Segmentation and estimation of the histological composition of the tumor mass in computed tomographic images of neuroblastoma

The problem that we investigate in the present paper is the improvement of the analysis of the primary tumor mass, in patients with advanced neuroblastoma, using X-ray computed tomography (CT) examinations. To achieve this goal, we propose a methodology for the estimation of the histological content of the mass that comprises a technique for semi-automatic segmentation of the primary tumor mass in CT images of neuroblastoma and a statistical method to estimate, from segmented CT images, the histological composition of the primary tumor. The results of the method are compared with the results of histological analysis of surgically resected tumor mass.