Opportunities and challenges of machine learning in anticaner drug therapies

Abstract

Antitumor drug therapies encounter substantial costs and intricate challenges, imposing a financial strain on patients and potentially leading to serious adverse effects. These issues have prompted a shift towards personalized precision medicine, although the increased workload for clinicians limits its full implementation. Machine learning (ML) offers innovative solutions to these challenges. By effectively integrating and analysing large clinical datasets, ML can develop models to predict potential treatment-related risks for patients and optimize dosing regimens, thereby improving efficacy and reducing adverse effects. Additionally, ML can evaluate drug efficacy, providing empirical support for personalized treatments. This review highlights the research progress in ML for antitumor drug therapies and examines its crucial role in advancing personalized precision medicine. It is expected that ML will deliver more accurate, efficient, and cost-effective treatment options for patients while providing strong support for clinicians in refining treatment decisions, making it an essential tool in cancer therapy.