A new approach for improved tumor volumetry

Abstract

The tumor volume is an important parameter in the process of tumor diagnostics and therapy. In the ICRU 50 Report [1] the gross tumor volume (GTV) is defined as the visible or palpable volume of a tumor. In practice the GTV is calculated from the delineated tumor contours in CT or MRT image data sets. We showed in a previous study [2] that there are semiautomatic segmentation algorithms, which provide an acceptable accuracy for the determination of tumor volumes. The process of delineation incorporates several error sources, which prevent the exact calculation of the GTV. On the one hand, the result of a tumor volumetry is dependent on the user, who carries out the delineation of the tumor. On the other hand there are several quality factors of the image data (like image noise or tumor contrast), which influence the accuracy of the determination process. The neglect of this error can lead to misinterpretation and decisions based on wrong information. Since the error of the determination cannot be avoided, we developed a new approach to determine the accuracy of a tumor volumetry dependent on different influencing parameters.