A New Simulation Technique

1
A New Simulation Technique Using Virtual
Reality Visualization to Optimize Beam Geometry
in Prostate Cancer IMRT
Tim R. Williams1, Charles Y. Shang1, Andrew
Beavis2, James Ward3, Roger Phillips3, Colin
Sims4 (1) Lynn Cancer Institute, Boca Raton
Community Hospital, Boca Raton, FL, (2) Princess
Royal Hospital, Hull, GB, (3) University of
Hull, Hull, GB, (4) Computerized Med Sys, Inc.,
St Louis, MO ASTRO 2006
2
Purpose Introduction
In prostate cancer IMRT the use of non-coplanar
beam orientations, while reported1, has been
limited. The simulation process as currently
practiced is suited for non-reciprocal, coplanar
beam geometries, but using the z axis is
difficult due to geometric limitations. We report
a new method of simulation using a virtual
reality simulator (VRS), in which the physician
views the patient as if he were in the treatment
position, on the treatment table. The
presentation platform is an advanced technology
LCD screen. The physician actually sees depth in
the LCD screen, without goggles or other user
mounted devices (fig.1). With essential functions
such as anti-collision (fig.2), real time beams
eye view (fig.3), and image/contour enhancement
functions (fig.4), the VRS has demonstrates its
unique potential.
3
Materials Methods
We randomly selected twelve prostate cancer
patients currently receiving intensity modulated
radiation therapy (IMRT) for VRS (RTStar,
provided by U. of Hull, UK). The standard beam
geometry for IMRT consists of seven coplanar,
non-reciprocal ports. Their initial plans for 45
Gy were then compared to test IMRT plans with
beam geometries derived from the VRS. The
specific geometry varied from patient to patient,
but generally involved two anterior oblique beams
angled inferiorly and four oblique lateral beams
tilted more anteriorly, depending on anatomy of
the patient. Comparable IMRT plans were
calculated and compared for PTV dose homogeneity,
rectal D10cc , bladder D30cc and their mean doses.
4
Results Discussion
Tables 1-3 display the dose comparisons and the
test results. The seven-field prostate IMRT
setups used at most centers is a seven
evenly-angled beam arrangement along the axial
plane, in which the BEV of most posterior oblique
port often reveals more than 75-100 of the
rectum overlapping with PTV (fig.5, left). In
VRS, non-coplanar settings are an option to
separate adjacent beams. We created a pair of
non-coplanar oblique beams by tilting the couch,
depending on the patient by 30-40º and rotating
the gantry laterally to 240º (fig.6).
Consequently, all other beams were rotated more
anteriorly in comparison with those in a typical
coplanar setting. Thus, the reductions in organ
volume inclusion displayed with the Beams Eye
View (BEV) were evident - for bladder from the
inferiorly tilted beams and for rectum from
posterior oblique beams. The result was improved
dosimetry.
5
CONCLUSION
Virtual reality simulation, with its enhanced
presentation of 3D visualization and
anti-collision functionality, benefited the
process of optimizing the beams arrangement in
prostate IMRT. A deliverable, non-coplanar
treatment plan, improved dose homogeneity of PTV,
dose sparing to the bladder, and reduced high
rectal dose were all seen in the VRS plans for
prostate IMRT. Further study will include other
anatomic sites which require more complex IMRT
setups and unique 3D dose evaluation with virtual
reality technology (fig. 7).
6
Table 1. PTV Dose Homogeneities
A significant dose reduction in both Dmax and D5
of PTV was observed in the test group, indicating
a better dose homogeneity.
7
Table 2. Bladder Dose Comparison
Bladder Dmean and D30cc show a significant
reduction in the test group, implying a lower
risk of GU side effect.
8
Table 3. Rectal Dose Comparison
Both rectal Dmean and D10cc does show a
significant reduction in the test group.
9
Figure 1. VRS 3D Displays
Display Layout of Virtual Reality Simulator
A non-goggles based Virtual Reality Simulation
system operated with RTStar consists of a 3D
stereo-scopic LCD monitor for virtual treatment
room view (left) and a 2D display for beam eye
view and menus bars (right).
10
Figure 2. VRS Anti-collision Function
An anti-collision algorithm is embedded into the
VRS that forbids undeliverable beam geometries.
11
Figure 3. Beam Eye View in VRS
PTV Bladder Rectum
Real Time Beam Eye View
Conjunction with corresponding beams eye view,
the VRS allows assessment of the juxtaposition of
the prostate PTV and critical normal tissues.
12
Figure 4. Image/Contour Enhancement
Selective Color and Transparency for Contour
Displays
CT Images Overlap
PTV Bladder Rectum
Image and contour enhancement function in virtual
reality simulation allows more innovation in
clinical applications .
13
Figure 5. Two Geometries in RPO Port
RPO in Control Group
RPO in Test Group
PTV Bladder Rectum
RPO port geometry is modified during VRS to
minimize the overlap of PTV and rectum.
14
Figure 6. Non-Coplanar Beam Setting
The new non-coplanar 7 port setting includes a
pair of inferiorly tilted anterior oblique beams,
which decreases the bladder volume receiving
high exposure.
15
Figure 7. VR 3D Dose Evaluation
16
ACKNOWLEDGEMENT The author would like to express
special gratitude to Boca Raton Community
Hospital (who funded the VRS project), all of the
computer scientists at the University of Hull who
contributed to the RTStar program, CMS, Inc., and
the physics team at the Lynn Regional Cancer
Center, and particularly Dr. Shang, for their
invaluable contribution to this study.
References 1. Price Jr RA, Hanks GE,
McNeeley SW, Horwitz EM, Pinover WH Advantages
of using noncoplanar vs. axial beam arrangements
when treating prostate cancer with
intensity-modulated radiation therapy and the
step-and-shoot delivery method Int J Rad Onc
Biol Phys 2002, Vol. 53236-243