Ultrasound Augmented Laparoscopy

Abstract

Laparoscopic surgery is perhaps the most common minimally invasive procedure for many diseases in the abdomen and thorax. Since the laparoscopic camera provides only the surface view of the internal organs, in many procedures, surgeons use laparoscopic ultrasound (LUS) to visualize deep-seated surgical targets. Conventionally, the 2D LUS image is visualized in a display spatially separate from that which displays the laparoscopic video. Therefore, reasoning about the geometry of hidden targets requires mentally solving the spatial alignment and resolving the modality differences, which are cognitively very challenging. Moreover, the mental representation of hidden targets in space acquired through such cognitive mediation may be error prone, and cause incorrect actions to be performed. To remedy this, advanced visualization strategies are required where the US information is visualized in the context of the laparoscopic video. To realize such visualization schemes, efficient computational methods are required (i) to accurately align the US image coordinate system with that centred in the camera, (ii) to accurately represent 3D information of hidden targets from a series of 2D US images, and (iii) to blend the US information with the surface image provided by the camera such that the surgeons perceive the geometry of hidden targets accurately. In this thesis, a complete pipeline is described to visualize hidden targets, imaged by a LUS probe, in 3D in the context of the laparoscopic video. A novel method to register US images with a coordinate system centred in the camera is detailed with an experimental investigation into accuracy bounds in representing an imaged target in this coordinate system. This method eliminates the requirement for extrinsic tracking devices in the operating room (OR), significantly reducing both the financial and logistical overhead. An improved method to blend US information with the surface view provided by the camera is also presented with an experimental investigation into the accuracy of perception of the target locations in space. The work presented here, together with concurrent development in related fields, will enable image-guidance in laparoscopic soft-tissue surgery. The suggested improvements will