Note: Descriptions are shown in the official language in which they were submitted.
CA 02331108 2001-O1-16
METHOD AND APPARATUS FOR IMPROVING IMAGE QUALITY
IN POSITRON EMISSION TOMOGRAPHY
FIELD OF INVENTION
The present invention relates to scintillation cameras. In particular, the
invention
relates to a method and apparatus for improving the quality of images produced
during
positron emission tomography.
BACKGROUND OF THE INVENTION
In the human body, increased metabolic activity is associated with an increase
in
emitted radiation. In the field of nuclear medicine, increased metabolic
activity within a
patient is detected using a radiation detector such as a scintillation camera.
Scintillation cameras are well known in the art, and are used for medical
diagnostics.
A patient ingests, inhales or is injected with a small quantity of a
radioactive isotope. The
radioactive isotope emits gamma rays that are detected by a scintillation
medium in the
scintillation camera. The scintillation medium is commonly a sodium iodide
crystal, BGO
or other. The scintillation medium emits a small flash or scintillation of
light, in response to
stimulating radiation, such as from a patient. The intensity of the
scintillation of light is
proportional to the energy of the stimulating photon, such as a gamma photon.
Note that the
relationship between the intensity of the scintillation of light and the gamma
ray is not linear.
A conventional scintillation camera such as a gamma camera includes a detector
which
converts into electrical signals gamma rays emitted from a patient after
radioisotope has been
administered to the patient. The detector includes a scintillator and
photomultiplier tubes.
The gamma rays are directed to the scintillator which absorbs the radiation
and produces, in
1
CA 02331108 2001-O1-16
response, a very small flash of light. An array of photodetectors, which are
placed in optical
communication with the scintillation crystal, converts these flashes into
electrical signals
which are subsequently processed. The processing enables the camera to produce
an image
of the distribution of the radioisol;ope within the patient.
Scintillation cameras are used to take four basic types of pictures: spot
views, whole
body views, partial whole body viiews, SPECT views, and whole body SPECT
views.
A spot view is an image of a part of a patient. The area of the spot view is
less than
or equal to the size of the field of view of the gamma camera. In order to be
able to achieve
a full range of spot views, a gamma camera must be positionable at any
location relative to
a patient.
One type of whole body view is a series of spot views fitted together such
that the
whole body of the patient may be viewed at one time. Another type of whole
body view is
a continuous scan of the whole body of the patient. A partial whole body view
is simply a
whole body view that covers only part of the body of the patient. In order to
be able to
achieve a whole body view, a gamma camera must be positionable at any location
relative to
a patient in an automated sequence of views.
The acronym "SPELT" stands for single photon emission computerized
tornography.
A SPELT view is a series of slice-like images of the patient. The slice-like
images are often,
but not necessarily, transversely oriented with respect to the patient. Each
slice-like image
is made up of multiple views taken at different angles around the patient, the
data from the
various views being combined to :form the slice-like image. In order to be
able to achieve a
SPELT view, a scintillation camera must be rotatable around a patient, with
the direction of
the detector head of the scintillation camera pointing in a series of known
and precise
directions such that reprojection of the data can be accurately undertaken.
A whole body SPELT view is a series of parallel slice-like transverse images
of a
patient. Typically, a whole body SPELT view consists of sixty four spaced
apart SPELT
2
CA 02331108 2001-O1-16
views. A whole body SPECT view results from the simultaneous generation of
whole body
and SPECT image data. In order to be able to achieve a whole body SPECT view,
a
scintillation camera must be rotatable around a patient, with the direction of
the detector head
of the scintillation camera pointing in a series of known and precise
directions such that
reprojection of the data can be accurately undertaken.
Therefore, in order that the radiation detector be capable of achieving the
above four
basic views, the support structure for the radiation detector must be capable
of positioning the
radiation detector in any position relative to the patient. Furthermore, the
support structure
must be capable of moving the radiation detector relative to the patient in a
controlled manner
along any path.
In order to operate a scintillation camera as described above, the patient
should be
supported horizontally on a patient support or stretcher.
A certain design of gantry or support structure for a scintillation camera
includes a
frame upon which a vertically oriented annular support rotates. Extending out
from the
rotating support is an elongate support. The elongate generally comprises a
pair of arms. The
pair of arms generally extends through a corresponding pair of apertures in
the rotating
support. One end of the pair of arms supports the detector head on one side of
the annular
support. The other end of the pair of arms supports a counter balance weight.
Thus, the
elongate support is counterbalanced with a counterweight on the opposite side
of the detector
head.
With such a design of support structure for a scintillation camera, a patient
must lie
on a horizontally oriented patient ;support. The patient support must be
cantilevered so that
the detector head can pass underneath the patient. If the detector head must
pass underneath
only one end of the patient, such as the patient's head, the cantilevered
portion of the patient
support is not long enough to cause serious difficulties in the design ofthe
cantilevered patient
support. However, if the camera must be able to pass under the entire length
of the patient,
the entire patient must be supported by the cantilevered portion of the
patient support. As the
3
CA 02331108 2001-O1-16
cantilevered portion of the patif,nt support must be thin so as not to
interfere with the
generation of images by the scintillation camera, serious design difficulties
are encountered.
Among the advantages associated with such as design of support structure is
that a
patient may be partially pass through the orifice defined by the annular
support so that the pair
of arms need not be as long. However, the patient support must be able to
support the patient
in this position relative to the annular support, must be accurately
positionable relative to the
annular support, and must not interfere either with the rotation ofthe annular
support or with
the cables which will inevitably e:~tend from the detector head to a nearby
computer or other
user control.
The photomultiplier tubes in a scintillation camera generate electric signals.
The
signals are processed, and images are created corresponding to the radiation
emitted by the
patient.
From time to time, images. are generated that contain one or more artifacts or
flaws.
Artifacts are often caused by one or more malfunctioning photomultiplier
tubes. A
malfunctioning photomultiplier tube may be generating incorrect signals, may
be generating
no signal at all, or the processing o f the signals from a particular
photomultiplier tube may not
be proper.
To determine the cause of the artifact and then correct the artifact, it is
important to
identify all malfunctioning photomultiplier tubes. However, inspecting and
testing
photomultiplier tubes is difficult, time consuming and expensive.
From time to time, images of poor quality are also generated. Of particular
concern
are the images produced by Position Emission Tomography. Position Emission
Tomography
(PET) is a practice common in the art wherein two detectors are placed with
their fields of
view at 180 ° to one another. After' the patient ingests the isotope,
positrons are emitted from
areas where is isotope has gathered) in the body. The positrons that are
released from the body
in opposite directions collide with electrons in the body and effectively form
two gamma rays.
4
CA 02331108 2001-O1-16
The gamma rays are detected by 'the detectors and as mentioned above are used
to generate
images. However, in PET, only gamma rays originating from a collision between
a positron
and an electron that are detected at 180 ° (referred to as coincidence)
from one another are
considered true events. Preferably only true events are used to generate
images.
Unfortunately what sometime occurs is that the gamma ray will ricochet off a
second
electron in the body before being emitted and the angle is changed. The two
gamma rays will
not be detected at 180 ° from one another, resulting in a "random"
event. Random events are
really just noise signals that when used to generate an image, cause poor
quality imagery. It
is known in the art that an increase in area (of field of view) results in an
increase in the
probability of random events. Since conventional PET cameras use relatively
large detectors
with large fields of view and they commonly use the total data values for the
entire detector
head, the chance of using random events to generate an image is high. As well,
since data
from a large field of view must be processed, the time frame window during
which data is
analysed is large resulting in yet a. higher probability of detecting random
events.
In Constant Fraction Discrimination (CFDs) cameras, the probability of random
events
is also relatively high, resulting in poorer quality images. Figure 1
illustrates the data obtained
from a Constant Fraction Discriminator. Constant Fraction Discriminators use a
constant
fraction (or percentage) of the input pulse to precisely determine the timing
of an event.
Inaccuracies occur when two events are detected in such a short time frame
such as to create
overlap. In the data when two or more events overlay, it is impossible to
separate them to
obtain before an event in order to separate the data. As seen in Figure l, the
data from areas
A, B and C can be separated in order to analyse the individual events l and 2.
SUMMARY OF THE INVENTION
An object of the invention is to provide a method and apparatus for improving
a PET
image quality. This is achieved by analysing individual photomultiplier tubes
for true events
and by providing time stamps to photomultiplier tube signals. Analysing data
from individual
photomultiplier tubes as opposed to entire detector field of views results in
smaller areas and
smaller amounts of data to be processed. This then permits smaller time frame
windows to
5
CA 02331108 2001-O1-16
be used. The use of time stamps also permits data before and after a
particular event to be
kept as record.
The invention relates to an apparatus for improving the quality of images
produced by
a scintillation camera during positron emission tomography wherein both true
events and
random events occur, comprising: a photomultiplier tube for generating a
photomultiplier tube
signal; means for generating a code signal identifying the photomultiplier
tube; a clock for
generating a clock signal providing; a time stamp for the photomultiplier
tube; a bus buffer for
transmitting an encoded signal comprising the photomultiplier tube signal
followed by the
code signal and the time stamp; a data analyser for determining whether the
encoded signal
represents a true event; a position computing device for calculating the
position of a true
event; an image computer for generating an image of the events from a
plurality of encoded
signals and the positions of their corresponding events; and a display for
displaying the image.
The invention also relates to a method for improving the image produced by a
scintillation camera comprising an array of photomultiplier tubes, comprising
the steps of:
generating a photomultiplier tube signal after an event; generating a code
signal identifying
the photomultiplier tube; generating a clock signal providing a time stamp for
the
photomultiplier tube; generating an encoded signal comprising the
photomultiplier tube
signal followed by the code signal .and the time stamp; determining whether
the event is a true
event; calculating the position of the event; generating an image from a
plurality of encoded
signals; and displaying the image.
One embodiment relates to an apparatus for improving the image produced a
scintillation camera comprising an array of photomultiplier tubes, comprising:
means for
generating a photomultiplier tube signal after an event; means for generating
a code signal
identifying the photomultiplier tube; means for generating a clock signal
providing a time
stamp for the photomultiplier tube;; means for generating an encoded signal
comprising the
photomultiplier tube signal followed by the code signal and the time stamp;
means for
determining whether the event is a true event; means for calculating the
position of the event;
6
CA 02331108 2001-O1-16
means for generating an image from a plurality of encoded signals; and means
for displaying
the image.
Other advantages, obj ects and features of the present invention will be
readily apparent
to those skilled in the art from a review of the following detailed
description of preferred
embodiments in conjunction with the accompanying drawings and claims.
BRIEF DESCRIPTION OF THE DRAWINGS
The embodiments of the invention will now be described with reference to the
accompanying drawings, in which:
Figure 1 illustrates the data obtained with a CFD;
Figure 2 illustrates the basics of PET;
Figure 3 is a drawing of an embodiment of the photomultiplier tube identifier
of the
present invention;
Figure 4 is a drawing of the bus buffer of the embodiment of Figure 3; and
Figure 5 is a flowchart illustrating the operation of the data analyser.
Similar references are used in different figures to denote similar components.
DETAILED DESCRIPTION OF 'THE INVENTION
Figure 2 illustrates the basics of PET. Briefly, when a collision occurs in
the body,
two gamma rays are emitted and detected by the detector (known as events). If
it is
determined that the events are true events (as detailed below), they are used
in image
generation. However, if one gamma ray, for example gamma ray 2, ricochets to
create event
7
CA 02331108 2001-O1-16
3 rather than true event 2, it causes a random or scattered event and is
preferably not used in
image generation.
Figures 3 and 4 illustrate an array of photomultiplier tubes 405 in a
scintillation
camera. A photomultiplier tube identifier 410 is an apparatus for identifying
a photomultiplier
tube in the array of photomultiplic~r tubes 405.
The photomultiplier tube identifier 410 includes amplifier/integrators 415,
analog to
digital converters (ADCs) 420, bus buffers 425, pull-up resistors 430, a bus
435, a position
computing device 440, an image computer 445, a user display 450 and a clock
42fi.
Output signals from individual photomultiplier tubes in the array of
photomultiplier
tubes 405 are amplified and integrated by the amplifier/integrators 415. The
output signals
from the amplifier/integrators 415 are then digitized in the analog to digital
converters 420.
The output signal from a digital to analog converter 420 corresponds to the
strength of the
signal from an individual photomultiplier tube in the array of photomultiplier
tubes 405.
The bus buffers 425 receive output signals from the digital to analog
converters 420.
Some of the gates of the bus buffers 425 are also connected to the pull up
resistors 430. The
gates of the bus buffer are set by the pull up resistors 430 to a logic high
or logic low which
correspond to the identities of the individual photomultiplier tubes from
which signals have
been obtained. To each output signal from the digital to analog converters
420, the bus
buffers 425 add a code below the least significant bits identifying the
photomultiplier tube
from which the signal was obtained. Thus, the output signals from the bus
buffers 425
corresponds to the strength of the signals received from the array
ofphotomultiplier tubes 405
plus a code identifying the photomultiplier tube from which the signals were
obtained.
In addition, the clock 426 provides clock signals providing a continuously
running
clock or stream of time stamps to~ each photomultiplier tube identifier. The
clock signals
provide the time stamp for each photomultiplier tube output signal at a
predetermined clock
increment. The stream of time stamps maintain records of when events have
taken place.
8
CA 02331108 2001-O1-16
In a preferred embodiment, the clock increments in cycles from 0 to 256. That
is, each
cycle produces 256 time stamps, but any suitable number could be used
depending upon the
accuracy required.
In a preferred embodiment, time stamps are generated every two nanoseconds,
but
another suitable length of time ca~~ be chosen.
Figure 4 illustrates a bit bus buffer 425. Output signals 455 from a digital
to analog
converter 420, in this case twelve most significant bits of signal data, are
received by the bus
buffer 425. The twelve bit output signals 455 correspond to the specific
photomultiplier tube
providing the output signal. Logic values 460 from pull up resistors 430, in
this case 6 bits
of data, provide a hard wired code corresponding to the identity of the
specific photomultiplier
tube. In this case, as the pull up resistors provide six bits of data, the
signals from sixty four
different photomultiplier tubes 40.'i may be encoded. As well, approximately
ten bits of clock
signals 461, are also written into l;he bus buffer and encoded. While ten bits
of time stamp
data is preferable, any number of lbits could be used.
Upon receipt of the enable command at 475, the data (the data signal values,
the
photomultiplier tube identifier andl time stamps) from the bus buffer is read
onto the bus 435.
The signal values 465, that is, the: first twelve bits of data correspond to
the output signal
received from the digital to analog, converter 415. The code values 470, that
is, the next four
bits of data, provide the code identifying the specific photomultiplier tube
405 providing the
information. The time stamp values 428 provides the time data from the clock
signals 461.
The signals 460 in Figure 4 provide a code of 010011, ground being represented
by 0 and
VCC being represented by 1. If more codes are required, a larger bus buffer
can be used, such
as a twenty or thirty two bit bus buffer.
The first twelve bits of each encoded signal 480 are the signals values 465,
and six
bits of each encoded signal 480 are, the code values 470 while the remaining
bits are the time
stamp values 428. The encoded silmals 480 are received by a processing unit.
Since the code
values 470 are in the low part of t:he encoded signal 480 or data word used by
the position
9
CA 02331108 2001-O1-16
computing device 440, the change in value created by adding the code values
470 to the signal
values 470 is negligible. Therefore, the code values 470 do not need to be
removed before
the encoded signal 480 is used by the position computing device 440. For
example, the
encoded signal may represent the value 1,001,325.238. The final two digits,
that is, eight and
three, may be the code identifying the thirty eighth photomultiplier tube in
the array. The
0.038 value and the time stamp data could be removed from the encoded signal
480 prior to
processing by the position computing device 440 and reattached to the signal
480 afterwards.
However, such a calculation would not be beneficial as the 0.038 a negligible
value compared
with the value 1,001,325.238. If an artifact appears on the generated image,
and the artifact
can be traced to the data value 1,001,325.238, then photomultiplier tube
number thirty eight
can be repaired or replaced. Similarly, if an artifact appears on the
generated image, and
fewer data values traceable to photomultiplier tube number thirty eight than
are statistically
expected, then photomultiplier tube number thirty eight may need repairing or
replacing.
Encoded signals 480, including the time stamp, are read onto the bus buffer
425. This
data for each multiplier tube is then fed across the bus 435 and may be stored
in a temporary
memory 428. The data coming from a particular photornultiplier tube can be
analysed by a
data analyser 441. If there is an event, the data before that event, and after
the event is
recorded. In the case of CFOs, this allows overlapping event signals to be
separated into
individual true event signals. In other words, if data from two events have
overlapped, the
data values for one event can be subtracted or removed from the data values
for the second
event. This is known in the art as deconvalving the events.
Similarly, the signals for all the photomultiplier tube outputs can be
analysed for
photomultiplier tubes that are at 180 degrees to one another. From this data,
it can be
determined whether an event is within a certain time window, and whether those
photomultiplier tubes are in coincidence. This is accomplished by analysing
the data for two
photomultiplier tubes at 180 degrees within a very small time window, for
example, two
nanoseconds. The true events data is then transferred to a main memory 442 and
then to
processing and image generation. 'the other data (random data) is effectively
useless and may
be purged. In this way, the position computing device 440 can transmit
information to the
CA 02331108 2001-O1-16
image computer 445 and then the display 450 quickly and inexpensivelywhile
retaining intact
information identifying the specif c photomultiplier tubes corresponding the
specific data.
Refernng to Figure 5, therefore, first individual tube values are analysed to
determine whether
an events are in coincidence and then to determine the location of the event.
Prior art systems typically operate in the following manner: when events
occur, the
location of the events are determined, and then whether the events are in
coincidence is
determined using the total data values from the entire detector heads.
As mentioned above, quality of PET imagery is affected by two factors: the
probability
of random events and the size of the time window.
Since the probability of random events increases as the field of view area
increases,
it is desirable to have less area, to improve the PET images. Therefore,
individual
1 S photomultiplier tubes are placed i:n coincidence which reduces the area,
and the probability
of random events is minimized. 'the data from individual photomultiplier tubes
is used to
determine coincidence as opposed to the data from the entire detector head.
Note that it may
be possible to have photomultiplier tubes that are skewed because it is where
the events occur
in the crystal that determine whether they are in coincidence.
Another way to improve PET images is to have smaller time windows during which
data is analysed such that the time to pick up random events is reduced.
Encoding a time
stamp to each photomultiplier tuba at predetermined times produces a stream of
time stamps
for each tube. Then each stream can be analysed to determined which tubes are
in
coincidence. Tubes in coincidence will have the same time stamp, or match a
time stamp
within a predetermined time window. By analysing individual photomultiplier
tube data,
smaller amounts of data are processed allowing a smaller time window to be
used.
Numerous modifications, 'variations and adaptations may be made to the
particular
embodiments of the invention described above without departing from the scope
of the
invention, which is defined in the claims.
11
