Convergence Stability of Depth-Depth-Matching-Based Steepest Descent Method in Simulated Liver Surgery

Abstract

We recently established that our digital potential function was globally stable at the point where a virtual liver coincided with its real counterpart. In particular, because three rotational degrees of freedom are frequently used in a surgical operation on a real liver, stability of the potential function concerning three rotational degrees of freedom was carefully verified in the laboratory, using fluorescent lamps and sunlight. We achieved the same stability for several simulated liver operations using a 3D printed viscoelastic liver in a surgical operating room equipped with two light-emitting diode shadowless lamps. As a result, with increasing number of lamps, stability of our depth-depth matching in the steepest descendent algorithm improved because the lamps did not emit an infrared spectrum such as the one emitted by our depth camera. Furthermore, the slower the angular velocity in a surgical sequence, the more overall stability improved.