ENHANCING THE ROLE OF ARTIFICIAL INTELLIGENCE IN THE SURGICAL MANAGEMENT OF PATIENTS WITH CHRONIC MAXILLARY SINUSITIS

Abstract

Chronic maxillary sinusitis represents a clinically important subtype of chronic rhinosinusitis and is characterized by persistent sinonasal inflammation lasting more than 12 weeks, typically confirmed by symptoms together with objective endoscopic and radiologic evidence. Endoscopic sinus surgery remains the standard surgical option for patients with disease refractory to appropriate medical treatment, but preoperative interpretation of computed tomography, anatomical variation analysis, disease extent assessment, and postoperative monitoring remain partly subjective and operator-dependent. In recent years, artificial intelligence has emerged as a promising adjunct in chronic rhinosinusitis through automated image interpretation, lesion segmentation, prognostic modeling, and standardized postoperative assessment. Systematic reviews report that AI models for CT and CBCT-based maxillary sinus pathology detection generally achieve accuracy values in the range of 85% to 97%, sensitivity from 87% to 100%, and specificity from 87.2% to 99.7%, while a recent CT-based AI study for chronic maxillary sinusitis reported sensitivity of 0.9796, specificity of 0.8636, accuracy of 0.9247, and an AUC of 0.94 on the test set. The aim of this modeled study was to evaluate whether an AI-assisted workflow could improve surgical planning, intraoperative efficiency, and early clinical outcomes in 150 patients with chronic maxillary sinusitis treated at the TDTU ENT Department. The modeled protocol combined clinical assessment, nasal endoscopy, CT-based Lund-Mackay scoring, AI-assisted image analysis, automatic maxillary sinus segmentation, anatomy-aware surgical route selection, and AI-supported postoperative endoscopic evaluation. Synthetic results suggested that AI support improved preoperative diagnostic concordance, reduced time to surgical decision-making, shortened operating time, decreased intraoperative change of plan, and improved 12-month symptom control and revision-free follow-up. These findings support the concept that AI may enhance, but not replace, surgeon-led decision-making in chronic maxillary sinusitis.