ENHANCING SURGICAL EFFICIENCY THROUGH ARTIFICIAL INTELLIGENCE IN THE MANAGEMENT OF CHRONIC RHINOSINUSITIS WITH NASAL POLYPS

Abstract

Chronic rhinosinusitis with nasal polyps, or CRSwNP, is a chronic inflammatory disease of the nose and paranasal sinuses characterized by persistent symptoms for more than 12 weeks together with objective evidence on nasal endoscopy and/or computed tomography. Contemporary guidance consistently supports nasal endoscopy, CT imaging, and patient-reported outcome tools such as SNOT-22 as the core framework for disease assessment, while endoscopic sinus surgery remains the standard surgical option for patients whose disease is insufficiently controlled by appropriate medical treatment. Recent studies have shown that artificial intelligence can support CRSwNP care by improving CT-based endotype prediction, postoperative outcome prediction, endoscopic image analysis, and radiologic workflow standardization. A 2024 CT-based deep-learning study reported a testing AUC of 0.963 for predicting eosinophilic versus non-eosinophilic CRSwNP, while a 2024 prospective machine-learning study predicted postoperative control, partial control, or relapse at 18 months after ESS with accuracies ranging from 69.23% using noninvasive variables to 84.62% when microRNAs were added. In addition, a 2024 pilot study showed that AI can automatically detect and segment nasal polyps from endoscopy videos, and a 2025 multicenter study demonstrated the feasibility of AI-based postoperative endoscopic outcome analysis in chronic rhinosinusitis. The aim of this modeled study was to evaluate whether an artificial-intelligence-assisted surgical workflow could improve preoperative planning, intraoperative efficiency, and 12-month outcomes in 142 patients with CRSwNP treated at the TDTU ENT Department. The modeled protocol combined standard clinical assessment, nasal endoscopy, CT-based Lund-Mackay scoring, AI-driven imaging interpretation, polyp burden mapping, anatomy-aware surgical route planning, and AI-supported postoperative endoscopic analysis. Synthetic results suggested that AI support improved diagnostic concordance, shortened the time to operative planning, reduced intraoperative changes in surgical strategy, lowered operating time, and improved 12-month disease control and revision-free follow-up.