BACKGROUND OF THE INVENTION
The subject matter disclosed herein relates generally to imaging systems, and
more particularly to a method and system for performing attenuation correction of medical
images.
Multi-modality imaging systems scan using different modalities, for example,
Computed Tomography (CT), Magnetic Resonance Imaging (MRI), Positron Emission
Tomography (PET), and Single Photon Emission Computed Tomography (SPECT). During
operation, image quality may be affected by various factors. One such factor is patient motion.
Another factor is inaccurate attenuation correction between images acquired using two different
imaging modalities caused by the patient motion.
Accordingly, at least one known PET-CT system utilizes data that is generated
by the CT system to generate an attenuation correction of the PET scan data. Specifically, a
plurality of emission attenuation correction factors are derived from CT data that is generated
during a CT scan, wherein the CT system is specifically configured to generate data to be
utilized for the CT attenuation correction factors. More specifically, the CT information is
utilized to generate a linear attenuation map at 511 keV, which may then be applied to
attenuation correct the PET information.
Moreover, at least one known PET-MR system utilizes data that is generated by
the MR system to generate an attenuation correction of the PET scan data. However, utilizing
the MR data to generate a linear attenuation map at 511 keV, which may then be applied to
attenuation correct the PET information may result in unwanted artifacts or inaccurate PET
quantitation and therefore may reduce the diagnostic value of images generated using the PET
scan data.
BRIEF DESCRIPTION OF THE INVENTION
In one embodiment, a method for correcting a positron emission tomography
(PET) image is provided. The method includes obtaining a magnetic resonance (MR) image
dataset, classifying at least one object in the MR image as a bone, generating MR-derived PET
attenuation correction factors based on the object classified as the bone, and attenuation
correcting a plurality of positron emission tomography (PET) emission data using the MRderived
PET attenuation correction factors. A medical imaging system and a non-transitory
computer readable medium are also described herein
In another embodiment, a medical imaging system is provided. The medical
imaging system includes a MRI system, a PET imaging system, and a computer coupled to the
MRl system and the PET system. The computer is programmed to obtain a MR image dataset,
classify at least one object in the MR image as a bone, generate MR-derived PET attenuation
correction factors based on the object classified as the bone, and attenuation correct a plurality of
positron emission tomography (PET) emission data using the MR-derived PET attenuation
correction factors.
In a further embodiment, a non-transitory computer readable medium is
provided. The non-transitory computer readable medium is encoded with a program
programmed to instruct a computer to obtain a MR image dataset, classify at least one object in
the MR image as a bone, generate MR-derived PET attenuation correction factors based on the
object classified as the bone, and attenuation correct a plurality of positron emission tomography
(PET) emission data using the MR-derived PET attenuation correction factors.
BRIEF DESCRIPTION OF THE DRAWINGS
Figure 1 is a pictorial view of an exemplary imaging system formed in
accordance with various embodiments.
Figure 2 is a flowchart illustrating a method for attenuation correcting positron
emission tomography (PET) emission data in accordance with various embodiments.
Figure 3 is an exemplary image that may be generated in accordance with
various embodiments.
Figure 4 is a portion of a density map that may be generated in accordance with
various embodiments.
Figure 5 is an exploded view of the density map shown in Figure 4.
Figure 6 is a portion of another density map that may be generated in
accordance with various embodiments.
Figure 7 is an exploded view of the density map shown in Figure 6.
Figure 8 is a portion of another density map that may be generated in
accordance with various embodiments.
Figure 9 is an exploded view of the density map shown in Figure 8.
Figure 10 is a portion of still another density map that may be generated in
accordance with various embodiments.
Figure 11 is an exploded view of the density map shown in Figure 10.
Figure 12 is an attenuation correction map that may be generated in accordance
with various embodiments.
Figure 13 is another attenuation correction map that may be generated in
accordance with various embodiments.
Figure 14 is a block schematic diagram of the first modality unit shown in
Figure 1 in accordance with various embodiments.
Figure 15 is a block schematic diagram of the second modality unit shown in
Figure 1 in accordance with various embodiments.
DETAILED DESCRIPTION OF THE INVENTION
The foregoing summary, as well as the following detailed description of various
embodiments, will be better understood when read in conjunction with the appended drawings.
To the extent that the figures illustrate diagrams of the functional blocks of the various
embodiments, the functional blocks are not necessarily indicative of the division between
hardware circuitry. Thus, for example, one or more of the functional blocks (e.g., processors or
memories) may be implemented in a single piece of hardware (e.g., a general purpose signal
processor or a block of random access memory, hard disk, or the like) or multiple pieces of
hardware. Similarly, the programs may be stand alone programs, may be incorporated as
subroutines in an operating system, may be functions in an installed software package, and the
like. It should be understood that the various embodiments are not limited to the arrangements
and instrumentality shown in the drawings.
As used herein, an element or step recited in the singular and proceeded with the
word "a" or "an" should be understood as not excluding plural of said elements or steps, unless
such exclusion is explicitly stated. Furthermore, references to "one embodiment" of the present
invention are not intended to be interpreted as excluding the existence of additional embodiments
that also incorporate the recited features. Moreover, unless explicitly stated to the contrary,
embodiments "comprising" or "having" an element or a plurality of elements having a particular
property may include additional elements not having that property.
Also as used herein, the phrase "reconstructing an image" is not intended to
exclude embodiments in which data representing an image is generated, but a viewable image is
not. Therefore, as used herein the term "image" broadly refers to both viewable images and data
representing a viewable image. However, many embodiments generate, or are configured to
generate, at least one viewable image.
Various embodiments described herein provide an imaging system 10 as shown
in Figure 1. The imaging system 10 is a multi-modality imaging system that includes different
types of imaging modalities, such as Positron Emission Tomography (PET), Single Photon
Emission Computed Tomography (SPECT), Computed Tomography (CT), ultrasound. Magnetic
Resonance Imaging (MRI) or any other system capable of generating diagnostic images. In the
illustrated embodiment, the imaging system 10 is a PET/MRI system. It should be realized that
the various embodiments are not limited to multi-modality medical imaging systems, but may be
used on a single modality medical imaging system such as a stand-alone PET imaging system or
a stand-alone MRI system, for example. Moreover, the various embodiments are not limited to
medical imaging systems for imaging human subjects, but may include veterinary or nonmedical
systems for imaging non-human objects, etc.
Referring to Figure 1, the multi-modality imaging system 10 includes a first
modality unit 12 and a second modality unit 14. These units may be aligned along an axis, as
shown in 10, or may co-habit a common space surrounding the patient such as having 14 inside
12 or vice versa. The two modality units enable the multi-modality imaging system 10 to scan
an object or subject 16 in a first modality using the first modality unit 12 and to scan the subject
16 in a second modality using the second modality unit 14. The scans may optionally, in the cohabited
modality case, be simultaneous. The multi-modality imaging system 10 allows for
multiple scans in different modalities to facilitate an increased diagnostic capability over single
modality systems. In the illustrated embodiment, the first modality 12 is a PET imaging system
and the second modality 14 is a MRI system. The imaging system 10 is shown as including a
gantry 18 that is associated with the PET imaging system 12 and a gantry 20 that is associated
with the MRI system 14. During operation, the subject 16 is positioned within a central opening
22, defined through the imaging system 10, using, for example, a motorized table 24.
The imaging system 10 also includes an operator workstation 30. During
operation, the motorized table 24 moves the subject 16 into the central opening 22 of the gantry
18 and/or 20 in response to one or more commands received from the operator workstation 30.
The workstation 30 then operates the first and/or second modalities 12 and 14 to both scan the
subject 16 and to acquire emission data and/or MRI data of the subject 16. The workstation 30
may be embodied as a personal computer (PC) that is positioned near the imaging system 10 and
hard-wired to the imaging system 10 via a communication link 32. The workstation 30 may also
be embodied as a portable computer such as a laptop computer or a hand-held computer that
transmits information to, and receives information from the imaging system 10. Optionally, the
communication link 32 may be a wireless communication link that enables information to be
transmitted to and/or from the workstation 30 to the imaging system 10 wirelessly. In operation,
the workstation 30 is configured to control the operation of the imaging system 10 in real-time.
The workstation 30 is also programmed to perform medical image diagnostic acquisition and
reconstruction processes described herein.
The operator workstation 30 includes a central processing unit (CPU) or
computer 34, a display 36, and an input device 38. As used herein, the term "computer" may
include any . processor-based or microprocessor-based system including systems using
microcontrollers, reduced instruction set computers (RISC), application specific integrated
circuits (ASICs), field programmable gate array (FPGAs), logic circuits, and any other circuit or
processor capable of executing the functions described herein. The above examples are
exemplary only, and are thus not intended to limit in any way the definition and/or meaning of
the term "computer". In the exemplary embodiment, the computer 34 executes a set of
instructions that are stored in one or more storage elements or memories, in order to process
information received from the first and second modalities 12 and 14. The storage elements may
also store data or other information as desired or needed. The storage element may be in the
form of an information source or a physical memory element located within the computer 34.
The imaging system 10 also includes an attenuation correction module 40 that is
configured to implement various methods described herein. In general, in many areas of the
human body, bone density may be modeled as a dense outer shell having a less dense inner core,
depending on the location and size of the bone structure. For example, the human skull tends to
have a larger density but thinner total dimension. Whereas, a pelvic bone, a femoral bone, the
spinal cord, and/or other large bone structures may have a dense outer shell, but also may have
marrow in the center core with a lower overall density. Accordingly, for MR-based attenuation
correction of 5llkeV PET data, the attenuation correction module 40 is configured to estimate
the linear attenuation coefficient for 51 IkeV gamma rays for at least one of the bones in the MRI
image. More specifically, the attenuation correction module 40 is configured to convert the MR
images based upon classification of different image features, such as, for example, a bone as is
described in more detail below.
The attenuation correction module 40 may be implemented as a piece of
hardware that is installed in the computer 34. Optionally, the attenuation correction module 40
may be implemented as a set of instructions that are installed on the computer 34. The set of
instructions may be stand alone programs, may be incorporated as subroutines in an operating
system installed on the computer 34, may be fiinctions in an installed software package on the
computer 34, and the like. It should be understood that the various embodiments are not limited
to the arrangements and instrumentality shown in the drawings.
The set of instructions may include various commands that instruct the
computer 34 as a processing machine to perform specific operations such as the methods and
processes of the various embodiments described herein. The set of instructions may be in the
form of a software program. As used herein, the terms "software" and "firmware" are
interchangeable, and include any computer program stored in memory for execution by a
computer, including RAM memory, ROM memory, EPROM memory, EEPROM memory, and
non-volatile RAM (NVRAM) memory. The above memory types are exemplary only, and are
thus not limiting as to the types of memory usable for storage of a computer program.
Figure 2 is a flowchart of an exemplary method 100 for attenuation correcting
PET emission data. In various embodiments, the method 100 may be implemented using for
example, the computer 34 and/or the attenuation correction module 40. At 102, an MRI dataset
50 is acquired using, for example, the MRI system 14 shown in Figure 1. The MRI dataset 50
may be obtained by performing a scan of the subject 16 to produce the MRI dataset 50.
Optionally, the MRI dataset 50 may be obtained from data collected during a previous scan of
the subject 16, wherein the MRI dataset 50 has been stored in a memory device, such as a
memory device 42 (shown in Figure 1). The MRI dataset 50 may be stored in any format. The
MRI dataset 50 may be obtained during real-time scanning of the subject 16. For example, the
methods described herein may be performed on MRI data as the MRI dataset 50 is received from
the MRI system 14 during a real-time examination of the subject 16. In various embodiments,
the MRI dataset 50 includes at least one bone. For example. Figure 3 is an exemplary image 200
of the subject 16, including an exemplary object 202. In various embodiments, the object 202
may be a first bone type or classification 204 which has a substantially uniform density, such as
for example, a skull bone. In various other embodiments, the object 202 may be a second bone
type or classification 206 bone having an exterior portion having a higher density, and an interior
portion having a lower density than the exterior portion, such as for example, a femur, a pelvic
bone, etc.
At 104, a PET emission dataset 52, or sinograms, are acquired using, for
example, the PET system 12 shown in Figure 1. The PET emission dataset 52 may be obtained
by performing a scan of the subject 16 to produce the PET emission dataset 52. Optionally, the
PET emission dataset 52 may be obtained from data collected during a previous scan of the
subject 16, wherein the PET emission dataset 52 has been stored in a memory device, such as a
memory device 42 (shown in Figure 1). The PET emission dataset 52 may be stored in any
format. The PET emission dataset 52 may be obtained during real-time scanning of the subject
16. For example, the methods described herein may be performed on PET emission data as the
PET emission dataset 52 is received from the PET system 12 during a real-time examination of
the subject 16.
At 106, at least one object 202 in the MRI dataset 50 is classified as a bone.
More specifically, and as shown in Figure 3, the voxels representing the object 202 have a higher
intensity and thus appear as brighter voxels. Whereas, soft tissue, organs, etc., surrounding the
object 202, have a lower density and thus appear as darker voxels. Accordingly, in various
embodiments, to classify the object 202 as a bone, an intensify based segmentation is performed
on the MRI dataset 50. In operation, a segmentation algorithm, which may be installed on the
attenuation correction module 40, is configured to locate objects of interest, such as the bone
202, and separate image data of the bone 202 from image data of surrounding objects of lesser or
no interest.
The segmentation algorithm uses a principle, whereby it is generally assumed
that the object 202 may be differentiated from other anatomical features by determining the
intensify of each voxel in the image data. Based on the intensify values of each of the voxels, the
bone 202 may be distinguished from the other anatomical features. Accordingly, at 106 the
segmentation algorithm automatically compares the intensify values for each voxel in the MRI
dataset 50 to a predetermined intensify value, using for example, a thresholding process. In the
exemplary embodiment, the predetermined intensify value may be a range of predetermined
intensity values. The predetermined intensify value range may be automatically set based on a
priori information of the bone, for example. Optionally, the predetermined range may be
manually input by the operator. In one embodiment, if the intensify value of a voxel representing
the object 202 is within the predetermined range, the voxel is classified as bone. Otherwise, the
voxel is classified as not belonging to the bone. It should be realized that the segmentation
algorithm may also be utilized with other segmentation techniques to identify the bone.
Additionally, as should be appreciated, other suitable segmentation algorithms may be used.
In various embodiments, the image data in the MRI dataset 50, for example,
voxel information that is identified using the segmentation algorithm, may be utilized to generate
a three-dimensional densify map, also referred to herein as a label map, wherein voxels identified
as bone are assigned a first label map value, and voxels that are not bone are assigned a second
10
label map value. For example, Figure 4 illustrates a portion of an exemplary three-dimensional
(3D) density map 220 that may be generated at 106 and Figure 5 is an exploded view of the label
map 220 shown in Figure 4. In the illustrated embodiments, the density map 220 includes
twenty-seven voxels 222. More specifically, the density map 220 illustrates a central voxel i that
is surrounded by twenty-six voxels. In the illustrated embodiment, the twenty-six voxels are
adjacent to the voxel i and are therefore referred to herein as "neighbors". It should be
appreciated that in operation, the density map 220 may include thousands of voxels 222 and
twenty-seven voxels 222 are shown to explain various embodiments described herein. In the
illustrated embodiment, the voxels 222 identified as bone are assigned a first label map value,
such as 1, and the voxels 222 that are not bone are assigned a second label map value, such as 0.
Accordingly, in various embodiments, the object 202 may be classified as a bone by performing
a segmentation of the MRI dataset 50 and assigning each voxel 222 a label map value that
indicates that the voxel 222 is either bone or not bone.
Additionally, in various embodiments, the object 202 may be classified as a
bone by manually comparing the object 202 to a plurality of images in an atlas. An atlas, as used
herein, is a electronic or hardcopy file that includes one or more images of various portions of
the human anatomy. Accordingly, in operation, the user may manually observe the object 202
and manually locate an image within the atlas that is substantially the same as the observed
object. In various other embodiments, the object 202 may be classified as a bone based on a
priori knowledge. More specifically, the operator may have a priori knowledge based on the
operator's experience that the object 202 is a skull, a femur, etc. The label map values may then
be manually entered into the density map 200 by the operator.
Referring again to Figure 2, at 108, MR-derived PET attenuation correction
factors are generated based on the label map values assigned at 106. In various embodiments,
generating MR-derived PET attenuation correction factors includes selecting at 120 a reference
bone voxel, identifying at 122 neighbor voxels having the same label map value as a reference
bone voxel, and generating at 124 a MR-derived PET attenuation correction factor for the
II
reference bone voxel based on the number of neighbor bone voxels. At 126, steps 120-124 are
repeated for each voxel classified as bone in the MRI dataset 50. The attenuation correction
factors are then utilized to attenuation correct the PET images.
More specifically, as discussed above, large bones often have a dense compact
bone shell surrounding a lower density trabecular bone center with a mixture of bone and
marrow. Accordingly, the interior of the bone may be distinguished from the exterior of the
bone by selecting a reference bone voxel and then calculating a quantity of neighbor voxels, e.g.
adjacent voxels that are also classified as bone based on the label map values assigned at 106. In
general, a "surface" bone portion is assigned a higher density value and an "interior" bone
portion is assigned a lower density value for use during PET attenuation correction.
For example, and referring again to Figures 4 and 5 a reference bone voxel 250,
such as the voxel i, is initially selected. In the illustrated embodiment, the reference bone voxel
250 has been previously identified and assigned a label map value of 1, as described above.
Each of the neighbor voxels 252 or connected voxels are then identified. Neighbor, as used
herein, means a voxel that is adjacent to the reference bone voxel 250 in an x-direction, a ydirection,
and a z-direction. Accordingly, in the illustrated embodiment, the reference bone
voxel 250 has twenty-six neighbor voxels 252. However, it should be realized that the method
described herein may be implemented with more than twenty-six neighbor voxels. For example,
various connected voxels may also be selected. Connected voxel, as used herein, means a voxel
that is adjacent to a neighbor voxel. For example, and referring to Figure 4, assume that a voxel
256 is selected as a reference voxel, then voxels 258 would be neighbor voxels and voxels 260
would be a connected voxel. It should be appreciated that the neighbor voxels 252 may be
positioned in each of the x-direction, the y-direction, and the z-direction from the reference voxel
250. Moreover, the connected voxels, such as the voxels 260 may also be positioned in each of
the x-direction, the y-direction, and the z-direction fi"om the reference voxel 250. Accordingly,
in the illustrated embodiment, the neighbor voxels 252 represent a 3x3x3 box of neighbor voxels
252 that surround the reference bone voxel 250.
12
After the neighbor voxels 252 and/or connected voxels are identified, the
neighbor voxels 252 and/or connected voxels classified as bone voxels are counted at 122. More
specifically, neighbor voxels 252 having a label map value of 1 are counted. In the illustrated
embodiment of Figures 4 and 5, the reference bone voxel 250 has twenty-six neighbor voxels
252. Moreover, twenty-five of the neighbor voxels 252 are classified as non-bone voxels (0) and
are therefore not counted. Accordingly, in the embodiment illustrated in Figures 4 and 5,
because all the neighbor voxels 252 are not bone, this signifies that the reference voxel 250 and
the neighbor voxels 252 represent a 'stand-alone' bone voxel, and hence the attenuation
correction factor, i.e., a density value assigned to the reference voxel 250 is left at a default
density value.
Figure 6 illustrates a portion of another exemplary three-dimensional (3D)
density map 270 that may be generated at 106 and Figure 7 is an exploded view of the label map
270 shown in Figure 6. In the illustrated embodiment, the density map 270 includes twentyseven
voxels 222 wherein the reference voxel 250 again has twenty-six neighbor voxels 252.
Moreover, nine of the neighbor voxels 252 are classified as bone voxels (1) and seventeen of the
neighbor voxels 252 are classified as non-bone voxels and are therefore not counted.
Accordingly, in the embodiment illustrated in Figures 6 and 7, because the distribution of the
neighbor voxels 252 exhibits reference voxel 250 as an edge voxel, the reference voxel 250 is
assigned an attenuation correction factor, i.e., a density value, that is higher than the intensity
value assigned to the reference voxel 250 in Figures 4 and 5. It should be realized that the
density value assigned to the reference voxel 250 is based on the quantity and distribution of
neighbor voxels 250 that are classified as bone. For example, assume that the density value
assigned to reference pixel is between a range of 0 and 1. Thus, if none of the neighbor voxels
252 are classified as bone, the density value assigned to the reference voxel is 1. Optionally, if
all the neighbor voxels 252 are classified as bone, the density value assigned to the reference
voxel is decreased. Accordingly, the reference voxel 250 is assigned a density value based on
the quantity of neighbor voxels 252 and/or coimected voxels that are classified as bone. Thus,
the embodiment illustrated in Figures 6 and 7 because nine of the neighbor voxels 252 are
13
classified as bone and are in one direction from reference voxel 250, the density value assigned
to the reference voxel 250 may be increased for example, approximately 1.3.
Figure 8 illustrates a portion of another exemplary three-dimensional (3D)
density map 300 that may be generated at 106 and Figure 9 is an exploded view of the label map
300 shown in Figure 8. In the illustrated embodiment, the density map 300 includes twentyseven
voxels 222 wherein the reference voxel 250 again has twenty-six neighbor voxels 252.
Moreover, seventeen of the neighbor voxels 252 are classified as bone voxels (1) and nine of the
neighbor voxels 252 are classified as non-bone voxels and are therefore not counted. In this
example, as compared to the example in Figures 6 and 7, the reference voxel 250 is less
'exterior' to the bone, and for example may have a density value assigned as 1.1 of the default
bone attenuation correction factor value.
Figure 10 illustrates a portion of another exemplary three-dimensional (3D)
density map 310 that may be generated at 106 and Figure II is an exploded view of the label
map 310 shown in Figure 10. In the illustrated embodiment, the density map 310 includes
twenty-seven voxels 222 wherein the reference voxel 250 again has twenty-six neighbor voxels
252. Moreover, twenty-six of the neighbor voxels 252 are classified as bone voxels (1) and none
of the neighbor voxels 252 are classified as non-bone voxels and are therefore not counted.
Accordingly, in the embodiment illustrated in Figures 10 and II, because all of the neighbor
voxels 252 are classified as bone, i.e. approximately 100%, the density value assigned to the
reference voxel 250 may be decreased for example, approximately 0.9 of the default bone
attenuation correction factor value.
In general, the methods described herein are configured to identify interior and
exterior bone voxels by counting the number of neighbor voxels that are the same, e.g. are bone
voxels. For example, if a reference bone voxel is bone, a determination is made as to how many
neighbor voxels are also bone. Accordingly, in various embodiments, the attenuation may be
modulated based on the number of neighbor voxels that are classified as bone. In one
14
embodiment, if all the neighbor voxels 252 are bone, this signifies that the reference voxel 250
and the neighbor voxels 252 represent an interior portion of the bone and are assigned a lower
attenuation correction factor. For example. Figure 12 is an exemplary attenuation correction
map 320 that illustrates the reference voxel 250 and the nearest neighbor voxels 252. As shown
in Figure 12, the reference voxel 250 is substantially surrounded by neighbor voxels that are also
bone. Thus, the reference voxel 250 is located in the interior portion of the bone and is assigned
a lower attenuation correction factor.
However, in other embodiments, if approximately half of the neighbor voxels
252 are bone voxels and the other half are not bone voxels and the distribution of bone/not bone
has the shape of an edge surface (0/1 are distributed left/right, up/down, etc.), the reference bone
voxel is assigned a higher attenuation correction factor. For example. Figure 13 is an exemplary
attenuation correction map 330 that illustrates a reference voxel 260 and the nearest neighbor
voxels 262 wherein approximately half the nearest neighbor voxels 262 are bone voxels and half
are not bone voxels. Thus, the reference voxel 260 is located in the exterior portion of the bone
and is assigned a higher attenuation correction factor. The attenuation correction factor is based
on the quantity of neighboring voxels identified as bone. In various embodiments, the
attenuation correction factor may be within a predetermined range. For example, in one
embodiment, if all of the nearest neighbor voxels are bone voxels, the reference bone voxel may
be assigned an attenuation correction factor of 0.9 of the default bone attenuation correction
factor value. Optionally, if only one of the neighbor voxels is a bone voxel, the reference bone
voxel may be assigned an attenuation correction factor of approximately 1. In various
embodiments, the attenuation correction factor scaling assigned to each reference voxel is based
upon a linear scale of between, for example, 0 and 1. Thus, if half the neighbor voxels are bone
voxels and are distributed in one direction from the reference voxel, and half the neighbor voxels
are not bone voxels and distributed in the opposite direction from the reference voxel, the
reference bone voxel may be assigned an attenuation correction scaling factor of approximately
1.1.
15
Described herein are methods and systems that utilize MR information to
provide attenuation correction of PET images. More specifically, various embodiments identify
thicker bones generally having a "soft" (lower attenuation) center of marrow (spine, pelvis). In
operation, MR images are utilized to find the bone class. The number of neighbor voxels within
the bone class for each voxel classified as bone is then determined. A lower attenuation
correction factor is assigned to bone voxels based upon an increasing or fixed large number of
neighbor voxels since the more internal a voxel identified as bone is, the more likely that the
voxel is less dense for some types of human bone. It should be realized that the methods
described herein may also be run "backwards". More specifically, a reference non-bone voxel
may be identified in the MR image dataset, a number and distribution of neighbor non-bone
voxels for the reference bone voxel may be counted, a MR-derived PET attenuation correction
factor scaling for the reference non-bone bone voxel based on the number and distribution of
neighbor non-bone voxels may be generated generating; and a MR-derived PET attenuation
correction factor scaling for the reference non-bone voxel based on the number and distribution
of neighbor non-bone voxels may be generated.
Various embodiments of the methods described herein may be provided as part
of, or used with, a medical imaging system, such as a dual-modality imaging system 10 as shown
in Figure 1. Figure 14 is a block schematic diagram of the first modality unit 12, e.g. the PET
imaging system, shown in Figure 1. Figure 15 is a block schematic diagram of the second
modality unit 14, e.g. the MRI system, shown in Figure 1.
As shown in Figure 14, the PET system 12 includes a detector array 400 that is
arranged as ring assembly of individual detector modules 402. The detector array 10 also
includes the central opening 22, in which an object or patient, such as the subject 16 may be
positioned, using, for example, the motorized table 24 (shown in Figure 1). The motorized table
24 is aligned with the central axis of the detector array 400. During operation, the motorized
table 24 moves the subject 16 into the central opening 22 of the detector array 400 in response to
one or more commands received from the operator workstation 30. More specifically, a PET
16
scanner controller 410 responds to the commands received from the operator workstation 30
through the communication link 32. Therefore, the scanning operation is controlled from the
operator workstation 30 through PET scanner controller 410.
During operation, when a photon collides with a scintillator on the detector
array 400, the photon collision produces a scintilla on the scintillator. The scintillator produces
an analog signal that is transmitted to an electronics section (not shown) that may form part of
the detector array 400. The electronics section outputs an analog signal when a scintillation
event occurs. A set of acquisition circuits 420 is provided to receive these analog signals. The
acquisition circuits 420 process the analog signals to identify each valid event and provide a set
of digital numbers or values indicative of the identified event. For example, this information
indicates when the event took place and the position of the scintillation scintillator that detected
the event.
The digital signals are transmitted through a communication link, for example, a
cable, to a data acquisition controller 422. The data acquisition processor 422 is adapted to
perform the scatter correction and/or various other operations based on the received signals. The
PET system 12 may also include an image reconstruction processor 424 that is interconnected
via a communication link 426 to the data acquisition controller 422. During operation, the image
reconstruction processor 424 performs various image enhancing techniques on the digital signals
and generates an image of the subject 16.
As shown in Figure 15 the MRI system 14 includes a superconducting magnet
assembly 500 that includes a superconducting magnet 502. The superconducting magnet 502 is
formed from a plurality of magnetic coils supported on a magnet coil support or coil former. In
one embodiment, the superconducting magnet assembly 500 may also include a thermal shield
504. A vessel 506 (also referred to as a cryostat) surrounds the superconducting magnet 502, and
the thermal shield 504 surrounds the vessel 506. The vessel 506 is typically filled with liquid
helium to cool the coils of the superconducting magnet 502. A thermal insulation (not shown)
17
may be provided surrounding the outer surface of the vessel 506. The MRI system 14 also
includes a main gradient coil 520, a shield gradient coil 522, and an RF transmit coil 524. The
MRI system 14 also generally includes a controller 530, a main magnetic field control 532, a
gradient field control 534, the memory device 42, the display device 36, a transmit-receive (T-R)
switch 540, an RF transmitter 542 and a receiver 544.
In operation, a body of an object, such as the subject 16 (shown in Figure 1) is
placed in the opening 22 on a suitable support, for example, the motorized table 24 (shown in
Figure 1). The superconducting magnet 502 produces a uniform and static main magnetic field
Bo across the opening 22. The strength of the electromagnetic field in the opening 22 and
correspondingly in the patient, is controlled by the controller 530 via the main magnetic field
control 532, which also controls a supply of energizing current to the superconducting magnet
502.
The main gradient coil 520, which may include one or more gradient coil
elements, is provided so that a magnetic gradient can be imposed on the magnetic field Bo in the
opening 22 in any one or more of three orthogonal directions x, y, and z. The main gradient coil
520 is energized by the gradient field control 534 and is also controlled by the controller 530.
The RF coil assembly 524 is arranged to transmit magnetic pulses and/or
optionally simultaneously detect MR signals from the patient, if receive coil elements are also
provided. The RF coil assembly 524 may be selectably interconnected to one of the RF
transmitter 542 or receiver 544, respectively, by the T-R switch 540. The RF transmitter 542 and
T-R switch 540 are controlled by the controller 530 such that RF field pulses or signals are
generated by the RF transmitter 542 and selectively applied to the patient for excitation of
magnetic resonance in the patient.
Following application of the RF pulses, the T-R switch 540 is again actuated to
decouple the RF coil assembly 524 fi-om the RF transmitter 542. The detected MR signals are in
turn communicated to the controller 530. The controller 530 may include a processor 554 that
18
controls the processing of the MR signals to produce signals representative of an image of the
subject 16. The processed signals representative of the image are also transmitted to the display
device 36 to provide a visual display of the image. Specifically, the MR signals fill or form a kspace
that is Fourier transformed to obtain a viewable image which may be viewed on the
display device 36.
As used herein, a set of instructions may include various commands that instruct
the computer or processor as a processing machine to perform specific operations such as the
methods and processes of the various embodiments of the invention. The set of instructions may
be in the form of a software program, which may form part of a tangible non-transitory computer
readable medium or media. The software may be in various forms such as system software or
application software. Further, the software may be in the form of a collection of separate
programs or modules, a program module within a larger program or a portion of a program
module. The software also may include modular programming in the form of object-oriented
programming. The processing of input data by the processing machine may be in response to
operator commands, or in response to results of previous processing, or in response to a request
made by another processing machine.
As used herein, the terms "software" and "firmware" may include any computer
program stored in memory for execution by a computer, including RAM memory, ROM
memory, EPROM memory, EEPROM memory, and non-volatile RAM (NVRAM) memory.
The above memory types are exemplary only, and are thus not limiting as to the types of memory
usable for storage of a computer program.
It is to be understood that the above description is intended to be illustrative, and
not restrictive. For example, the above-described embodiments (and/or aspects thereof) may be
used in combination with each other. In addition, many modifications may be made to adapt a
particular situation or material to the teachings of the various embodiments without departing
from their scope. While the dimensions and types of materials described herein are intended to
19
define the parameters of the various embodiments, they are by no means limiting and are merely
exemplary. Many other embodiments will be apparent to those of skill in the art upon reviewing
the above description. The scope of the various embodiments should, therefore, be determined
with reference to the appended claims, along with the full scope of equivalents to which such
claims are entitled. In the appended claims, the terms "including" and "in which" are used as the
plain-English equivalents of the respective terms "comprising" and "wherein." Moreover, in the
following claims, the terms "first," "second," and "third," etc. are used merely as labels, and are
not intended to impose numerical requirements on their objects. Further, the limitations of the
following claims are not written in means-plus-function format and are not intended to be
interpreted based on 35 U.S.C. § 112, sixth paragraph, unless and until such claim limitations
expressly use the phrase "means for" followed by a statement of function void of further
structure.
This written description uses examples to disclose the various embodiments,
including the best mode, and also to enable any person skilled in the art to practice the various
embodiments, including making and using any devices or systems and performing any
incorporated methods. The patentable scope of the various embodiments is defined by the
claims, and may include other examples that occur to those skilled in the art. Such other
examples are intended to be within the scope of the claims if the examples have structural
elements that do not differ from the literal language of the claims, or the examples include
equivalent structural elements with insubstantial differences from the literal languages of the
claims.
20
Parts List
Imaging system 10
First modality imit 12
Second modality unit 14
Object or patient 16
Gantry 18
Gantry 20
Central opening 22
Table 24
Woricstation 30
Communication link 32
Processor 34
Display 36
Input device 38
Attenuation correction module 40
Memory device 42
MRIdataset 50
PET emission dataset 52
Method , 100
At 102
At 104
At 106
At 108
At 120
At 122
At 124
At 126
Image 200
Object 202
Bone type or classification 204
Bone type or classification 206
Density map 220
Voxels 222
21
Reference voxel 250
Nearest neighbors 252
Reference voxel 260
Nearest neighbors 262
Attenuation correction map 300
Attenuation correction map 310
Detector array 400
Detector modules 402
PET scanner controller 410
Acquisition circuits 420
Data acquisition processor 422
Image reconstruction processor 424
Communication link 426
Superconducting magnet assembly 500
Superconducting magnet 502
Thermal shield 504
Vessel 506
Main gradient coil 520
Shield gradient coil 522
RF coil assembly 524
Controller 530
Main magnetic field control 532
Gradient field control 534
T-R switch 540
RF transmitter 542
Receiver 544
Processor 554
22

We Claims :
1. A medical imaging system (10) comprising:
a magnetic resonance imaging (MRI) system (14);
a positron emission tomography (PET) imaging system (12); and
a computer (34) coupled to the MRI system and the PET system, said computer being
programmed to:
obtain (102) a MR image dataset (50);
classify (106) at least one object( 202) in the MR image (50) as a bone;
identify (120) a reference bone voxel (250) in the MR image dataset;
count (122) a number of neighbor bone voxels (252) and their distribution for the
reference bone voxel;
generate (124) a MR-derived PET attenuation correction factor scaling for the
reference bone voxel based on the number and distribution of neighbor bone voxels; and
attenuation correct (108) a plurality of positron emission tomography (PET)
emission data using the MR-derived PET attenuation correction factors.
2. The medical imaging system of Claim 8, wherein the computer is further
programmed to perform an intensity based segmentation of the MR image dataset to identify the
bone.
3. The medical imaging system (10) of Claim 8, wherein the computer (34) is further
programmed to:
compare the object (202) to a plurality of images in an atlas; and
23
classify the bone based on the comparison.
4. The medical imaging system (10) of Claim 8, wherein the computer (34) is further
programmed to:
assign a first label map value to voxels representing bone; and
assign a second label map value to voxels that are not bone.
5. The medical imaging system (10) of Claim 8, wherein the computer (34) is further
programmed to:
identify (120) a reference bone voxel (250) in the MR image dataset (50);
count (122) a number of connecting bone voxels (252) and their distribution for the
reference bone voxel; and
generate (124)a MR-derived PET attenuation correction factor scaling for the reference
bone voxel based on the number of connecting bone voxels and neighbor bone voxels.
6. The medical imaging system (10) of Claim 8, wherein the computer (34)is further
programmed to:
assign a label map value to voxels representing bone (250);
identify neighbor voxels (252)and their distribution having the same label map value as a
reference bone voxel; and
generate a MR-derived PET attenuation correction factor scaling for the reference bone
voxel based on the number of neighbor bone voxels.
7. The medical imaging system (10) of Claim 8, wherein the computer is further
programmed to:
24
determine (122) a number of connecting bone voxels and their distribution around a
reference bone voxel (250); and
generate (124)the MR-derived PET attenuation correction factors based on the
determined number and distribution of connecting bone voxels and neighbor bone voxels.
8. A non-transitory computer readable medium encoded with a program
programmed to instruct a computer (34) to:
obtain (102) a magnetic resonance (MR) image dataset (50);
classily (156)at least one object (202) in the MR image as a bone;
identify (120) a reference bone voxel (250) in the MR image dataset;
count (122) a number of neighbor bone voxels (252) for the reference bone voxel;
generate (124) a MR-derived PET attenuation correction factor for the reference bone
voxel based on the number of neighbor bone voxels; and
attenuation correct (108) a plurality of positron emission tomography (PET) emission
data using the MR-derived PET attenuation correction factors.
9. The non-transitory computer readable medium of Claim 15, wherein the program
is further programmed to instruct the computer (34) to perform an intensity based segmentation
of the MR image dataset (50) to identify the bone.
10. The non-transitory computer readable medium of Claim 15, wherein the program
is further programmed to instruct the computer (34) to:
compare the object (202)to a plurality of images in an atlas; and
classify the bone based on the comparison.