Design, fabrication, and testing of a new soft-pouch robot with 6 degrees of freedom to expand the reach of open and endonasal skull base approaches.

Objective: Most robots currently used in neurosurgery aid surgeons in placing spinal hardware and guiding electrodes and biopsy probes toward brain targets. These robots are inflexible, cannot turn corners, and exert excessive force when dissecting and retracting brain tissue, limiting their applicability in cranial base surgery. In this study, the authors present a novel soft-pouch robot prototype driven by compressed air and capable of gentle tissue manipulation. The robot is manufactured with technology developed by the authors, with multiple bidirectional bending points and a miniature camera running through the robot's central channel.

Methods: A soft, pneumatically driven pouch manipulator was created using a novel rapid and scalable system (integrated multilayer pouch robots with inkjet-patterned thin films). Made from 4 layers of thin, low-density polyethylene films, the manipulator has a thin deflated profile (152 µm) and contains 5 independent bidirectional joints with 50° range in each direction, as well as a wrapping end-effector. The robot carries a camera through its central channel. Four cadaveric models were used to demonstrate the robotic prototype being maneuvered inside different anatomical structures during simulated endonasal and posterior fossa approaches, with a manually positioned robot base and manually controlled air pressures.

Results: The robot is a pneumatically driven, soft-continuum manipulator with 12 control inputs and 6 independently controllable degrees of freedom. This design enables in-plane obstacle avoidance and orientation control. The robot is trapezoidal-shaped, with a total weight of 0.4 g, a 10-mm-wide distal end, and a length of 138 mm. The variable production cost (materials cost) of the manipulator is approximately $1. The manipulator is maneuvered to enter the maxillary sinus and through the endonasal corridor, demonstrating its potential use for anterior skull base approaches. It is also successfully maneuvered around the pons in a simulated retrosigmoid approach.

Conclusions: This robot offers a promising solution for safely maneuvering through narrow surgical windows encountered during skull base approaches. The multiple bending points of the robot, combined with its passive deformation capacity, allow it to turn around immovable structures, expanding the reach of surgical openings. The cost-effectiveness, rapid production, and scalability of the robot represent additional advantages.