Skip to main content
Fig. 4 | Cancer Imaging

Fig. 4

From: Interpretability of radiomics models is improved when using feature group selection strategies for predicting molecular and clinical targets in clear-cell renal cell carcinoma: insights from the TRACERx Renal study

Fig. 4

Violin plots showing the distribution in AUROCs across repeated cross-validation obtained for different targets with different pipelines (Conventional and Proposed) and sub-segmentation feature sets (whole, whole/high/low, whole/core/rim). The horizontal lines are the mean AUROCs and in almost all cases the difference in the mean value between the Conventional and Proposed pipelines is small compared to the variation across the cross-validation splits

Back to article page