Towards the Design of a Patient-Specific Virtual Tumour

Figure 6

Tumour evolution (case M4-L). (a) The evolution of the tumour fluorescence distribution is assessed over 4 time points (days 3, 7, 14, and 28). The horizontal axis represents the fluorescence intensity of the images from dark pixels (with value 0) to bright pixels (with value 1). The range of pixel intensity values has been normalized for each image between the minimum value (darker pixel equals 0) and the maximum value (brighter pixel equals 1) since they are not necessarily the same from one image to another. The vertical axis stands for the number of pixels for each fluorescence intensity normalized by the total number of tumour pixels (the integral of each curve equals 1). The close-up compares day 3 with day 28 to exhibit the switch in the tumour fluorescence profile. (b) The evolution of the tumour area is first evaluated from the experimental images (green curves with bullets) using both a manual Yen-guided filter (plain line) and the Default ImageJ filter (dotted line). The curves are then compared with the areas measured on the corresponding virtual tumour (blue curves with squares), where the effective area is the area effectively occupied by the tumour cells (plain line) and the apparent area is the area which is delineated by the tumour’s edge (dotted line).

(a)

(b)