Scholarlink

Purpose <p>This study aims to integrate CT (Computed Tomography) radiomic features, gene expression profiles, and clinical data to identify radiogenomic biomarkers and improve overall survival prediction in gastric cancer (GC) patients.</p> <p>Procedures.</p> <p>Quantitative radiomic analysis was performed on 37 GC CT images, alongside gene expression and clinical data, to identify biomarkers associated with overall survival. Tumor segmentation and radiomic feature extraction were followed by Pearson correlation for feature selection. Gene Set Enrichment Analysis (GSEA) identified pathways linking gene expression changes with radiomic features. Regression models were applied to explore the relationships between these pathways, radiomic features, and clinical data in survival prediction.</p> Results <p>A total of 107 radiomic features were extracted, with 46 radiomic features, 1,032 genes, and one clinical feature (age) selected for further analysis. GSEA identified 29 significant KEGG pathways, mainly involving immune, signal transduction, and catabolism pathways. In survival analysis, the SVM model performed best, identifying age, genes CSF1R and CXCL12, and image features ShortRunHighGrayLevelEmphasis and Idn (Inverse Difference Normalized) as independent predictors.</p> Conclusion <p>This study highlights the potential of integrating imaging, genomics, and clinical data for prognosis in GC patients, with identified genes suggesting new radiogenomic biomarker candidates for future evaluation.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Radiogenomic Profiling for Survival Analysis in Gastric Cancer: Integrating CT Imaging, Gene Expression, and Clinical Data

  • 期刊论文

  • 发表日期：

    2025年5月15日

查看全文
  • Anju R. Nath,
  • Kiruthika Thenmozhi,
  • Jeyakumar Natarajan

摘要

Purpose

This study aims to integrate CT (Computed Tomography) radiomic features, gene expression profiles, and clinical data to identify radiogenomic biomarkers and improve overall survival prediction in gastric cancer (GC) patients.

Procedures.

Quantitative radiomic analysis was performed on 37 GC CT images, alongside gene expression and clinical data, to identify biomarkers associated with overall survival. Tumor segmentation and radiomic feature extraction were followed by Pearson correlation for feature selection. Gene Set Enrichment Analysis (GSEA) identified pathways linking gene expression changes with radiomic features. Regression models were applied to explore the relationships between these pathways, radiomic features, and clinical data in survival prediction.

Results

A total of 107 radiomic features were extracted, with 46 radiomic features, 1,032 genes, and one clinical feature (age) selected for further analysis. GSEA identified 29 significant KEGG pathways, mainly involving immune, signal transduction, and catabolism pathways. In survival analysis, the SVM model performed best, identifying age, genes CSF1R and CXCL12, and image features ShortRunHighGrayLevelEmphasis and Idn (Inverse Difference Normalized) as independent predictors.

Conclusion

This study highlights the potential of integrating imaging, genomics, and clinical data for prognosis in GC patients, with identified genes suggesting new radiogenomic biomarker candidates for future evaluation.