Introduction - Microsurgery: Advances, Simulations, and Applications - page 13

1.2.4 Navigation Systems
As the treatment targets of minimally of MIS may be reached through
a natural hole with a trans-luminal approach, inserting a needle
to a precise location, or through small incisions in the body, it is
desirable to use preoperative data or intra-surgery imaging systems
information to guide the physicians and reach the target faster, easier
and safer. An augmented reality environment allowing navigation is
obtained using a computer generated three-dimensional image of
the target obtained from medical imaging systems. After registration
of this image with landmarks in the patient body; this image is
superimposed to intra-surgical images to provide guidance. As the
intra-surgical images are obtained from a variable location, the
superimposed image must change according to the relative motion
of camera that provides the intra-surgical image and the landmarks
on the patient's body.
In laparoscopic surgery, the motion capture of the laparoscope
is done using optical trackers [42], magnetic trackers, or a hybrid
system combining them [12, 43]. The target image generation is
done conventionally using preoperative data of the target such as a
tomogram or magnetic resonance scan and kinematics calculations.
Optical trackers have the property to be the most accurate motion
capture sensors, but they have the problem that visual contact is
needed between the object to be tracked and the sensor. During the
surgery, occlusion may occur due to the constant activity of surgeons
around the patient, which makes it dificult to have a continuous
motion capture of all the landmarks during the surgery. Those
trackers need at least three landmarks to be attached to the object to
be tracked.
Magnetic trackers, however, offer a constant tracking even when
there are non-ferrous objects between the transmitter and the
receiver. But those trackers have a distorted output, due to ferrous
objects in the working space and intrinsic properties of the ield
generated by their transmitter. Such sensors are also affected by
oscillating magnetic ields such as produced by cathode ray tube
monitors. This distortion is a barrier for the use of preoperative
data with magnetic trackers, because it needs to be corrected in
order to have both models in a non-distorted frame of reference
[44-45]. Hybrid systems, such as the one discussed in [42], correct
the distortion of the magnetic trackers using an optical tracker and
numerical methods.