Advancements in artificial intelligence for cancer diagnosis and prognosis prediction: current applications and...

Cancer continues to be a leading cause of mortality worldwide, presenting substantial challenges to public health systems. The traditional approaches to cancer diagnosis and prognosis prediction exhibit certain limitations with respect to accuracy, comprehensiveness, dynamic monitoring, and personalization. With the advancement of artificial intelligence (AI) technologies, novel diagnostic and predictive methods are increasingly addressing these shortcomings. This review provides a comprehensive overview of the primary AI algorithms applied in oncology, including machine learning, deep learning, and large language models. It further examines the distinctive characteristics and appropriate use cases of AI algorithms, highlighting their specific roles in cancer screening, diagnostic accuracy, and outcome forecasting. Additionally, the review discusses emerging trends and persistent challenges, aiming to provide actionable insights that support clinical decision-making and advance scientific innovation in this rapidly evolving field. In conclusion, this review systematically outlines recent advances in AI applications for cancer diagnosis and prognostic prediction, with the objective of facilitating a transformative shift in oncology from experience-based practices toward data-driven precision medicine.