Abstrakt

The evolution of Artificial Intelligence (AI) heralds a profound technological shift, fundamentally transforming numerous sectors, from administration to medicine. AI is an umbrella term for technologies that emulate human cognition and problem-solving, particularly utilizing Machine Learning and Deep Learning via neural networks to derive insights from vast datasets. In medicine, AI is gaining increasing traction, for instance, in medical imaging (e.g., X-rays) and pattern recognition within electronic health records (EHRs). Particularly in sleep medicine, AI promises a paradigm shift: it is poised to enhance diagnostic accuracy and optimize therapeutic regimens for sleep disorders. Progress is evident in ambulatory diagnostics, where wearable sensors and AI algorithms facilitate more reliable detection of sleep architecture and sleep-related breathing disorders (SRBD), even when managing massive data volumes. In the sleep laboratory, AI supports diagnostics conducted via polysomnography (PSG). By analyzing immense volumes of biosignal data, AI can make sleep staging (the scoring of sleep phases) more precise, rapid, and standardized, thereby reducing the high inter-rater variability inherent in manual scoring. Future applications of AI promise personalized risk stratification for sequelae of sleep disorders, moving beyond the sole Apnea-Hypopnea Index (AHI) to guide tailored therapeutic interventions. Despite challenges such as data privacy and the explainability of AI decisions, the ultimate responsibility and liability rest with the clinician. Nevertheless, the trajectory is clear: AI is becoming an indispensable tool for a more precise, efficient, and personalized sleep medicine, without intending to supersede the sleep specialist.

Autoreninformation

Autoren

• G. Nilius

Abteilungen

• Klinikum Dortmund gGmbH, Dortmund

Adresse

Prof. Dr. Georg Nilius
Klinikum Dortmund gGmbH
Münsterstr. 240
44145 Dortmund
Email: [email protected]

Citation