Modern approaches to Minimally Invasive Surgery (MIS) such as Single Port Surgery (SLS), laparoscopic surgery, Natural Orifice Trans-luminal Endoscopic Surgery (NOTES), are associated with significant limitations (lack of tactile information, loss of depth perception and complex hand-eye coordination), imposing considerable challenges to the surgeon’s performance. Hence, the requirement for human-efficient and intuitive surgical support systems to provide more robust and predictable MIS becomes apparent. The prerequisite for modelling the surgical environment is reliable tissue deformation recovery and accurate camera pose estimation to enable the recovery of 3D anatomical structures. Direct application of the commonly used vision techniques for pose estimation during MIS has significant problems due to the paucity of reliable salient features to track coupled with changing visual appearance of the surgical environment and what is often a narrow baseline during surgical navigation.

In this presentation, a robust framework for intra-operative free-form deformation recovery based on SFM has been proposed. Unlike previous approaches, the proposed framework does not impose explicit constraints on tissue deformation, allowing realistic free-form deformation recovery. A novel adaptive Unscented Kalman Filter (UKF) parameterization scheme is proposed to fuse vision information with data from an Inertial Measurement Unit (IMU). The method is built on a compact scene representation scheme suitable for both surgical episode identification and instrument-tissue motion modelling. Results derived from validation on synthetic and phantom data demonstrate the intrinsic accuracy achievable and the potential clinical value of the technique.