Answers are grounded in this paper and cite their sources.
A literature review of Russian and international studies on radiomics and radiogenomics in lung cancer was conducted, focusing on computed tomography and positron emission tomography imaging. Thе analyzed methodological stages included image acquisition, tumor segmentation, feature extraction, machine learning and artificial intelligence applications, as well as model validation. Radiogenomic models demonstrate varying predictive performance across different mutations. The highest accuracy has been reported for EGFR and ALK mutations, whereas results for KRAS remain less reproducible. Integration of radiomics with clinical and pathological data, as well as the use of deep learning, can substantially improve prediction accuracy, but challenges remain regarding model interpretability and lack of standardization. Overall, radiogenomics is a promising non-invasive tool for risk stratification, treatment response monitoring, and clinical decision support in non-small cell lung cancer. However, clinical implementation requires methodological standardization, large multicenter studies, and external validation.
No takes yet. Share an insight, caveat, or question.
Pavlov et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: