Content based modified reaction-diffusion equation for modeling tumor growth of low grade glioma

摘要

This paper presents a content based modified reaction diffusion (RD) equation for modeling glioma growth. The reaction diffusion equation is modified by a weighted parameter that measures the white matter proportion in a small window. Given two MRI time-points scans of the same patient, the manually segmented tumor of the first scan is used as an initial seed to the proposed method while the second scan is used as the ground truth to measure the accuracy of the simulated results. For healthy tissues segmentation around the initial seed, spatial fuzzy C-means algorithm that accounts for neighborhood information of the image is used. As a proof of concept, the proposed method is tested on one low grade glioma case with 7 month difference between the two scans. The preliminary results of the modified RD equation show higher accuracy as compared with the standard RD equation.