Abstract
With the rise in the prevalence of insomnia, cognitive behavioral therapy for insomnia (CBT-I) has become an important non-pharmacological approach to the treatment of insomnia patients. Traditional CBT-I faces challenges such as low patient acceptance and long treatment cycles. To address these issues, Internet-delivered Cognitive Behavioral Therapy for Insomnia (eCBT-I) emerged as a digital treatment modality with greater flexibility, accessibility and cost-effectiveness. However, eCBT-I still faces issues such as digital literacy and patient engagement. With the development of artificial intelligence (AI) technology, especially the application of large language models (LLM), AI-driven CBT-I is becoming an important direction for future therapy. The LLM is able to provide personalized treatment recommendations based on patient feedback, improving treatment outcomes and patient engagement. This paper reviews the application status and challenges of CBT-I and eCBT-I, focuses on the potential and prospect of LLM in CBT-I therapy, and proposes future research directions, including multi-source data fusion and privacy protection issues, to promote the innovation and development of CBT-I.