Our project is part of the "Industry on Campus" which is embedded in the Inter-University Center for Medical Technologies Stuttgart - Tuebingen. In this project we aim to develop a system to simplify the orientation of a surgeon during a minimally invasive procedure, especially in laparoscopic abdominal surgery. The orientation during such a surgery is difficult, because of various reasons. The three most important problems are an intraoperativly altered environment, the dependence on images without depth information and the lack of fixed markers, like bones, in the laparoscopic images. This it is not only a problem with respect to orientation, it is also problematic to combine preoperative data, like CT-data, with the actual surgery data. A navigation system which solves these issues will not only speed up the surgery it will also enhance the postoperative results and patient safety.
Today's tracking systems have the disadvantage of not being able to reference the laparoscopic images to a fixed coordinate system. In addition to this, a fusion of the intraoperative data with the preoperative data is necessary in order to guide the surgeon. Because of the time difference between CT and laparoscopy and different location of the patient this fusion is not straightforward. In this project we will try to solve this by using the absolute position of the tool and an inertial navigation system.
Fig. 1: Endoscope system provided by Aesculap AG
Extraction of visual features for localisation
Fig. 2: A laparoscopy image with marked features.
Currently we are focusing on the detection of features in laparoscopic images. Besides the realtime requirement of the detection, the features need to be robustly found in an environment which is very self similar and very fast changing.
The distance between the tracked features and the camera has to be determined as a precursor to image data fusion
Fusion of image data
The features of the laparoscopic image have to be fused with data from CT.
Aesculap AG University of Stuttgart, Institute for System Dynamics University of Stuttgart, Institute of Applied Optics University Department of Urology Tübingen
|This work is funded by the Baden-Wuerttemberg Ministry of Science and Arts in the context of the founding program "Industry on Campus" in the Inter-University Center for Medical Technologies Stuttgart-Tuebingen.|
C. Robert Pech, Phone: (+49/0) 7071 / 29 70436, Email: robert.pech at uni-tuebingen.de