U.S. patent application number 16/697722 was filed with the patent office on 2021-07-01 for methods and apparatus for improving the image resolution and sensitivity of whole-body positron emission tomography (pet) imaging.
This patent application is currently assigned to Washington University. The applicant listed for this patent is Ke Li, Joseph O'Sullivan, Yuan-Chuan Tai, Qiang Wang. Invention is credited to Ke Li, Joseph O'Sullivan, Yuan-Chuan Tai, Qiang Wang.
Application Number | 20210196211 16/697722 |
Document ID | / |
Family ID | 1000005650093 |
Filed Date | 2021-07-01 |
United States Patent
Application |
20210196211 |
Kind Code |
A9 |
Tai; Yuan-Chuan ; et
al. |
July 1, 2021 |
METHODS AND APPARATUS FOR IMPROVING THE IMAGE RESOLUTION AND
SENSITIVITY OF WHOLE-BODY POSITRON EMISSION TOMOGRAPHY (PET)
IMAGING
Abstract
A positron emission tomography (PET) technique that can enhance
the image resolution and system sensitivity of a clinical PET/CT
scanner for imaging a whole body or a target region of a subject is
provided. The system includes a detector array and a detector
panel. The detector array includes an array of gamma ray detectors
defining a field of view of a scanner and configured to detect at
least one coincidence event. The detector panel includes an array
of gamma ray detectors having a higher intrinsic spatial resolution
than the detector array and positioned in closer proximity to a
patient table than the detector array. The detector panel is
positioned outside the field of view defined by the detector array
during at least a portion of scanning by the PET system. The
detector panel is configured to detect at least one coincidence
event in cooperation with the detector array. The control unit is
configured to control the detector array, the detector panel, and
the patient bed to operate in cooperation with each other.
Inventors: |
Tai; Yuan-Chuan; (St. Louis,
MO) ; O'Sullivan; Joseph; (St. Louis, MO) ;
Li; Ke; (St. Louis, MO) ; Wang; Qiang; (St.
Louis, MO) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Tai; Yuan-Chuan
O'Sullivan; Joseph
Li; Ke
Wang; Qiang |
St. Louis
St. Louis
St. Louis
St. Louis |
MO
MO
MO
MO |
US
US
US
US |
|
|
Assignee: |
Washington University
St. Louis
MO
|
Prior
Publication: |
|
Document Identifier |
Publication Date |
|
US 20200170588 A1 |
June 4, 2020 |
|
|
Family ID: |
1000005650093 |
Appl. No.: |
16/697722 |
Filed: |
November 27, 2019 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
62773595 |
Nov 30, 2018 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
A61B 6/4233 20130101;
A61B 6/037 20130101; G01T 1/2985 20130101 |
International
Class: |
A61B 6/03 20060101
A61B006/03; G01T 1/29 20060101 G01T001/29; A61B 6/00 20060101
A61B006/00 |
Goverment Interests
GOVERNMENTAL SUPPORT
[0002] This invention was made with government support under grant
number RO1 CA136554 awarded by National Institutes of Health. The
government has certain rights in the invention.
Claims
1. A positron emission tomography (PET) system for imaging a target
region of a subject, the system comprising: a detector array
comprising an array of gamma ray detectors defining a field of view
and configured to detect at least one coincidence event; a detector
panel comprising an array of gamma ray detectors having higher
resolutions than the detector array and positioned in closer
proximity to a patient table than the detector array, the detector
panel positioned outside the field of view defined by the detector
array during at least a portion of scanning by the PET system, the
detector panel configured to detect at least one coincidence event
in cooperation with the detector array; and a control unit
configured to be in communication with the detector panel and the
detector array and configured to control the detector array and the
detector panel to operate in cooperation with each other.
2. The system of claim 1, wherein the detector panel further
comprises an outsert positioned outside the field of view defined
by the detector array during entire scanning by the PET system.
3. The system of claim 2, further comprising at least one
additional detector panel positioned in closer proximity to the
patient table than the detector array, wherein the at least one
additional detector panel is in communication with the control unit
and is configured to detect at least one additional coincidence
event in cooperation with the detector array and the detector
panel.
4. The system of claim 1, wherein the detector panel is movable
across the field of view defined by the detector array during
scanning by the PET system.
5. The system of claim 4, wherein the detector panel is positioned
in closer proximity to a target region in the subject than the
detector array, and the detector array remains stationary relative
to the target region during the scanning by the PET system.
6. The system of claim 1, further comprising a patient table on
which the subject lies during scanning by the PET system, wherein
the detector panel is positioned below the patient table during the
scanning by the PET system.
7. The system of claim 1, wherein the control unit is configured to
generate an image of the target region based on the at least one
coincidence event detected by the detector array and the at least
one coincidence event detected by the detector panel.
8. A device to enhance an image resolution of a positron emission
tomography (PET) system, the PET system configured to image a
target region of a subject, the device comprising: a detector panel
comprising an array of gamma ray detectors having higher
resolutions than detectors of a detector array of the PET system
and positioned in closer proximity to a patient table than the
detector array, the detector panel to be positioned outside a field
of view defined by the detector array during at least a portion of
scanning by the PET system and configured to detect at least one
coincidence event in cooperation with the detector array, wherein
the detector panel is configured to be in communication with a
control unit of the PET system, the control unit configured to be
in communication also with the detector array and configured to
control the detector array and the detector panel to operate in
cooperation with each other.
9. The device of claim 8, wherein the detector panel further
comprises an outsert to be positioned outside a field of view
defined by the detector array during entire scanning by the PET
system.
10. The device of claim 9, further comprising at least one
additional detector panel to be positioned in closer proximity to
the patient table than the detector array, wherein the at least one
additional detector panel is configured to be in communication with
the control unit and is configured to detect at least one
additional coincidence event in cooperation with the detector array
and the detector panel.
11. The device of claim 8, wherein the detector panel is configured
to be movable across the field of view defined by the detector
array during scanning by the PET system.
12. The device of claim 11, wherein the detector panel is
configured to be positioned in closer proximity to the target
region in the subject than the detector array while the detector
array remains stationary relative to the target region during the
scanning by the PET system.
13. The device of claim 8, wherein the detector panel is configured
to be positioned below a patient table on which the subject lies
during the scanning by the PET system.
14. The device of claim 8, wherein a surface of the detector panel
that is to be positioned facing the subject during scanning by the
PET system is substantially flat.
15. A method for enhancing a resolution of an image of a target
region within a subject obtained using a positron emission
tomography (PET) system, wherein the PET system includes a detector
array including an array of gamma ray detectors that define a field
of view and configured to detect at least one coincidence event and
a control unit configured to be in communication with the detector
array and configured to control the detector array, the method
comprising: providing a detector panel comprising an array of gamma
ray detectors having higher resolutions than the gamma ray
detectors of the detector array, the detector panel configured to
be positioned in closer proximity to a patient table than the
detector array and positioned outside the field of view defined by
the detector array during at least a portion of scanning by the PET
system, the detector panel configured to detect at least one
coincidence event in cooperation with the detector array, and the
detector panel configured to be in communication with the control
unit and to be controlled by the control unit to operate in
cooperation with the detector array.
16. The method of claim 15, further comprising: positioning the
detector panel in closer proximity to the patient table than the
detector array and outside the field of view defined by the
detector array during at least a portion of scanning by the PET
system; detecting the at least one coincidence event using the
detector array and the at least one coincidence event using the
detector panel in cooperation with the detector array; and
reconstructing an image of the target region of the subject using
the at least one coincidence event detected by the detector array
combined with the at least one coincidence event detected by the
detector panel.
17. The method of claim 16, wherein providing a detector panel
further comprises providing the detector panel that includes an
outsert configured to be positioned outside the field of view
defined by the detector array during entire scanning by the PET
system.
18. The method of claim 17, further comprising: providing at least
one additional detector panel including an additional array of
high-resolution gamma ray detectors; positioning the additional
detector panel in closer proximity to the patient table than the
detector array and outside the field of view defined by the
detector array during entire scanning by the PET system; detecting
at least one additional coincidence event using the at least one
additional detector panel in cooperation with the detector array
and the detector panel; and reconstructing an image of the target
region of the subject using the at least one additional coincidence
event combined with the at least one coincidence event detected by
the detector array and the at least one coincidence event detected
by the detector panel.
19. The method of claim 16, further comprising moving the detector
panel across the field of view defined by the detector array during
scanning by the PET system.
20. The method of claim 19, wherein positioning the detector panel
further comprises positioning the detector panel in closer
proximity to the target region than the detector array, and moving
the detector panel further comprises moving the detector panel and
the patient table such that the detector panel remains stationary
relative to the target region during the scanning by the PET
system.
Description
CROSS-REFERENCE TO RELATED APPLICATION
[0001] This application claims the benefit of U.S. Provisional
Application Ser. No. 62/773,595, filed Nov. 30, 2018, entitled
"Methods and Apparatus for Improving the Image Resolution and
Sensitivity of Whole-Body PET Imaging," which is hereby
incorporated herein by reference in its entirety. invention.
BACKGROUND
[0003] The field of the disclosure relates generally to imaging
systems and methods, and more particularly to systems and methods
for improving image resolution and sensitivity of a whole body
positron emission tomography (PET) system by including one or more
detector panel in addition to the detector array of the PET
system.
[0004] Cancer continues to be a leading cause of death in the US
and worldwide. Patients with early stage cancer are treated with
surgery and patients with advanced stage disease or lymph node
metastases are treated with chemotherapy or radiation therapy.
Surgery is the preferred treatment for early stage disease because
it is curative. If patients undergo surgery and are found to have
metastatic cancer, the patients will require additional
chemotherapy or radiation therapy post-operatively. A treatment of
combining surgery and post-operative chemo- or radiation-therapy
incurs much greater morbidity than treatment with either treatment
alone. Therefore, accurate staging of the disease is crucial for
the treatment planning and disease management for cancer
patients.
[0005] To guide therapy choice, pre-treatment imaging is obtained
with combinations of computed tomography (CT), positron emission
tomography/CT (PET/CT), or magnetic resonance imaging (MRI).
Fluorodeoxyglucose-PET (FDG-PET) generally outperforms CT and MRI
for metastatic cancer detection. Unfortunately, the sensitivity of
PET/CT imaging for detection of lymph node metastasis may decrease
significantly in early stage disease. For example, the sensitivity
of PET/CT for detecting pelvic lymph node metastases is only
approximately 50% in patients with stages Ib to IIa diseases in
cervical cancer patients. That is, approximately one-half of
patients in this population with positive pelvic lymph nodes will
be under-staged and undergo needless surgery resulting in increased
rates of severe toxicity by receiving both surgery and
post-operative chemo-radiation. Accordingly, there is an unmet
clinical need for technologies that improve the detection of
metastatic lymph nodes in order to decrease unnecessary morbidity
and improve patient management.
[0006] Combined PET and CT is a proven technique for the detection
of primary and metastatic cancers. It is used clinically for
staging, restaging, and monitoring of treatment response of many
types of cancers. The detection of lesions of unknown location is
typically done by a whole-body imaging protocol scanning a patient
from the skull apex to upper thighs in multiple bed positions. The
actual scan duration is often limited to 2-3 minutes per position
due to factors such as a patient's ability to remain still, an
amount of attenuation by patient's body mass, clinical throughput,
and the cost-benefit ratio of extended imaging time. With such
short scan duration, the quality of whole-body PET images is often
limited by counting statistics rather than the intrinsic spatial
resolution of a scanner. Therefore, there is a need of systems and
methods that enhance the signal-to-noise ratio of images (e.g.,
time-of-flight (TOF) PET) with increased counting statistics.
BRIEF DESCRIPTION
[0007] In one aspect, a positron emission tomography (PET) system
of imaging a target region of a subject is provided. The system
includes a detector array and a detector panel. The detector array
includes an array of gamma ray detectors defining a field of view
and configured to detect at least one coincidence event. The
detector panel includes an array of gamma ray detectors having
higher resolutions than the detector array and positioned in closer
proximity to a patient table than the detector array. The detector
panel is positioned outside the field of view defined by the
detector array during at least a portion of scanning by the PET
system. The detector panel is configured to detect at least one
coincidence event in cooperation with the detector array. The
control unit is configured to be in communication with the detector
panel and the detector array, and configured to control the
detector array and the detector panel to operate in cooperation
with each other.
[0008] In another aspect, a device to enhance an image resolution
of a PET system is provided. The PET system is configured to image
a target region of a subject. The device includes a detector panel
including an array of gamma ray detectors having higher resolutions
than a detector array of the PET system and positioned in closer
proximity to a patient table than the detector array. The detector
panel is positioned outside a field of view defined by the detector
array during at least a portion of scanning by the PET system and
configured to detect at least one coincidence event in cooperation
with the detector array. The detector panel is configured to be in
communication with a control unit of the PET system, the control
unit configured to be in communication also with the detector array
and configured to control the detector array and the detector panel
to operate in cooperation with each other.
[0009] In yet another aspect, a method for enhancing a resolution
of an image of a target region within a subject obtained using a
PET system is provided. The PET system includes a detector array
including an array of gamma ray detectors that define a field of
view and configured to detect at least one coincidence event and a
control unit configured to be in communication with the detector
array and configured to control the detector array. The method
includes providing a detector panel including an array of gamma ray
detectors having higher resolutions than the detector array, the
detector panel configured to be positioned in closer proximity to a
patient table than the detector array and positioned outside the
field of view defined by the detector array during at least a
portion of scanning by the PET system. The detector panel is
configured to detect at least one coincidence event in cooperation
with the detector array, and the control unit configured to be in
communication also with the detector panel and configured to
control the detector array and the detector panel to operate in
cooperation with each other.
BRIEF DESCRIPTION OF THE DRAWINGS
[0010] The patent or application file includes at least one drawing
executed in color. Copies of this patent or patent application
publication with color drawing(s) will be provided by the Office
upon request and payment of the necessary fee.
[0011] The drawings described below illustrate various aspects of
the disclosure.
[0012] FIG. 1A is a block diagram of an example positron emission
tomography (PET) system.
[0013] FIG. 1B is a block diagram of an example imaging system
using the PET system shown in FIG. 1A.
[0014] FIG. 2 is a block diagram of an example computing
device.
[0015] FIG. 3A is a schematic diagram of an example of augmented
whole-body scanning via magnifying PET (AWSM-PET) scanner system
that includes an example outsert.
[0016] FIG. 3B is an illustration of fields of view (FOVs) of the
scanner detectors and of detectors of the outsert shown in FIG.
3A.
[0017] FIG. 3C is an illustration of the FOVs of the scanner
detectors and the detectors of the outsert shown in FIG. 3A after
the patient table and a patient's body have been moved along the
axial direction of the scanner.
[0018] FIG. 3D is an illustration of the scanner system shown in
FIG. 3A including two outserts and the FOVs covered by their
detectors.
[0019] FIG. 4A is a rear perspective view of a PET scanner having
virtual-pinhole magnifying PET (VP-PET) insert.
[0020] FIG. 4B is a front perspective view of the PET scanner shown
in FIG. 4A.
[0021] FIG. 5A is an illustration of images of .sup.68Ge line
sources acquired using scanner detectors of the PET scanner shown
in FIGS. 4A-4B.
[0022] FIG. 5B is an illustration of images of .sup.68Ge line
sources acquired using scanner detectors of the PET scanner shown
in FIG. 4A and the detectors of the VP-PET shown in FIG. 4A.
[0023] FIG. 5C is an illustration of profiles along lines in the
images shown in FIGS. 5A and 5B.
[0024] FIG. 6A is an illustration of sensitivity profiles of the
scanner system shown in FIGS. 3A-3D with one outsert.
[0025] FIG. 6B is an illustration of sensitivity profiles of the
scanner system shown in FIGS. 3A-3D with two outserts.
[0026] FIG. 7A is an illustration of a decay map of detected
coincidence events from a body phantom.
[0027] FIG. 7B is an image of the phantom shown in FIG. 7A that is
acquired by a Biograph 40.TM. PET scanner.
[0028] FIG. 7C is an image of the phantom shown in FIG. 7A that is
acquired by a Biograph Vision.TM. PET scanner.
[0029] FIG. 7D is an image of the phantom shown in FIG. 7A that is
acquired by a Biograph Vision.TM. PET scanner including one
outsert.
[0030] FIG. 7E is an image of the phantom shown in FIG. 7A that is
acquired by a Biograph Vision.TM. PET scanner including two
outserts.
[0031] FIG. 8 is an illustration of contrast recovery coefficient
(CRC) measured by a Biograph Vision.TM. PET scanner and the
Biograph Vision.TM. PET scanner including one or two outserts.
[0032] FIG. 9A is an illustration of geometry of a Monte Carlo
simulation study using a VP magnifying PET insert for a simulated
breast and body imaging.
[0033] FIG. 9B is a magnified view of the section of the simulated
breast region as marked in FIG. 9A.
[0034] FIG. 9C are images acquired by the scanner detectors.
[0035] FIG. 9D shows images acquired by a VP-PET scanner.
[0036] FIG. 10 is a schema of an example human study.
[0037] FIG. 11A is an illustration of a torso phantom that includes
a breast phantom.
[0038] FIG. 11B is an illustration of the breast phantom of the
torso phantom shown in FIG. 11A.
[0039] FIG. 11C is an image of the torso phantom shown in FIG. 11A
that is acquired by scanner detectors.
[0040] FIG. 11D is an image of the torso phantom shown in FIG. 11D
that is acquired by a VP-PET scanner.
[0041] FIG. 12A is a partial rear perspective view of another
example PET scanner system that uses targeted VP-PET (TVP-PET)
technology.
[0042] FIG. 12B is an illustration of the FOVs covered by scanner
detectors and high-resolution detectors of the system shown in FIG.
12A as a target region moves through the FOV of the scanner
detectors.
[0043] FIG. 13A is an illustration of an example a compact PET
detector module.
[0044] FIG. 13B is a flood image of the detector module shown in
FIG. 13B that is read out by PET/CT scanner electronics.
[0045] FIG. 13C is a front view of the detector module shown in
FIG. 13A with its cover removed and a cell phone placed next to for
size reference.
[0046] FIG. 14A is an illustration of a setup of a torso phantom
imaging experiment with a VP-PET insert placed underneath the
patient table.
[0047] FIG. 14B is an illustration of the torso phantom shown in
FIG. 14 having fillable lesions of various sizes.
[0048] FIG. 15A is an illustration of scanner setups of a torso
phantom imaging study.
[0049] FIG. 15B are images corresponding to the scanner setups
shown in FIG. 15A.
[0050] FIG. 16A is an illustration of the mean and standard
deviation of CRC as a function of tumor sizes of the phantom used
for FIGS. 15A and 15B.
[0051] FIG. 16B is an illustration of the ratio of CRC with and
without the VP-PET technology.
[0052] FIG. 17A is an image of the phantom shown in FIG. 14A that
is acquired by scanner detectors of a PET/CT scanner.
[0053] FIG. 17B is an image of the phantom shown in FIG. 14A that
is acquired by the PET/CT scanner equipped with a VP-PET technology
having flat-panel detectors placed below a patient table.
[0054] FIG. 17C is an image of the phantom shown in FIG. 14A that
is acquired by the PET/CT scanner equipped with the VP-PET
technology used for FIG. 17B and using a TVP-PET protocol.
[0055] FIG. 18A is an illustration of the mean and standard
deviation of CRC as function of tumor sizes of the phantom shown in
FIG. 14A.
[0056] FIG. 18B is an illustration of the ratio of CRC for VP-PET
and TVP-PET relative to the native whole-body PET/CT scanner.
DETAILED DESCRIPTION
[0057] The systems and methods described herein relate to imaging
systems and methods, and more specifically, systems and methods of
a positron emission tomography (PET) system having a detector panel
of high-resolution gamma-ray detectors positioned in closer
proximity to a patient table than the detector array of the scanner
detectors for imaging a subject. As used herein, a subject may be a
human or an animal, or part of a human or an animal. With the
detector panel, the signal-to-noise ratio of the PET is
increased.
[0058] The suboptimal performance of whole-body Fluorodeoxyglucose
(FDG) PET/computed tomography (PET/CT) for detecting small lesions
is partially due to the limited counting statistics of the data and
partially due to the limited spatial resolution of clinical
scanners. To find distant metastasis at unknown location(s), a
patient is scanned from the base of skull to thighs in multiple
steps, typically using 2-3 minutes per bed position
non-discriminatorily. With this short scan duration, the
statistical uncertainty increases the difficulty for an already
challenging task of delineating a small uptake in a limited number
of cancerous cells from overwhelming background noise due to
non-specific uptakes in surrounding tissue and blood. To
accommodate a patient's body, clinical PET scanners include large
rings of .gamma.-ray detectors with a ring diameter typically
greater than 80 cm. The image resolution of PET is known to be
limited by positron range, photon acolinearity and detector
intrinsic spatial resolution. A large ring diameter limits the
image resolution of a whole-body PET scanner to be no better than
approximately 2 mm full-width-at-half-maximum (FWHM) by the photon
acolinearity effect, where the two .gamma. rays from a positron
annihilation are not emitted along a straight line. This resolution
limit of PET precludes .gamma.-ray detectors having further
increased resolution in a whole-body scanner design because the
higher-cost detectors do not significantly improve the overall
image resolution.
[0059] The systems and methods disclosed herein include a detector
panel having high-resolution gamma-ray detectors in addition to
scanner detectors to increase the signal-to-noise ratio of a PET
system. The high-resolution detectors have higher resolutions than
the detector array of the scanner detectors and are positioned in
closer proximity to a patient table than the detector array.
[0060] In some embodiments, an augmented whole-body scanning via
magnifying PET (AWSM-PET) system is used. A detector panel of
high-resolution detectors is positioned outside of the field of
view (FOV) covered by the scanner detectors such that the detector
panel detects incidents outside the FOV of the scanner
detectors.
[0061] In some embodiment, a targeted virtual-pinhole PET (TVP-PET)
system is used. A detector panel of high-resolution detectors is
positioned in close proximity of a target region in a subject and
follows the target region as the patient table moves during a PET
scanning. As a result, the system provides increased incident
detection of the target region. A target region may be anatomy in a
patient in which a physician is interested, such as a cervix of a
subject.
[0062] FIG. 1A is a block diagram of an example PET system 100.
System 100 includes detectors 102 having a plurality of
scintillators, photomultiplier tubes (PMTs) or silicon
photomultipliers (SiPM) 104, a control unit 106, a signal detection
and acquisition unit 108, a signal and image processing unit 110,
and a patient bed unit 112. Scintillators of detectors 102 include
scintillation materials such as cerium-doped lutetium
oxyorthosilicate (LSO) that convert .gamma.-rays emitted from a
subject to photons. Two photons are observed at roughly at the same
time (in coincidence) in a detector ring. An annihilation event,
i.e., the radioactive tracer, can be located somewhere on the line
connecting the two photon-detection points. The detected photons
are therefore called coincidence events and used to reconstruct an
image of the radioactive tracer that reflects the function of the
anatomy of the subject at the location of the radioactive tracer.
During scanning or detection of coincidence events by system 100,
detectors 102 scan around a subject. PMTs or SiPM 104 convert the
photons to electrical signals, which are then acquired, processed,
and converted to digital signals through signal detection and
acquisition unit 108. Images of the subject are generated based on
the outputted digital signals using signal and image processing
unit 110. To image a section of a patient's body that is longer
than the scanner's imaging field-of-view, control unit 106 may
control patient bed 112 to move a patient through the scanner's
imaging field-of-view either continuously or step-by-step while the
coincidence events are collected. Control unit 106 is in
communication with and controls detectors 102, PMTs or SiPM 104,
signal detection and acquisition unit 108, signal and image
processing unit 110, and patient bed 112 such that system 100
detects coincidence events and reconstruct an image based on the
coincidence events.
[0063] FIG. 1B is a block diagram of an example imaging system 150
using an imaging device 152. Imaging device may be PET system 100
shown in FIG. 1A, a PET/CT system, or a PET/magnetic resonance
imaging (PET/MRI) system. System 150 further includes a computing
device 154 to receive imaging data from the imaging device 152.
Computing device 154 may be configured to control the imaging
device 152.
[0064] System 150 further includes a data management system 158
that is coupled to computing device 154 via a network 159. In some
embodiment, the computing device 154 includes a data management
system 158. Data management system 158 may be any device capable of
accessing network 159 including, without limitation, a desktop
computer, a laptop computer, or other web-based connectable
equipment. More specifically, in the example embodiment, data
management system 158 includes a database 160 that includes
previously acquired data of other subjects. In the example
embodiment, database 160 can be fully or partially implemented in a
cloud computing environment such that data from the database is
received from one or more computers (not shown) within system 150
or remote from system 150. In the example embodiment, the
previously acquired data of the other subjects may include, for
example, a plurality of measurements of lesion region of other
subjects. The information about lesion location from other subjects
who have similar diseases can be used to direct the scanning of
patients using the systems and methods described herein. Database
160 can also include any additional information of each of the
subjects that enables system 150 to function as described
herein.
[0065] In the example embodiment, in system 150, computing device
154 is coupled to imaging device 152 via a data conduit 156. It
should be noted that, as used herein, the term "couple" is not
limited to a direct mechanical, electrical, and/or communication
connection between components, but may also include an indirect
mechanical, electrical, and/or communication connection between
multiple components. Imaging device 152 may communicate with
computing device 154 using a wired network connection (e.g.,
Ethernet or an optical fiber), a wireless communication means, such
as radio frequency (RF), e.g., FM radio and/or digital audio
broadcasting, an Institute of Electrical and Electronics Engineers
(IEEE.RTM.) 802.11 standard (e.g., 802.11(g) or 802.11(n)), the
Worldwide Interoperability for Microwave Access (WIMAX.RTM.)
standard, a short-range wireless communication channel such as
BLUETOOTH.RTM., a cellular phone technology (e.g., the Global
Standard for Mobile communication (GSM)), a satellite communication
link, and/or any other suitable communication means. IEEE is a
registered trademark of the Institute of Electrical and Electronics
Engineers, Inc., of New York, N.Y. WIMAX is a registered trademark
of WiMax Forum, of Beaverton, Oreg. BLUETOOTH is a registered
trademark of Bluetooth SIG, Inc. of Kirkland, Wash.
[0066] In various embodiments, data management system 158
communicates with computing device 154 using a wired network
connection (e.g., Ethernet or an optical fiber), a wireless
communication means, such as, but not limited to radio frequency
(RF), e.g., FM radio and/or digital audio broadcasting, an
Institute of Electrical and Electronics Engineers (IEEE.RTM.)
802.11 standard (e.g., 802.11(g) or 802.11(n)), the Worldwide
Interoperability for Microwave Access (WIMAX.RTM.) standard, a
cellular phone technology (e.g., the Global Standard for Mobile
communication (GSM)), a satellite communication link, and/or any
other suitable communication means. More specifically, in the
example embodiment, data management system 158 transmits the data
for the subjects to computing device 154. While the data is shown
as being stored in database 160 within data management system 158,
it should be noted that the data of the subjects may be stored in
another system and/or device. For example, computing device 154 may
store the data therein.
[0067] FIG. 2 is a block diagram of computing device 154 such as
control unit 106. In the example embodiment, computing device 154
includes a user interface 204 that receives at least one input from
a user, such as an operator of imaging device 152 or system 150.
User interface 204 may include a keyboard 206 that enables the user
to input pertinent information. Additionally or alternatively, user
interface 204 may include, for example, a pointing device, a mouse,
a stylus, a touch sensitive panel (e.g., a touch pad or a touch
screen), a gyroscope, an accelerometer, a position detector, and/or
an audio input interface (e.g., including a microphone).
[0068] Moreover, in the example embodiment, computing device 154
includes a presentation interface 207 that presents information,
such as input events and/or validation results, to the user.
Presentation interface 207 may include a display adapter 208 that
is coupled to at least one display device 210. More specifically,
in the example embodiment, display device 210 may be a visual
display device, such as a cathode ray tube (CRT), a liquid crystal
display (LCD), a light emitting diode (LED) display, an organic LED
(OLED) display, and/or an "electronic ink" display. Alternatively,
presentation interface 207 may include an audio output device
(e.g., an audio adapter and/or a speaker) and/or a printer.
[0069] Computing device 154 also includes a processor 214 and a
memory device 218. Processor 214 is coupled to user interface 204,
presentation interface 207, and to memory device 218 via a system
bus 220. In the example embodiment, processor 214 communicates with
the user, such as by prompting the user via presentation interface
207 and/or by receiving user inputs via user interface 204. The
term "processor" refers generally to any programmable system
including systems and microcontrollers, reduced instruction set
circuits (RISC), application specific integrated circuits (ASIC),
programmable logic circuits (PLC), and any other circuit or
processor capable of executing the functions described herein. The
above examples are exemplary only, and thus are not intended to
limit in any way the definition and/or meaning of the term
"processor."
[0070] In the example embodiment, memory device 218 includes one or
more devices that enable information, such as executable
instructions and/or other data, to be stored and retrieved.
Moreover, memory device 218 includes one or more computer readable
media, such as, without limitation, dynamic random access memory
(DRAM), static random access memory (SRAM), a solid state disk,
and/or a hard disk. In the example embodiment, memory device 218
stores, without limitation, application source code, application
object code, configuration data, additional input events,
application states, assertion statements, validation results,
and/or any other type of data. Computing device 154, in the example
embodiment, may also include a communication interface 230 that is
coupled to processor 214 via system bus 220. Moreover,
communication interface 230 is communicatively coupled to imaging
device 152 and data management system 158.
[0071] In the example embodiment, processor 214 may be programmed
by encoding an operation using one or more executable instructions
and providing the executable instructions in memory device 218. In
the example embodiment, processor 214 is programmed to select a
plurality of measurements that are received from imaging device
152. The plurality of measurements may include, for example, a
plurality of voxels of at least one image of the subject, wherein
the image may be generated by processor 214 within computing device
154. The image may also be generated by an imaging device (not
shown) that may be coupled to computing device 154 and imaging
device 152, wherein the imaging device may generate the image based
on the data received from imaging device 152 and then the imaging
device may transmit the image to computing device 154 for storage
within memory device 218. Alternatively, the plurality of
measurements may include any other type measurement of the lesion
region that enables system 150 to function as described herein.
[0072] Although, in some of the examples provided below, the
systems and methods disclosed herein are used on certain part of
the body or certain types of lesions, the systems and methods are
not limited to that part of human or animal body or that type of
lesions. Further, method aspects will be in part apparent and in
part explicitly discussed in the following description.
EXAMPLES
Example 1
Augmented Whole-Body Scanning via Magnifying PET (AWSM-PET)
[0073] FIGS. 3A-3D are illustrations of an example system 300
including the AWSM-PET technology. System 300 is a whole-body PET
scanner augmented with or including additional detectors besides
detectors of the whole-body scanner to increase the resolution and
sensitivity of the system. In the exemplary embodiment, system 300
includes a detector array 302 of the scanner and a detector panel
304 additional to scanner detectors 302. Detector array 302
includes an array of gamma-ray detectors configured to detect at
least one coincidence event. Detector panel 304 includes an array
of high-resolution gamma ray detectors positioned in closer
proximity to a patient table 112 relative to detector array 302.
FIG. 3A is schematic diagram of a prototype AWSM-PET device
integrated with a Siemens Biograph Vision.TM. PET/CT scanner. A
group of high-resolution PET detectors are placed close to a
patient's body but outside the scanner's imaging FOV in the axial
direction. These added-on detectors are sometimes referred to
herein as "outsert" in contrast to an "insert" where
high-resolution detectors are placed inside a scanner's imaging
FOV. That is, during the entire scanning by system 300, detector
panel 304 is positioned outside the FOV of detector array 302.
Being outside of the scanner, the presence of these outsert
detectors does not interfere with the operation of the native
scanner. Therefore, standard whole-body PET/CT images can still be
acquired independent of the add-on AWSM-PET device. FIGS. 3B and 3C
show a group of high-resolution "outsert" detectors are placed
outside of the scanner's axial FOV. Coincidence detection between
the outsert detectors and scanner detectors creates an augmented
scan zone that collects additional events simultaneously when a
patient undergoes a whole-body scan. A section of the body is first
scanned by the scanner's native FOV (in gray) and then by the
augmented imaging FOV (in blue) when a patient is moved from one
bed position (FIG. 3B) to the next position (3C). FIG. 3D shows a
dual-panel AWSM-PET device that generates 2 additional augmented
scan zones 2 and 3, besides scan zone 1 with a single additional
panel to further improve the overall resolution and
sensitivity.
[0074] Establishing coincidence detection between the outsert
detectors and the scanner detectors permits the system to detect
additional coincidence events that are otherwise missed by the
native scanner. For example, without an outsert, annihilation
.gamma. rays originated from the patient's body in the blue zone
(augmented scan zone marked in FIG. 3B) may be undetected by the
scanner if one of the two .gamma. rays travels outwards and hit the
outsert detectors. These extra coincidence events, when augmented
to the native scanner data for joint image reconstruction, increase
the overall counting statistics and therefore improve the noise
characteristics of the PET images. Further, the outsert's
high-resolution detectors are placed close to a patient's body for
zoom-in imaging using a virtual-pinhole PET geometry. The image
resolution of the native scanner can be improved when the
higher-resolution events are added to jointly reconstruct PET
images. This is an extension and new application of the
virtual-pinhole PET insert technology that we previously developed
and validated, as will be described below. As a patient's body is
moved across the scanner's imaging FOV during a whole-body imaging
session, the patient is also being scanned by the AWSM-PET device
simultaneously. An optimized imaging protocol and accurate image
reconstruction algorithm allows AWSM-PET technology to enhance the
performance of a PET scanner without increasing scan time for
whole-body cancer imaging.
1.1. Prior Prototype Virtual-Pinhole (VP) PET System
[0075] A proof-of-principle virtual-pinhole magnifying PET insert
system (shown in FIGS. 4A and 4B) was previously built using 28 PET
detector modules and arranged them to form two half rings. FIGS. 4A
and 4B show front (FIG. 4A) and rear (FIG. 4B) views of a human
PET/CT scanner integrated with the insert. Each detector module
includes 13.times.13 lutetium oxyorthosilicate (LSO) crystals read
out by a multi-anode PMT. The LSO crystals measure
2.times.2.times.5 mm each. The system was centered in the imaging
FOV in a Siemens Biograph 40.TM. PET/CT scanner. One of the PET
detector rings (out of a 4-ring system) was disabled and its
associated electronics were used to process the VP-PET detector
signal for coincidence detection. This integrated system has a
transaxial FOV of approximately 24 cm in diameter and an axial FOV
of 16.4 cm, with approximately 6 cm of the axial FOV in the center
having higher resolution images.
[0076] The image resolution of the integrated system was evaluated
using a sealed .sup.68Ge line source placed at 9 different
locations in the imaging FOV. .sup.68Ge line source images from the
scanner are shown in FIG. 5A, and images from the augmented system
are shown in FIG. 5B. FIG. 5C shows profiles of the line source
images along a line 502, 504 depicted in FIGS. 5A and 5B. The
images are reconstructed using the filtered back-projection
algorithm with a ramp filter and no smoothing. The resolution of
the original PET scanner ranges from 4.3 to 5.5 mm FWHM after the
source dimension was corrected and subtracted. This agrees well
with the reported performance of the Biograp.TM. PET/CT scanner.
The resolution of the integrated VP-PET insert system ranges from
2.4 to 3.0 mm FWHM. This agrees well with theoretical calculation
using Equation (1) below, which predicts 2.3 mm FWHM at the center
of FOV.
R img .apprxeq. 1.25 R src 2 + [ 0.0088 d 1 d 2 ( d 1 + d 2 ) ] 2 +
[ d 2 w 1 + d 1 w 2 + d 2 w 1 - d 1 w 2 2 ( d 1 + d 2 ) ] 2 , ( 1 )
##EQU00001##
where d.sub.1=123 mm and d.sub.2=428 mm are the radii of the
detector rings in the virtual-pinhole PET device and the scanner,
respectively, and w.sub.1=2 mm and w.sub.2=4 mm are the width of
the LSO crystals in the virtual-pinhole PET and scanner,
respectively.
1.2. Estimated Performance of a Prototype AWSM-PET System
[0077] An example AWSM-PET technology was evaluated using a Siemens
Biograph Vision.TM. scanner through Monte Carlo simulation method.
Siemens Biograph Vision.TM. is a clinical PET/CT that includes a
128-slice CT and an 8-ring PET scanner to offer 25.6 cm axial FOV.
The PET scanner includes a total of 608 block detectors each made
of a LSO crystal array (10.times.10 elements of 3.2 mm pitches) and
a silicon photo-multiplier (SiPM) array (8.times.8 elements of 4 mm
pitches). The 608 detectors are divided into 19 groups of 32
detectors each. Each group of 32 block detectors are arranged to
form an eight (axially) by four (transaxially) array and supported
by a detector electronics assembly (DEA) that determines the
position, energy and timing of each event. Qualified single events
are packaged and transmitted to a Gantry Interface Module (GIM).
The GIM compares the time stamps of events from all 19 DEAs to
identify coincidence events. Valid coincidence events are packaged
and transmitted to a host computer as a stream of list mode
data.
[0078] Image resolution: The Siemens Biograph Vision.TM. scanner
has an excellent intrinsic spatial resolution (from 3.2 mm pitched
LSO crystals). Using the equation (1), and substituting w.sub.2=3.2
mm for the Biograph Vision.TM. scanner, the image resolution of an
AWSM-PET system is estimated to be slightly over 2.0 mm FWHM at
best if 1.6 mm LSO crystals are used for two outsert detectors. In
contrast, the native image resolution of the Biograph Vision.TM.
scanner is estimated to be from approximately 3.2 mm to more than 4
mm FWHM. If 0.8 mm LSO crystals are used for a dual-panel AWSM-PET
system, the image resolution of the system is estimated to be
better than 2 mm FWHM for a significant portion of the imaging FOV.
This will overcome the theoretical resolution limit of 2 mm FWHM
for whole-body PET imaging and may improve the diagnostic accuracy
of whole-body PET/CT image for the detection of small metastatic
lesions. Thus, the outserts and the technique disclosed herein
provide a unique and cost-effective upgrade option to enhance the
performance of the current state-of-the-art clinical whole-body
PET/CT scanners further.
[0079] System sensitivity: The gain in system sensitivity is
evaluated by simulating a Biograph Vision.TM. scanner with AWSM-PET
technology using Monte Carlo technique. The AWSM-PET device
includes either one or two outsert detectors. Each outsert detector
was simulated as 32 block detectors, where each block detector was
made of 40.times.40 LSO crystals each in the dimension of
0.8.times.0.8.times.7 mm. The timing resolution of the detectors
was assumed to be 250 ps FWHM. For a single-panel AWSM-PET system,
the outsert detector was positioned below the patient bed and
directly outside of the PET scanner's axial FOV (same geometry as
depicted in FIG. 3A). For a dual-panel AWSM-PET system, a second
outsert detector was positioned above the patient. A .sup.22Na
point source was stepped across the scanner's imaging FOV along the
axial direction at three different vertical positions (Y=-120 mm,
0, or 120 mm) from the center of FOV. FIG. 6A and 6B show the
sensitivity profiles of the simulated single-panel and dual-panel
AWSM-PET systems, respectively. Coincidence events measured by the
scanner (SS counts) or by the add-on detectors (OS counts) are
shown in Table 1 below.
TABLE-US-00001 TABLE 1 Biograph Vision .TM. + 1 Outsert
Y-coordinate SS counts OS counts % gain -120 mm 1916024 464560
24.25% 0 mm 1064610 213004 20.01% 120 mm 1914747 143924 7.52%
Biograph Vision .TM. + 2 Outserts Y-coordinate SS counts OS counts
OO counts % gain -120 mm 1904817 583121 84291 35.04% 0 mm 1057855
393156 115330 48.07% 120 mm 1908202 582068 84129 34.91%
[0080] The sensitivity gain of the scanner is computed from the
number of coincidence events detected by the outsert detectors
(outsert-to-scanner or OS counts, plus outsert-to-outsert or OO
counts) divided by the number of events detected by the native
scanner (SS counts), expressed in percentage. A scanner equipped
with a dual-panel AWSM-PET device could gain up to 48% more counts
than the native scanner for tissues near the center of the FOV. A
scanner equipped with a single-panel AWSM-PET device would gain 7%
to 24% more counts than the native scanner, with higher gain for
tissues near the outsert detector.
[0081] The improvement in counting statistics by AWSM-PET
technology is significant, especially with the dual-panel device. A
native Biograph Vision.TM. scanner includes 19 DEA to support a
total of 608 block detectors. An AWSM-PET device with two outsert
detectors will require two external DEA and will increases the
number of block detectors in the system by 10.5%. In return, the
overall system sensitivity is increased by 35% to 48% for
whole-body imaging applications. Therefore, the AWSM-PET technology
is not only an innovative solution to improve the image resolution
of a whole-body PET scanner, but also a cost-efficient strategy to
improve the sensitivity of a scanner.
[0082] Contrast recovery and lesion detectability: To evaluate the
benefits of the AWSM-PET technology for whole-body imaging, PET
scanners were simulated with the following configurations. The
object being imaged is a torso phantom with small lesions (4 mm in
diameter) of low contrast (tumor to background activity
concentration ratio (T/B)=4). Lesions that are hard to detect by
current clinical scanners are purposely chosen. The activity
concentration in the background is 5.3 kBq/mL assuming 10 mCi FDG
is uniformly distributed in a 70 kg body. The acquisition time is 3
minutes per position using a step size of 15 cm for whole-body
imaging. FIG. 7A shows the map of origins of coincidence events
detected by the scanners, where few counts originated deep inside a
body can escape and be detected. This illustrates the challenge of
detecting small lesions deep in a body. Four following scanner
configurations were simulated: Siemens Biograph 40.TM. (4 mm LSO
crystals, no time-of-flight PET (TOF-PET) capability), Biograph
Vision.TM. (3.2 mm LSO crystals, TOF-PET with <250 ps CRT),
Biograph Vision.TM. plus a single-panel AWSM-PET device (i.e., an
outsert), and Biograph Vision.TM. plus a dual-panel AWSM-PET device
(e.g., two outserts). The location and composition of the outsert
detectors are identical to those used to simulate the system
sensitivity in FIGS. 6A and 6B. The images were reconstructed with
1 mm.sup.3 voxel size using a list-mode image reconstruction
program. No regularization was applied during the reconstruction. A
post-reconstruction smoothing with a 3-point averaging (along all
three dimensions) was applied to the images before data
analysis.
[0083] Results in FIG. 7B show that none of the lesions is
detectable by the Biograph 40.TM., which has no TOF-PET feature and
uses 4 mm LSO detectors. FIG. 7C shows that a few lesions that are
close to the body surface may be detectable by the Biograph
Vision.TM., although still with little confidence. FIG. 7D and 7E
show that more lesions become detectable when an AWSM-PET device is
augmented to the Biograph Vision.TM. scanner.
[0084] FIG. 8 shows the contrast recovery coefficient of lesions in
FIGS. 7C, 7D and 7E. Lesions were grouped based on their
Y-coordinates, which affect the distance between the lesions and
the outsert detectors. As expected, the contrast recovery
coefficient was improved more significantly for lesions that are
closer to the outsert detectors. For lesions that are far away from
the outsert (e.g., the top row of lesions are 270 mm away from the
outsert in the single-panel AWSM-PET device), the contrast recovery
coefficient approaches that of the native scanner. In the case of a
dual-panel AWSM-PET system, contrast recovery coefficient is
enhanced more uniformly throughout the imaging FOV although still
spatially variant. Since the AWSM-PET feature does not interfere
with the operation of the native scanner, two sets of images are
generated and can be available to physicians for review, one set of
standard PET/CT images from the native scanner and one set of
jointly reconstructed images using the extra AWSM-PET events.
Therefore, the lesion detectability may still be improved by the
technology despite the level of enhancement may not be spatially
uniform.
[0085] Other design consideration: Based on the results in FIGS.
6A, 6B, and 8, a dual-panel AWSM-PET device may have more
significant and more uniform improvement in system sensitivity,
image resolution, and contrast recovery than a single-panel device.
However, an outsert detector in close proximity above a patient's
body may raise safety concerns regarding collision of a detector
with the patient. The position of the upper outsert detector,
therefore, should be adjusted automatically to conform to an
individual patient's body geometry. This will increase enhancement
in image resolution and system sensitivity.
[0086] 1.3. An Image Reconstruction and Data Correction Framework
for AWSM-PET System
[0087] A generalized image reconstruction framework was developed
that models PET systems of arbitrary geometries. This framework
includes computing a system model for PET scanners that are made of
cuboidal block detectors of arbitrary geometries, automatically
identifying symmetric lines of response that can share the same
forward operator during image reconstruction to save computation
and storage burden, and incorporating all appropriate data
correction techniques for statistical image reconstruction. This
reconstruction framework was used to support the development of
prototype virtual-pinhole magnifying PET insert systems where the
high-resolution add-on device was placed inside the scanner's
imaging FOV to zoom in on an organ of interest. With the
high-density detectors in the imaging FOV, several data correction
techniques used for PET were re-developed and validated. In the
framework, the attenuation correction is based on forward
projection of a composite attenuation map using CT images of the
patient and the known geometry of the virtual-pinhole magnifying
detectors. Scatter contribution is estimated using a single scatter
simulation technique. Random coincidences are corrected using the
standard delayed-window technique offered on the PET/CT scanner.
Normalization is based on a statistically-estimated model-based
technique. All these algorithms have been implemented and fully
validated.
[0088] To support the development of AWSM-PET technology,
additional improvements to the image reconstruction framework were
made. These improvements include: (1) switching from sinogram mode
to fast list-mode reconstruction; (2) implementing GPU based image
reconstruction and real-time system matrix computation; and (3)
adding the support for TOF-PET image reconstruction. These changes
may be included because some embodiments of the AWSM-PET system
have nearly twice the number of crystal elements as the original
scanner. Smaller image pixel size is used to reconstruct high
resolution images. These changes result in a system matrix size
that ranges from several hundred gigabytes to over a trillion
bytes. Switching to list-mode image reconstruction and computing
the system matrix in real-time by GPU reduce overall computational
time by a factor of more than 10.
[0089] In one embodiment, new algorithms are developed that use the
native scanner images as a prior when reconstructing the jointly
estimated AWSM-PET images. This further reduces the reconstruction
time. Data correction algorithms described above support the
AWSM-PET geometry. The quantitative accuracy of this image
reconstruction framework is validated using both Monte Carlo
simulation and actual phantom experiments.
1.4. Benefits of Resolution Enhancement by the AWSM-PET
technology
[0090] A Monte Carlo simulation study was conducted to evaluate how
one may use the virtual-pinhole magnifying PET technology to
improve cancer imaging. FIG. 9A shows a simplified torso phantom
with a breast attachment being imaged by a prototype
virtual-pinhole PET insert system (as shown in FIG. 4A and 4B).
FIG. 9B is a magnified view of the breast region showing 6 groups
of spherical tumors (2, 3, 4, 6, 8, and 12 mm in diameter (O)) were
placed in the breast and torso. The T/B in the phantom was varied
as 3, 6, 9, or 12, respectively. The simulated acquisition time was
2.26 or 6.78 minutes. The activity concentration in the body
background was 5.3 kBq/mL (143 nCi/mL), based on the assumption of
10 mCi of FDG uniformly distributed in a 70 kg patient. The PET/CT
scanner simulated was a Siemens Biograph-40.TM. with or without the
prototype virtual-pinhole magnifying PET insert attached.
[0091] FIG. 9C and 9D show images acquired by the native scanner
and by the same scanner equipped with virtual-pinhole magnifying
PET technology, respectively, under various of T/B conditions.
Cross comparison of these images reveals many benefits enabled by
the virtual-pinhole magnifying PET technology. When T/B=3, it is
difficult to detect even the biggest tumors (12 mm O) using the
native PET scanner regardless of the scan time (2.26 or 6.78-min).
In contrast, the scanner augmented with the virtual-pinhole
magnifying PET can detect the 12 mm O tumors (marked with a red
triangle) when the counting statistics are high (e.g. a 6.78-min
scan) due to its improved resolution and contrast recovery. When
T/B=6, the native scanner can detect 8 mm tumors (marked with a
green triangle) in a 6.78 min scan, but not in a 2.26 min scan. The
addition of virtual-pinhole magnifying PET enables the detection of
the 8 mm O tumors (marked with a green triangle) in a 2.26 min scan
(i.e. a whole-body scan). When the T/B=9, with high counting
statistics (scan time=6.78 min), the native scanner can detect 6 mm
O tumors (marked with a blue triangle). The addition of
virtual-pinhole magnifying PET enables the detection of the same
sized tumors when the T/B=6 (marked with a blue triangle). When
T/B=9 or 12, the addition of virtual-pinhole magnifying PET can
resolve 6 or 4 mm .circle-solid. tumors in a 2.26 min scan with
more confidence than the native PET scanner (marked with yellow
triangles). Under all conditions, a PET scanner augmented with the
virtual-pinhole magnifying PET outperforms the native PET scanner
itself
[0092] This study demonstrates that the virtual-pinhole magnifying
PET technology can enhance the performance of a clinical PET/CT
scanner for lesion detection even when the acquisition time is only
2-3 min/bed position. However, given that there are multiple ways
to benefit from this technology, an imaging protocol that will
achieve the best image quality for whole-body cancer imaging
without disruption to the current clinical workflow will be
desirable.
[0093] Continuous-bed-motion is a technology used for clinical
whole-body PET/CT imaging. Instead of stop-and-go motion, patient's
body is moved and scanned continuously during the whole-body PET
imaging protocol. A continuous bed motion will alleviate the effect
of the mismatched imaging FOV between the native scanner and the
augmented outsert detectors, as illustrated in FIGS. 3A-3D. If the
conventional stop-and-go motion is used, some tissues may be inside
the augmented scanning zone longer than other tissues, resulting in
non-uniform sensitivity profiles. Using the continuous-bed-motion
technology, the body will move through the augmented scanning zone
at a constant speed, resulting in a more uniform sensitivity
profile. It also produces PET images with more uniform noise
characteristic. Therefore the AWSM-PET imaging protocol may be
implemented using the continuous-bed-motion to have those
benefits.
1.5. Human Imaging Study Using AWSM-PET Technology
[0094] The AWSM-PET technology will increase the diagnostic
sensitivity of FDG-PET/CT for the detection of metastatic lymph
node, particularly for lesions that are smaller than the detection
limit of the standard whole-body FDG-PET/CT exam. In a planned
initial human imaging study, the performance of the AWSM-PET
technology will be evaluated against the standard whole-body PET/CT
images as to the detection of metastatic lymph nodes. The commonly
accepted gold standard to evaluate the diagnostic accuracy of an
imaging technology for metastatic lymph node detection is the
pathology report. Therefore, the study is designed to image
patients who will receive pathologic analysis of surgically
dissected lymph nodes as part of their standard-of-care. For
patients who have early stage diseases and are candidates for
radical surgery, they typically receive a pre-surgery FDG-PET/CT
scan to rule out metastases. If there is not metastatic disease
based on FDG-PET/CT images, they may receive surgery and
lymphadenectomy. The pathology report of the dissected lymph nodes
will serve as the reference standard to assess the incremental
diagnostic sensitivity of FDG-PET/CT for the detection of
metastatic lymph nodes when the AWSM-PET technology is added. The
population to be studied will be cancer patients who are surgical
candidates and their standard of care treatment may include
lymphadenectomy, and that they are scheduled to receive a
whole-body PET/CT scan as standard of care to rule out metastatic
diseases. The objective of the study is to measure the incremental
diagnostic sensitivity of the FDG-PET/CT when the AWSM-PET
technology is included for the detection of metastatic cancer.
[0095] The overall design of a human imaging study is illustrated
in the schema (FIG. 10). Cancer patients will be recruited to
participate in a study that includes two imaging sessions, a
standard whole-body FDG-PET/CT scan and an AWSM-PET/CT scan.
Patients will continue with their standard-of-care treatment based
on the result of the standard FDG PET/CT exam. Pathology reports of
dissected lymph nodes from patients who receive surgery will serve
as reference standard. Data acquired by the AWSM-PET protocol will
be reconstructed in the following two ways. Coincidence events from
the native scanner will be reconstructed using standard Siemens
software to produce standard whole-body PET/CT images (designated
as image 1). Coincidence events acquired by AWSM-PET outsert
detectors will be added to the scanner data to jointly reconstruct
AWSM-PET images (designated as image 2). Image 1 will be reviewed
by two nuclear medicine physicians to make diagnosis. After a
wash-out period of at least three months, images 1 and 2 will be
reviewed together by the same nuclear medicine physicians to make a
second diagnosis. The final evaluation will be determined by a
third reader when necessary. The number of metastatic lymph nodes
that are misclassified as false negative by the first read but
correctly identified as true positive after the AWSM-PET images are
included for the second read are counted. The results will be used
to estimate the incremental diagnostic sensitivity of FDG-PET/CT
for the detection of metastatic lymph nodes when the AWSM-PET
technology is used. The incremental diagnostic sensitivity will be
calculated on per-patient basis. Additionally, the number of sites
of probable metastatic disease detected by the AWSM-PET technology
will also be reported.
Example 2
Targeted Virtual-Pinhole PET (TVP-PET)
[0096] The example virtual-pinhole PET (VP-PET) systems bring the
benefits of organ-specific PET to a clinical PET/CT scanner. In
such systems, by establishing coincidence detection between a set
of high-resolution PET detectors and a PET scanner, each
high-resolution detector can act as an electronic pinhole
collimator to improve the image resolution of a whole-body PET
scanner. Several prototype VP-PET insert devices were developed,
which demonstrated the feasibility of improving the image
resolution of a clinical PET/CT scanner without compromising its
body imaging capability. FIG. 11A shows a torso phantom with a
breast phantom attachment positioned in the field of view of a
Siemens Biograph 40 PET/CT scanner. A prototype virtual-pinhole
magnifying PET insert system, as illustrated in FIGS. 4A-4D, was
positioned near the breast phantom attachment to enhance the image
resolution of the organ-of-interest, which is a breast in this
case. FIG. 11B shows a breast phantom by itself The breast phantom
includes spherical tumors. The breast compartment includes
spherical tumors of various sizes ranging from 3.59 mm to 11.4 mm
in diameter and filled with radioactivity having T/B=6.
Additionally, there are spherical tumors in the torso phantom with
sizes ranging from 3.59 mm to 11.4 mm in diameter. The body imaging
capability was preserved while improving the image resolution and
tumor contrast using the virtual-pinhole magnifying PET insert
technology. FIG. 11C shows an image acquired by the scanner. FIG.
11D shows an image at the same location as FIG. 11C acquired by
virtual-pinhole magnifying PET. Results in FIGS. 11C and 11D
demonstrate that the virtual-pinhole magnifying PET technology
improves image resolution within the breast section of the phantom.
The level of improvement is consistent with those observed in the
Monte Carlo simulation study in FIGS. 9C and 9D. The
virtual-pinhole magnifying PET technology enhances tumor contrast
in not only the breast compartment but also the torso despite the
level of improvement diminishes for tissues farther away from the
insert detectors. This advantage over organ-specific PET imagers
allows the technology to be extended for whole-body cancer imaging
using targeted virtual-pinhole PET (TVP-PET).
[0097] A second-generation VP-PET insert system or targeted VP-PET
system 1400 (FIGS. 12A-12B) was constructed with several technical
advances that may make it more suitable for, among other things,
human imaging studies. System 1400 includes detector array 302 and
detector panel 304. Different from PET system 300, in system 1400,
detector panel 304 is positioned proximate to a target region 1402
of a subject and follows target region 1402 during the scanning by
PET system 1400. As a result, detector panel 304 is outside the FOV
covered by detector array 302 during at least a portion of scanning
by system 1400. FIG. 12A is a second-generation VP-PET prototype
with an integrated robotic arm that can position a flat-panel
detector at an arbitrary location around a patient's body. FIG. 12B
is a schematic drawing showing the principle of the Targeted VP-PET
(TVP-PEP) imaging. A patient is moved across the scanner's imaging
FOV (illustrated by the dash lines) in multiple bed positions for
whole-body imaging. The high-resolution detectors in a VP-PET
device are moved by a robotic arm controlled by control unit 106 to
follow the target region as the patient is moved across the
scanner. In some embodiments, the high-resolution detectors may be
moved manually. In some other embodiments, the high-resolution
detectors may be embedded in the patient bed and moved across the
scanner FOV along with the patient's body during whole-body PET/CT
imaging. The shaded blue areas show the imaging FOV of the VP-PET
device. With the combination of high-resolution detectors with
detectors of the whole-body scanner, the target region is scanned
by the VP-PET detectors from multiple angles, effectively boosting
the counting statistics from the target region by several
folds.
[0098] Taking advantage of the fully integrated robotic arm in this
second-generation VP-PET device, a specific region of a body is
targeted and followed during a whole-body PET/CT scan. This
strategy not only enhances the image resolution but also increases
the counting statistics of the data from a high-risk target region,
such as the pelvis for cervical cancer patients, or the axilla
and/or contralateral breast for breast cancer patients. FIG. 12B is
an illustration of the principle of the TVP-PET. The
high-resolution PET detectors may be placed close to a patient's
body to "zoom in" on a user-selected target region. The TVP-PET
detectors follow the target region as the patient is moved across
the scanner's imaging FOV during a whole-body imaging protocol. As
the target region approaches and leaves the native scanner's
imaging FOV, the VP-PET detectors are placed just outside the
scanner's detector rings in the axial direction (e.g., at bed
positions #1 and #3 in FIG. 12B). When the target region is in the
scanner's imaging FOV, the VP-PET detectors are placed inside the
scanner as a typical VP-PET insert device (e.g., at bed position
#2). Using this targeted imaging approach, the VP-PET detectors
will collect extra coincidence events originated from the target
region when the patient's body is at bed positions #1 and #3. All
coincidence events detected by the VP-PET detectors will be
incorporated with the native scanner's data for joint image
reconstruction. The results will be a set of whole-body PET images
with significantly improved image resolution, counting statistics,
and noise characteristics in the target region. These benefits can
be gained without incurring any additional scan time or slowing
down the clinical workflow when the TVP-PET technology is adopted
by PET/CT manufacturers.
[0099] The TVP-PET technology complements PET, PET/CT, or PET/MRI
scanners. It works hand-in-hand with any PET/CT scanner (including
TOF-PET and digital PET that employ advanced reconstruction
algorithms) to further improve their native image resolution and
sensitivity. It should be noted that the TVP-PET technology may be
implemented by a manufacturer without a complicated robotic arm
design. A robotic arm is used to position high-resolution detectors
because it offers the maximum flexibility and allows us to use the
prototype VP-PET device to explore different applications. The
high-resolution detectors may be embedded in the patient bed and
made moveable along the axial direction relative to a patient's
body in order to implement the proposed target-tracking feature.
This will avoid the complexity and cost of a robotic arm while
maintaining the TVP-PET functionality.
[0100] The effect of improved detector spatial resolution versus
timing resolution on lesion detectability under a whole-body
imaging condition was analyzed. The results demonstrate that even
under limited counting statistics when using the whole-body imaging
protocol (i.e., 2-3 minutes per bed position) one can still improve
lesion detectability by enhancing image resolution using detectors
of higher spatial resolution or by enhancing the signal-to-noise
ratio (S/N) using detectors of better timing resolution. An
increased lesion detectability is achieved when one improves both
spatial resolution of a scanner and the S/N ratio of the data.
Therefore, technologies such as the TVP-PET that can overcome the
fundamental resolution limit of whole-body PET (approximately 2 mm
FWHM) and also enhance the S/N of PET images may be valuable for
clinical PET applications.
[0101] The TVP-PET imaging technology was validated using a
prototype VP-PET insert device and a research PET/CT scanner
(Biograph 40.TM.). The device has an integrated robotic arm (FIG.
12A) that can position the high-resolution detector panel at
arbitrary locations around a patient's body.
[0102] A flat-panel VP-PET device with compact PET detectors of
sub-millimeter intrinsic resolution. FIGS. 13A-13C show an example
high-resolution PET detector. FIG. 13A is an illustration of a
SiPM-based high-resolution compact PET detector module 1502. Top
(top left of FIG. 13A) and perspective (Bottom left of FIG. 13A)
views of detector module 1502 are provided in FIG. 13A. FIG. 13B
shows a representative flood image of a detector module read out by
PET/CT scanner electronics. FIG. 13C shows the flat panel detector
1500 with its carbon fiber cover removed. A cell phone having a
5.5'' (14.0 cm) screen is placed next to the detector 1500 as a
size reference. The new VP-PET device employs a flat-panel geometry
to allow more flexibility in positioning the detectors around a
patient's body.
[0103] In the example embodiment, the flat-panel detector 1500
includes 32 PET detector modules 1502 arranged in a 4.times.8 array
(FIG. 13C). Detector module 1502 includes a 16.times.16 lutetium
yttrium orthosilicate (LYSO) array 1504, an acrylic light guide
1506, a custom silicon photomultiplier (SiPM) array 1508, and a
printed circuit board 1510 for signal multiplexing with a
charge-division resistor network and connectors. The LYSO array
1504 includes 16.times.16 elements of 0.92.times.0.92.times.3 mm
crystal, arranged in 1 mm pitches to provide sub-millimeter
intrinsic spatial resolution. FIG. 13B shows the flood image of the
detector when it is exposed to a uniform flux of 511 keV
.gamma.-rays. All crystal elements are clearly resolved. For
example, the energy resolution of the detector module is
10.2.+-.0.7% FWHM for 511 keV .gamma.-rays. The timing resolution
of the detector is 0.95.+-.0.08 ns FWHM when measured against a
reference fast detector. These characteristics match well with that
of a Siemens Biograph 40.TM. PET/CT scanner. Moreover, these
SiPM-based PET detectors are compact (less than 15 mm total
thickness) and can be packaged to form large sensing areas with
minimal gaps between modules (<1 mrn in this device). The
enclosure of the flat panel is made of carbon fiber plates to
minimize the weight and attenuation to .gamma.-ray signal. The
overall dimension of the flat-panel enclosure is
150.times.81.times.23 mm (L.times.W.times.H). The sensitive area of
the flat-panel device 1500 is 135.times.67 nm (L.times.W). The
image resolution of the PET scanner equipped with the new VP-PET
device is estimated to be equal to or better than 2 mm FWHM within
17.5 cm from the front surface of the flat panel detector using Eq.
1 shown in Example 1 above. In Eq. 1, d.sub.1 is the distance from
target to the panel detector; d.sub.2 is the distance from the
target to the scanner detectors; w.sub.1=1 mm and w.sub.2=4 mm are
the crystal width of the VP-PET and scanner crystals, respectively.
At a distance farther than 17.5 cm, the resolution is worse than 2
mm FWHM, but still better than the native resolution of the PET
scanner.
[0104] Expanded system electronics to support the augmented
functionality. New add-on electronics and firmware were developed
to support the VP-PET technology using a previous generation PET/CT
scanner (Siemens Biograph 40). A set of switch boards were added to
the scanner to support additional electronics to decode signals
from VP-PET detectors, where the system may include up to 96 PET
detector modules. New firmware can be loaded to the scanner
on-demand to enable coincidence detection between the VP-PET device
and the scanner. These expansions allow full preservation of the
sensitivity and functionality of the native PET/CT scanner, with an
option to turn ON or OFF the additional VP-PET capability using a
simple batch command. With additional detectors in the VP-PET
device, the augmented PET/CT scanner will detect more annihilation
.gamma.-rays than the original PET/CT scanner. The VP-PET
technology can be easily translated to scanners that have reserved
additional electronics channels, such as Siemens Biograph
Vision.TM. PET/CT or scanners that have expandable system
architecture.
[0105] Body phantom study by the second-generation VP-PET device. A
body phantom with small lesions of various sizes was imaged to
evaluate the enhancement of the contrast recovery coefficient (CRC)
by the second-generation VP-PET device. The phantom is a Data
Spectrum Elliptical Lung-Spine Torso Phantom that measures
316.times.228.times.162 mm.sup.3, as shown in FIG. 14A. A panel of
fillable spherical glass tumors (FIG. 14B) was mounted inside the
torso phantom. The diameters of the 6 clusters of tumors are 3.3,
4.3, 6.0, 8.0, 9.6, and 11.4 mm, respectively. Only three spheres
in each size group were filled with radioactivity and blue dyes,
and other spheres in the size group contain water only.
[0106] The torso phantom and the spherical tumors were filled with
.sup.18F solutions that include an initial activity concentration
of 3.63 kBq/mL and 21.78 kBq/mL, respectively, to obtain a
tumor-to-background ratio of 6. The flat-panel detector were placed
at 4 different locations (as illustrated in FIG. 15A) and scanned
the torso phantom 300 s at each location (adjusted for the decay of
.sup.18F). The VP-PET detectors were either unattached or
positioned around the torso phantom at 4 different locations (FIG.
15A). FIG. 15B shows corresponding reconstructed images using an
acquisition time equivalent to 200 s during a typical clinical FDG
scan. This is equivalent to a 200 s body scan at each panel
location when 370 MBq FDG is uniformly distributed in a 70 kg
patient. Subsequently, the VP-PET insert device was removed and the
phantom was imaged again by the native PET scanner for a duration
of 200 s-equivalent after decay correction. A CT scan of the
phantom was also acquired for attenuation correction.
[0107] All images were reconstructed using the above graphic
processing unit (GPU)-based reconstruction platform. FIG. 15B shows
the reconstructed images from the 5 configurations after 30
iterations. Visible improvement is observed in tumor contrast when
the high-resolution VP-PET insert was employed, especially for the
smaller tumors. A spherical volume-of-interest (VOI) of the same
diameter as each tumor was used to estimate the mean count density
C.sub.T,i for each tumor. The center location of each tumor was
determined from the co-registered CT images. The background count
density C.sub.B,i was determined as the mean number of counts in a
large square VOI over 20 adjacent slices. The CRC.sub.i for the
i-th tumor was calculated according to the NEMA NU2-2001 definition
as shown below:
CRC i = 100 % .times. ( c T , i c B , i - 1 ) ( uptake - 1 ) , ( 2
) ##EQU00002##
where uptake is 6 in this study. FIG. 16A shows the mean and
standard deviation of CRC as a function of tumor sizes of the
phantom shown in FIGS. 14A-14B. FIG. 16B shows the ratio of CRC
with and without the VP-PET technology. The CRC is averaged over of
the tumors of the same diameter and presented as a function of
tumor sizes in FIG. 16A.
[0108] From the native scanner images, the average CRCs are 0.82%,
2.90%, 5.25%, and 13.63% for tumors with diameters of 3.3, 4.3,
6.0, and 8.0 mm, respectively. The 3.3 mm and 4.3 mm diameter
tumors have the best average CRC of 2.73% and 6.21%, respectively,
when the VP-PET detectors are at position 1. The 6.0 mm and 8.0 mm
diameter tumors have the best average CRC of 11.92% and 21.81%,
respectively, when the VP-PET detectors are at position 3. For
tumors with larger diameters (e.g. 9.6 mm and 11.4 mm), the
enhancement in average CRC is less significant regardless where the
VP-PET detectors are placed. This is because the image resolution
of the native scanner is adequate in recovering the contrast of the
larger tumors.
[0109] The results demonstrate the improvement in tumor contrast
improvement by the VP-PET technology, particularly for smaller
lesions. Comparison of results from different panel-locations show
that the resolution and CRC enhancement depend on both the source
location and the VP-PET panel location. When the panel is placed
far away from the target region (e.g., position 4 in FIG. 15A), the
level of enhancement is reduced. Placing the panel detectors below
a patient's body (position #1 in FIG. 15A) is a logical choice
considering that cancers may metastasize to either side of a
patient's body. Based on FIG. 16B, the panel location 1 also offers
significant enhancement to the CRC for lesions at most locations.
As a result, the VP-PET detectors is placed below a patient's body
(position #1 in FIG. 15A) to evaluate the TVP-PET technology.
1. Functionality and Reproducibility of the TVP-PET Technology
[0110] To implement the proposed TVP-PET technology, the same torso
phantom with spherical lesions is scanned using the
second-generation VP-PET device based on the targeted imaging
protocol as illustrated in FIG. 12B. The torso phantom will be
moved across the scanner's native imaging FOV and imaged at
multiple bed positions (minimum of 3 beds). The robotic arm is used
to place the flat-panel detector under the patient bed and move the
detector along the axial direction to follow the section of the
phantom that includes the clusters of spherical lesions ("target").
The scan time per bed position is kept the same as that of the
standard-of-care whole-body imaging protocol. With the custom
firmware and software, the high-resolution coincidence events are
simultaneously acquired using the VP-PET detectors as the scanner
acquires regular whole-body PET imaging data. This tracking and
imaging strategy is expected to improve the counting statistics
from the target region by approximately a factor of three if the
protocol in FIG. 12B is used when compared to the original VP-PET
technology. Coincidence events from the native scanner and the
high-resolution events from all three bed positions are combined
for joint image reconstruction. This should further improve the
image resolution and CRC of tumors when compared to the images
shown in FIG. 15B.
[0111] In a preliminary test, the torso phantom was filled with
.sup.18F solution using a tumor-to-background ratio of 6 for the
spherical lesions. The phantom was first imaged using the native
PET/CT scanner. Then the phantom was imaged with the VP-PET
detectors centered in the scanner's imaging FOV (i.e., the original
VP-PET technology). Finally, the phantom was imaged using the
TVP-PET protocol in FIG. 12B. Results are shown in FIGS. 17A-17C.
FIGS. 17A-17C are images of the torso phantom with spherical tumors
imaged by a PET/CT scanner, the same scanner equipped with VP-PET,
and the native scanner with the flat-panel detectors placed under
the patient bed (i.e., the position as shown in FIG. 12A) using a
TVP-PET protocol as shown in FIG. 12B, respectively.
[0112] The average CRC were computed for each group of tumors of
the same size. FIG. 18A shows the average CRC as a function of
tumor size for each of the 3 scanner configurations. FIG. 18B shows
the ratio of the CRC from the VP-PET images (or TVP-PET images)
over CRC from the standard PET/CT images for each size of tumors.
As shown, smaller lesions suffer more from partial volume effect
and have poorer CRC. The VP-PET technology improves the CRC when
compared to the native scanner. The TVP-PET technology further
improves the CRC when compared to the VP-PET technology.
[0113] When introducing elements of aspects of the invention or the
embodiments thereof, the articles "a," "an," "the," and "said" are
intended to mean that there are one or more of the elements. The
terms "comprising," "including," and "having" are intended to be
inclusive and mean that there may be additional elements other than
the listed elements.
[0114] Although described in connection with an example computing
system environment, embodiments of the invention are operational
with numerous other general purpose or special purpose computing
system environments or configurations. The computing system
environment is not intended to suggest any limitation as to the
scope of use or functionality of any aspect of the invention.
[0115] Embodiments of the invention may be described in the general
context of computer-executable instructions, such as program
modules, executed by one or more computers or other devices. The
computer-executable instructions may be organized into one or more
computer-executable components or modules. Generally, program
modules include, but are not limited to, routines, programs,
objects, components, and data structures that perform particular
tasks or implement particular abstract data types. Aspects of the
invention may be implemented with any number and organization of
such components or modules. For example, aspects of the invention
are not limited to the specific computer-executable instructions or
the specific components or modules illustrated in the figures and
described herein. Other embodiments of the invention may include
different computer-executable instructions or components having
more or less functionality than illustrated and described herein.
Aspects of the invention may also be practiced in distributed
computing environments where tasks are performed by remote
processing devices that are linked through a communications
network. In a distributed computing environment, program modules
may be located in both local and remote computer storage media
including memory storage devices.
[0116] While example embodiments of components, assemblies and
systems are described, variations of the components, assemblies and
systems are possible to achieve similar advantages and effects.
Specifically, the shape and the geometry of the components and
assemblies, and the relative locations of the components in the
assembly, may be varied from that described and depicted without
departing from inventive concepts described. In addition, in
certain embodiments certain components in the assemblies described
may be omitted to accommodate particular applications and
installations, while still providing improved systems.
[0117] This written description uses examples to disclose the
invention, including the best mode, and also to enable any person
skilled in the art to practice the invention, including making and
using any devices or systems and performing any incorporated
methods. The patentable scope of the invention 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 they have structural elements that do not differ
from the literal language of the claims, or if they include
equivalent structural elements with insubstantial differences from
the literal languages of the claims.
* * * * *