ENHANCING SURGICAL EFFICIENCY THROUGH ARTIFICIAL INTELLIGENCE IN THE SURGICAL TREATMENT OF OTOSCLEROSIS

Abstract

Otosclerosis is a localized disorder of otic capsule bone remodeling that commonly causes progressive conductive or mixed hearing loss through stapes footplate fixation. High-resolution temporal bone CT is the principal imaging study used to support diagnosis and surgical planning, but its diagnostic yield is variable, with umbrella-review data showing CT sensitivity of 60% to 95% and specificity of 75% to 100%. Deep-learning tools for temporal bone CT now show promising diagnostic performance, and recent studies suggest that AI can support otosclerosis detection, improve image quality, assist middle-ear diagnostic interpretation, and predict postoperative hearing after stapedotomy. The aim of this modeled study was to evaluate whether an AI-assisted workflow could improve preoperative decision-making, intraoperative efficiency, and 12-month hearing outcomes in 122 patients with otosclerosis treated surgically at the TDTU ENT Department. The modeled protocol combined standard otologic assessment with deep-learning-enhanced CT review, AI-based otosclerosis focus detection, machine-learning interpretation of wideband immittance data, and a preoperative hearing-outcome prediction model. Synthetic results suggested that AI support improved diagnostic concordance, shortened time to final operative planning, reduced operative time and intraoperative plan changes, and improved the proportion of patients achieving favorable air-bone gap closure at follow-up.