ENHANCING SURGICAL EFFICIENCY THROUGH ARTIFICIAL INTELLIGENCE IN THE SURGICAL MANAGEMENT OF MIDDLE EAR CHOLESTEATOMA

Abstract

Middle ear cholesteatoma is a destructive epithelial lesion of the middle ear and mastoid that may lead to ossicular erosion, labyrinthine fistula, facial nerve injury, intracranial complications, and recurrent disease if inadequately treated. Surgery remains the mainstay of treatment, but preoperative imaging interpretation, disease mapping, selection of the optimal surgical approach, and postoperative surveillance continue to depend heavily on expert judgment. Recent evidence shows that artificial intelligence can improve cholesteatoma care in several domains, including CT-based diagnosis, differentiation from chronic suppurative otitis media, intraoperative residual lesion detection, and prediction of postoperative hearing recovery. A 2025 systematic review found that most published AI studies in cholesteatoma have focused on temporal bone CT, while a 2024 explainable 3D CNN model achieved expert-level performance in CT-based cholesteatoma identification and contributed to clinical decision-making in 90.1% of cases. In parallel, non-echo-planar diffusion-weighted MRI remains highly accurate for residual or recurrent cholesteatoma and may reduce unnecessary second-look surgery. The aim of this modeled study was to evaluate whether an AI-assisted workflow could improve preoperative planning, intraoperative efficiency, and 12-month outcomes in 122 patients with middle ear cholesteatoma treated surgically at the TDTU ENT Department. The modeled protocol combined standard otologic assessment with AI-assisted temporal bone CT analysis, machine-learning-based risk stratification, automated lesion extension mapping, AI-supported intraoperative residual-disease screening, and structured postoperative imaging follow-up. Synthetic results suggested that AI support improved diagnostic concordance, reduced time to final operative planning, shortened operating time, lowered intraoperative strategy changes, and improved hearing and recurrence-related outcomes at 12 months.