Correlation between CT based radiomics features and gene expression data in non-small cell lung cancer

Area of Science:

• Radiogenomics
• Computational Biology
• Medical Imaging Analysis

Background:

• Radiogenomics connects imaging phenotypes with gene expression.
• Non-small cell lung cancer (NSCLC) requires advanced diagnostic tools.

Purpose of the Study:

• To explore associations between CT radiomics features and gene expression in NSCLC.
• To evaluate the potential of radiomics in reflecting biological information for NSCLC.

Main Methods:

• Extracted and selected radiomics features from CT scans of 89 NSCLC patients.
• Clustered co-expressed genes into metagenes and performed statistical analysis.
• Built predictive models and conducted metagene enrichment analysis.

Main Results:

• Identified 187 significant correlations between CT radiomics features and NSCLC metagenes.
• Eighteen metagenes were annotated with Gene Ontology and KEGG terms.
• Achieved prediction accuracies for metagenes ranging from 41.89% to 89.93%.

Conclusions:

• CT radiomics features are associated with NSCLC co-expressed gene sets.
• Radiomics can reflect significant biological information in NSCLC.
• Findings support the clinical utility of CT radiomics for decision support in NSCLC.