Fig. 2
Predicted response to radiation therapy for a representative animal receiving 20 Gy. Results are shown over the entire tumor volume for a representative animal receiving a single fraction of 20 Gy. The measured and model predicted \( {\phi}_T\left(\overline{x},t\right) \) and \( {\phi}_V\left(\overline{x},t\right) \) are shown for the RT1 model at t6 and t7, respectively. The fused image depicts areas where there exists perfect agreement between the model and measurement (white areas), where the model exists where the measurement does not (green areas), and where the measurement exists and the model does not (pink areas). The blue contours represent the prediction of the RT2 model. The distributions of \( {\phi}_T\left(\overline{x},t\right) \) and \( {\phi}_V\left(\overline{x},t\right) \) are normalized to the maximum value of \( {\phi}_T\left(\overline{x},t\right) \) and \( {\phi}_V\left(\overline{x},t\right) \) observed in the measurement, and thus share a colorbar ranging from 0 to 1. This particular animal has a relatively spatially homogenous distribution of tumor cells. That is, it appears that no necrosis or low cell density regions have developed despite, poorly vascularized regions observed on the measured \( {\phi}_V\left(\overline{x},t\right) \). The RT2 model overestimates tumor growth most noticeably in slices 1, 5, 6, and 7. Visually, the RT1 model resulted in accurate predictions of low and high \( {\phi}_V\left(\overline{x},t\right) \) regions at both t6 and t7. The highest level agreement (white areas) was observed for the high \( {\phi}_V\left(\overline{x},t\right) \) region near the brain-skull boundary. Less agreement regions were observed away from the brain-skull boundary