MOAMEN ABDELFADIL ISMAIL, NAWAF AHMAD ALMUTAIRI, HASSAN ALI H. ALALAWI, TAIF KHALID HAKAMI, RAHAF KHALID ABDULSAMED ALHAJJI, HAWRAA FADHEL ALSAYRAFI ,ABDULRAHMAN KHALED ALSHUAIB, HASSAN MANSOUR ALBARAKATI, LAMA KHALID ALHARBI, ABDULRAHMAN IBRAHIM ALTOWAIRQI, ZAINAB ALI ALKHAWAJAH, ILAL YASER RASHED ALHUTHAIL
DOI: https://doi.org/Background: Hearing loss represents an underrecognized complication of diabetes mellitus (DM), often attributed to microvascular and neuropathic changes affecting the auditory system. With the rise of artificial intelligence (AI), predictive tools can potentially facilitate earlier detection of hearing deterioration, optimizing intervention and rehabilitation.
Objective: This systematic review aimed to synthesize existing empirical evidence on AI and machine learning (ML) applications for predicting hearing deterioration, focusing on diabetic and sudden sensorineural hearing loss (SSNHL) populations.
Methods: Following PRISMA 2020 guidelines, eleven peer-reviewed studies published between 2015 and 2025 were reviewed across PubMed, Scopus, and Web of Science. Extracted data included study design, AI model type, predictors, and outcome performance metrics. Quality assessment used the Newcastle–Ottawa Scale and Cochrane RoB 2 tool.
Results: AI models such as AdaBoost, LightGBM, and multiple imputation frameworks demonstrated predictive accuracies of 80–83% for SSNHL recovery and prognosis. Across diabetic cohorts, hearing loss prevalence ranged from 30% to 75%, strongly correlating with higher HbA1c levels, diabetes duration, and hypertension. Studies integrating metabolic and audiological parameters achieved superior predictive accuracy compared to traditional methods.
Conclusions: AI models show substantial potential for early detection of auditory decline in diabetic patients, with integration of metabolic data improving predictive sensitivity. However, real-world implementation requires standardized datasets, cross-validation, and interdisciplinary collaboration between endocrinology and audiology.
