A method for controlled diagnosis-oriented enhancement of selected regions of interest in intraoral radiographs is presented. Image enhancement is accomplished by adaptive non-linear grey scale transformation depending on the result of objective quality measurement. In order to assure reliable image duality measurement as well as controlled image enhancement, automatic image segmentation is applied to avoid the influence of disturbing factors (e.g. metallic restorations) on quality measurement and image enhancement. Based on existing a-priori knowledge about object structure and composition of the selected regions of interest in intraoral radiographs, different image segmentation algorithms and image enhancement procedures were developed for different types of diagnosis. (C) 1998 Elsevier Science Ireland Ltd. All rights reserved.

A system for three-dimensional oral implant treatment planning is presented. Virtual reality technologies are used in order to improve the human image interpretation and planning performance. The methods described are based on computer tomography (CT) data of the mandible and of the maxilla. A novel approach to volume rendering and voxel based modelling of implants is introduced which allows interactive three-dimensional manipulation of the anatomic model and real-time manipulation of virtual implants. A spline-based reconstruction method is described to assess the implant site in a clinically oriented view with regard to bone structures and angulation. Two parameters are deduced which represent the bone properties at the surface of implants. While an implant is navigated with six degrees of freedom, these parameters are acoustically rendered which is a novel approach to exploration of spatial bone properties in a CT data set.