Predictive Performance of Artificial Intelligence Models for ≥Grade 3 Radiotherapy Toxicity in Patients with Non–Small Cell Lung Cancer: A Scoping Review

Severe toxicities (≥ grade 3) from radiotherapy in patients with non-small cell lung cancer (NSCLC) can significantly reduce treatment benefit and patient quality of life. Artificial intelligence (AI) is increasingly being used to develop models that predict severe toxicities in patients. If AI models are able to better predict the onset of severe toxicity than current non-AI methods, radiotherapy treatments can be optimized to limit toxicities and improve patient outcomes. Thus, this scoping review aims to understand the current evidence on the ability of AI predictive models to predict ≥ grade 3 treatment toxicity, compared with non-AI models in patients with NSCLC.