Evaluating deep neural network for automated sleep staging in real-life scenarios

Dingbang Cao; Congjia Hu; Jiaming Hu; Dailun Li; Wenkang Li

doi:10.1117/12.2640943

3 October 2022 Evaluating deep neural network for automated sleep staging in real-life scenarios

Dingbang Cao, Congjia Hu, Jiaming Hu, Dailun Li, Wenkang Li

Proceedings Volume 12290, International Conference on Computer Network Security and Software Engineering (CNSSE 2022); 122900I (2022) https://doi.org/10.1117/12.2640943
Event: International Conference on Computer Network Security and Software Engineering (CNSSE 2022), 2022, Zhuhai, China

Abstract

Using artificial intelligence in automated sleep staging has become very popular. In this paper, we evaluate a deep learning model, Tiny Sleep Net, to illustrate the limitations of the deep learning model in the task of sleep staging predication. During the experiment, we changed the hyperparameter to achieve various study rates, changed the number of epochs and the balance of datasets. The results showed that the actual result of this model is quite acceptable, which is 82.3%. In contrast, the whole model is relatively fragile when some parameters are changed, witnessing an incredibly low accuracy when predicting. In addition, compared with the overall accuracy, the difference in the per-class accuracy can be drastic. Therefore, these results showed that the neural network is susceptible to the hyperparameter and the exact number of epochs generated. Therefore, in exact situations when the input dataset is not balanced or hyperparameters have minus changes, the final prediction accuracy will decrease dramatically.

Citation Download Citation

Dingbang Cao, Congjia Hu, Jiaming Hu, Dailun Li, and Wenkang Li "Evaluating deep neural network for automated sleep staging in real-life scenarios", Proc. SPIE 12290, International Conference on Computer Network Security and Software Engineering (CNSSE 2022), 122900I (3 October 2022); https://doi.org/10.1117/12.2640943

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
10 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Electroencephalography

Data modeling

Neural networks

Performance modeling

Brain

Feature extraction

Transform theory

Show All Keywords

Keywords/Phrases

Search In:

Publication Years