Skip to main content
Fig. 4 | Biomarker Research

Fig. 4

From: Exploring non-invasive precision treatment in non-small cell lung cancer patients through deep learning radiomics across imaging features and molecular phenotypes

Fig. 4

Correlation and significance analysis of different imaging features. a. Correlation heatmap of deep network features and radiomic score (RadScore) features for various molecular alterations. b. Distribution of the contributions of deep network features and RadScore features in predicting various molecular alterations. ‘Tumor’ represents the RadScore for the tumor region and ‘Peritumoral’ for the peritumoral region

Back to article page

Biomarker Research

ISSN: 2050-7771

Contact us