A points-to-surfaces matching technique for the application of augmented reality during spine surgery

An augmented reality system has been developed and is currently under evaluation to document changes induced by surgical correction of a spine deformity. A pre-operative measurement technique is developed that permits the matching between a patient and a virtual 3-D medical image of normally hidden anatomy. The matching algorithm tested on a phantom gives a RMS error of respectively 1.1 mm and 2.0 mm with 1.0 and 2.0 mm random noises added to the measured points.