|
|
||||||||||||
![]() |
|
|||||||||||
|
|
||||||||||||
|
|
||||||||||||
|
Anatomy Modeling in Less Time
Improved anatomy modeling tools let physicians share the workload—with their software
Radiation oncologists are in short supply in many parts of the world. In the
United States and other countries, they work long hours to keep up with the
growing number of cancer patients resulting from the aging of the baby boom
generation.
Sitting at a computer screen and tracing the targets and the organs at risk onto
CT images, slice by slice, can be time-consuming work for physicians. Often
taking hours for complex cases, it is the longest step in the radiation
treatment planning workflow. “Our radiation oncology residents often spend four or even six hours on target
definition and anatomy modeling of head and neck cases or other cases where the
contrast is low,” reports Andrew Lukban, chief physicist at New York’s SUNY Downstate Medical Center. “Anatomy modeling is the gating factor in our department productivity.”
To relieve the bottleneck and free up doctors’ time, dosimetrists or medical physicists in some departments may do preliminary
outlining of high-contrast structures, such as the liver, prostate, and skin,
before the radiation oncologist takes over. However, in 2003 the ASTRO
workforce committee reported a shortage of these radiation oncology
professionals as well, and many practices are feeling the impact of that
shortage today. Regardless of attempts to share the workload of anatomy
modeling, the heavy lifting of delineating targets and contouring low-contrast
structures, such as shadowy lymph nodes, remains the responsibility of very
busy radiation oncologists.
Improved anatomy modeling tools are needed to speed the treatment planning
workflow, increase department productivity, and free doctors to spend more time
with patients. “The largest problem is the lack of time for the oncologists,” says Ulf Isacsson, PhD, a medical physicist at Sweden’s Uppsala University Hospital. “So I should say that better tools for defining targets and organs at risk
quicker would be a solution that should help us a lot.” Uppsala has four linacs, a high-dose-rate (HDR) brachytherapy afterloader, and
a proton facility. Clinicians there treat approximately 1,200 patients a year
with radiation therapy, and they use Oncentra to plan both external beam and
brachytherapy treatments.
A “pearl” of a contouring tool
Lukban points to Oncentra Anatomy as a new-generation contouring solution. SUNY
Downstate uses the platform-independent Oncentra treatment planning system to
plan intensity-modulated radiation therapy (IMRT) and other external beam
treatments, as well as HDR brachytherapy. “We have a residency program, and our residents work at five hospitals in New
York City, where they get to experience different treatment planning systems.
It was our residents who were the main proponents for Oncentra Anatomy because
they found that it helped them to contour faster,” recalls Lukban. “We have been using the new Oncentra Anatomy tool for about seven months, and the
residents are doing modeling in half the time.”
To speed up contouring, Oncentra Anatomy provides a “pearl” tool, a round contouring cursor that is popular with users because it is so
easy to use. The boundary of the pearl can be expanded close to the size of the
organ being contoured. “If you hold the mouse button down and move the mouse, the edge of the ball will
trace around no matter where you go,” explains Lukban. “It allows you to fill in the entire structure very quickly. You can expand the
diameter of the pearl with the arrow keys and almost just wobble it to fill in
the boundaries of the structure instead of marking point by point.” This is just one of the many streamlined modeling features of Oncentra Anatomy,
which can be configured as a standalone module and connected to virtually any
DICOM-compliant treatment planning system.
Automation is key to speed and consistency
“It’s good to have more efficient drawing tools, but if you can get the job done
more automatically, you can make the whole process go faster,” says Lukban. That is the impetus behind model-based segmentation (MBS), a new
feature of Oncentra Anatomy. Oncentra MBS lets planners start with an organ
model and parameters selected from a library. Then the automatically generated
contours can be stretched asymmetrically in three dimensions with a single
mouse operation. “To correct the contours in three dimensions in this way is a much faster way
than we could do with previous tools,” says Kristina Nilsson, MD, radiation oncologist at Uppsala University Hospital.
“I have seen this demonstrated for the prostate, and I think it would be very
useful.”
Lukban agrees. “With model-based segmentation you can produce the contours automatically; then
all the physician needs to do is to review and make corrections. This would
save a lot of time for physicians.”
Lukban sees another advantage of model-based segmentation: consistency. Not all
radiation oncologists contour anatomy and targets in the same way. Some plot a
series of points and let the planning software connect the dots. Many prefer to
draw in the contours themselves because it’s faster. However, drawing with the computer mouse can easily create
inconsistencies and anomalies that confound the dose calculation engine.
Consistency is important in another way as well, points out Isacsson. “Consistency in anatomy contouring is desirable for clinical studies and for
following up patients,” he explains. “It is important to know that every patient has received consistent treatment.
While a tumor varies, other structures, such as lymph nodes, are defined in the
same way in every patient. To evaluate patients, you want to know that they
have been treated in the same way. Model-based segmentation is a step in the
right direction.”
Oncentra Anatomy, with advanced tools for contouring and model-based
segmentation, is now available. Radiation oncology centers worldwide use
Oncentra, the independent treatment planning system, to plan brachytherapy,
IMRT, IGRT, and proton beam treatments. For more information about Oncentra
innovations, contact your Nucletron representative or visit http://www.oncentra.info/.
|
|
|||||||||||
![]() |
|
|||||||||||
|
|
||||||||||||
|
Oncentra MBS
Model-based segmentation speeds contouring and improves plan consistency.
Oncentra MBS provides:
• Modeling of the organ surface
based on a library of organ contours and parameters
• Interactive tools for manual
adjustment of the 3D organ surface
• An expandable library of organ
models
• Integration with Oncentra
Anatomy |
|
|||||||||||
|
|
||||||||||||
![]() |
|
|||||||||||
|
|
|
|||||||||||
|
|
||||||||||||
|
|
||||||||||||
|
“The largest problem is the lack of time for the oncologists.”
Ulf Isacsson, PhD, of Sweden’s Uppsala University Hospital
|
|
|||||||||||
|
|
||||||||||||
![]() |
|
|||||||||||
|
|
||||||||||||
|
“To correct the contours in three dimensions in this way is a much faster way
than we could do with previous tools. I have seen this demon-strated for the
prostate, and I think it would be very useful.”
Kristina Nilsson, MD, of Uppsala University Hospital
|
|
|||||||||||
|
|
||||||||||||
|
|
||||||||||||
|
|
|
|
|
|
|
|
|
|
|
|
|
|