
The same procedure is shown in Figure 6c with
only the CNN features. A similar behavior is present,
as sleep stages seem to group in certain areas with
some overlappings. Judging by the coloring of the
manually annotated sleep stages, the CNN features
can group the sleep stages in a relatively accurate
way. Finally, Figure 6c enhances the explainability
of the model, since it shows which classes are most
frequently misclassified and in which areas.
5 CONCLUSIONS
We introduce an advanced sleep monitoring system
which combines AI-based analysis and interactive vi-
sualization tools. Three key components—a spin-
dle detection method, a sleep stage identification
model and a two-dimensional embedding of sleep in-
tervals—combined with raw signal visualization in
an interactive dashboard, enable the system to im-
plement a multi-view approach. The detection of
spindles on raw EEG data is a powerful tool that
can enhance the capabilities of sleep analysis. The
stage classification model demonstrated varying per-
formance across sleep stages which reflects the inher-
ent complexity of sleep classification. Moreover, t-
SNE visualization with large datasets can place lim-
itations due to its high computational cost. Future
work includes the installation of the system at the
CUB premises, and a thorough assessment of its use-
fulness and usability with standardized questionnaires
(e.g. SUS scale) after a pilot usage. In addition, we
will focus on quantitative evaluation of spindle detec-
tion using expert ground truth and extending the 2D
visualization framework for apnea analysis.
ACKNOWLEDGMENT
This work has been supported by the EU H2020
project ODIN (H2020-DT-ICT-12-2020, grant agree-
ment no. 101017331).
REFERENCES
Caballero, H., Westenberg, M., Gebre, B., and Wijk, J.
(2019). V-awake: A visual analytics approach for cor-
recting sleep predictions from deep learning models.
Computer Graphics Forum, 38:1–12.
Chawla, N. V., Bowyer, K. W., Hall, L. O., and Kegelmeyer,
W. P. (2002). Smote: Synthetic minority over-
sampling technique. Journal of Artificial Intelligence
Research, 16:321–357.
Combrisson, E., Vallat, R., Eichenlaub, J.-B., O’Reilly, C.,
Lajnef, T., Guillot, A., Ruby, P. M., and Jerbi, K.
Sleep: An open-source python software for visualiza-
tion, analysis, and staging of sleep data.
Fogel, S. M. and Smith, C. T. (2011). The function of the
sleep spindle: A physiological index of intelligence
and a mechanism for sleep-dependent memory con-
solidation. Neuroscience & Biobehavioral Reviews,
35(5):1154–1165.
Guo, Y., Nowakowski, M., and Dai, W. (2024). Flexsleep-
transformer: a transformer-based sleep staging model
with flexible input channel configurations. Scientific
Reports, 14.
Hochreiter, S. and Schmidhuber, J. (1997). Long short-term
memory. Neural computation, 9:1735–80.
Kaulen, L., Schwabedal, J. T. C., Schneider, J., Ritter, P.,
and Bialonski, S. (2022). Advanced sleep spindle
identification with neural networks. Scientific Reports,
12(1).
Lacourse, K., Delfrate, J., Beaudry, J., Peppard, P., and
Warby, S. C. (2019). A sleep spindle detection al-
gorithm that emulates human expert spindle scoring.
Journal of Neuroscience Methods, 316:3–11. Methods
and models in sleep research: A Tribute to Vincenzo
Crunelli.
Lema
ˆ
ıtre, G., Nogueira, F., and Aridas, C. K. (2017).
Imbalanced-learn: A python toolbox to tackle the
curse of imbalanced datasets in machine learning.
Journal of Machine Learning Research, 18(17):1–5.
Liang, Z., Ploderer, B., Liu, W., Nagata, Y., Bailey, J., Ku-
lik, L., and Li, Y. (2016). Sleepexplorer: A visualiza-
tion tool to make sense of correlations between per-
sonal sleep data and contextual factors. Personal and
Ubiquitous Computing, 20:985–1000.
Loh, H. W., Ooi, C. P., Vicnesh, J., Oh, S. L., Faust, O., Ger-
tych, A., and Acharya, U. R. (2020). Automated de-
tection of sleep stages using deep learning techniques:
A systematic review of the last decade (2010–2020).
Applied Sciences.
Markun, L. C. and Sampat, A. (2023). Clinician-
focused overview and developments in polysomnog-
raphy. Journal of Clinical Sleep Medicine.
Morokuma, S., Hayashi, T., Kanegae, M., Mizukami, Y.,
Asano, S., Kimura, I., Tateizumi, Y., Ueno, H., Ikeda,
S., and Niizeki, K. (2023). Deep learning-based
sleep stage classification with cardiorespiratory and
body movement activities in individuals with sus-
pected sleep disorders. Scientific Reports, 13.
Supratak, A., Dong, H., Wu, C., and Guo, Y. (2017). Deep-
sleepnet: A model for automatic sleep stage scoring
based on raw single-channel eeg. IEEE Transactions
on Neural Systems and Rehabilitation Engineering,
25(11):1998–2008.
Tapia, N. I. and Est
´
evez, P. A. (2020). Red: Deep recur-
rent neural networks for sleep eeg event detection. In
2020 International Joint Conference on Neural Net-
works (IJCNN), pages 1–8.
van der Maaten, L. and Hinton, G. (2008). Visualizing data
using t-sne. Journal of Machine Learning Research,
9(86):2579–2605.
IVAPP 2025 - 16th International Conference on Information Visualization Theory and Applications
990