Neural network analysis of sleep stages enables efficient diagnosis of narcolepsy

• Published in: Nature communications Vol. 9; no. 1; pp. 5229 - 15
• Main Authors: Stephansen, Jens B., Olesen, Alexander N., Olsen, Mads, Ambati, Aditya, Leary, Eileen B., Moore, Hyatt E., Carrillo, Oscar, Lin, Ling, Han, Fang, Yan, Han, Sun, Yun L., Dauvilliers, Yves, Scholz, Sabine, Barateau, Lucie, Hogl, Birgit, Stefani, Ambra, Hong, Seung Chul, Kim, Tae Won, Pizza, Fabio, Plazzi, Giuseppe, Vandi, Stefano, Antelmi, Elena, Perrin, Dimitri, Kuna, Samuel T., Schweitzer, Paula K., Kushida, Clete, Peppard, Paul E., Sorensen, Helge B. D., Jennum, Poul, Mignot, Emmanuel
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 06.12.2018; Nature Publishing Group; Nature Portfolio
• Subjects: 631/114/1305; 631/378/1385; 631/378/1385/519; 692/617/375/1816; 9/26; Automation; Diagnosis; Human health and pathology; Humanities and Social Sciences; Inspection; Life Sciences; multidisciplinary; Narcolepsy; Network analysis; Neural networks; Neurons and Cognition; Probability distribution; Science; Science (multidisciplinary); Sleep; Sleep disorders; Technicians; Tissue typing
• ISSN: 2041-1723; 2041-1723
• DOI: 10.1038/s41467-018-07229-3

Analysis of sleep for the diagnosis of sleep disorders such as Type-1 Narcolepsy (T1N) currently requires visual inspection of polysomnography records by trained scoring technicians. Here, we used neural networks in approximately 3,000 normal and abnormal sleep recordings to automate sleep stage scoring, producing a hypnodensity graph—a probability distribution conveying more information than classical hypnograms. Accuracy of sleep stage scoring was validated in 70 subjects assessed by six scorers. The best model performed better than any individual scorer (87% versus consensus). It also reliably scores sleep down to 5 s instead of 30 s scoring epochs. A T1N marker based on unusual sleep stage overlaps achieved a specificity of 96% and a sensitivity of 91%, validated in independent datasets. Addition of HLA-DQB1*06:02 typing increased specificity to 99%. Our method can reduce time spent in sleep clinics and automates T1N diagnosis. It also opens the possibility of diagnosing T1N using home sleep studies. The diagnosis of sleep disorders such as narcolepsy and insomnia currently requires experts to interpret sleep recordings (polysomnography). Here, the authors introduce a neural network analysis method for polysomnography that could reduce time spent in sleep clinics and automate narcolepsy diagnosis.