Artificial intelligence and machine learning in cochlear implant rehabilitation: a narrative review of outcome prediction, speech enhancement, and therapy-support applications

Artificial intelligence (AI) and machine learning (ML) are increasingly influencing clinical decision-making and rehabilitation strategies for cochlear implant (CI) users. This narrative review synthesizes representative research examining how AI- and ML-based approaches have been applied in CI rehabilitation, with particular emphasis on prediction modeling, speech enhancement, and therapy-oriented digital tools relevant to hearing and speech outcomes. Existing literature suggests that ML-based prediction models can support estimation of CI outcomes, offering opportunities for more personalized counseling and rehabilitation planning. Advances in deep-learning–based speech enhancement have demonstrated consistent improvements in speech intelligibility, particularly in noisy listening environments, which may facilitate the generalization of therapy gains to everyday communication. Additionally, tele-rehabilitation tools and AI-assisted therapy platforms have the potential to improve practice intensity, engagement, and accessibility for both adult and pediatric CI users. Despite these developments, the evidence base remains heterogeneous, and many approaches lack validation in real-world clinical settings. Overall, current research highlights the potential of AI to augment traditional speech therapy in CI rehabilitation, while underscoring the need for more standardized research designs and clinically grounded studies to support effective implementation.