Starting with a mathematical algorithm they had previously developed to help surgeons locate specific vertebrae during spine surgery, the team adapted the method to the task of surgical navigation. When they tested the method on cadavers, they found a level of accuracy better than 2 millimeters and consistently better than a conventional surgical tracker, which has 2 to 4 millimeters of accuracy in surgical settings.
"The breakthrough came when we discovered how much geometric information could be extracted from just one or two X-ray images of the patient," says Ali Uneri, a graduate student in the Department of Computer Science in the Johns Hopkins University Whiting School of Engineering. "From just a single frame, we achieved better than 3 millimeters of accuracy, and with two frames acquired with a small angular separation, we could provide surgical navigation more accurately than a conventional tracker."
The team investigated how small the angle between the two images could be without compromising accuracy and found that as little as 15 degrees was sufficient to provide better than 2 millimeters of accuracy.
An additional advantage of the system, Uneri says, is that the two X-ray images can be acquired at extremely low dose of radiation far below what is needed for a visually clear image, but enough for the algorithm to extract accurate geometric information.
The team is translating the method to a system suitable for clinical studies. While the system could potentially be used in a wide range of procedures, Siewerdsen expects it to be most useful in minimally invasive surgeries, such as spinal and intracranial neurosurgery.
A. Jay Khanna, M.D., an associate professor of orthopaedic surgery and biomedical engineering at the Johns Hopkins University School of Medicine, evaluated the system in its first application to spine surgery. "Accurate surgical navigation is essential to hi
|Contact: Catherine Kolf|
Johns Hopkins Medicine