patent
Application |
20090227871 |
Kind Code |
A2 |
Srinivasan; Seshadri ; et
al. |
September 10, 2009 |
Automatic Identification of Orientation in Medical Diagnostic
Ultrasound
Abstract
Orientation is automatically identified in medical diagnostic
ultrasound image. An area or volume based process, such as region
shrinking or using locations associated with flow or tissue
structure, determines a direction or orientation. Real-time imaging
in B-mode and/or flow mode may be enhanced based on the direction
or orientation.
Inventors: |
Srinivasan; Seshadri;
(Mountain View, CA) ; Rao; Bimba; (San Jose,
CA) ; Guracar; Ismayil; (Redwood City, CA) |
Correspondence
Address: |
SIEMENS CORPORATION;INTELLECTUAL PROPERTY DEPARTMENT
170 WOOD AVENUE SOUTH
ISELIN
NJ
08830
UNITED STATES
732-321-3026
|
Assignee: |
Siemens Medical Solutions USA,
Inc.
51 Valley Stream Parkway
Malvern
PA
19355-1406
|
Prior
Publication: |
|
Document Identifier |
Publication Date |
|
US 20080097212 A1 |
April 24, 2008 |
|
|
Family ID: |
39318863 |
Appl. No.: |
11/505560 |
Filed: |
August 16, 2006 |
Current U.S.
Class: |
600/453 |
Current CPC
Class: |
G06T 2207/30004
20130101; G06T 2207/10132 20130101; G06T 7/155 20170101; A61B 8/488
20130101; G01S 15/8988 20130101; G01S 15/8979 20130101; A61B 8/13
20130101; G06T 7/13 20170101; A61B 8/06 20130101; A61B 8/08
20130101 |
Class at
Publication: |
600/453 |
International
Class: |
A61B 8/00 20060101
A61B008/00 |
Claims
1. A method for automatic identification of orientation in medical
diagnostic ultrasound, the method comprising: obtaining ultrasound
data representing a region associated with a tissue structure;
determining locations associated with the region as a function of
the ultrasound data; and determining an orientation of the tissue
structure as a function of the locations.
2. The method of claim 1 wherein the ultrasound data represents
flow and wherein obtaining comprises: performing: suppressing bleed
of the ultrasound data representing flow onto locations associated
with tissue boundaries as a function of B-mode information;
increasing values of the ultrasound data representing flow below a
display threshold to values above the display threshold suppressing
flash in the ultrasound data representing flow, the suppressing
being a function of B-mode information; or combinations
thereof.
3. The method of claim 1 wherein determining locations comprises
generating a binary image as a function of a threshold applied to
the ultrasound data, and wherein determining the orientation
comprises area or volume based processing of the binary image.
4. The method of claim 1 wherein obtaining comprises decimating the
ultrasound data, the determining the locations and determining the
orientations based on the decimated ultrasound data.
5. The method of claim 1 wherein determining the orientation
comprises region shrinking as a function of the locations.
6. The method of claim 1 wherein determining the orientation
comprises: iteratively performing a morphological skeleton
operation on the locations, and identifying a skeletal line from
the output of the performing.
7. The method of claim 6 further comprising: selecting the skeletal
line from a plurality of lines as a function of a number of the
locations associated with each of the plurality of lines; and
identifying a straight line as a function of the selected skeletal
line.
8. The method of claim 1 further comprising: positioning a spectral
Doppler gate as a function of the orientation.
9. The method of claim 1 further comprising setting a scan line
angle as a function of the orientation.
10. The method of claim 1 further comprising: performing:
determining a lateral flow velocity as a function of the
orientation; correcting a sign of the lateral flow velocity, an
axial flow velocity, or another velocity as a function of a line
associated with the orientation; and combinations thereof.
11. The method of claim 1 wherein determining the orientation
comprises determining tissue edges as a function of B-mode
information; further comprising: selecting tissue edges associated
with the tissue structure; and filtering the B-mode information as
a function of the selected tissue edges.
12. The method of claim 1 wherein determining the orientation
comprises determining tissue edges as a function of B-mode
information; further comprising: selecting tissue edges associated
with the tissue structure; and setting demodulation frequency,
depth-gain curve or both as a function of the tissue edges.
13. The method of claim 1 further comprising: determining tissue
edges as a function of B-mode information; selecting tissue edges
associated with the tissue structure; and altering the velocity
data as a function of the selected tissue edges and the
orientation.
14. The method of claim 1 wherein determining the orientation
comprises locating a regional center of mass.
15. The method of claim 1 further comprising: identifying a largest
flow region.
16. In a computer readable storage medium having stored therein
data representing instructions executable by a programmed processor
for automatic identification of orientation in medical diagnostic
ultrasound, the storage medium comprising instructions for:
obtaining ultrasound data representing a region; region shrinking
as a function of the ultrasound data; and determining a direction
as a function of the region shrinking.
17. The instructions of claim 16 further comprising generating a
binary image as a function of a threshold applied to the ultrasound
data, wherein region shrinking comprises region shrinking the
binary image; and wherein determining the direction comprises
determining a line from a convergence of the region shrinking.
18. The instructions of claim 16 wherein region shrinking comprises
iteratively performing a morphological skeleton operation on the
region, and wherein determining the flow direction comprises
identifying a skeletal line from the output of the morphological
skeleton operation.
19. The instructions of claim 16 wherein the ultrasound data
represents flow and the direction is a flow direction; further
comprising performing: positioning a spectral Doppler gate as a
function of the flow direction; setting a scan line angle as a
function of the flow direction; determining a lateral flow velocity
as a function of the flow direction; correcting a sign of the
lateral flow velocity, an axial flow velocity, or another velocity
as a function of a line associated with the flow direction; or
combinations thereof.
20. The instructions of claim 16 wherein determining a direction
comprises determining tissue edges as a function of B-mode
information; further comprising: selecting tissue edges associated
with the region; and filtering the B-mode information as a function
of the selected tissue edges.
21. The instructions of claim 16 further comprising: determining
tissue edges as a function of B-mode information; selecting tissue
edges associated with the region; and performing: setting
demodulation frequency, depth-gain curve or both as a function of
the tissue edges; altering velocity data as a function of the
selected tissue edges and the flow direction; or combinations
thereof.
22. A system for automatic identification of orientation in medical
diagnostic ultrasound, the system comprising: a memory operable to
store ultrasound data comprising velocity data, energy data or
both; a processor operable to apply an area or volume based process
to the ultrasound data, the processor operable to identify a line
as a function of the area or volume based process and free of a
determination of a location of maximum velocity, energy or
both.
23. The system of claim 22 wherein the processor is operable to
alter B-mode information as a function of the line, and wherein the
processor is operable to alter the ultrasound data as a function of
B-mode edges.
Description
BACKGROUND
[0001] The present embodiments relate to identification of
orientation. In particular, orientation of flow or structure is
determined for medical diagnostic ultrasound imaging.
[0002] Medical diagnostic ultrasound imaging may be used to show
flow with respect to tissue, such as flow mode image overlaid with
a B-mode image. B-mode data determines whether a given location is
associated with tissue or with flow. A B-mode value above a
threshold indicates a location associated with tissue. Color bleed
from flow data over vessel walls represented by B-mode data occurs
since B-mode amplitude alone may not be an adequate identifier of
vessel walls. At low velocity scales, significant flash may occur
in flow imaging, since B-mode amplitude may not sufficiently
identify valid tissue and anatomical structures. Further
thresholding may be provided, such as removing flow information
below a velocity and/or energy threshold. Removal may create holes
in the flow representation.
[0003] The B-mode and flow mode data are acquired along a plurality
of scan lines. For flow mode, velocity information represents
motion along the scan lines. Depending on the location of the
structure scanned, the flow may not be only along the scan lines.
The velocity information may be inaccurate since flow perpendicular
to the scan line may not be detected.
BRIEF SUMMARY
[0004] By way of introduction, the preferred embodiments described
below include methods, systems and computer readable media for
automatic identification of orientation in medical diagnostic
ultrasound. An area or volume based process, such as region
shrinking or processing locations associated with flow or tissue
structure, determines a direction or orientation. Real-time imaging
in B-mode and/or flow mode may be enhanced based on the direction
or orientation.
[0005] In a first aspect, a method is provided for automatic
identification of orientation in medical diagnostic ultrasound.
Ultrasound data representing flow in a region of a tissue structure
is obtained. Locations associated with the region are determined as
a function of the ultrasound data representing flow. An orientation
of the tissue structure is determined as a function of the
locations.
[0006] In a second aspect, a computer readable storage medium has
stored therein data representing instructions executable by a
programmed processor for automatic identification of orientation in
medical diagnostic ultrasound. The storage medium includes
instructions for: obtaining ultrasound data representing flow in a
region; region shrinking as a function of the ultrasound data
representing flow; and determining a flow direction as a function
of the region shrinking.
[0007] In a third aspect, a system is provided for automatic
identification of orientation in medical diagnostic ultrasound. A
memory is operable to store ultrasound data comprising velocity
data, energy data or both. A processor is operable to apply an area
or volume based process to the ultrasound data. The processor is
operable to identify a line as a function of the area or volume
based process and free of a determination of a location of maximum
velocity, energy or both.
[0008] The present invention is defined by the following claims,
and nothing in this section should be taken as a limitation on
those claims. Further aspects and advantages of the invention are
discussed below in conjunction with the preferred embodiments.
BRIEF DESCRIPTION OF THE DRAWINGS
[0009] The components and the figures are not necessarily to scale,
emphasis instead being placed upon illustrating the principles of
the invention. Moreover, in the figures, like reference numerals
designate corresponding parts throughout the different views.
[0010] FIG. 1 is a flow chart diagram of one embodiment of a method
for automatic identification of orientation in medical diagnostic
ultrasound;
[0011] FIG. 2 is a graphical representation of one embodiment of a
region associated with flow; and
[0012] FIG. 3 is a block diagram of one embodiment of a system for
automatic identification of orientation in medical diagnostic
ultrasound.
DETAILED DESCRIPTION OF THE DRAWINGS AND PRESENTLY PREFERRED
EMBODIMENTS
[0013] Area- or volume-based image enhancement is applied to color
flow and/or B-mode images. Area-based processing, such as
morphological processing, on B-mode and/or flow images identifies
an orientation of flow or tissue structure (e.g., vessel walls).
The orientation information may be used to a) reduce artifacts in
color flow imaging, such as holes, bleed and flash, b)
automatically correct velocities, c) steer scan-lines, d)
automatically identify a spectral Doppler window position,
orientation and size, and/or e) enhance B-mode and/or flow mode
images. The identification of a direction allows adaptive filtering
of the B-mode images.
[0014] FIG. 1 shows a method for automatic identification of
orientation in medical diagnostic ultrasound. The method is
implemented by a medical diagnostic ultrasound imaging system, a
review station, a workstation, a computer, a PACS station, a
server, combinations thereof, or other device for image processing
ultrasound data. For example, the system or computer readable media
shown in FIG. 3 implements the method. The method is implemented in
the order shown or a different order. Additional, different, or
fewer acts may be performed. For example, acts 22, 24, 26 and 28
are optional. As another example, act 16 is performed without acts
14, 18 and/or 20. In another example, act 14 is performed without
acts 16, 18 and/or 20.
[0015] In act 12, ultrasound data is obtained. The ultrasound data
is obtained during real-time imaging, such as obtaining data as a
patient is scanned. Alternatively, the ultrasound data is obtained
from a memory, such as obtaining the data from a database of
images, a patient record or a transfer from another device. The
ultrasound data is formatted for display on a display device or in
another format, such as a Polar coordinate or other acquisition
format. The data is obtained by requesting the data, loading the
data, receiving the data, recording the data, or other process.
[0016] The ultrasound data represents flow in a region. For
example, the data is velocity values, energy values, variance
values or combinations thereof. Other Doppler or flow information
may be used, such as motion information determined from correlation
of B-mode data.
[0017] The region corresponds to a blood pool, vessel, artery,
other cardiovascular structure, or other fluid containing
structure. In an organ, flow of vessels and veins may perfuse the
organ. The region may be associated with the digestive system or
other systems. The region is two or three dimensional, such as an
area or a volume. Other regions may be used. The ultrasound data
includes samples of the area or volume along any regular or
irregular grid.
[0018] Data associated with tissue structure with or without flow
may be obtained. For example, B-mode data representing tissue
structures containing flow is obtained. The tissue structures may
be vessel walls, organ tissue or other structures. The region
represented by the obtained data includes only flow information or
flow and tissue information.
[0019] The ultrasound data is obtained without processing, such as
raw data output by a receive beamformer. Some processing may be
provided, such as receiving data output by a detector or estimator
without further processing. Alternatively, the ultrasound data is
processed, such as spatially or temporally filtered.
[0020] In one embodiment, bleed of the ultrasound data representing
flow onto locations associated with tissue boundaries is suppressed
or reduced. Due to resolution differences, flow data (e.g.,
velocity or energy) may overlay desired tissue data at boundaries,
such as a vessel wall. Any function may be used to reduce the color
(i.e., flow) artifact. The bleed may be suppressed as a function of
B-mode information. An enhanced B-mode image (B.sub.enh) (e.g.,
speckle reduced B-mode image) and an edge image (B.sub.edge) are
computed. In one embodiment, the enhanced B-mode image is formed by
any speckle reduction process, such as frequency or spatial
filtering. Alternatively, the raw B-mode image is used for the
enhanced B-mode image. The edge image is a gradient image of the
B-mode data. For example, the edge image is determined as
B.sub.edge=[B(i,j)-B(i,j-1)]/2+[B(i,j)-B(i-1,j)]/2, where B(i,j) is
the B-mode data.
[0021] Spatial locations which should not include flow information
are identified from the enhanced and edge images. A threshold,
Edge.sub.th is applied to the edge image, B.sub.edge, and another
threshold, a B-mode dominance threshold (B.sub.th), is applied to
the enhanced B-mode image, B.sub.enh. Where the B-mode information
exceeds the threshold of either test, the flow information is set
to zero to suppress bleed (e.g., if B.sub.edge<Edge.sub.th, then
V=0 and/or E=0; if B.sub.enh<B.sub.th, V=0 and/or E=0). The
thresholds may be any value, such as about 48 or 64 on a scale of
0-255 for the Edge.sub.th and 96 or 64 for B.sub.th. The thresholds
identify spatial locations with sufficiently strong B-mode
information to indicate tissue rather than fluid. Other processes
with or without B-mode processing may be used to reduce bleed.
[0022] In another embodiment, holes in flow information are filled.
Values of the ultrasound data representing flow that are below a
display threshold are increased to values above the display
threshold for the flow region. Velocity (V) and energy (E)
estimates that are non-zero and less than the display threshold
(display.sub.th) are weighted or replaced. For example, if
0<abs(V)<display.sub.th, V=sign(V)*display.sub.th, and if
0<E<display.sub.th then E=display.sub.th. The display
threshold for the velocity is the same or different than for
energy, and may be any value, such as 8-12 for a scale of 0-255.
Since velocity is directional with positive and negative portions
of the scale, a lower threshold may be used. Where flow is
determined not to exist, such as where velocity and/or energy are
set to 0, fill is not provided.
[0023] In another embodiment, flash in the ultrasound data
representing flow is suppressed. The suppression is a function of
B-mode information, but may be performed without B-mode information
in other embodiments. Any suppression may be used. For example, in
a current frame (N), V and E estimates are modulated with the
B-mode values as: E1=E(N)*(1-.gamma.*(B.sub.enh(N)+B.sub.enh(N-1)))
and V1=V(N)*(1-.gamma.*(B.sub.enh(N)+B.sub.enh(N-1))) where .gamma.
is a fraction between 0 and 1/510 for a B-mode range of 0-511. The
modulated V1 and E1 estimates are compared with a display threshold
to determine locations without flow. If E1<display.sub.th, then
E=0, and if V1<display.sub.th, then V=0. The display threshold
is 8-12 for a range of 0-255 values, but may be another number.
After thresholding, the remaining E and V values are persisted as a
function of the B-mode information. For example,
V(N)=(1-.alpha.(N))*V(N)+.alpha.(N)*V(N-1) and
E(N)=(1-.alpha.(N))*E(N)+.alpha.(N)*E(N-1) where
.alpha.(N)=(1-.beta.)*B.sub.enh(N)/255+.beta.*B.sub.enh(N-1)/255
and .beta. is a fraction between 0 and 1, such as 1/2. Other
persistence or modulations may be used. Infinite impulse response
is used in this example, but finite impulse response processes may
be performed.
[0024] The ultrasound data representing flow may be processed
according to other processes. Combinations of two or more processes
may be performed, such as reducing bleed, filling and flash
suppression being performed sequentially on velocity and energy
data.
[0025] The ultrasound data representing the region is used to
determine orientation. In one embodiment, the flow data
representing an area or volume is used, such as all of the flow
information for a region. In other embodiments, the ultrasound data
is decimated. Any amount of decimation may be used, such as using
every second, third or fourth data point or spatial location. More
than one decimation operation may be used, such as generating two
frames of data associated with a same or different amount of
decimation. For a same amount of decimation, the different frames
may correspond to different starting locations for decimation. The
ultrasound data representing flow is decimated, but ultrasound data
representing tissue may be decimated. The decimation is provided
with or without filtering.
[0026] In act 14, locations associated with flow are determined. In
one embodiment, the processes used for reducing bleed, flash
suppression, and/or filling define locations associated with flow.
In another embodiment, additional or different thresholding is
applied to determine locations associated with a desired level of
flow. Energy, velocity, variance or combinations thereof may be
thresholded. In other embodiments, display thresholds, automatic
border detection, a region of interest, or user tracing define, at
least in part, locations associated with flow.
[0027] The locations rather than the values are selected for
determining orientation. For example, a binary image is generated.
A 1 value represents spatial locations associated with flow, and a
0 value represents other locations (or vice versa). Other
indications of location, such as a table of coordinates, may be
used.
[0028] The orientation identification is based on flow locations,
rather than B-mode information. Alternatively, B-mode information
is used to determine, at least in part, the orientation.
[0029] An orientation of a tissue structure or flow region is
determined in act 16. For example, the orientation of the Carotid
artery, thyroid, kidney, liver, heart, vessel, or other structure
is determined. The orientation corresponds to the flow or tissue.
For example, locations associated with flow are used to determine
an orientation of the flow. The orientation of flow provides or is
the same as the orientation of the associated organ or tissue
structure.
[0030] The orientation is determined as a function of the
locations. The orientation is directly based on the locations, not
on the values at the locations. Indirectly, the values are used to
determine the locations. However, the locations without value
information are used to determine the orientation. In alternative
embodiments, the values are also used to determine the
orientation.
[0031] The orientation is determined from the locations by area or
volume based processing. For example, the binary image of locations
is used to determine the orientation. Any now known or later
developed area or volume based processing may be used. For example,
a regional or local center of mass is determined with or without
region shrinking.
[0032] In one embodiment represented by act 18, region shrinking is
performed. The region shrinking or other area-based method
generates or identifies a line representing or following the local
or regional center of mass. The region shrinking is a function of
the locations or of ultrasound data representing flow. For example,
the binary image or other location representation is shrunk. Any
region shrinking may be used, such as removing outer layers to
define an internal line. For example, a morphological skeleton
operation reduces area or volume objects to lines. A kernel is
applied to the locations. Any size kernel may be used, such as a
3.times.3, other symmetric or asymmetric kernel. The kernel is
applied to non-overlapping portions of the binary image. For each
kernel location, a median is determined. All of the data (e.g., 9
locations for a 3.times.3 kernel) for a given kernel location is
replaced by the median. The morphological skeleton operation is
iteratively repeated by repeating the process a plurality of times.
Eventually, the region does not converge any further or reaches a
threshold number of iterations or amount of region shrinking. The
result is a line or region having a direction. The line is straight
or curved. More than one line may result, such as lines with
branches and/or associated with different regions.
[0033] If two or more frames of data associated with different
decimation are provided, skeletons or lines are determined
separately for each frame. Skeletons or lines not present from all
or some of the frames are rejected, and lines present in some or
all of the frames are maintained. Small color regions that can
distort the skeleton orientations may be removed by rejection of
the skeletons. The region shrinking process may or may not be
repeated after removal of the rejected lines and corresponding flow
information.
[0034] FIG. 2 shows a two-dimensional image of a vessel with an
associated flow region 42 having vessel walls 41. Tissue 40 is
around the flow region 42. The binary image provides for flow
locations within the flow region 42 and not within the tissue 40.
The flow region 42 is shrunk to identify the line 44.
[0035] In act 20, a flow direction is determined as a function of
the region shrinking. For example, the line 44 is determined from
the convergence of the region shrinking. This skeletal line 44 from
the output of the morphological skeleton operation indicates a
direction of flow.
[0036] In other embodiments, the line is filtered or processed to
determine the flow direction. For example, a straight line 46
associated with the entirety of or a portion of the skeleton line
44 is determined. The straight line 46 may be an average slope of
the skeleton line 44. As another example, a Hough or Radon
transform is applied to the skeleton line 44 to identify the slope
and location of the straight line 46 with the highest energy. Any
kernel size may be used, such as an 8.times.8 kernel.
Alternatively, a piecewise line-fit is performed on the skeletal
line 44 to identify orientation. Each line 44 is treated separately
for line fitting. Alternatively, a bank of filters with specific
orientations is used for identifying the orientation of small
(e.g., 8.times.8 or 4.times.4) regions in the skeletal image. By
filtering for different orientations, one of the oriented filters
identifies the straight line 46 and other filters filter out the
skeleton line 44. Other processes or filtering may be used.
[0037] More than one skeleton line 44 may exist, such as for images
of an organ or other structure with branching or different flow
regions. One or more of the skeleton lines 44 or other lines 46 are
selected. Any selection criteria may be used, such as manual input,
longest line, straightest line, or line closest to a center of an
image. For example, the skeletal line 44 associated with a greatest
number of flow locations is selected. The number of pixels or
locations rejected or replaced during region shrinking indicates a
number of flow locations associated with a given line 44. The
locations included in the count are contiguous with the skeletal
line 44. The skeletal line 44 that corresponds to the largest color
region within an image is selected.
[0038] The selected skeletal line 44 or a corresponding processed
line 46 (e.g., a straight line) indicate the orientation of the
flow, flow region, and/or associated tissue structure. The
orientation may or may not be displayed to the user. The
orientation may or may not be used for further processing and/or to
set imaging parameters. Acts 22, 24, 26, and 28 correspond to
further processes using the orientation.
[0039] In act 22, a spectral Doppler gate is positioned as a
function of the orientation. The center location, width, and/or
orientation of the gate are positioned based on the orientation.
The spectral Doppler window is positioned at an ultrasound scan
line that is closest to the central pixel of the skeletal line 44
or closest to an intersection of the skeletal line 44 with a
particular location or other line. The window may be oriented along
the same orientation as the skeletal line 44. Perpendicular to the
skeletal line 44, the extent of the flow region is determined. The
perpendicular extent or distance is determined from the center
location of the gate. The width of the gate is set to or less than
the distance. The length of the gate is set automatically to assure
that the entire gate is within the flow region. Any reasonable
window size for spectral Doppler that does not exceed the extent of
the flow region at a given location may be used.
[0040] In act 24, a scan line angle is set as a function of the
orientation. The scan lines for color or flow imaging may be
oriented to be more along the direction of flow. The orientation,
such as the mean orientation within a region of interest, is
determined. A steering angle or angles available with a given
transducer at a given position on the patient are selected. The
angle is selected to match, to the extent possible, the mean
orientation. The scan lines are as parallel as possible to the
orientation so that the velocity determined by Doppler or flow
processing more closely matches the actual velocity. B-mode scan
lines may be positioned to be as perpendicular as possible. Tissue
structures may more strongly reflect acoustic energy with a
perpendicular relationship. Scan lines for B-mode, flow mode, or
both may be adjusted or set as a function of the orientation.
[0041] In act 26, the velocity values are corrected or altered as a
function of the orientation. For example, a lateral flow velocity
is determined as a function of the orientation. The velocity values
represent velocity along a scan line. The scan lines may not be
parallel with the flow direction. The magnitude of velocity along
the flow direction is different than the detected velocity. The
detected velocity may include both lateral and axial flow
information to different extents. For each velocity along a given
ultrasound line, a corresponding closest skeletal line location is
determined. The orientation of the skeletal line locations relative
to the scan line is determined. This angle information allows
determination of the component of the velocity along the vessel
wall (V.sub.wall) or flow direction. The velocity is corrected
based on the orientation of the skeletal pixel with respect to the
ultrasound line as: V.sub.wall(i,j)=V(i,j)/cos .theta.(i,j). Other
angle corrections may be used, such as to identify turbulent flow
in directions other than the flow direction.
[0042] As another example of correction of velocity values, a sign
of the lateral flow velocity, an axial flow velocity, or another
velocity is corrected. Due to curvature of the flow region and/or
orientation of the scan lines through the flow region, some
portions of the region may show negative flow and other portions
positive flow where all of the portions should be positive or
negative. The aliasing of the velocity is corrected as a function
of a line 44 associated with the orientation. The mean or median
sign (S.sub.mean) of the velocities for locations along the
skeletal line 44 is determined. For velocities within the flow
region whose sign does not match S.sub.mean, the sign is corrected.
A local neighborhood (e.g., 3.times.3) may be filtered to correct
the magnitude of the axial and lateral velocities.
[0043] Both sign and angle correction may be performed. One or the
other may be performed. Other velocity corrections may be
performed.
[0044] In act 28, other image enhancements are performed as a
function of the orientation. B-mode and/or flow mode (e.g.,
velocity or energy) images are enhanced. For example, the
orientation is compared to an orientation derived from a B-mode
image. Tissue edges are determined as a function of B-mode
information. Boundary detection may be used. In one embodiment, the
B-mode edge image determined as discussed above for bleeding
reduction is used. The edge image is or is not decimated. Any
number of decimation frames may be generated, such as one or two
levels. Using locations, area or volume based processing determines
lines associated with edges. A morphological skeleton operation is
performed on the decimated or entire B-mode edge image
(B.sub.edge). For example, region shrinking of a binary B-mode edge
image provides B-mode edge locations. The B-mode edge locations are
compared with the color pixel coordinates to reduce color bleed.
Other edge detection may be used.
[0045] Tissue edges associated with the tissue structure or flow
orientation are selected. The edge orientation and location are
compared with the flow orientation to identify vessel walls in the
B-mode image. If the coordinates of a flow skeleton line 44 lie
within the coordinates of two B-mode edges, then the B-mode edges
represent the vessel walls or tissue regions associated with vessel
walls defined by the extent of the flow region.
[0046] The B-mode information may be filtered as a function of the
selected tissue edges. Filtering may make the B-mode edges more
continuous and smooth along vessel walls. A directional low-pass
filter, such as with a 3.times.3 kernel whose coefficients are
strong along the vessel orientation, may highlight the tissue
edges. For regions in the B-mode image that do not contain strong
edges, a symmetric smoothing kernel may reduce speckle.
[0047] As an alternative or in addition to filtering based on the
tissue edges, the demodulation frequency, depth-gain curve or both
are set as a function of the selected tissue edges. The
demodulation frequencies and the depth-gain compensation (dgc)
curves are modified according to the location and size of the flow
region. The attenuation caused by tissue is different than caused
by flow. The demodulation frequency and depth-gain curve may
account, at least in part, for the different attenuation. By using
the orientation information, the extent of the flow region and the
tissue regions is determined. Modifying for different attenuation
may improve the B-mode registration and image quality over depth.
For example, the location and number of breakpoints for
demodulation frequencies and depth-gain curve can be varied
dynamically.
[0048] As another example of act 28, the flow image is enhanced by
artifact reduction. The flow is enhanced as a function of B-mode
images, such as an enhanced B-mode image as discussed above or
another B-mode image. The velocity data is altered as a function of
the selected tissue edges and the orientation of the flow region.
The B-mode edge orientations are compared with the color
orientations to identify the extent of the flow region. If
coordinates of a flow orientation line lie within the coordinates
of two B-mode edges, the flow orientation is part of a flow region.
Other flow orientations may be discarded. For each flow region,
aliasing is reduced using the vessel orientation and the flow
direction. Filling or reduction of holes in the flow is performed
using local spatial filtering or correcting/re-estimating
velocities in the holes using the flow direction information.
[0049] FIG. 3 shows a system 50 for automatic identification of
orientation in medical diagnostic ultrasound. The system 50 is a
medical diagnostic ultrasound imaging system, but may be a
computer, workstation, database, server or other system. The system
50 includes a processor 52, a memory 54, and a display 56.
Additional, different, or fewer components may be provided.
[0050] The memory 54 is a buffer, cache, RAM, removeable media,
hard drive, magnetic, optical, or other now known or later
developed memory. The memory 54 is a single device or group of two
or more devices.
[0051] The memory 54 stores the ultrasound data. For example, the
memory 54 stores flow (e.g., velocity, energy or both) and/or
B-mode data. Alternatively, the ultrasound data is transferred to
the processor 52 from another device.
[0052] The memory 54 is additionally or alternatively a computer
readable storage medium with processing instructions. Data
representing instructions executable by the programmed processor 54
for automatic identification of orientation in medical diagnostic
ultrasound is stored in the memory 54. The instructions for
implementing the processes, methods and/or techniques discussed
above are provided on computer-readable storage media or memories,
such as a cache, buffer, RAM, removable media, hard drive or other
computer readable storage media. Computer readable storage media
include various types of volatile and nonvolatile storage media.
The functions, acts or tasks illustrated in the figures or
described herein are executed in response to one or more sets of
instructions stored in or on computer readable storage media. The
functions, acts or tasks are independent of the particular type of
instructions set, storage media, processor or processing strategy
and may be performed by software, hardware, integrated circuits,
filmware, micro code and the like, operating alone or in
combination. Likewise, processing strategies may include
multiprocessing, multitasking, parallel processing and the like. In
one embodiment, the instructions are stored on a removable media
device for reading by local or remote systems. In other
embodiments, the instructions are stored in a remote location for
transfer through a computer network or over telephone lines. In yet
other embodiments, the instructions are stored within a given
computer, CPU, GPU or system.
[0053] The processor 52 is a general processor, digital signal
processor, application specific integrated circuit, field
programmable gate array, digital circuit, analog circuit or other
now known or later developed device for processing ultrasound data.
The processor 52 is a single device, a plurality of devices, or a
network. For more than one device, parallel or sequential division
of processing may be used.
[0054] The processor 52 is operable to perform one or more of the
acts discussed above. For example, the processor 52 applies area or
volume based processing to ultrasound data, such as flow and/or
tissue data. The processor 52 identifies a line as a function of
the area or volume based process and free of a determination of a
location of maximum velocity, energy or both. For example, the line
is determined from locations rather than flow values. The
orientation may be used for any purpose. For example, the processor
52 alters B-mode information as a function of the orientation or
line associated with a flow region. As another example, the
processor 52 alters ultrasound data as a function of B-mode
edges.
[0055] While the invention has been described above by reference to
various embodiments, it should be understood that many changes and
modifications can be made without departing from the scope of the
invention. It is therefore intended that the foregoing detailed
description be regarded as illustrative rather than limiting, and
that it be understood that it is the following claims, including
all equivalents, that are intended to define the spirit and scope
of this invention.
* * * * *