FIGURE 7 from A Pilot Study on Patient-specific Computational Forecasting of Prostate Cancer Growth during Active Surveillance Using an Imaging-informed Biomechanistic Model

Summary of metrics assessing the performance of the biomechanistic model in the fitting-forecasting scenario. A and B, Unity plots to assess the agreement between the model estimation and mpMRI measurement of the tumor volume (VT) and the total tumor cell volume (VN) across the patient cohort (n = 7). These global tumor metrics were calculated at the dates of the second and third mpMRI scans (blue and green points). The unity plots in A and B further report the corresponding PCC and CCC for the global tumor metrics at the dates of the second and third mpMRI scans. C–F, Distributions of four local metrics assessing the agreement between the tumor cell density map calculated with our biomechanistic model and extracted from mpMRI data across the patient cohort in the global calibration scenario: Dice similarity coefficient (DSC), root mean squared error (RMSE), Pearson correlation coefficient (PCC), and concordance correlation coefficient (CCC). These local metrics were calculated at the dates of the second and third mpMRI scans (blue and green boxplots).