Fig. 3
From: Toward real-time margin assessment in breast-conserving surgery with hyperspectral imaging
Classification performance. (A) ROC curves of the classification algorithm to distinguish healthy tissue from tumor tissue within a margin of 0 (yellow) and 2 mm (blue) from the entire resection surface. The circular markers indicate the selected cut-off points for which the results are reported. (B-C) Confusion matrices based on the cut-off points on the ROC curves, showing the predicted and true number of patients with a negative (healthy) or positive (tumor) margin.
