Skip to main content
Fig. 1 | Cancer Imaging

Fig. 1

From: Preoperative CT-based radiomic prognostic index to predict the benefit of postoperative radiotherapy in patients with non-small cell lung cancer: a multicenter study

Fig. 1

Comprehensive outline and pipeline of the study.The initial step encompassed the identification and annotation of the primary nodule evident in the CT scan. Pyradiomics was employed to extract intra-tumoural textural features. Using the LASSO-Cox method, top features were selected and applied in the construction of RPI. The LRPI was constructed utilising prognostic clinical features alongside RPI. The prognostic performance of RPI and the predictive potential of LRPI for the benefit of PORT were subsequently validated. Furthermore, the relationships between RPI features and the tumour-stroma ratio, as seen on whole slide imaging, were evaluated. Correlations between RPI and mRNA data were also explored to investigate the underlying biological pathways. LASSO, least absolute shrinkage and selection operator; RPI, radiomic prognostic index; LRPI, lymph-radiomic prognostic index; PORT, postoperative radiotherapy

Back to article page

Cancer Imaging

ISSN: 1470-7330

Contact us