Investigation of an Augmented Reality based Neurosurgical Navigation System for Preoperative Learning

Description

Augmented reality (AR) has become a valuable intraoperative aid for surgical operations. Traditional navigation systems provide a two-dimensional (2-D) observation resulting in a lack of total spatial information for the operating field. Moreover, the current navigation system is often separated from the sterile operating area, demanding the operating neurosurgeons to perform tiresome intermittent switching of his/her field of vision between the 2-D representations of the brain scans on a computer screen and the surgical field, i.e. the patient’s brain. In this project, we aim to develop a 3-D display of the brain model and a tracking system for preoperative learning. We also propose to enhance the accuracy of the registration by adopting an unscented Kalman filter and deep learning method, and implement the system on a mobile device.