Toward Interpretable Sleep Stage Classification Using Cross-Modal Transformers

Jathurshan Pradeepkumar; Mithunjha Anandakumar; Vinith Kugathasan; Dhinesh Suntharalingham; Simon L. Kappel; Anjula C. De Silva; Chamira U.S. Edussooriya

doi:10.1109/TNSRE.2024.3438610

Toward Interpretable Sleep Stage Classification Using Cross-Modal Transformers

Jathurshan Pradeepkumar, Mithunjha Anandakumar, Vinith Kugathasan, Dhinesh Suntharalingham, Simon L. Kappel, Anjula C. De Silva, Chamira U.S. Edussooriya^*

^*Corresponding author for this work

Department of Electrical and Computer Engineering - Biomedical Engineering

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaper › Journal article › Research › peer-review

5 Citations (Scopus)

Abstract

Accurate sleep stage classification is significant for sleep health assessment. In recent years, several machine-learning based sleep staging algorithms have been developed, and in particular, deep-learning based algorithms have achieved performance on par with human annotation. Despite improved performance, a limitation of most deep-learning based algorithms is their black-box behavior, which have limited their use in clinical settings. Here, we propose a cross-modal transformer, which is a transformer-based method for sleep stage classification. The proposed cross-modal transformer consists of a cross-modal transformer encoder architecture along with a multi-scale one-dimensional convolutional neural network for automatic representation learning. The performance of our method is on-par with the state-of-the-art methods and eliminates the black-box behavior of deep-learning models by utilizing the interpretability aspect of the attention modules. Furthermore, our method provides considerable reductions in the number of parameters and training time compared to the state-of-the-art methods. Our code is available at https://github.com/Jathurshan0330/Cross-Modal-Transformer. A demo of our work can be found at https://bit.ly/Cross_modal_transformer_demo.

Original language	English
Journal	IEEE Transactions on Neural Systems and Rehabilitation Engineering
Volume	32
Pages (from-to)	2893-2904
Number of pages	12
ISSN	1534-4320
DOIs	https://doi.org/10.1109/TNSRE.2024.3438610
Publication status	Published - 2024

Keywords

Automatic sleep stage classification
deep neural networks
interpretable deep learning
transformers

Access to Document

10.1109/TNSRE.2024.3438610Licence: CC BY

Cite this

@article{d8de235501064c5f829931e1e123343e,

title = "Toward Interpretable Sleep Stage Classification Using Cross-Modal Transformers",

abstract = "Accurate sleep stage classification is significant for sleep health assessment. In recent years, several machine-learning based sleep staging algorithms have been developed, and in particular, deep-learning based algorithms have achieved performance on par with human annotation. Despite improved performance, a limitation of most deep-learning based algorithms is their black-box behavior, which have limited their use in clinical settings. Here, we propose a cross-modal transformer, which is a transformer-based method for sleep stage classification. The proposed cross-modal transformer consists of a cross-modal transformer encoder architecture along with a multi-scale one-dimensional convolutional neural network for automatic representation learning. The performance of our method is on-par with the state-of-the-art methods and eliminates the black-box behavior of deep-learning models by utilizing the interpretability aspect of the attention modules. Furthermore, our method provides considerable reductions in the number of parameters and training time compared to the state-of-the-art methods. Our code is available at https://github.com/Jathurshan0330/Cross-Modal-Transformer. A demo of our work can be found at https://bit.ly/Cross_modal_transformer_demo.",

keywords = "Automatic sleep stage classification, deep neural networks, interpretable deep learning, transformers",

author = "Jathurshan Pradeepkumar and Mithunjha Anandakumar and Vinith Kugathasan and Dhinesh Suntharalingham and Kappel, {Simon L.} and {De Silva}, {Anjula C.} and Edussooriya, {Chamira U.S.}",

note = "Publisher Copyright: {\textcopyright} 2001-2011 IEEE.",

year = "2024",

doi = "10.1109/TNSRE.2024.3438610",

language = "English",

volume = "32",

pages = "2893--2904",

journal = "IEEE Transactions on Neural Systems and Rehabilitation Engineering",

issn = "1534-4320",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

Toward Interpretable Sleep Stage Classification Using Cross-Modal Transformers. / Pradeepkumar, Jathurshan; Anandakumar, Mithunjha; Kugathasan, Vinith et al.
In: IEEE Transactions on Neural Systems and Rehabilitation Engineering, Vol. 32, 2024, p. 2893-2904.

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaper › Journal article › Research › peer-review

TY - JOUR

T1 - Toward Interpretable Sleep Stage Classification Using Cross-Modal Transformers

AU - Pradeepkumar, Jathurshan

AU - Anandakumar, Mithunjha

AU - Kugathasan, Vinith

AU - Suntharalingham, Dhinesh

AU - Kappel, Simon L.

AU - De Silva, Anjula C.

AU - Edussooriya, Chamira U.S.

PY - 2024

Y1 - 2024

N2 - Accurate sleep stage classification is significant for sleep health assessment. In recent years, several machine-learning based sleep staging algorithms have been developed, and in particular, deep-learning based algorithms have achieved performance on par with human annotation. Despite improved performance, a limitation of most deep-learning based algorithms is their black-box behavior, which have limited their use in clinical settings. Here, we propose a cross-modal transformer, which is a transformer-based method for sleep stage classification. The proposed cross-modal transformer consists of a cross-modal transformer encoder architecture along with a multi-scale one-dimensional convolutional neural network for automatic representation learning. The performance of our method is on-par with the state-of-the-art methods and eliminates the black-box behavior of deep-learning models by utilizing the interpretability aspect of the attention modules. Furthermore, our method provides considerable reductions in the number of parameters and training time compared to the state-of-the-art methods. Our code is available at https://github.com/Jathurshan0330/Cross-Modal-Transformer. A demo of our work can be found at https://bit.ly/Cross_modal_transformer_demo.

AB - Accurate sleep stage classification is significant for sleep health assessment. In recent years, several machine-learning based sleep staging algorithms have been developed, and in particular, deep-learning based algorithms have achieved performance on par with human annotation. Despite improved performance, a limitation of most deep-learning based algorithms is their black-box behavior, which have limited their use in clinical settings. Here, we propose a cross-modal transformer, which is a transformer-based method for sleep stage classification. The proposed cross-modal transformer consists of a cross-modal transformer encoder architecture along with a multi-scale one-dimensional convolutional neural network for automatic representation learning. The performance of our method is on-par with the state-of-the-art methods and eliminates the black-box behavior of deep-learning models by utilizing the interpretability aspect of the attention modules. Furthermore, our method provides considerable reductions in the number of parameters and training time compared to the state-of-the-art methods. Our code is available at https://github.com/Jathurshan0330/Cross-Modal-Transformer. A demo of our work can be found at https://bit.ly/Cross_modal_transformer_demo.

KW - Automatic sleep stage classification

KW - deep neural networks

KW - interpretable deep learning

KW - transformers

UR - http://www.scopus.com/inward/record.url?scp=85200824552&partnerID=8YFLogxK

U2 - 10.1109/TNSRE.2024.3438610

DO - 10.1109/TNSRE.2024.3438610

M3 - Journal article

C2 - 39102323

AN - SCOPUS:85200824552

SN - 1534-4320

VL - 32

SP - 2893

EP - 2904

JO - IEEE Transactions on Neural Systems and Rehabilitation Engineering

JF - IEEE Transactions on Neural Systems and Rehabilitation Engineering

ER -

Toward Interpretable Sleep Stage Classification Using Cross-Modal Transformers

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this