Mixed Reality Improves Point Accuracy in Electrophysiology Procedures

The SentiAR engine uses a mixed reality head-mounted display (Microsoft HoloLens) for the display of real time digital stereoscopic 3-dimensional electroanatomic models of patient-specific geometry with catheter locations within the body. The purpose of this study is to describe the first-in-human usability and accuracy assessment using the SentiAR system compared to a traditional electroanatomic mapping system (EAMS; EnSite Velocity).

Pediatric patients scheduled for electrophysiology (EP) studies were enrolled in the study. During the procedural waiting phase, physicians were asked to perform a series of time-limited tasks under 2 conditions: (1) EAMS, (2) SentiAR. Tasks included: (1) creation of a cardiac chamber (maximum time 5 minutes) and (2) point-by-point navigation (maximum time 60 seconds/point). During point-by-point navigation, a series of 5 spatially disparate points were placed within the geometry. Physicians were asked to sequentially navigate to each point with markers placed at the successful site of navigation defined by the physician. Target points were defined as ground truth. Coordinate x,y,z locations for target and navigation points were taken from the EAMS and used to calculate the Euclidian distance. Area, volume and sphericity were calculated.

Nine patients were enrolled. There was no difference in time to geometry creation between the conditions (p=0.1) or time in point by point navigation (p=0.9). When comparing geometries, there was no significant difference in area (p=1), volume (p=0.8) or sphericity (p=0.5) between the models. When using the EAMS, the average distance of marker lesions from the intended target point was 4.6±4.1mm; when using the SentiAR system, the average distance from the intended target point was 2.9±1.8mm (p=0.006).

Use of a mixed reality system, providing the physician with stereoscopic visualization as well as control of model views allowed for more point accurate navigation when compared to navigation using EAMS.