IEC 61675-1:1998

INTERNATIONAL
IEC
STANDARD
61675-1
First edition
1998-02
Radionuclide imaging devices –
Characteristics and test conditions –
Part 1:
Positron emission tomographs
Dispositifs d’imagerie par radionucléides –
Caractéristiques et conditions d’essai –
Partie 1:
Tomographes à émission de positrons
Reference number
Numbering
As from 1 January 1997 all IEC publications are issued with a designation in the 60000 series.
Consolidated publications
Consolidated versions of some IEC publications including amendments are available. For example, edition
numbers 1.0, 1.1 and 1.2 refer, respectively, to the base publication, the base publication incorporating
amendment 1 and the base publication incorporating amendments 1 and 2.
Validity of this publication
The technical content of IEC publications is kept under constant review by the IEC, thus ensuring that the content
reflects current technology.
Information relating to the date of the reconfirmation of the publication is available in the IEC catalogue.
Information on the revision work, the issue of revised editions and amendments may be obtained from
IEC National Committees and from the following IEC sources:
• IEC Bulletin
• IEC Yearbook
On-line access*
•
Catalogue of IEC publications
Published yearly with regular updates
(On-line access)*
Terminology, graphical and letter
symbols
For general terminology, readers are referred to IEC 60050: International Electrotechnical Vocabulary (IEV).
For graphical symbols, and letter symbols and signs approved by the IEC for general use, readers are referred to
publications IEC 60027: Letter symbols to be used in electrical technology, IEC 60417: Graphical symbols for use
on equipment. Index, survey and compilation of the single sheets and IEC 60617: Graphical symbols for diagrams.
IEC publications prepared by the same
technical committee
The attention of readers is drawn to the end pages of this publication which list the IEC publications issued by the
technical committee which has prepared the present publication.
* See web site address on title page.

INTERNATIONAL
IEC
STANDARD
61675-1
First edition
1998-02
Radionuclide imaging devices –
Characteristics and test conditions –
Part 1:
Positron emission tomographs
Dispositifs d’imagerie par radionucléides –
Caractéristiques et conditions d’essai –
Partie 1:
Tomographes à émission de positrons
 IEC 1998 Droits de reproduction réservés  Copyright - all rights reserved
Aucune partie de cette publication ne peut être reproduite ni No part of this publication may be reproduced or utilized in
utilisée sous quelque forme que ce soit et par aucun any form or by any means, electronic or mechanical,
procédé, électronique ou mécanique, y compris la photo- including photocopying and microfilm, without permission in
copie et les microfilms, sans l'accord écrit de l'éditeur. writing from the publisher.
International Electrotechnical Commission 3, rue de Varembé Geneva, Switzerland
Telefax: +41 22 919 0300 e-mail: inmail@iec.ch IEC web site http: //www.iec.ch
CODE PRIX
Commission Electrotechnique Internationale
PRICE CODE W
International Electrotechnical Commission
Pour prix, voir catalogue en vigueur
For price, see current catalogue

– 2 – 61675-1 © IEC:1998(E)
CONTENTS
Page
FOREWORD . 3
Clause
1 General . 4
1.1 Scope and object . 4
1.2 Normative reference . 4
2 Terminology and definitions . 4
3 Test methods. 10
3.1 SPATIAL RESOLUTION . 10
3.2 RECOVERY COEFFICIENT . 13
3.3 Tomographic sensitivity. 14
3.4 Uniformity . 16
3.5 COUNT RATE CHARACTERISTIC . 16
3.6 Scatter measurement. 19
3.7 ATTENUATION correction. 21
4ACCOMPANYING DOCUMENTS . 23
Table 1 – RADIONUCLIDES to be used in performance measurements . 25
Figures
1 Cylindrical head phantom. 26
2 Cross-section of body phantom . 27
3 Arm phantom . 27
4 Phantom insert with hollow spheres . 28
5 Phantom insert with holders for the scatter source . 29
6 Phantom insert for the evaluation of ATTENUATION correction . 30
7 Phantom configuration for COUNT RATE measurements according to 3.5.3.1.2
(cardiac imaging) . 31
8 Scheme of the evaluation of COUNT LOSS correction. 31
9 Evaluation of ATTENUATION correction. 32
10 Evaluation of SCATTER FRACTION. 32
11 Evaluation of FWHM . 33
12 Evaluation of EQUIVALENT WIDTH (EW) . 34
Annex A (informative) Index of defined terms. 35

61675-1 © IEC:1998(E) – 3 –
INTERNATIONAL ELECTROTECHNICAL COMMISSION
___________
RADIONUCLIDE IMAGING DEVICES –
CHARACTERISTICS AND TEST CONDITIONS –
Part 1: Positron emission tomographs
FOREWORD
1) The IEC (International Electrotechnical Commission) is a worldwide organization for standardization comprising
all national electrotechnical committees (IEC National Committees). The object of the IEC is to promote
international co-operation on all questions concerning standardization in the electrical and electronic fields. To
this end and in addition to other activities, the IEC publishes International Standards. Their preparation is
entrusted to technical committees; any IEC National Committee interested in the subject dealt with may
participate in this preparatory work. International, governmental and non-governmental organizations liaising
with the IEC also participate in this preparation. The IEC collaborates closely with the International Organization
for Standardization (ISO) in accordance with conditions determined by agreement between the two
organizations.
2) The formal decisions or agreements of the IEC on technical matters express, as nearly as possible, an
international consensus of opinion on the relevant subjects since each technical committee has representation
from all interested National Committees.
3) The documents produced have the form of recommendations for international use and are published in the form
of standards, technical reports or guides and they are accepted by the National Committees in that sense.
4) In order to promote international unification, IEC National Committees undertake to apply IEC International
Standards transparently to the maximum extent possible in their national and regional standards. Any
divergence between the IEC Standard and the corresponding national or regional standard shall be clearly
indicated in the latter.
5) The IEC provides no marking procedure to indicate its approval and cannot be rendered responsible for any
equipment declared to be in conformity with one of its standards.
6) Attention is drawn to the possibility that some of the elements of this International Standard may be the subject
of patent rights. The IEC shall not be held responsible for identifying any or all such patent rights.
International Standard IEC 61675-1 has been prepared by subcommittee 62C: Equipment for
radiotherapy, nuclear medicine and radiation dosimetry, of IEC technical committee 62:
Electrical equipment in medical practice.
The text of this standard is based on the following documents:
FDIS Report on voting
62C/205/FDIS 62C/214/RVD
Full information on the voting for the approval of this standard can be found in the report on
voting indicated in the above table.
In this standard, the following print types are used:
– TERMS DEFINED IN CLAUSE 2 OF THIS STANDARD OR LISTED IN ANNEX A: SMALL CAPITALS.
The requirements are followed by specifications for the relevant tests.
Annex A is for information only.
A bilingual version of this standard may be issued at a later date.

– 4 – 61675-1 © IEC:1998(E)
RADIONUCLIDE IMAGING DEVICES –
CHARACTERISTICS AND TEST CONDITIONS –
Part 1: Positron emission tomographs
1 General
1.1 Scope and object
This part of IEC 61675 specifies terminology and test methods for declaring the characteristics
POSITRON EMISSION TOMOGRAPHS OSITRON EMISSION TOMOGRAPHS ANNIHILATION
of . P detect the
RADIATION of positron emitting RADIONUCLIDEs by COINCIDENCE DETECTION.
The test methods specified in this part of IEC 61675 have been selected to reflect as much as
possible the clinical use of POSITRON EMISSION TOMOGRAPHS. It is intended that the test
methods be carried out by manufacturers, thereby enabling them to declare the characteristics
POSITRON EMISSION TOMOGRAPHS ACCOMPANYING
of . So, the specifications given in the
DOCUMENTS shall be in accordance with this standard. This standard does not imply which tests
will be performed by the manufacturer on an individual tomograph.
No test has been specified to characterize the uniformity of reconstructed images, because all
methods known so far will mostly reflect the noise in the image.
1.2 Normative reference
The following normative document contains provisions which, through reference in this text,
constitute provisions of this part of IEC 61675. At the time of publication, the edition indicated
was valid. All normative documents are subject to revision, and parties to agreements based
on this part of IEC 61675 are encouraged to investigate the possibility of applying the most
recent edition of the normative document indicated below. Members of IEC and ISO maintain
registers of currently valid International Standards.
IEC 60788:1984, Medical radiology – Terminology
2 Terminology and definitions
For the purpose of this part of IEC 61675, the definitions given in IEC 60788 (see annex A) and
the following definitions apply.
Defined terms are printed in small capitals.
2.1 TOMOGRAPHY (see annex A)
2.1.1
TRANSVERSE TOMOGRAPHY
in TRANSVERSE TOMOGRAPHY the three-dimensional object is sliced by physical methods, for
example collimation, into a stack of OBJECT SLICES, which are considered as being two-
dimensional and independent from each other. The transverse IMAGE PLANES are perpendicular
to the SYSTEM AXIS.
61675-1 © IEC:1998(E) – 5 –
2.1.2
EMISSION COMPUTED TOMOGRAPHY (ECT)
imaging method for the representation of the spatial distribution of incorporated RADIONUCLIDEs
in selected two-dimensional slices through the object
2.1.2.1
PROJECTION
transformation of a three-dimensional object into its two-dimensional image or of a two-
dimensional object into its one-dimensional image, by integrating the physical property which
determines the image along the direction of the PROJECTION BEAM
NOTE – This process is mathematically described by line integrals in the direction of projection (along the LINE OF
RESPONSE) and called Radon-transform.
2.1.2.2
PROJECTION BEAM
determines the smallest possible volume in which the physical property which determines the
image is integrated during the measurement process. Its shape is limited by SPATIAL
RESOLUTION in all three dimensions.
NOTE – The PROJECTION BEAM mostly has the shape of a long thin cylinder or cone. In POSITRON EMISSION
TOMOGRAPHY, it is the sensitive volume between two detector elements operated in coincidence.
2.1.2.3
PROJECTION ANGLE
angle at which the PROJECTION is measured or acquired
2.1.2.4
SINOGRAM
two-dimensional display of all one-dimensional PROJECTIONs of an OBJECT SLICE, as a function
of the PROJECTION ANGLE. The PROJECTION ANGLE is displayed on the ordinate, the linear
PROJECTION coordinate is displayed on the abscissa.
2.1.2.5
OBJECT SLICE
slice in the object. The physical property of this slice, that determines the measured
information, is displayed in the tomographic image.
2.1.2.6
IMAGE PLANE
a plane assigned to a plane in the OBJECT SLICE
NOTE – Usually the IMAGE PLANE is the midplane of the corresponding OBJECT SLICE.
2.1.2.7
SYSTEM AXIS
axis of symmetry, characterized by geometrical and physical properties of the arrangement of
the system
NOTE – For a circular POSITRON EMISSION TOMOGRAPH, the SYSTEM AXIS is the axis through the centre of the detector
ring. For tomographs with rotating detectors it is the axis of rotation.
2.1.2.8
TOMOGRAPHIC VOLUME
juxtaposition of all volume elements which contribute to the measured PROJECTIONs for all
PROJECTION ANGLES
– 6 – 61675-1 © IEC:1998(E)
2.1.2.8.1
TRANSVERSE FIELD OF VIEW
dimensions of a slice through the TOMOGRAPHIC VOLUME, perpendicular to the SYSTEM AXIS. For
a circular TRANSVERSE FIELD OF VIEW, it is described by its diameter
NOTE – For non-cylindrical TOMOGRAPHIC VOLUMES the TRANSVERSE FIELD OF VIEW may depend on the axial position
of the slice.
2.1.2.8.2
AXIAL FIELD OF VIEW
dimensions of a slice through the TOMOGRAPHIC VOLUME, parallel to and including the SYSTEM
AXIS. In practice, it is specified only by its axial dimension, given by the distance between the
IMAGE PLANE AXIAL SLICE WIDTH
centre of the outmost defined s plus the average of the measured
2.1.2.8.3
TOTAL FIELD OF VIEW
dimensions (three-dimensional) of the TOMOGRAPHIC VOLUME
2.1.3
POSITRON EMISSION TOMOGRAPHY (PET)
EMISSION COMPUTED TOMOGRAPHY utilizing the ANNIHILATION RADIATION of positron emitting
RADIONUCLIDES by COINCIDENCE DETECTION
2.1.3.1
POSITRON EMISSION TOMOGRAPH
tomographic device, which detects the ANNIHILATION RADIATION of positron emitting
RADIONUCLIDES by COINCIDENCE DETECTION
2.1.3.2
ANNIHILATION RADIATION
ionizing radiation that is produced when a particle and its antiparticle interact and cease to exist
2.1.3.3
COINCIDENCE DETECTION
a method which checks whether two opposing detectors have detected one photon each
simultaneously. By this method the two photons are concatenated into one event.
NOTE – The COINCIDENCE DETECTION between two opposing detector elements serves as an electronic collimation
to define the corresponding PROJECTION BEAM or LINE OF RESPONSE (LOR), respectively.
2.1.3.4
COINCIDENCE WINDOW
time interval during which two detected photons are considered being simultaneous
2.1.3.5
LINE OF RESPONSE (LOR)
the axis of the PROJECTION BEAM
NOTE – In PET, it is the line connecting the centres of two opposing detector elements operated in coincidence.
2.1.3.6
TOTAL COINCIDENCES
sum of all coincidences detected
2.1.3.6.1
TRUE COINCIDENCE
result of COINCIDENCE DETECTION of two gamma events originating from the same positron
annihilation
61675-1 © IEC:1998(E) – 7 –
2.1.3.6.2
SCATTERED TRUE COINCIDENCE
TRUE COINCIDENCE where at least one participating photon was scattered before the
COINCIDENCE DETECTION
2.1.3.6.3
UNSCATTERED TRUE COINCIDENCE
the difference between TRUE COINCIDENCES and SCATTERED TRUE COINCIDENCES
2.1.3.6.4
RANDOM COINCIDENCE
result of COINCIDENCE DETECTION in which both participating photons emerge from different
positron annihilations
2.1.3.7
SINGLES RATE
COUNT RATE measured without COINCIDENCE DETECTION, but with energy discrimination
2.1.4
Reconstruction
2.1.4.1
TWO-DIMENSIONAL RECONSTRUCTION
in TWO-DIMENSIONAL RECONSTRUCTION, the data are rebinned prior to reconstruction into
SINOGRAMS, which are the PROJECTION data of transverse slices, which are considered being
independent of each other and being perpendicular to the SYSTEM AXIS. So, each event will be
assigned, in the axial direction, to that transverse slice passing the midpoint of the
corresponding LINE OF RESPONSE. Any deviation from perpendicularity to the SYSTEM AXIS is
neglected. The data are then reconstructed by two-dimensional methods, i.e. each slice is
reconstructed from its associated SINOGRAM, independent of the rest of the data set.
NOTE – This is the standard method of reconstruction for POSITRON EMISSION TOMOGRAPHS using small axial
acceptance angles, i.e. utilizing septa. For POSITRON EMISSION TOMOGRAPHS using large axial acceptance angles,
i.e. without septa, this method is also called ‘single slice rebinning’.
2.1.4.2
THREE-DIMENSIONAL RECONSTRUCTION
in THREE-DIMENSIONAL RECONSTRUCTION, the LINES OF RESPONSE are not restricted to being
SYSTEM AXIS LINE OF RESPONSE
perpendicular to the . So, a may pass several transverse slices.
Consequently, transverse slices cannot be reconstructed independent of each other. Each slice
has to be reconstructed utilizing the full three-dimensional data set.
2.2
IMAGE MATRIX
arrangement of MATRIX ELEMENTs in a preferentially cartesian coordinate system
2.2.1
MATRIX ELEMENT
smallest unit of an IMAGE MATRIX, which is assigned in location and size to a certain volume
element of the object (VOXEL)
2.2.1.1
PIXEL
matrix element in a two-dimensional IMAGE MATRIX
2.2.1.2
TRIXEL
matrix element in a three-dimensional IMAGE MATRIX

– 8 – 61675-1 © IEC:1998(E)
2.2.2
VOXEL
volume element in the object which is assigned to a MATRIX ELEMENT in the IMAGE MATRIX (two-
dimensional or three-dimensional). The dimensions of the VOXEL are determined by the
dimensions of the corresponding MATRIX ELEMENT via the appropriate scale factors and by the
systems SPATIAL RESOLUTION in all three dimensions
2.3
POINT SPREAD FUNCTION (PSF)
scintigraphic image of a POINT SOURCE
2.3.1
PHYSICAL POINT SPREAD FUNCTION
for tomographs, a two-dimensional POINT SPREAD FUNCTION in planes perpendicular to the
PROJECTION BEAM at specified distances from the detector
NOTE – The PHYSICAL POINT SPREAD FUNCTION characterizes the purely physical (intrinsic) imaging performance of
the tomographic device and is independent of for example sampling, image reconstruction and image processing. A
PROJECTION BEAM is characterized by the entirety of all PHYSICAL POINT SPREAD FUNCTIONs as a function of distance
along its axis.
2.3.2
AXIAL POINT SPREAD FUNCTION
profile passing through the peak of the PHYSICAL POINT SPREAD FUNCTION in a plane parallel to
the sYSTEM AXIS
2.3.3
TRANSVERSE POINT SPREAD FUNCTION
reconstructed two-dimensional POINT SPREAD FUNCTION in a tomographic IMAGE PLANE
NOTE – In TOMOGRAPHY, the TRANSVERSE POINT SPREAD FUNCTION can also be obtained from a LINE SOURCE located
parallel to the SYSTEM AXIS.
2.4
SPATIAL RESOLUTION
ability to concentrate the count density distribution in the image of a POINT SOURCE to a point
2.4.1
TRANSVERSE RESOLUTION
SPATIAL RESOLUTION in a reconstructed plane perpendicular to the SYSTEM AXIS
2.4.1.1
RADIAL RESOLUTION
TRANSVERSE RESOLUTION along a line passing through the position of the source and the
SYSTEM AXIS
2.4.1.2
TANGENTIAL RESOLUTION
TRANSVERSE RESOLUTION in the direction orthogonal to the direction of RADIAL RESOLUTION
2.4.2
AXIAL RESOLUTION
for tomographs with sufficiently fine axial sampling fulfilling the sampling theorem, SPATIAL
RESOLUTION along a line parallel to the SYSTEM AXIS
2.4.3
AXIAL SLICE WIDTH
for tomographs, the width of the AXIAL POINT SPREAD FUNCTION

61675-1 © IEC:1998(E) – 9 –
2.4.4
EQUIVALENT WIDTH (EW)
width of that rectangle, having the same area and the same height as the response function,
for example the POINT SPREAD FUNCTION
2.4.5
FULL WIDTH AT HALF MAXIMUM (FWHM)
(see annex A)
2.5
RECOVERY COEFFICIENT
measured (image) ACTIVITY concentration of an active volume divided by the true ACTIVITY
concentration of that volume, neglecting ACTIVITY calibration factors
NOTE – For the actual measurement, the true ACTIVITY concentration is replaced by the measured ACTIVITY
concentration in a large volume.
2.6
Tomographic sensitivity
2.6.1
SLICE SENSITIVITY
ratio of COUNT RATE as measured on the SINOGRAM to the ACTIVITY concentration in the
phantom
NOTE – In PET, the measured counts are numerically corrected for scatter by subtracting the SCATTER FRACTION.
2.6.1.1
NORMALIZED SLICE SENSITIVITY
SLICE SENSITIVITY divided by the AXIAL SLICE WIDTH (EW) for that slice
2.6.2
VOLUME SENSITIVITY
sum of the individual SLICE SENSITIVITIES
2.7
COUNT RATE CHARACTERISTIC (see annex A)
2.7.1
COUNT LOSS
difference between measured COUNT RATE and TRUE COUNT RATE, which is caused by the finite
RESOLVING TIME of the instrument
2.7.2
COUNT RATE
number of counts per unit of time
2.7.3
TRUE COUNT RATE (see annex A)
2.7.4
ADDRESS PILE UP
for imaging devices false address calculation of an artificial event which passes the PULSE
AMPLITUDE ANALYZER WINDOW, but is formed from two or more events by the PILE UP EFFECT
2.7.4.1
PILE UP EFFECT
false measurement of the pulse amplitude, due to the absorption of two or more gamma rays,
reaching the same radiation detector within the RESOLVING TIME

– 10 – 61675-1 © IEC:1998(E)
2.8
SCATTER FRACTION (SF)
ratio between SCATTERED TRUE COINCIDENCES and the sum of SCATTERED plus UNSCATTERED
TRUE COINCIDENCES for a given experimental set-up
2.9
POINT SOURCE
RADIOACTIVE SOURCE approximating a δ-function in all three dimensions
2.10
LINE SOURCE
straight RADIOACTIVE SOURCE approximating a δ-function in two dimensions and being constant
(uniform) in the third dimension
3 Test methods
For all measurements, the tomograph shall be set up according to its normal mode of
operation, i.e. it shall not be adjusted specially for the measurement of specific parameters. If
the tomograph is specified to operate in different modes influencing the performance
parameters, for example with different axial acceptance angles, with and without septa, with
TWO-DIMENSIONAL RECONSTRUCTION and THREE-DIMENSIONAL RECONSTRUCTION, the test results
shall be reported in addition. The tomographic configuration (e.g. energy thresholds, axial
acceptance angle, reconstruction algorithm) shall be chosen according to the manufacturer’s
recommendation and clearly stated. If any test cannot be carried out exactly as specified in this
standard, the reason for the deviation and the exact conditions under which the test was
performed shall be stated clearly.
The test phantoms shall be centred within the tomographs’ AXIAL FIELD OF VIEW, if not specified
otherwise.
NOTE – For tomographs with an AXIAL FIELD OF VIEW greater than 16,5 cm, this centring will only produce
performance estimates for the central part. However, if the phantoms were displaced axially in order to cover the
entire AXIAL FIELD OF VIEW, false results could be obtained for the central planes, if the axial acceptance angle of the
detectors is not fully covered with ACTIVITY.
3.1 SPATIAL RESOLUTION
3.1.1 General
SPATIAL RESOLUTION measurements describe partly the ability of a tomograph to reproduce the
spatial distribution of a tracer in an object in a reconstructed image. The measurement is
performed by imaging POINT (or LINE) SOURCEs in air and reconstructing images, using a sharp
reconstruction filter. Although this does not represent the condition of imaging a patient, where
tissue scatter is present and limited statistics require the use of a smooth reconstruction filter,
the measured SPATIAL RESOLUTION provides a best-case comparison between tomographs,
indicating the highest achievable performance.
3.1.2 Purpose
The purpose of this measurement is to characterize the ability of the tomograph to recover
small objects by characterizing the width of the reconstructed TRANSVERSE POINT SPREAD
FUNCTIONs of radioactive POINT SOURCEs or of extended LINE SOURCEs placed perpendicular to
the direction of measurement. The width of the spread function is measured by the FULL WIDTH
AT HALF MAXIMUM (FWHM) and the EQUIVALENT WIDTH (EW).
To define how well objects can be reproduced in the axial direction, the AXIAL SLICE WIDTH
(commonly referred to as the slice thickness) is used. It is measured with a POINT SOURCE
which is stepped through the tomographs TRANSVERSE FIELD OF VIEW axially in small increments
and is characterized by the EW and the FWHM of the AXIAL POINT SPREAD FUNCTION for each
individual slice.
61675-1 © IEC:1998(E) – 11 –
The AXIAL RESOLUTION is defined for tomographs with sufficiently fine axial sampling (volume
detectors) and could be measured with a stationary POINT SOURCE. For these systems the AXIAL
RESOLUTION (EW and FWHM) is equivalent to the AXIAL SLICE WIDTH. These systems (fulfilling the
sampling theorem in the axial direction) are characterized by the fact, that the AXIAL POINT
SPREAD FUNCTION of a stationary POINT SOURCE would not vary, if the position of the source is
varied in the axial direction for half the axial sampling distance.
3.1.3 Method
For all systems, the SPATIAL RESOLUTION shall be measured in the transverse IMAGE PLANE in
two directions (i.e. radially and tangentially). In addition, for those systems having sufficiently
fine axial sampling, an AXIAL RESOLUTION also shall be measured.
The TRANSVERSE FIELD OF VIEW and the IMAGE MATRIX size determine the PIXEL size in the
transverse IMAGE PLANE. In order to measure accurately the width of the spread function, its
FWHM should span at least ten PIXELs. A typical imaging study of a brain, however, requires a
260 mm TRANSVERSE FIELD OF VIEW, which together with a 128 × 128 IMAGE MATRIX and 6 mm
SPATIAL RESOLUTION, results in a FWHM of only three PIXELs. The width of the response may be
incorrect if there are fewer than ten PIXELs in the FWHM. Therefore, if possible, the PIXEL size
should be made close to one-tenth of the expected FWHM during reconstruction and should be
indicated as ancillary data for the TRANSVERSE RESOLUTION measurement. For volume imaging
systems, the TRIXEL size, in both the transverse and axial dimensions, should be made close to
one-tenth the expected FWHM, and should be indicated as ancillary data for the SPATIAL
RESOLUTION measurement. For all systems, the AXIAL SLICE WIDTH is measured by moving the
source in fine steps to sample the response function adequately. For the AXIAL SLICE WIDTH
measurement, the step size should be close to one-tenth the expected EW. It is assumed that a
computer controlled bed will be used for accurate positioning of the RADIOACTIVE SOURCE.
3.1.3.1 RADIONUCLIDE
The RADIONUCLIDE for the measurement shall be F, with an ACTIVITY such that the percent
COUNT LOSS is less than 5 % and the RANDOM COINCIDENCE rate is less than 5 % of the TOTAL
COINCIDENCE
rate.
3.1.3.2 RADIOACTIVE SOURCE distribution
POINT SOURCES and LINE SOURCEs as defined in 2.9 shall be used.
3.1.3.2.1 TRANSVERSE RESOLUTION
Tomographs shall use LINE SOURCEs, suspended in air to minimize scatter, for measurements
of TRANSVERSE RESOLUTION. The sources shall be kept parallel to the long axis of the
tomograph and shall be positioned radially at 50 mm intervals along Cartesian axes in a plane
perpendicular to the long axis of the tomograph i.e. r = 10 mm, 50 mm, 100 mm, 150 mm . up
to the edge of the TRANSVERSE FIELD OF VIEW. The last position shall be not more than 20 mm
from the edge and shall be stated. Each of these positions yields two measurements of
TRANSVERSE RESOLUTION, which shall be distinguished by being in the radial or tangential
direction.
SPATIAL RESOLUTION
NOTE – The at r = 0 mm may yield artificial values due to sampling, so this measurement is
done at the position r = 10 mm.
3.1.3.2.2 AXIAL SLICE WIDTH
The AXIAL POINT SPREAD FUNCTION for POINT SOURCEs suspended in air shall be measured for all
systems. The POINT SOURCEs shall be moved in fine increments along the axial direction over
the length of the tomograph, at radial positions of r = 0 mm, 50 mm, 100 mm, . in 50 mm
steps up to the edge of the TRANSVERSE FIELD OF VIEW. The last position shall be not more than
20 mm from the edge and shall be stated. The source is stepped in the axial direction by one-
tenth of the expected EW of the axial response function. For each radial position, the measured
values shall be corrected for decay. This measurement does not apply to THREE-DIMENSIONAL
RECONSTRUCTION.
– 12 – 61675-1 © IEC:1998(E)
3.1.3.2.3 AXIAL RESOLUTION
For systems having axial sampling at least three times smaller than the FWHM of the AXIAL
POINT SPREAD FUNCTION the measurement of AXIAL RESOLUTION can be made with stationary
POINT SOURCEs. POINT SOURCEs suspended in air are positioned at radial intervals of 50 mm,
starting at the centre and extending to a distance which depends on the TRANSVERSE FIELD OF
VIEW, as described in the measurement of AXIAL SLICE WIDTH (3.1.3.2.2.). Each POINT SOURCE
shall be imaged at axial intervals of 20 mm, starting at the centre of the tomograph and
extending to within 10 mm from the edge of the AXIAL FIELD OF VIEW.
3.1.3.3 Data collection
Data shall be collected for all sources in all of the positions specified above, either singly or in
groups of multiple sources, to minimize the data acquisition time. At least fifty thousand counts
shall be acquired in each response function, as defined below.
3.1.3.4 Data processing
Reconstruction using a ramp filter with the cutoff at the Nyquist frequency of the PROJECTION
data, shall be employed for all SPATIAL RESOLUTION data.
3.1.4 Analysis
RADIAL RESOLUTION TANGENTIAL RESOLUTION
The and the shall be determined by forming one-
dimensional response functions, which result from taking profiles through the TRANSVERSE
POINT SPREAD FUNCTION in radial and tangential directions, passing through the peak of the
distribution.
The AXIAL RESOLUTION of the POINT SOURCE measurements is determined by forming one-
AXIAL POINT SPREAD FUNCTION
dimensional response functions ( s), which result from taking
profiles through the volume image in the axial direction, passing through the peak of the
distribution in the slice nearest the source.
The AXIAL SLICE WIDTH is determined by forming one-dimensional response functions (AXIAL
POINT SPREAD FUNCTIONs), which result from summing the counts per slice collected for each
slice at each axial location of each radial source location.
Each FWHM shall be determined by linear interpolation between adjacent PIXELs at half the
maximum PIXEL value, which is the peak of the response function (see figure 11). Values shall
be converted to millimetre units by multiplication with the appropriate PIXEL size.
Each EQUIVALENT WIDTH (EW) shall be measured from the corresponding response function.
EW is calculated from the formula
CP x W
i
EW =
∑
C
i m
where
C is the sum of the counts in the profile between the limits defined by 1/20 C on either
∑
i
m
side of the peak;
C is the maximum PIXEL value;
m
PW is the PIXEL width (or axial increment in the case of the AXIAL SLICE WIDTH) in millimetres
(see figure 12).
61675-1 © IEC:1998(E) – 13 –
3.1.5 Report
RADIAL and TANGENTIAL RESOLUTIONs (FWHM and EW) for each radius, averaged over all slices,
shall be calculated and reported as TRANSVERSE RESOLUTION values. AXIAL SLICE WIDTHs (EW
and FWHM) for each radius, averaged over all slices for each type (e.g. odd, even) shall be
reported. Transverse PIXEL dimensions and axial step size shall also be reported.
For systems, where AXIAL RESOLUTION is to be measured, AXIAL RESOLUTION (FWHM and EW),
averaged over all slices, shall be reported. For these systems, the axial PIXEL dimension in
millimetres shall also be reported.
For systems utilizing THREE-DIMENSIONAL RECONSTRUCTION, RESOLUTION data as listed above
shall not be averaged. Graphs of TRANSVERSE RESOLUTION and AXIAL RESOLUTION shall be
reported, showing the RESOLUTION values (RADIAL RESOLUTION, TANGENTIAL RESOLUTION, and
AXIAL RESOLUTION) for each radius as a function of slice number.
3.2 RECOVERY COEFFICIENT
3.2.1 General
The finite resolution of a tomograph leads to a spreading of image counts beyond the
geometrical boundaries of the object. This effect becomes more important as the object size
decreases. The RECOVERY COEFFICIENT provides an assessment of the ability of the tomograph
to quantify the ACTIVITY concentration as a function of the object size.
3.2.2 Purpose
The objective of the following procedures is to quantify the apparent decrease in tracer
concentration in a region of interest (ROI) of an image of spherical sources of different
diameters.
3.2.3 Method
A number of hollow spheres, filled with an ACTIVITY concentration of F from a stock solution,
are placed in the water-filled head phantom (see figures 1 and 4) which is in turn placed in the
centre of the TRANSVERSE FIELD OF VIEW. The phantom shall be held in position without
introducing additional attenuating material. At least two samples from this solution are counted
in a well counter. The spheres are arranged to be coplanar.
For discrete ring systems, utilizing TWO-DIMENSIONAL RECONSTRUCTION, separate measure-
ments shall be made with the spheres centred over each representative type of slice derived
from different ring combinations (e.g. direct and cross, or odd and even). A measurement shall
also be taken halfway in between slices in order to see the worst case of recovery in addition to
the best case. The measurements are taken near the axial centre of the tomograph.
For systems utilizing THREE-DIMENSIONAL RECONSTRUCTION, the measurements shall be done at
the axial centre of the tomograph and halfway between the axial centre and the edge of the
AXIAL FIELD OF VIEW.
After data acquisition, the spheres are removed and the cylinder filled with a uniform solution of
F from which at least two samples are taken for well counting.
3.2.4 Data collection
The data collection shall be carried out at low COUNT RATEs such that the COUNT LOSS is less
than 10 % and the RANDOM COINCIDENCE rate is less than 10 % of the TOTAL COINCIDENCE rate.
Care should be taken to acquire sufficient counts so that statistical variations do not
significantly affect the result. So, for the slice containing the spheres, at least 2 000 000 counts
shall be acquired. COUNT RATEs and scanning times shall be stated.

– 14 – 61675-1 © IEC:1998(E)
3.2.5 Data processing and analysis
Reconstruction shall be performed using a ramp filter with a cut-off at the Nyquist frequency
and with all corrections applied. The method of ATTENUATION correction shall be by an
analytical calculation. The ATTENUATION coefficient used shall be reported. The scatter
correction method used shall be clearly described.
Circular ROIs of diameter as close as possible to the FWHM as measured in section 3.1.3.2.1
are defined centrally on the image of each sphere. The precise ROI diameter should be stated.
A large ROI (diameter: 150 mm) is centred on the image of the uniform cylinder. Calculation of
the RECOVERY COEFFICIENT (RC ) for each sphere is obtained from the equation:
si
 C 
si
 
SM
 s 
RC =
si
 
C
u
 
SM
 
u
where
C are the ROI counts/pixel/s for sphere i;
si
SM are the sample counts/s/cm (stock solution spheres);
s
C are the ROI counts/pixel/s (head phantom);
u
SM are the sample counts/s/cm (head phantom);
u
C /SM represents a calibration factor for a large reference object.
u u
Care shall be taken to correct for any dead-time and sample volume effects in the well counter.
RC is then plotted against sphere diameter to give recovery curves.
si
3.2.6 Report
Graphs of RECOVERY COEFFICIENTs for each axial position described in 3.2.3 shall be reported.
The scatter correction method used shall be clearly described, as well as the attenuation
coefficient used.
3.3 Tomographic sensitivity
3.3.1 General
Tomographic sensitivity is a parameter that characterizes the rate at which coincidence events
are detected in the presence of a RADIOACTIVE SOURCE in the limit of low ACTIVITY where COUNT
LOSSES and RANDOM COINCIDENCES are negligible. The measured rate of TRUE COINCIDENCE
EVENTS for a given distribution of the RADIOACTIVE SOURCE depends upon many factors,
including the detector material, size, and packing fraction, tomograph ring diameter, axial
acceptance window and septa geometry, ATTENUATION, scatter, dead-time, and energy
thresholds.
3.3.2 Purpose
The purpose of this measurement is to determine the detected rate of TRUE COINCIDENCE
events per unit of ACTIVITY concentration for a standard volume source, i.e. a cylindrical
phantom of given dimensions.
3.3.3 Method
The tomographic sensitivity test places a specified volume of radioactive solution of known
ACTIVITY concentration in the TOTAL FIELD OF VIEW of the POSITRON EMISSION TOMOGRAPH and
observes the resulting COUNT RATE. The systems sensitivity is calculated from these values.

61675-1 © IEC:1998(E) – 15 –
The test is critically dependent upon accurate assays of radioactivity as measured in a dose
calibrator or well counter. It is difficult to maintain an absolute calibration with such devices to
accuracies finer than 10 %. Absolute reference standards using positron emitters should be
considered if higher degrees of accuracy are required.
3.3.3.1 RADIONUCLIDE
The RADIONUCLIDE used for these measurements shall be F. The amount of ACTIVITY used
shall be such that the percentage of COUNT LOSSES is less than 2 % and the RANDOM
COINCIDENCE rate is less than 2 % of the TOTAL COINCIDENCE rate.
3.3.3.2 RADIOACTIVE SOURCE distribution
The head phantom (figure 1) shall be filled with a homogeneous solution of known ACTIVITY
concentration. The phantom shall be held in position without introducing additional attenuating
material. It shall be centred both axially and transaxially in the TOTAL FIELD OF VIEW.
3.3.3.3 Data collection
Each coincident event between individual detectors shall be taken into account only once. Data
shall be assembled into SINOGRAMs. All events will be assigned to the transverse slice passing
the midpoint of the corresponding LINE OF RESPONSE.
At least 200 000 counts shall be acquired for each slice within the lesser of the AXIAL FIELD OF
VIEW or the central 16,5 cm where the phantom was placed.
3.3.3.4 Data processing
The ACTIVITY concentration in the phantom shall be corrected for decay to determine the
average ACTIVITY concentration, a , during the data acquisition time, T , by the following
ave acq
equation:
 T 
 
A 1 T TT− 
acq
cal 1/2 cal 0
a = exp ln2 1−−exp  ln2
 
ave  
V ln2 T T T
 
acq  1/ 2   1/ 2 
 
where
V is the volume of the phantom;
A is the ACTIVITY times branching ratio ("positron activity") measured at time T ;
cal cal
T is the acquisition start time;
T is the HALF LIFE of the RADIONUCLIDE.
1/2
It is not necessary to reconstruct these data. No corrections for detector normalization, COUNT
LOSS, scatter, and ATTENUATION shall be applied. The data shall be corrected for RANDOM
COINCIDENCES.
3.3.4 Analysis
The total counts C on each slice i shall be obtained by summing all PIXELs in the
i,tot,120mm
corresponding SINOGRAM within a radius of 120 mm. The SLICE SENSITIVITY S for unscattered
i
events shall be found by the following:
C ()1−SF
i,tot,120mm i
S =
i
T a
acq ave
where SF is the corresponding SCATTER FRACTION (see 3.6).
i
– 16 – 61675-1 © IEC:1998(E)
The NORMALIZED SLICE SENSITIVITY for each slice nS shall be calculated as follows:
i
S
i
nS =
i
EW
a,i
where EW is the AXIAL SLICE WIDTH for slice i (see 3.1.4).
a,i
NOTE – The NORMALIZED SLICE SENSITIVITY allows for comparison of tomographs with different AXIAL SLICE WIDTH.
The VOLUME SENSITIVITY, S , shall be the sum of S over all slices of the tomograph within the
tot i
central 16,5 cm or the AXIAL FIELD OF VIEW, whichever is smaller.
NOTE – This will yield only the VOLUME SENSITIVITY for the central part of the tomograph, if the AXIAL FIELD OF VIEW
is greater than 16,5 cm.
3.3.5 Report
For each slice i, tabulate the values of S and nS . The VOLUME SENSITIVITY S shall also be
i i tot
reported.
3.4 Uniformity
No test has been specified to characterize the uniformity of reconstructed images, because all
methods known so far will mostly reflect the noise in the image.
3.5 COUNT RATE CHARACTERISTIC
3.5.1 General
PET COUNT RATE performance depends in a complex manner on the spatial distribution of
ACTIVITY and scattering materials, which we will refer to as the different scatter conditions
(see 3.5.3.1). The COUNT RATE CHARACTERISTIC of the TRUE COINCIDENCE COUNT RATE is highly
dependent on the trues-to-singles ratio and on the COUNT RATE CHARACTERISTIC of the SINGLES
RATE and consequently on the set up of the measurements conditions, which therefore should
simulate the range of clinical imaging situations. In addition, COUNT RATE performance is
strongly influenced by the amount of RANDOM COINCIDENCEs and by the accuracy of the
subtraction of these events.
NOTE – As the TRUE COINCIDENCE COUNT RATE includes scattered events, the relative SCATTER FRACTION must be
considered when comparing tomographs with different design.
3.5.2 Purpose
The procedure described here is designed to evaluate deviations from the linear relationship
between TRUE COINCIDENCE COUNT RATE and ACTIVITY, caused by COUNT LOSSES, and the
evaluation of image distortions at high COUNT RATES, especially those leading to spatially
misplaced events by ADDRESS PILE UP. As modern PET tomographs are operated with COUNT
LOSS correction schemes, the accuracy of these correction algorithms is tested.
PET COUNT RATE performance means:
a) the relationship between measured TRUE COINCIDENCES (UNSCATTERED plus SCATTERED TRUE
COINCIDENCES) and ACTIVITY, i.e. the COUNT RATE CHARACTERISTIC of TRUE COINCIDENCE
COUNT RATE;
b) a test to determine address errors caused by ADDRESS PILE UP;
c) the evaluation of the accuracy of the COUNT LOSS correction scheme.

61675-1 © IEC:1998(E) – 17 –
3.5.3 Method
For dedicated brain tomographs, only the scatter condition described in 3.5.3.1.1 applies,
whereas for all other tomographs the scatter conditions described in 3.5.3.1.1 to 3.5.3.1.3
apply. For all tests the only correction to be applied is the subtraction of the multiple and the
RANDOM COINCIDENCEs (to calculate TRUE COINCIDENCE counts). No correction is made for
COUNT LOSSES, ATTENUATION, and scatter, unless otherwise stated. The ACTIVITY shall generally
be specified as the total amount of ACTIVITY within the phantom as specified in 3.5.3.1. As the
variation of ACTIVITY is normally achieved by radioactive decay, care should be taken with
respect to the radiochemical purity of the ACTIVITY used.
3.5.3.1 RADIOACTIVE SOURCE distribution
To describe various scatter conditions, three different experimental set-ups are to be used.
3.5.3.1.1 Head imaging
The head phantom (figure 1) filled homogeneously with ACTIVITY.
3.5.3.1.2 Cardiac imaging
The body phantom, figure 2, (head phantom inserted) with outer section and arms (figure 3) of
the phantom filled with water, inner section (head phant
...