Artificial intelligence in esophageal cancer radiotherapy: a literature review
Year: 2026
Type: article
Study design: Systematic Review
Abstract: Background and Objective: Radiotherapy is a key component in the management of esophageal cancer (EC). Artificial intelligence (AI) has emerged as a transformative tool for optimizing radiotherapy workflows. Numerous AI-based tools are being developed to improve efficiency and facilitate the personalization of radiation oncology. This study aimed to provide an up-to-date review of the literature on recent advances in AI applications across the EC radiotherapy workflow. Methods: The PubMed, Embase, and Web of Science databases were searched to retrieve original English-language articles published from January 1, 2020 to May 26, 2026. The following search terms were used: "Esophageal neoplasms", "esophageal cancer", "oesophageal cancer" AND "artificial intelligence", "AI" AND "radiation oncology", "radiotherapy", "radiation", "chemoradiotherapy". The included studies covered different steps in the radiotherapy workflow; that is, initial treatment decision-making, treatment planning and preparation, treatment delivery, and post-treatment follow-up. Key Content and Findings: A total of 105 studies covering the different stages of the radiotherapy workflow were included in the review. Most studies focused on initial treatment decision-making, primarily benefit-risk stratification through predictive modeling of treatment outcomes and treatment-related toxicities, followed by treatment planning and preparation, particularly automated segmentation and treatment planning. Conversely, few studies investigated AI applications in treatment plan approval and quality assurance, or treatment delivery. Notably, most of the included studies were retrospective, with substantial heterogeneity in study designs, methodologies, and reporting. Conclusions: AI demonstrates substantial potential to facilitate clinical decision-making while improving the efficiency and accuracy of radiotherapy workflows. Future research should prioritize multicenter prospective studies, standardized methodological and reporting frameworks, and explainable AI (XAI) to improve the reliability, transparency, and clinical translation of AI models.
Source: Journal of Thoracic Disease
Authors: Guangcai Wang, Gengshen Bai, Shiwen Huo
Institution: Jiangsu Hengrui Medicine (China)
Language: English
Subfield: Health Informatics
Field: Medicine
Domain: Health Sciences
Open Access status: diamond
OpenAlex ID: W7224137345