Abstract

Artificial intelligence (AI) is rapidly revolutionizing the landscape of oncological research and the advancement of personalized clinical interventions. Progress in three interconnected areas, including the development of methods and algorithms for training AI models, the evolution of specialized computing hardware, and increased access to large volumes of cancer data such as imaging, genomics, and clinical information, has converged, leading to promising new applications of AI in cancer research. AI applications are systematically organized according to specific cancer types and clinical domains, encompassing the elucidation and prediction of biological mechanisms, the identification and utilization of patterns within clinical data to improve patient outcomes, and the unraveling of the complexities inherent in epidemiological, behavioral, and real-world datasets. When applied in an ethical and scientifically rigorous manner, these AI-driven approaches hold the promise of accelerating progress in cancer research and ultimately fostering improved health outcomes for all populations. We review examples demonstrating the integration of AI within oncology, highlighting cases where deep learning has adeptly addressed challenges once deemed insurmountable, while also discussing the barriers that must be surmounted to facilitate broader adoption of these technologies.

Read the full article here