Multimodal AI and Large Language Models in Sleep Medicine and Neurodegeneration

The intersection of sleep medicine and neurodegenerative diseases has emerged as an area of increasing scientific and clinical importance. Sleep disorders, including obstructive sleep apnea, insomnia, narcolepsy, and circadian rhythm disruptions, have been shown to impact neurological health, contributing to the onset and progression of diseases such as Alzheimer’s, Parkinson’s, and vascular dementia. Current research recognizes multifactorial pathways—including hypoxia, neuroinflammation, and metabolic dysregulation—through which sleep disturbances exert systemic and brain-specific effects. Despite these advancements, substantial knowledge gaps remain in decoding the complex and bidirectional links between sleep pathologies and neurodegeneration, with most studies predominantly relying on unimodal assessments that capture only fragments of disease mechanisms.Traditional diagnostic and research techniques, such as polysomnography or individual neuroimaging modalities, are limited in their ability to holistically understand these intricate relationships. Recent breakthroughs in artificial intelligence, particularly multimodal AI frameworks and large language models (LLMs), open new possibilities for synthesizing diverse data types—including imaging, electrophysiology, genomics, and real-world clinical records—into unified analyses. Yet, there is a pressing need for novel computational strategies that can robustly integrate and interpret such heterogeneous datasets, support biomarker discovery, improve disease trajectory modeling, and enhance the translation of findings to clinical decision-making. Furthermore, the field faces unresolved challenges regarding interpretability, model generalizability, and ethical adoption in practice.This Research Topic aims to advance integrative, multimodal approaches in the study and management of sleep disorders and their neurological consequences. Bringing together expertise in neuroscience, biomedical engineering, data science, and clinical practice, the Research Topic seeks to foster the development and application of cutting-edge AI architectures for comprehensive analysis, actionable biomarker identification, and mechanistic insight into sleep-related neurodegenerative processes. Of particular interest are studies that leverage large-scale, cross-modal datasets, develop novel AI frameworks or large language models tailored to medical data, or offer translational solutions for diagnosis, monitoring, and intervention in diverse clinical contexts.To gather further insights in the comprehensive integration of multimodal AI and LLMs in sleep medicine and neurodegeneration, we welcome articles addressing, but not limited to, the following themes:• Multimodal data fusion and representation learning for sleep and neurological disorders• Novel computational frameworks and large language models for medical data integration• AI-driven discovery of disease patterns, biomarkers, and progression trajectories• Mechanistic analysis of shared and distinct pathways in sleep and neurodegenerative conditions• Personalized and remote monitoring approaches enabled by AI and sensor technologies• Clinical translation of AI-based systems for diagnosis and decision support• Strategies for interpretable, ethical, and generalizable AI deployment in healthcare Topic editors: Lifeng Li, MD ALTAF UL AMIN, Junaid Abdul Wahid Source: https://www.frontiersin.org/research-topics/76951/multimodal-ai-and-large-language-models-in-sleep-medicine-and-neurodegeneration