To diagnose sleep disorders, hours of sleep data from lots of different physiological sensors have to be analyzed. To do so, experts have to look through all the data which is time-consuming and error-prone. Automatic detection and classification of sleep related breathing disorders and arousals would significantly simplify this task. This years Physionet/CinC Challenge deals with this topic. This paper examines the use of a Long Short-Term Memory network for automatic arousal detection. On the test set, an AUPRC score of 0.14 was achieved.
CITATION STYLE
Schellenberger, S., Shi, K., Mai, M., Wiedemann, J. P., Steigleder, T., Eskofier, B., … Kolpin, A. (2018). Detecting Respiratory Effort-Related Arousals in Polysomnographic Data Using LSTM Networks. In Computing in Cardiology (Vol. 2018-September). IEEE Computer Society. https://doi.org/10.22489/CinC.2018.104
Mendeley helps you to discover research relevant for your work.