Search

The real-time requirement means that the simulation should be able to follow the actions of the user that may be moving in the virtual environment. The computer system should also store in its memory a three-dimensional (3D) model of the virtual environment. In that case a real-time virtual reality system will update the 3D graphic visualization as the user moves, so that up-to-date visualization is always shown on the computer screen. Upon completion of the tele-operation, the surgeon compares the preoperative and postoperative images and models of the operative field, and studies video records of the procedure itself. Using intraoperative records, animated images of the real tele-procedure performed can be designed. Virtual surgery offers the possibility of preoperative planning in rhinology. The intraoperative use of computer in real time requires development of appropriate hardware and software to connect medical instrumentarium with the computer and to operate the computer by thus connected instrumentarium and sophisticated multimedia interfaces.

The main goal of our work was to evaluate advantages and disadvantages of virtual endoscopy (VE) techniques in routinely diagnostic and preoperative management of patients with various sinus diseases or head traumas in our practice. Fly-through algorithm was performed using an Xeon based workstation on data sets created from axial CT images acquired from 320 patients with various paranasal sinus disorders. Images were created using Siemens Somatom Emotion 16 continiously rotating helical CT scanner and archived in DICOM format. In comparison with real endoscopy, the VE has several advantages. It is completely non-invasive. It is possible to repeat the same procedure several times, therefore it may be a valuable tool for training. Interactive control of all virtual camera parameters, including the field- -of-view is possible. Endoscopic viewing as opposed to real endoscopy is not restricted to the spaces defined by inner surfaces. The viewer may penetrate the walls and see the extent of lesions within and beyond the wall as well as the adjacent anatomic structures. Virtual endoscopy also has a potential to stage tumors by determining the location and the extent of transmural extension.

Conduction of tele-3D-computer assisted operations as well as other telemedicine procedures often requires highest possible quality of transmitted medical images and video. Unfortunately, those data types are always associated with high telecommunication and storage costs that sometimes prevent more frequent usage of such procedures. We present a novel algorithm for lossless compression of medical images that is extremely helpful in reducing the telecommunication and storage costs. The algorithm models the image properties around the current, unknown pixel and adjusts itself to the local image region. The main contribution of this work is the enhancement of the well known approach of predictor blends through highly adaptive determination of blending context on a pixel-by-pixel basis using classification technique. We show that this approach is well suited for medical image data compression. Results obtained with the proposed compression method on medical images are very encouraging, beating several well known lossless compression methods. The predictor proposed can also be used in other image processing applications such as segmentation and extraction of image regions.