This international standard was developed in accordance with internationally recognized principles on standardization established in the Decision on Principles

for the
Development of International Standards, Guides and Recommendations issued by the World Trade Organization Technical Barriers to Trade (TBT) Committee.

Designation: F2554 − 22

Standard Practice for

Measurement of Positional Accuracy of Computer-Assisted
Surgical Systems1
This standard is issued under the fixed designation F2554; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. A
superscript epsilon (´) indicates an editorial change since the last revision or reapproval.

1. Scope standard, except for angular measurements, which may be

1.1 This document provides procedures for measurement reported in terms of radians or degrees.
and reporting of basic static performance of surgical navigation 1.4 This standard does not purport to address all of the
and/or robotic positioning devices under defined conditions. safety concerns, if any, associated with its use. It is the
They can be performed on a subsystem (for example, tracking responsibility of the user of this standard to establish appro-
only) or a full computer-aided surgery system as would be used priate safety, health, and environmental practices and deter-
clinically. Testing a subsystem does not mean that the whole mine the applicability of regulatory limitations prior to use.
system has been tested. The functionality to be tested based on 1.5 This international standard was developed in accor-
this practice is limited to the performance (accuracy in terms of dance with internationally recognized principles on standard-
bias and precision) of the system regarding point localization in ization established in the Decision on Principles for the
space by means of a pointer. A point in space has no Development of International Standards, Guides and Recom-
orientation; only multidimensional objects have orientation. mendations issued by the World Trade Organization Technical
Therefore, orientation of objects is not within the scope of this Barriers to Trade (TBT) Committee.
practice. However, in localizing a point the different orienta-
tions of the pointer can produce errors. These errors and the 2. Referenced Documents
pointer orientation are within the scope of this practice. The 2.1 ASTM Standards:2
aim is to provide a standardized measurement of performance E456 Terminology Relating to Quality and Statistics
variables by which end users can compare within a system (for E2281 Practice for Process Capability and Performance
example, with different reference elements or pointers) and Measurement
between different systems (for example, from different manu-
facturers). Parameters to be evaluated include (based upon the 2.2 Other References:3
features of the system being evaluated): ISO 10360 Geometrical Product Specifications (GPS)—
(1) Accuracy of a single point relative to a coordinate Acceptance and Reverification Tests for Coordinate Mea-
system. suring Machines (CMM)
(2) Sensitivity of tracking accuracy due to changes in
pointer orientation. 3. Terminology
(3) Relative point-to-point accuracy. 3.1 Definitions:
1.1.1 This method covers all configurations of the evaluated 3.1.1 accuracy, n—the closeness of agreement between a
system as well as extreme placements across the measurement measurement result and an accepted reference value. E456
volume. [Link] Discussion—In the context of this standard, with the
1.2 This practice defines a standardized reporting format, definitions of bias and precision (see below), it can be
which includes definition of the coordinate systems to be used considered that the accuracy of a measurement of a point will
for reporting the measurements, and statistical measures (for be subject to some bias error and some precision error.
example, mean, RMS, and maximum error). 3.1.2 bias, n—the difference between the expectation of the
1.3 The values stated in SI units are to be regarded as measurement results and an accepted reference value. E456
standard. No other units of measurement are included in this

1 2
This practice is under the jurisdiction of ASTM Committee F04 on Medical and For referenced ASTM standards, visit the ASTM website, [Link], or
Surgical Materials and Devices and is the direct responsibility of Subcommittee contact ASTM Customer Service at service@[Link]. For Annual Book of ASTM
F04.38 on Computer Assisted Orthopaedic Surgical Systems. Standards volume information, refer to the standard’s Document Summary page on
Current edition approved Sept. 1, 2022. Published September 2022. Originally the ASTM website.
3
approved in 2010. Last previous edition approved in 2018 as F2554 – 18. DOI Available from American National Standards Institute (ANSI), 25 W. 43rd St.,
10.1520/F2554-22. 4th Floor, New York, NY 10036, [Link]

Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. United States

Copyright by ASTM Int'l (all rights reserved) Sat Jul 12 [Link] UTC 2025
1
Downloaded/printed by
University of British Columbia Library pursuant to License Agreement. No further reproductions authorized.
 F2554 − 22
[Link] Discussion—In the context of this standard, bias [Link] Discussion—In the context of this standard, preci-
represents the systematic error in a set of measurements of a sion represents scatter of a set of measurements of a point.
target reference point making their average deviate from the 3.1.19 range, R, n—the largest observation minus the small-
actual reference point with a certain magnitude and direction. est observation in a set of values or observations. E456, E2281
3.1.3 calibration, n—the pre- or intraoperative registration 3.1.20 reference element, n—an artificial item composed of
of an item or device to its reference element. rigidly bound markers in a unique and asymmetrical pattern
3.1.4 computer-assisted surgery (CAS), n—the use of com- recognizable by the tracker. While being rigidly attached to a
puters to facilitate or enhance surgical procedures via the use of therapeutic object, the position and orientation of the reference
three-dimensional space tracking of objects. element can be used to determine those of the therapeutic
3.1.5 coordinate measuring machine (CMM), n—measuring object after its calibration.
system with the means to move a stylus and capability to 3.1.21 registration, n—the determination of the spatial rela-
determine spatial coordinates on a work piece surface. ISO tionship between the referential frames of two coordinate
10360-1 systems. This may occur between two reference elements or
3.1.6 degree of freedom (DOF), n—set of independent between the fiducials and a reference element of a therapeutic
displacements that specify completely the displaced or de- object. The registration is rigid if it consists only of rotations
formed position of the body or system. and translations (six degrees of freedom) and non-rigid if it
also comprises scaling and/or local or global distortions (seven
3.1.7 dynamic reference base, n—the coordinate system of a
degrees of freedom and more).
reference element used for the tracking of other therapeutic
objects. 3.1.22 repeatability, n—precision under repeatability
3.1.8 fiducial, n—an artificial item (for example, a screw or conditions. E456
a sphere) rigidly attached to a therapeutic object to facilitate its 3.1.23 reproducibility, n—precision under reproducibility
calibration. conditions. E456
3.1.9 ground truth, n—short name for the accepted reference 3.1.24 robotic positioning system, n—use of an active me-
value (see 3.1.1). chanical (mechatronic) device to position an instrument guide
3.1.10 marker, n—a single 3-degree-of-freedom indicator on at a specified location in 3D space (up to six degrees of
a reference element or dynamic reference base. freedom).
3.1.11 maximum error, n—the largest distance between any 3.1.25 root mean square (RMS), n—means of estimation of
measured point and its ground truth for any trial during a the scatter of a set of values, which consists of the square root
testing procedure. of the average of the squared values.
3.1.12 mean, n—of a population, u, average or expected 3.1.26 therapeutic object, n—a surgical item or a part of the
value of a characteristic in a population; of a sample, x, sum of patient.
the observed values in the sample divided by the sample size. 3.1.27 tracker, n—a device that detects and locates fiducials
E456 and markers in its measurement volume. This can be achieved
3.1.13 measurement range, n—the interval of allowed val- by mechanical linkage or by analyzing signals of various types
ues for a specific degree of freedom while performing the (visible or infrared light, electromagnetic field, or ultrasound).
practice. 4. Summary of Practice
3.1.14 measurement volume, n—measuring range of a 4.1 This practice provides recommendations for the
tracker, stated as simultaneous limits on all spatial coordinates collection, analysis, and presentation of data regarding the
measured by the tracker. ISO 10360-1 positional accuracy (in terms of bias and precision) of surgical
3.1.15 navigation system, n—a set of devices consisting of a navigation and robotic positioning systems under repeatable
computer, its associated software, and a tracker capturing the conditions.
reference elements within the measurement volume. This
4.2 Data to be reported consists of all measurements, their
system provides real-time feedback of the state of the surgical
corresponding errors if applicable, their statistical analysis, the
scene under operation.
test conditions, and the system conditions.
3.1.16 phantom, n—standardized measurement object. See
Appendix X1 for details regarding the design of the phantom 5. Significance and Use
used in this practice. 5.1 The purpose of this practice is to provide data that can
3.1.17 pointer, n—the device offered by the evaluated sys- be used for evaluation of the accuracy of different CAS
tem to point and locate a position on any object including systems.
anatomical landmarks. The pointer is the whole device, includ- 5.2 The use of surgical navigation and robotic positioning
ing the stylus-like tip all the way to any reference element used systems is becoming increasingly common. In order to make
to track it in space. informed decisions about the suitability of such systems for a
3.1.18 precision, n—the closeness of agreement between given procedure, their accuracy capability needs to be evalu-
independent measurement results obtained under stipulated ated under clinical application and compared to the require-
conditions. E456 ments. As the performance of a whole system is constrained by

Copyright by ASTM Int'l (all rights reserved) Sat Jul 12 [Link] UTC 2025
2
Downloaded/printed by
University of British Columbia Library pursuant to License Agreement. No further reproductions authorized.
 F2554 − 22
those of its subparts, a preliminary step must be to objectively 8.3 System Conditions—The system is composed of various
characterize the accuracy of the tracking subsystem in a parts and all their references and configuration shall be
controlled environment under controlled conditions. provided, including firmware and software versions. Any
5.3 In order to make comparisons within and between changes to the system beyond what is provided and configured
systems, a standardized way of measuring and reporting by the manufacturer are to be reported and justified (for
accuracy is needed. Parameters such as coordinate system, example, using third-party markers or pointer). Specific details
units of measurement, terminology, and operational conditions of the phantom are also to be reported (for example, the divot
must be standardized. dimensions). The measured points are to be acquired only
through the firmware and software provided by the manufac-
6. Apparatus turer.
6.1 The system under test is considered to have at least
8.4 Phantom Placement and Registration—In the first series
some tracking functionality, a pointer and associated hardware,
of tests, place the system tracker nominally at the recom-
and software. If the system is provided by the manufacturer
mended distance from the phantom. At this location, a regis-
with various combinations of parts, the evaluation must be
performed at least with the combination known to present the tration of the phantom to the dynamic reference base may be
worst-case scenario in terms of accuracy. For example, the required for most systems. Any registration shall be performed
tester may use the longest pointer with the smallest reference as described by the manufacturer, simulating registration of a
elements. patient’s anatomy in the clinical environment. Registration can
only be done once to cover the sequence of steps 8.6 – 8.9, but
6.2 This practice relies on a phantom. See Appendix X1 for may be repeated in between.
design requirements. The phantom size and points have been
designed to approximate a typical surgical site on the human 8.5 Point Acquisition—In each trial, the tester locates the
body. All divots of the phantom shall be measured by a CMM individual labeled points on the phantom and acquires its
(or another measurement system of similar performance trace- position using the pointer following the system manufacturer’s
able to NIST, FDA, EU, and ISO standards). These phantom instructions for obtaining point data. This includes pointer
measurements will constitute the ground truth used for the orientation except for the rotation tests in 8.7.
accuracy assessment. Therefore, the accuracy of the CMM
must be better than that of the system being evaluated. 8.6 Test 1: Single Point Accuracy—Treating the phantom as
if it was part of the patient’s anatomy, this test requires the
6.3 If the evaluated system relies on a dynamic reference measurement of a designated point of the phantom multiple
base for its measurements, a reference element is attached to times, to compare the positions measured versus the actual
the phantom. This reference element and its attachment shall
position on the phantom relative to its local coordinate system.
replicate as close as possible those used in a surgical setting.
The single point measurement is then independently performed
6.4 If dedicated additional software functionality is used for 20 times on the central divot (#20 in Fig. X1.1). Bias is
assistance in performing the tasks outlined in this practice, or estimated by the difference between the average of the mea-
for statistical analysis of the measurements, this addition must sured points and the central divot. The result is a small error
not alter the way the measurements are made by the system to vector emanating from the target reference point. Calculate the
be used clinically. average of all the error vectors by vectorial summation then
7. Hazards dividing the length of the resulting vector by the number of
vectors. Report the average error vector and the length of the
7.1 None. longest error vector. For the determination of precision, start by
8. Procedure calculating the average point of all measurements, which
represents the system’s best estimate for the location of the
8.1 Set up the system to be evaluated, including the dynamic
target central divot. Calculate the distances of all the measure-
reference base, if used. The dynamic reference base shall not
ments from this average point, determine and report the RMS
be repositioned relative to the phantom during the procedure.
of these distances as well as their maximum.
Measured points shall be expressed in the coordinate system of
the phantom, to enable them to be directly compared to those 8.7 Pointer Rotation—These steps quantify the variation in
made by CMM. successive measurements of the same central divot (#20 Fig.
8.2 Test Conditions—The conditions used for this procedure X1.1) with various pointer orientations (tilt). The maximum
should be as close as possible to those typically found in the physically possible angular range depends on the axis of
surgical setting for which the system is designed, and those rotation, but most systems have some limit to this range
conditions shall be reported. They include all known factors beyond which tracking is no longer provided. The actual range
that may influence the performance of the evaluated system, for which the system provides tracking data should be tested at
such as temperature, humidity, lighting, as well as potential no fewer than ten uniformly spaced intervals. The actual range,
sources of interference (for example, infrared noise and reflec- all angular increments, and the corresponding measurements
tions for optical systems, large metal objects for electromag- shall be reported. The tester may use a protractor scale to
netic systems). Some of these factors may not be replicated if measure the pointer angles or any other method. For each test
their non-inclusion is reported and justified. in 8.7, the maximum distance between any two measurements

Copyright by ASTM Int'l (all rights reserved) Sat Jul 12 [Link] UTC 2025
3
Downloaded/printed by
University of British Columbia Library pursuant to License Agreement. No further reproductions authorized.
 F2554 − 22
shall be reported, as well as the RMS of the deviations between distributed at the outer boundaries of the measurable volume
all measurements and the average point measured in 8.6 (Test cross section at the furthest functional distance from the tracker
1). (see Fig. X1.6).
8.7.1 Test 2—The single point is measured under successive 9. Report
rotations of the pointer about its principal (shaft) axis from its
9.1 Report characteristics of the phantom including its
nominal orientation (see left picture in Fig. X1.5). For this
CMM measurements and divot geometry.
mode of rotation, the maximum theoretical angular range is
[–180°, 180°]. 9.2 Test conditions, as described in 8.2.
8.7.2 Test 3—The single point is measured during succes- 9.3 System conditions, as described in 8.3.
sive rotations about the line between the tracker and the central 9.4 For each step in 8.6 – 8.9, the report should include:
divot (see middle picture in Fig. X1.5). For this mode of 9.4.1 The orientation of the phantom relative to the tracker
rotation, the maximum theoretical angular range is [–90°, 90°], as described in 8.9 and expressed in the coordinate system of
limited further by various factors including the pointer leaving the tracker.
the divot or impacting the dynamic reference base. 9.4.2 The location of the phantom relative to the measurable
8.7.3 Test 4—The single point is measured during succes- volume as described in 8.10 and expressed in the coordinate
sive rotations about an axis perpendicular to the line between system of the tracker.
the tracker and the central divot (see right picture in Fig. X1.5). 9.4.3 The coordinates and label of all measured points with
For this mode of rotation, the maximum theoretical angular appropriate units. All measurements of points on the phantom
range is [–90°, 90°], limited further by various factors includ- should be expressed in a Cartesian coordinate system of the
ing the pointer leaving the divot or impacting the dynamic phantom with clearly described origin and axes.
reference base. 9.4.4 The resulting statistical analysis, as detailed for each
step in 8.6 – 8.8.
8.8 Test 5: Point-to-Point Accuracy—The purpose of this
test is to determine the accuracy of measuring the distances 10. Precision and Bias
between various points. A minimum of 20 points are to be 10.1 As with any measurement system, uncertainties and
measured, including divot #30, with at least two points per errors are present, and the purpose of this standard is to
plane of the phantom (four points for the slanted plane). The estimate them. In order for the accuracy of computer-aided
point-to-point distance shall be calculated for each possible orthopedic systems to be estimated in a reproducible and
combination of two measured points (for example, 20 mea- comparable manner among laboratories, it is essential that
sured points yield 190 distances). The errors are the differences uniform procedures be established as specified by this stan-
between these calculated distances and the corresponding dard. Sufficient data have not been produced using identical
distances from the ground truth based on the phantom CMM navigation or robotic systems in different laboratories to
measurements. The errors are to be reported, alongside the provide the precision and bias of this procedure.
maximum error and the RMS. 10.2 The publication of this test method is intended, in part,
to facilitate uniform testing and reporting of data from systems
8.9 Test 6: Phantom Rotation—During the course of a
used clinically. Some data (2–4) have been published to help
surgical operation using the system, the patient’s pose may be
toward this end. Validation of this methodology may be
changed and hence that of the dynamic reference base, or the
achieved through round-robin testing.
tracking system may move (for example, cameras). To measure
the impact of such movement on the system’s accuracy, the 10.3 As specified in 6.2, a CMM or other traceable mea-
phantom (assembly, included the reference base) shall be surement device of suitable precision must be used to establish
rotated to two extreme orientations about the vertical axis a ground truth of all the locations of the points on the phantom.
passing through the center of the phantom held horizontally CMM manufacturers typically claim accuracies which exceed
(see Fig. X1.7). Those extreme orientations represent the two the expected accuracies of surgical navigation systems by
opposite orientation limits where tracking of the phantom several times.
remains possible. The steps described in 8.6 shall then be 11. Keywords
performed again.
11.1 augmented reality assisted surgery; computer-assisted
8.10 Phantom Locations—Repeat 8.6 – 8.9 at four different surgery; computer navigation; imageless guided surgery; infra-
locations within the measurable volume, representing its func- red tracking system; magnetic tracking; optical tracking; ro-
tional extremes. The four locations should be uniformly botic surgery

Copyright by ASTM Int'l (all rights reserved) Sat Jul 12 [Link] UTC 2025
4
Downloaded/printed by
University of British Columbia Library pursuant to License Agreement. No further reproductions authorized.
 F2554 − 22
APPENDIX

(Nonmandatory Information)

X1. PHANTOM

X1.1 A phantom is used in this practice to evaluate the lines (#1 to #18) and (#1 to #19) as the x-axis and y-axis,
tracking accuracy of a system. To enable reliable comparisons respectively. The z-axis, defined as the cross product of the x-
between evaluated systems, all phantoms should provide the and y-axes, points vertically upwards from the phantom. The
functionality described hereafter. coordinates of the points in CAD, the points measured by
X1.1.1 The phantom has 47 marked divots arrayed on five CMM, and the points measured by the evaluated system are all
different machined faces with two orientations (see Fig. X1.1 be expressed in this coordinate system.
for illustration). The CAD coordinates of these points are listed
in Table X1.1 and blueprints of the phantom are provided in X1.3 The user can construct their own phantom (or pur-
Figs. X1.2 and X1.3. chase one from a third party), provided each individual
X1.1.2 The phantom also includes an interface (for phantom has all its points measured with a CMM or similar
example, threaded holes) that enables attachment of the dy- measurement device. It is recommended that the phantom is
namic reference base. As specified in 6.3, this reference made from a rigid and thermally stable material, so that its
element and its attachment are specific to the evaluated system shape and thus the relationship of the 3D positions of the divots
and intended to replicate those of a surgical setting. The details relative to each other remain stable between uses of the
of this interface are not shown in the drawings and are left for phantom. The detailed design of the divots can be chosen by
the standard user to decide. the user to suit the shape and dimensions of the pointer tip (see
X1.2 For consistency, the Cartesian coordinate system of Fig. X1.4 for an example of divot designed for a 1 mm
the phantom has point #1 set as the origin, the perpendicular diameter ball point tip).

FIG. X1.1 Photo and Diagram of Phantom with Designated Numbering of Divots

Copyright by ASTM Int'l (all rights reserved) Sat Jul 12 [Link] UTC 2025
5
Downloaded/printed by
University of British Columbia Library pursuant to License Agreement. No further reproductions authorized.
 F2554 − 22
TABLE X1.1 Example Table of CMM Measurements for the
Phantom in Figs. X1.2 and X1.3
x y z
Point #
(mm) (mm) (mm)
1 0.00 0.00 0.00
2 0.00 14.44 0.00
3 14.44 0.00 0.00
4 0.00 28.89 0.00
5 28.88 0.00 0.00
6 0.00 43.33 0.00
7 43.32 0.00 0.00
8 0.00 57.78 0.00
9 57.76 0.00 0.00
10 0.00 72.22 0.00
11 72.20 0.00 0.00
12 0.00 86.66 0.00
13 86.64 0.00 0.00
14 0.00 101.11 0.00
15 101.08 0.00 0.00
16 0.00 115.55 0.00
17 115.52 0.00 0.00
18 0.00 130.00 0.00
19 129.96 0.00 0.00
20 32.50 32.50 0.00
21 16.25 113.75 0.00
22 32.50 97.50 0.00
23 48.74 81.25 0.00
24 64.99 65.01 0.00
25 81.24 48.76 0.00
26 97.49 32.51 0.00
27 113.73 16.26 0.00
28 35.36 130.00 10.28
29 59.34 115.91 10.28
30 68.54 96.82 10.28
31 87.63 87.63 15.28
32 96.82 68.54 15.28
33 115.91 59.34 15.28
34 130.00 35.36 15.28
35 65.87 129.99 36.62
36 97.94 97.94 36.63
37 130.00 65.87 36.62
38 88.44 116.72 43.20
39 116.72 88.44 43.20
40 84.43 130.01 49.75
41 107.23 107.23 49.77
42 130.01 84.43 49.75
43 101.89 130.00 56.35
44 130.00 101.89 56.35
45 120.18 120.18 56.35
46 118.81 130.00 56.35
47 130.00 118.81 56.35

FIG. X1.2 Blueprint of Phantom and Position of Some Points (all dimensions in mm)

Copyright by ASTM Int'l (all rights reserved) Sat Jul 12 [Link] UTC 2025
6
Downloaded/printed by
University of British Columbia Library pursuant to License Agreement. No further reproductions authorized.
 F2554 − 22

FIG. X1.3 Position of the Other Points and Angle of the Slanted Surface (all dimensions in mm)

Copyright by ASTM Int'l (all rights reserved) Sat Jul 12 [Link] UTC 2025
7
Downloaded/printed by
University of British Columbia Library pursuant to License Agreement. No further reproductions authorized.
 F2554 − 22

FIG. X1.4 Example of Divot Geometry to Accommodate a Ball-Point Tip with 1 mm Diameter

FIG. X1.5 Illustrations of Rotation Tests (with reference to 8.7.1, 8.7.2, and 8.7.3 respectively)

FIG. X1.6 Illustration of Phantom Locations in a Measurable Volume (L0 is the default location and L1 to L4 are the extreme locations)

FIG. X1.7 Illustration of Two Opposite Extreme Vertical Rotations of the Phantom Seen from Above

Copyright by ASTM Int'l (all rights reserved) Sat Jul 12 [Link] UTC 2025
8
Downloaded/printed by
University of British Columbia Library pursuant to License Agreement. No further reproductions authorized.
 F2554 − 22

FIG. X1.8 Additional Photos Showing Different Aspects of a Phantom with a Reference Element Attached

RELATED MATERIAL

ASTM E691 Practice for Conducting an Interlaboratory Study to Deter- Retrieval and Handling (Derived from ASTM F561)
mine the Precision of a Test Method 21 CFR section 1020.30 Radiological Health—Performance Standards for
ISO 12891-1 Retrieval and Analysis of Surgical Implants—Part 1: Ionizing Radiation Emitting Products (FDA)

ASTM International takes no position respecting the validity of any patent rights asserted in connection with any item mentioned
in this standard. Users of this standard are expressly advised that determination of the validity of any such patent rights, and the risk
of infringement of such rights, are entirely their own responsibility.

This standard is subject to revision at any time by the responsible technical committee and must be reviewed every five years and
if not revised, either reapproved or withdrawn. Your comments are invited either for revision of this standard or for additional standards
and should be addressed to ASTM International Headquarters. Your comments will receive careful consideration at a meeting of the
responsible technical committee, which you may attend. If you feel that your comments have not received a fair hearing you should
make your views known to the ASTM Committee on Standards, at the address shown below.

This standard is copyrighted by ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959,
United States. Individual reprints (single or multiple copies) of this standard may be obtained by contacting ASTM at the above
address or at 610-832-9585 (phone), 610-832-9555 (fax), or service@[Link] (e-mail); or through the ASTM website
([Link]). Permission rights to photocopy the standard may also be secured from the Copyright Clearance Center, 222
Rosewood Drive, Danvers, MA 01923, Tel: (978) 646-2600; [Link]

Copyright by ASTM Int'l (all rights reserved) Sat Jul 12 [Link] UTC 2025
9
Downloaded/printed by
University of British Columbia Library pursuant to License Agreement. No further reproductions authorized.