Predicting cancer outcomes with radiomics and artificial intelligence in radiology

Prognostication of outcome across multiple cancers and prediction of response to various treatment modalities are among the next generation of challenges that artificial intelligence (AI) tools can solve using radiology images. The authors of this Perspective describe the evolution of AI-based approaches in oncology imaging and address the path to their adoption as decision-support tools in the clinic. The successful use of artificial intelligence (AI) for diagnostic purposes has prompted the application of AI-based cancer imaging analysis to address other, more complex, clinical needs. In this Perspective, we discuss the next generation of challenges in clinical decision-making that AI tools can solve using radiology images, such as prognostication of outcome across multiple cancers, prediction of response to various treatment modalities, discrimination of benign treatment confounders from true progression, identification of unusual response patterns and prediction of the mutational and molecular profile of tumours. We describe the evolution of and opportunities for AI in oncology imaging, focusing on hand-crafted radiomic approaches and deep learning-derived representations, with examples of their application for decision support. We also address the challenges faced on the path to clinical adoption, including data curation and annotation, interpretability, and regulatory and reimbursement issues. We hope to demystify AI in radiology for clinicians by helping them to understand its limitations and challenges, as well as the opportunities it provides as a decision-support tool in cancer management.

Automated summary

The successful use of artificial intelligence in oncology imaging for diagnostic purposes has prompted the exploration of its potential in addressing more complex clinical needs. This perspective discusses the evolution of AI tools in oncology imaging, focusing on challenges such as outcome prognostication across multiple cancers and response prediction to various treatment modalities. The authors also highlight the opportunities and challenges in the path to clinical adoption, aiming to demystify AI for clinicians and emphasize its role as a decision-support tool in cancer management.