Environmental and Genetic Interaction Models for Predicting Lung Cancer Risk Using Machine Learning: A Systematic Review and Meta-Analysis

Abstract

This systematic review and meta-analysis examined environmental and genetic interaction models for predicting lung cancer risk using machine learning techniques. The findings underscore the importance of considering both genetic and environmental factors to enhance predictive accuracy, with significant clinical implications. Among the models reviewed, the Rotation Forest model demonstrated superior performance, achieving an AUC of 0.993, reflecting excellent predictive capabilities. The mean AUC across all models was approximately 0.789, indicating moderate to good discrimination. These results hold promising implications for personalized medicine and clinical decision-making, potentially improving patient outcomes and reducing the global burden of lung cancer. The meta-analysis further highlighted strong performance, with an average TRI-performance (accuracy, precision, recall) of 88.1%, demonstrating robust predictive abilities. Integrating machine learning with multidimensional data deepens the understanding of the biological mechanisms underlying lung cancer and supports its application in precision oncology, paving the way for individualized interventions and improved clinical management.