Skip to main content
. 2019 Nov 22;9:1250. doi: 10.3389/fonc.2019.01250

Figure 9.

Figure 9

Decision curves for radiomics models in training (A), validation (B), and testing sets (C). The grey reference line indicates the assumption that all patients are diagnosed to be MSI CRC; black dashed reference line indicates the assumption that all patients are diagnosed to be MSS CRC. In training, validation, testing sets, the net benefit is 17.44, 15.40, and 13.43, respectively. A larger area under the decision curve indicates a better clinical utility. MSI, microsatellite instability; CRC, colorectal cancer; MSS, microsatellite stability.