Advanced sleep spindle identification with neural networks
Abstract Sleep spindles are neurophysiological phenomena that appear to be linked to memory formation and other functions of the central nervous system, and that can be observed in electroencephalographic recordings (EEG) during sleep. Manually identified spindle annotations in EEG recordings suffer...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2022-05-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-022-11210-y |
| _version_ | 1811253273373442048 |
|---|---|
| author | Lars Kaulen Justus T. C. Schwabedal Jules Schneider Philipp Ritter Stephan Bialonski |
| author_facet | Lars Kaulen Justus T. C. Schwabedal Jules Schneider Philipp Ritter Stephan Bialonski |
| author_sort | Lars Kaulen |
| collection | DOAJ |
| description | Abstract Sleep spindles are neurophysiological phenomena that appear to be linked to memory formation and other functions of the central nervous system, and that can be observed in electroencephalographic recordings (EEG) during sleep. Manually identified spindle annotations in EEG recordings suffer from substantial intra- and inter-rater variability, even if raters have been highly trained, which reduces the reliability of spindle measures as a research and diagnostic tool. The Massive Online Data Annotation (MODA) project has recently addressed this problem by forming a consensus from multiple such rating experts, thus providing a corpus of spindle annotations of enhanced quality. Based on this dataset, we present a U-Net-type deep neural network model to automatically detect sleep spindles. Our model’s performance exceeds that of the state-of-the-art detector and of most experts in the MODA dataset. We observed improved detection accuracy in subjects of all ages, including older individuals whose spindles are particularly challenging to detect reliably. Our results underline the potential of automated methods to do repetitive cumbersome tasks with super-human performance. |
| first_indexed | 2024-04-12T16:47:26Z |
| format | Article |
| id | doaj.art-ca9c83e84b3e4c3197806758855978c8 |
| institution | Directory Open Access Journal |
| issn | 2045-2322 |
| language | English |
| last_indexed | 2024-04-12T16:47:26Z |
| publishDate | 2022-05-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj.art-ca9c83e84b3e4c3197806758855978c82022-12-22T03:24:31ZengNature PortfolioScientific Reports2045-23222022-05-0112111010.1038/s41598-022-11210-yAdvanced sleep spindle identification with neural networksLars Kaulen0Justus T. C. Schwabedal1Jules Schneider2Philipp Ritter3Stephan Bialonski4Department of Medical Engineering and Technomathematics, FH Aachen University of Applied SciencesIndependent researcherDepartment of Psychiatry and Psychotherapy, University Hospital Carl Gustav Carus, Technische Universität DresdenDepartment of Psychiatry and Psychotherapy, University Hospital Carl Gustav Carus, Technische Universität DresdenDepartment of Medical Engineering and Technomathematics, FH Aachen University of Applied SciencesAbstract Sleep spindles are neurophysiological phenomena that appear to be linked to memory formation and other functions of the central nervous system, and that can be observed in electroencephalographic recordings (EEG) during sleep. Manually identified spindle annotations in EEG recordings suffer from substantial intra- and inter-rater variability, even if raters have been highly trained, which reduces the reliability of spindle measures as a research and diagnostic tool. The Massive Online Data Annotation (MODA) project has recently addressed this problem by forming a consensus from multiple such rating experts, thus providing a corpus of spindle annotations of enhanced quality. Based on this dataset, we present a U-Net-type deep neural network model to automatically detect sleep spindles. Our model’s performance exceeds that of the state-of-the-art detector and of most experts in the MODA dataset. We observed improved detection accuracy in subjects of all ages, including older individuals whose spindles are particularly challenging to detect reliably. Our results underline the potential of automated methods to do repetitive cumbersome tasks with super-human performance.https://doi.org/10.1038/s41598-022-11210-y |
| spellingShingle | Lars Kaulen Justus T. C. Schwabedal Jules Schneider Philipp Ritter Stephan Bialonski Advanced sleep spindle identification with neural networks Scientific Reports |
| title | Advanced sleep spindle identification with neural networks |
| title_full | Advanced sleep spindle identification with neural networks |
| title_fullStr | Advanced sleep spindle identification with neural networks |
| title_full_unstemmed | Advanced sleep spindle identification with neural networks |
| title_short | Advanced sleep spindle identification with neural networks |
| title_sort | advanced sleep spindle identification with neural networks |
| url | https://doi.org/10.1038/s41598-022-11210-y |
| work_keys_str_mv | AT larskaulen advancedsleepspindleidentificationwithneuralnetworks AT justustcschwabedal advancedsleepspindleidentificationwithneuralnetworks AT julesschneider advancedsleepspindleidentificationwithneuralnetworks AT philippritter advancedsleepspindleidentificationwithneuralnetworks AT stephanbialonski advancedsleepspindleidentificationwithneuralnetworks |