Abstract Archives of the RSNA, 2007
SSJ16-03
Co-visualization of Stereoscopic and Monoscopic Images for Instrument Navigation in the Interventional Room
Scientific Papers
Presented on November 27, 2007
Presented as part of SSJ16: Physics (Visualization, Displays, PACS)
Markus Kukuk PhD, Presenter: Employee, Siemens AG
Jarrett Rosenberg PhD, Abstract Co-Author: Nothing to Disclose
Sandy Napel PhD, Abstract Co-Author: Medical Advisory Board, Fovia, Inc
Medical Advisory Board, Vital Images, Inc
Stockholder, Hologic, Inc
Stockholder, General Electric Company
Procedures requiring endovascular instrument navigation do not yet benefit from stereo visualization, made possible by rotational acquisitions or C-arm CT imaging. The live instrument is imaged using monoscopic fluoroscopy and the conventional method of combining the stereoscopic vessel image with the monoscopic instrument image often places the vessel segment containing the instrument's tip at the wrong depth, creating depth confusion. We present a method that overcomes this
problem by tracking the instrument's tip and adjusting the stereo disparity of the vascular images such that the portion of the vessel containing the tip always appears at the same depth as the instrument's tip. We conducted a pilot study to evaluate subject's preference for either method.
We randomly presented 28 subjects (radiologists, 3D technologists, others) with 30 pairs of stereo images (anaglyph technique), representing the two methods side-by-side, with the simulated instrument in different vessel segments. Subjects responded to the question, "In which image is the tip of the instrument better depicted as being inside the vessel?" on a 7-point scale, where 0 indicated no preference. We correlated subjects' responses with the degree of depth adjustment made in our method, image contrast, and other factors.
Subjects had a preference only 56% of the time at depth adjustments <4 units, but 80% of the time at greater depths (p<.001). When they had a preference, they chose our method 75% of the time (95% CI: 72-79%). Subject consistency varied: 19 of 28 chose our method from 75-96% of the time, 6 chose it from 42-67% of the time, and 3 chose it from 16-36% of the time.
As expected, at small depth adjustments subjects did not prefer one method over the other. For larger differences in depth between the instrument's tip and the containing vessel, subjects typically had a preference and chose our method 75% of the time. We conclude that adjusting the depth of the vessel tree to the depth of the instrument is a promising technique for avoiding the limitations of the conventional method.
Stereoscopic roadmapping applications may greatly facilitate instrument navigation during complicated cases.
Kukuk, M,
Rosenberg, J,
Napel, S,
Co-visualization of Stereoscopic and Monoscopic Images for Instrument Navigation in the Interventional Room. Radiological Society of North America 2007 Scientific Assembly and Annual Meeting, November 25 - November 30, 2007 ,Chicago IL.
http://archive.rsna.org/2007/5009630.html