Fig. 5: Dot plots of DVH endpoint comparison between the ATP outputs and predicted doses across different disease sites and institutions. | Nature Communications

Fig. 5: Dot plots of DVH endpoint comparison between the ATP outputs and predicted doses across different disease sites and institutions.

From: Multicenter study on the versatility and adoption of AI-driven automated radiotherapy planning across cancer types

Fig. 5

For each site, n = 30, 10, 10 cases at Institution A, B, C, respectively. The dots and shaded bars in a–e represent the average and 95% CI bounds of the metric difference between the final plans and predicted doses in each institution. Significant differences (p < 0.05) are shown in red (lower than prediction) or blue (higher than prediction), while data points of no significance (p ≥ 0.05) are denoted in black. The p values have been corrected to account for multiple comparisons across numerous dosimetric endpoints. Source data are provided as a Source data file.

Back to article page