Automatic sleep stage classification with deep residual networks in a mixed-cohort setting
Research output: Contribution to journal › Journal article › Research › peer-review
Standard
Automatic sleep stage classification with deep residual networks in a mixed-cohort setting. / Olesen, Alexander Neergaard; Jørgen Jennum, Poul; Mignot, Emmanuel; Sorensen, Helge Bjarup Dissing.
In: Sleep, Vol. 44, No. 1, zsaa161, 01.01.2021.Research output: Contribution to journal › Journal article › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - JOUR
T1 - Automatic sleep stage classification with deep residual networks in a mixed-cohort setting
AU - Olesen, Alexander Neergaard
AU - Jørgen Jennum, Poul
AU - Mignot, Emmanuel
AU - Sorensen, Helge Bjarup Dissing
N1 - Publisher Copyright: © 2020 Sleep Research Society 2020. Published by Oxford University Press on behalf of the Sleep Research Society. All rights reserved. For permissions, please e-mail journals.permissions@oup.com.
PY - 2021/1/1
Y1 - 2021/1/1
N2 - Study Objectives: Sleep stage scoring is performed manually by sleep experts and is prone to subjective interpretation of scoring rules with low intra- and interscorer reliability. Many automatic systems rely on few small-scale databases for developing models, and generalizability to new datasets is thus unknown. We investigated a novel deep neural network to assess the generalizability of several large-scale cohorts. Methods: A deep neural network model was developed using 15,684 polysomnography studies from five different cohorts. We applied four different scenarios: (1) impact of varying timescales in the model; (2) performance of a single cohort on other cohorts of smaller, greater, or equal size relative to the performance of other cohorts on a single cohort; (3) varying the fraction of mixed-cohort training data compared with using single-origin data; and (4) comparing models trained on combinations of data from 2, 3, and 4 cohorts. Results: Overall classification accuracy improved with increasing fractions of training data (0.25%: 0.782 ± 0.097, 95% CI [0.777-0.787]; 100%: 0.869 ± 0.064, 95% CI [0.864-0.872]), and with increasing number of data sources (2: 0.788 ± 0.102, 95% CI [0.787-0.790]; 3: 0.808 ± 0.092, 95% CI [0.807-0.810]; 4: 0.821 ± 0.085, 95% CI [0.819-0.823]). Different cohorts show varying levels of generalization to other cohorts. Conclusions: Automatic sleep stage scoring systems based on deep learning algorithms should consider as much data as possible from as many sources available to ensure proper generalization. Public datasets for benchmarking should be made available for future research.
AB - Study Objectives: Sleep stage scoring is performed manually by sleep experts and is prone to subjective interpretation of scoring rules with low intra- and interscorer reliability. Many automatic systems rely on few small-scale databases for developing models, and generalizability to new datasets is thus unknown. We investigated a novel deep neural network to assess the generalizability of several large-scale cohorts. Methods: A deep neural network model was developed using 15,684 polysomnography studies from five different cohorts. We applied four different scenarios: (1) impact of varying timescales in the model; (2) performance of a single cohort on other cohorts of smaller, greater, or equal size relative to the performance of other cohorts on a single cohort; (3) varying the fraction of mixed-cohort training data compared with using single-origin data; and (4) comparing models trained on combinations of data from 2, 3, and 4 cohorts. Results: Overall classification accuracy improved with increasing fractions of training data (0.25%: 0.782 ± 0.097, 95% CI [0.777-0.787]; 100%: 0.869 ± 0.064, 95% CI [0.864-0.872]), and with increasing number of data sources (2: 0.788 ± 0.102, 95% CI [0.787-0.790]; 3: 0.808 ± 0.092, 95% CI [0.807-0.810]; 4: 0.821 ± 0.085, 95% CI [0.819-0.823]). Different cohorts show varying levels of generalization to other cohorts. Conclusions: Automatic sleep stage scoring systems based on deep learning algorithms should consider as much data as possible from as many sources available to ensure proper generalization. Public datasets for benchmarking should be made available for future research.
KW - automatic sleep stage classification
KW - computational sleep science
KW - deep learning
KW - machine learning
U2 - 10.1093/sleep/zsaa161
DO - 10.1093/sleep/zsaa161
M3 - Journal article
C2 - 32844179
AN - SCOPUS:85100280414
VL - 44
JO - Sleep (Online)
JF - Sleep (Online)
SN - 0161-8105
IS - 1
M1 - zsaa161
ER -
ID: 305025311