AI-Enabled Personalized Medicine in Lung Cancer: Diagnostics, Prognostics, and Targeted Drug Delivery- A Systematic Review

Introduction: With mutations in important proto-oncogenic drivers such as Kirsten rat sarcoma and epidermal growth factor receptor, non-small cell lung cancer has a diverse genetic background. Regardless of smoking history, Kras-activating mutations are present in about 40% of adenocarcinomas. The PI3K/Akt/mTOR pathway has been strongly linked to the development of tumors and the advancement of NSCLC. Methods: Various search engines are used for the investigation of reviews and research, like PubMed, Springer Nature, Bentham Science, PLOS One, and Google Scholar. The articles included data on artificial intelligence and machine learning in NSCLC. Result: AI-assisted technologies are now benefiting patients, such as CT radiomics, 18FFDG PET/CT, LASSO regression, and Random Forest classifiers. Methods of Artificial Intelligence –Machine learning-based models are also developed that help in the treatment of NSCLC. Discussion: AI-based techniques have broadened the scope of biomarker discovery by showcasing the effectiveness of combining multimodal data from pre-existing datasets to find novel meta-biomarkers. Few trials provided strong evidence for an urgent change in practice, even though most indicated potential for AI-based forecasting of immunotherapy outcomes. Conclusion: Regarding lung cancer-specific survival, the deep learning survival neural network model exhibits potential advantages in prognostic assessment and therapy recommendation. This innovative analytical method could offer reliable data on human survival and treatment suggestions. Personalized treatment plans, better treatment results, and a decrease in needless toxicities might all be made possible by the proper use of AI/ML models. DOI:10.2174/0118756921432491260401063552