Efficient application of 3D morphological operations in the framework of a computer-assisted diagnosis system

Pleural thickenings as biomarker of exposure to asbestos may evolve into malignant pleural mesothelioma. The diagnosis is based on a visual investigation of CT images, which is a time consuming and subjective procedure. Our image processing system identifies the pleural contours and detects pleural thickenings. In two algorithm steps, namely the detection of the thorax and the removal of air and trachea, 3D morphological operations were implemented and tested, in order to find an optimum of its application in the aspect of calculation expense and accuracy. After an evaluation, we obtained an experimental optimum for both algorithm steps.