Robotic Assisted Cleft Palate Repair Using Novel 3 mm Tools: A Reachability and Collision Analysis.

Abstract

Cleft palate repair is technically challenging: visualization is reduced, delicate dissection is required, and instrument reachability is limited within the infant oral cavity using standard instruments. A 3 mm diameter pin-jointed wristed tool was developed that couples to a da Vinci Research Kit (dVRK) to perform infant cleft palate surgery. A reachability simulation analysis was performed to determine the number of collisions (tool-tool and tool-cavity) for tool tip positions within the oral cavity and the impact of dVRK trocar height. Tool reachability was measured using a reachability ratio of collision-free to total poses considered for a given oral cavity target point. Three physical experiments were performed using a high-fidelity cleft palate simulator: (1) the 3 mm tools were compared to existing 8 mm tools assessing visual field obstruction and the number of collisions during suturing tasks; (2) an end-to-end robotic assisted cleft palate repair was performed on the simulator using the new tool assessing collisions; and (3) the impact of trocar height on tool performance was investigated to validate the reachability simulation analysis. The 3 mm tools demonstrated fifteen times fewer tool-cavity collisions and 21% less visual obstruction than the 8 mm tools. Average tool-cavity collisions per throw demonstrated an increasing trend posteriorly within the oral cavity (nasal mucosa: $R^{2}=0.344$, $p=0.013$; oral mucosa: $R^{2}=0.811$, $p< 0.001$); and a 31% increase ($R^{2}=0.988$, $p=0.069$) with increasing trocar height. Physical experiments validated tool functionality for cleft repair and the reachability analysis reveals increasing collision risk deeper within the oral cavity.