Advances in artificial intelligence-based radiogenomics for lung cancer precision medicine

Area of Science:

• Oncology
• Radiology
• Bioinformatics

Background:

• Precision medicine in lung cancer requires integrating diverse data types.
• Radiogenomics offers a non-invasive method to link imaging phenotypes with genomic information.
• Artificial intelligence (AI) is crucial for analyzing complex radiogenomic data.

Purpose of the Study:

• To explore the role of AI-based radiogenomics in lung cancer precision medicine.
• To investigate the prediction of oncogenic driver mutations using radiogenomics.
• To assess the potential of radiogenomics in developing prognostic and predictive models for lung cancer.

Main Methods:

• Integration of medical imaging, genomics, and clinical data.
• Application of machine learning and deep learning techniques for mutation status prediction.
• Analysis of correlations between radiomic features and gene expression/clusters.

Main Results:

• AI radiogenomics enables non-invasive prediction of key oncogenic mutations (e.g., EGFR, KRAS).
• Significant correlations found between radiomic features and gene clusters, revealing tumor heterogeneity.
• Radiogenomics shows potential in predicting treatment response, recurrence, and survival.

Conclusions:

• AI-driven radiogenomics is a promising tool for personalized lung cancer management.
• Further research is needed to address challenges like standardization, validation, interpretability, and data diversity.
• Advancements in AI and multi-omics databases will enhance the clinical utility of radiogenomic models.