Adaptive threshold method based on PET measured lesion-to-background ratio for the estimation of Metabolic Target Volume from F-18-FDG PET images

Most advanced radiotherapy planning can be optimized by incorporating the metabolic image of the tumour to be irradiated (Metabolic Target Volume). MTV can be properly obtained by in vivo molecular imaging studies as the volume of metabolically active tumour component. Positron Emission Tomography (PET) can provide MTV by in vivo measuring the glucose metabolism of a tumour. Aim of this work was to develop an innovative PET image segmentation method, based on an adaptive threshold of the image signal-to-noise ratio, for the estimation of the MTV. The proposed method was tested on a set of anthropomorphic phantoms proving its feasibility, operator-independency and accuracy in a clinical setting.