Artificial Intelligence in Rhinology: A State-of-the-Art Review of Clinical Readiness and Implementation Pathways.
Authors
Affiliations (2)
Affiliations (2)
- City St. George's University London School of Medicine, Program Delivered by University of Nicosia at the Chaim Sheba Medical Center, Ramat Gan, Israel.
- Department of Otolaryngology-Head and Neck Surgery, Houston Methodist Hospital, Houston, TX 77030, USA.
Abstract
To critically evaluate advances in artificial intelligence (AI) within rhinology, focusing on translational readiness, regulatory alignment, and clinical implementation pathways. PubMed and Scopus. PubMed and Scopus were searched using predefined Title/Abstract Boolean strategies with publication date (January 1, 2023-February 1, 2025), English-language, and human-study restrictions to support a structured state-of-the-art narrative synthesis. Major application domains were categorized as automated computed tomography (CT) analysis, endoscopic computer vision, phenotyping and endotyping in chronic rhinosinusitis, outcome prediction using patient-reported data, digital olfaction, and patient-facing language tools. Each domain was assessed for validation status, workflow feasibility, equity, and data-economics considerations, and alignment with regulatory pathways for software as a medical device. Automated sinus CT models have achieved multi-institutional external validation and appear closest to workflow translation. Endoscopic systems demonstrate promising performance in near real-time video but remain largely retrospective and require evaluation in live workflows. Predictive modeling using integrated clinical, molecular, or patient-reported data remains exploratory, while digital olfaction and language models lack standardized validation and regulatory oversight. Across domains, implementation barriers persist, including interoperability with electronic health records, economic disincentives to data aggregation, and risks of inequitable performance. AI in rhinology is progressing toward integration into clinical care, with automated imaging applications leading adoption. Responsible deployment requires prospective multicenter trials, clinician-supervised workflows, transparency in performance across demographic groups, and evidence of improved patient outcomes. A structured readiness framework may guide stakeholders in prioritizing regulatory-feasible tools that offer measurable clinical value.