Int J CARS (2012) 7:433–444
DOI 10.1007/s11548-011-0634-9
ORIGINAL ARTICLE
Volume slicing of cone-beam computed tomography images
for navigation of percutaneous scaphoid fixation
Erin J. Smith · Hisham A. Al-Sanawi ·
Braden Gammon · Paul J. St. John ·
David R. Pichora · Randy E. Ellis
Received: 10 January 2011 / Accepted: 7 June 2011 / Published online: 25 June 2011
© CARS 2011
Abstract
Purpose Percutaneous scaphoid fixation (PSF) is growing in
popularity as a treatment option for non-displaced fractures.
Success of this procedure demands high-precision screw
placement, which can be difficult to achieve with standard
2D imaging. This study aimed to develop and test a system
for computer-assisted navigation using volume slicing of 3D
cone-beam computed tomography (CBCT).
Methods The navigated technique involved a distinctive
workflow in which a 3D CBCT imager was calibrated preop-
eratively, circumventing the need for intraoperative patient-
based registration. Intraoperatively, a 3D CBCT image was
acquired for both preoperative planning and direct naviga-
tion using volume-rendered slices. An in vitro study was
conducted to compare the navigated approach to two con-
ventional fluoroscopic methods for volar PSF. The surgical
goal was to insert a guide wire to maximize both length and
central placement.
Results There was no significant difference in the mean
central placement of guide wire, although the variance in cen-
tral placement was significantly lower using VS navigation
E. J. Smith · R. E. Ellis
Department of Mechanical Engineering, Queen’s University,
Kingston, ON, Canada
e-mail: smith@me.queensu.ca
H. A. Al-Sanawi · B. Gammon · D. R. Pichora · R. E. Ellis
Department of Surgery, Queen’s University, Kingston,
ON, Canada
P. J. St. John · D. R. Pichora · R. E. Ellis
Human Mobility Research Centre, Queen’s University,
Kingston, ON, Canada
R. E. Ellis (B )
School of Computing, Queen’s University, Kingston, ON, Canada
e-mail: ellis@cs.queensu.ca
(P < 0.01). The lengths of the drill paths were significantly
longer for the VS-navigated group compared with one 2D
group (P < 0.1). Each navigated trial required only one
drilling attempt and resulted in less radiation exposure than
conventional C-arm (P < 0.01).
Conclusions Volume-sliced navigation achieved a more
repeatable and reliable central pin placement, with fewer
drilling attempts than conventional 2D techniques. Volume-
sliced navigation had a higher number of drill paths within
the optimal zone maximizing both length of the path and
depth from the surface.
Keywords Computer-aided surgery · Preoperative
planning · Navigation · Volume slicing · Percutaneous
fixation · Scaphoid fractures
Introduction
The scaphoid bone is the most commonly fractured of the
eight small carpal bones in the human wrist, usually the
result of a fall on an outstretched arm [1]. Although conven-
tional casting of a non-displaced scaphoid for approximately
10–12 weeks is usually an effective treatment option, cast-
ing can be poorly tolerated by active individuals, particularly
those dependent on their upper extremities for work or sports.
Furthermore, delayed unions and malunions following cast
immobilization are not uncommon, even with minimally dis-
placed or proximal pole fractures, and may lead to pain and
loss of function at the wrist. For these reasons, there is a grow-
ing popularity in offering internal fixation as a treatment for
minimally displaced and non-displaced fractures [2].
The scaphoid bone is tiny, measuring on average 26.0mm
in length and as little as 7.2 mm at its narrowest point, the
waist [3], so placement of standard fixation screw does not
leave much tolerance for error. Because of the complex
123