A Deep Learning Strategy for Automatic Sleep Staging Based on Two-Channel EEG Headband Data
Sleep disturbances are common in Alzheimer’s disease and other neurodegenerative disorders, and together represent a potential therapeutic target for disease modification. A major barrier for studying sleep in patients with dementia is the requirement for overnight polysomnography (PSG) to achieve f...
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-05-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/21/10/3316 |
| _version_ | 1797534623428771840 |
|---|---|
| author | Amelia A. Casciola Sebastiano K. Carlucci Brianne A. Kent Amanda M. Punch Michael A. Muszynski Daniel Zhou Alireza Kazemi Maryam S. Mirian Jason Valerio Martin J. McKeown Haakon B. Nygaard |
| author_facet | Amelia A. Casciola Sebastiano K. Carlucci Brianne A. Kent Amanda M. Punch Michael A. Muszynski Daniel Zhou Alireza Kazemi Maryam S. Mirian Jason Valerio Martin J. McKeown Haakon B. Nygaard |
| author_sort | Amelia A. Casciola |
| collection | DOAJ |
| description | Sleep disturbances are common in Alzheimer’s disease and other neurodegenerative disorders, and together represent a potential therapeutic target for disease modification. A major barrier for studying sleep in patients with dementia is the requirement for overnight polysomnography (PSG) to achieve formal sleep staging. This is not only costly, but also spending a night in a hospital setting is not always advisable in this patient group. As an alternative to PSG, portable electroencephalography (EEG) headbands (HB) have been developed, which reduce cost, increase patient comfort, and allow sleep recordings in a person’s home environment. However, naïve applications of current automated sleep staging systems tend to perform inadequately with HB data, due to their relatively lower quality. Here we present a deep learning (DL) model for automated sleep staging of HB EEG data to overcome these critical limitations. The solution includes a simple band-pass filtering, a data augmentation step, and a model using convolutional (CNN) and long short-term memory (LSTM) layers. With this model, we have achieved 74% (±10%) validation accuracy on low-quality two-channel EEG headband data and 77% (±10%) on gold-standard PSG. Our results suggest that DL approaches achieve robust sleep staging of both portable and in-hospital EEG recordings, and may allow for more widespread use of ambulatory sleep assessments across clinical conditions, including neurodegenerative disorders. |
| first_indexed | 2024-03-10T11:33:09Z |
| format | Article |
| id | doaj.art-d8b85c4367cb452b82e60bdf3a1961d9 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T11:33:09Z |
| publishDate | 2021-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-d8b85c4367cb452b82e60bdf3a1961d92023-11-21T19:07:40ZengMDPI AGSensors1424-82202021-05-012110331610.3390/s21103316A Deep Learning Strategy for Automatic Sleep Staging Based on Two-Channel EEG Headband DataAmelia A. Casciola0Sebastiano K. Carlucci1Brianne A. Kent2Amanda M. Punch3Michael A. Muszynski4Daniel Zhou5Alireza Kazemi6Maryam S. Mirian7Jason Valerio8Martin J. McKeown9Haakon B. Nygaard10Department of Electrical and Computer Engineering Capstone, University of British Columbia, Vancouver, BC V6T 1Z4, CanadaDepartment of Electrical and Computer Engineering Capstone, University of British Columbia, Vancouver, BC V6T 1Z4, CanadaDjavad Mowafaghian Centre for Brain Health, Division of Neurology, University of British Columbia, Vancouver, BC V6T 1Z3, CanadaDepartment of Electrical and Computer Engineering Capstone, University of British Columbia, Vancouver, BC V6T 1Z4, CanadaDepartment of Electrical and Computer Engineering Capstone, University of British Columbia, Vancouver, BC V6T 1Z4, CanadaDepartment of Electrical and Computer Engineering Capstone, University of British Columbia, Vancouver, BC V6T 1Z4, CanadaCenter for Mind and Brain, Department of Psychology, University of California, Davis, CA 95618, USADjavad Mowafaghian Centre for Brain Health, Division of Neurology, University of British Columbia, Vancouver, BC V6T 1Z3, CanadaDjavad Mowafaghian Centre for Brain Health, Division of Neurology, University of British Columbia, Vancouver, BC V6T 1Z3, CanadaDjavad Mowafaghian Centre for Brain Health, Division of Neurology, University of British Columbia, Vancouver, BC V6T 1Z3, CanadaDjavad Mowafaghian Centre for Brain Health, Division of Neurology, University of British Columbia, Vancouver, BC V6T 1Z3, CanadaSleep disturbances are common in Alzheimer’s disease and other neurodegenerative disorders, and together represent a potential therapeutic target for disease modification. A major barrier for studying sleep in patients with dementia is the requirement for overnight polysomnography (PSG) to achieve formal sleep staging. This is not only costly, but also spending a night in a hospital setting is not always advisable in this patient group. As an alternative to PSG, portable electroencephalography (EEG) headbands (HB) have been developed, which reduce cost, increase patient comfort, and allow sleep recordings in a person’s home environment. However, naïve applications of current automated sleep staging systems tend to perform inadequately with HB data, due to their relatively lower quality. Here we present a deep learning (DL) model for automated sleep staging of HB EEG data to overcome these critical limitations. The solution includes a simple band-pass filtering, a data augmentation step, and a model using convolutional (CNN) and long short-term memory (LSTM) layers. With this model, we have achieved 74% (±10%) validation accuracy on low-quality two-channel EEG headband data and 77% (±10%) on gold-standard PSG. Our results suggest that DL approaches achieve robust sleep staging of both portable and in-hospital EEG recordings, and may allow for more widespread use of ambulatory sleep assessments across clinical conditions, including neurodegenerative disorders.https://www.mdpi.com/1424-8220/21/10/3316deep learningEEG headbandsleep stagingmachine learningneurodegenerative diseasesleep |
| spellingShingle | Amelia A. Casciola Sebastiano K. Carlucci Brianne A. Kent Amanda M. Punch Michael A. Muszynski Daniel Zhou Alireza Kazemi Maryam S. Mirian Jason Valerio Martin J. McKeown Haakon B. Nygaard A Deep Learning Strategy for Automatic Sleep Staging Based on Two-Channel EEG Headband Data Sensors deep learning EEG headband sleep staging machine learning neurodegenerative disease sleep |
| title | A Deep Learning Strategy for Automatic Sleep Staging Based on Two-Channel EEG Headband Data |
| title_full | A Deep Learning Strategy for Automatic Sleep Staging Based on Two-Channel EEG Headband Data |
| title_fullStr | A Deep Learning Strategy for Automatic Sleep Staging Based on Two-Channel EEG Headband Data |
| title_full_unstemmed | A Deep Learning Strategy for Automatic Sleep Staging Based on Two-Channel EEG Headband Data |
| title_short | A Deep Learning Strategy for Automatic Sleep Staging Based on Two-Channel EEG Headband Data |
| title_sort | deep learning strategy for automatic sleep staging based on two channel eeg headband data |
| topic | deep learning EEG headband sleep staging machine learning neurodegenerative disease sleep |
| url | https://www.mdpi.com/1424-8220/21/10/3316 |
| work_keys_str_mv | AT ameliaacasciola adeeplearningstrategyforautomaticsleepstagingbasedontwochanneleegheadbanddata AT sebastianokcarlucci adeeplearningstrategyforautomaticsleepstagingbasedontwochanneleegheadbanddata AT brianneakent adeeplearningstrategyforautomaticsleepstagingbasedontwochanneleegheadbanddata AT amandampunch adeeplearningstrategyforautomaticsleepstagingbasedontwochanneleegheadbanddata AT michaelamuszynski adeeplearningstrategyforautomaticsleepstagingbasedontwochanneleegheadbanddata AT danielzhou adeeplearningstrategyforautomaticsleepstagingbasedontwochanneleegheadbanddata AT alirezakazemi adeeplearningstrategyforautomaticsleepstagingbasedontwochanneleegheadbanddata AT maryamsmirian adeeplearningstrategyforautomaticsleepstagingbasedontwochanneleegheadbanddata AT jasonvalerio adeeplearningstrategyforautomaticsleepstagingbasedontwochanneleegheadbanddata AT martinjmckeown adeeplearningstrategyforautomaticsleepstagingbasedontwochanneleegheadbanddata AT haakonbnygaard adeeplearningstrategyforautomaticsleepstagingbasedontwochanneleegheadbanddata AT ameliaacasciola deeplearningstrategyforautomaticsleepstagingbasedontwochanneleegheadbanddata AT sebastianokcarlucci deeplearningstrategyforautomaticsleepstagingbasedontwochanneleegheadbanddata AT brianneakent deeplearningstrategyforautomaticsleepstagingbasedontwochanneleegheadbanddata AT amandampunch deeplearningstrategyforautomaticsleepstagingbasedontwochanneleegheadbanddata AT michaelamuszynski deeplearningstrategyforautomaticsleepstagingbasedontwochanneleegheadbanddata AT danielzhou deeplearningstrategyforautomaticsleepstagingbasedontwochanneleegheadbanddata AT alirezakazemi deeplearningstrategyforautomaticsleepstagingbasedontwochanneleegheadbanddata AT maryamsmirian deeplearningstrategyforautomaticsleepstagingbasedontwochanneleegheadbanddata AT jasonvalerio deeplearningstrategyforautomaticsleepstagingbasedontwochanneleegheadbanddata AT martinjmckeown deeplearningstrategyforautomaticsleepstagingbasedontwochanneleegheadbanddata AT haakonbnygaard deeplearningstrategyforautomaticsleepstagingbasedontwochanneleegheadbanddata |