Artificial Intelligence and Machine Learning in Audiology and Hearing Disorders: A Scoping Review with Bibliometric and Thematic Mapping (1995–2025)

Abstract

<b>Background and Objectives:</b> Artificial intelligence (AI) and machine learning (ML) are increasingly integrated into audiology, supporting diagnosis, screening, rehabilitation, and digital health. Despite rapid growth, the literature remains methodologically and clinically heterogeneous, limiting a consolidated view of research trajectories and translational readiness. This scoping review examined the evolution of AI and ML applications in audiology and hearing disorders, focusing on thematic development, research productivity, collaboration patterns, and clinical orientation. <b>Methods:</b> A scoping review was conducted using the Web of Science Core Collection (Science Citation Index Expanded). Original and review articles published between 1995 and 2025 were included. Bibliometric and thematic mapping were applied to analyze publication trends, citation patterns, keyword evolution, and collaboration networks. A structured translational categorization assessed clinical domains and validation maturity. Findings reflect the Web of Science-indexed segment of the literature. <b>Results:</b> A total of 127 publications were analyzed. Research output increased markedly after 2020, with an estimated doubling time of approximately 2.1 years. China, the United States, and South Korea contributed the highest publication volumes, although citation impact did not consistently parallel productivity. Thematic analyses revealed a shift toward AI-driven methodological frameworks, particularly in machine learning, deep learning, and cochlear implant-related applications. Most studies remain at proof-of-concept or internally validated stages, with limited external validation. Emerging areas include tele-audiology and personalized hearing aid optimization. <b>Conclusions:</b> AI and ML research in audiology is increasingly application-oriented; however, broader external validation and prospective implementation are required to support routine clinical integration.

Ähnliche Arbeiten