While medical imaging data has traditionally been viewed on 2D displays, medical augmented
reality (AR) allows physicians to project the medical imaging data on patient's bodies.
An important application of medical AR is intra-operative surgical guidance by providing
the physician with the ability to “see through” the patient's skin and locate important
anatomy.
We present a medical AR application to support the retrosigmoid approach, an important
approach to access the internal auditory canal. During the retrosigmoid approach,
the craniotomy window is placed directly behind the sigmoid sinus, a large venous
blood drainage running inside the posterior cranial fossa adjacent to the skull. The
current standard is to use a surgical navigation system and anatomical surface landmarks
to guide the surgeon during the procedure. However, surgical navigation systems require
a long setup time and lack intuitiveness in presenting reformatted oblique planes
to the surgeon while surface landmarks lack anatomical accuracy. As a simple and accurate
alternative, we propose the use of an AR application that augments the surgeon's vision
to guide the targeting procedure.
In this work, two surgeons performed a retrosigmoid approach 14 times on 8 cadaver
heads. In each case the surgeon manually aligned a CT-derived virtual rendering of
the sigmoid sinus on the real cadaveric heads using a see-through AR display, allowing
the surgeon to plan and perform the craniotomy accordingly ([Fig. 1]). After the craniotomy, we performed repeat CT scans to measure the margins between
the osteotomy border and the dural sinuses to see how accurately the surgeon performed
the craniotomy with respect to its intended location.
We showed that the two surgeons had a mean margin of davg = 0.6 ± 4.7 mm and davg = 3.7 ± 2.3 mm over all their cases, respectively. For 12 of the 14 cases, the surgeons
only had positive margins, successfully avoiding the sigmoid sinus during surgery.
The relatively small and consistent margins found in this pilot study translate favorably
for what may be expected at the time of actual surgery. These findings suggest that
this approach has the potential to be a valuable tool to facilitate planning a variety
of similar skull-base procedures.
