A new method for classifying target sleep arousal regions is presented in a recent paper by Nox Research at Nox Medical. The paper was published in 2018 Computing in Cardiology (CinC) Proceedings, hosted in the Netherlands as part of the Computing in Cardiology conference last year.
The paper presents an AI based method for classifying target sleep arousal regions of polysomnographies, using recurrent artificial neural networks. The method was validated on PhysioNet Challenge 2018 dataset, which consisted of a training set of 994 subjects and a hidden test set of 989 subjects.
The identification of arousals is important for the evaluation of sleep continuity and for diagnosis of various sleep disorders [1]. However, manual scoring of these events is costly due to the huge amount of data recorded per night, and difficult due to variance across patients and technicians experience [2] [3]. Automation of the detection procedure is therefore important and different works have explored different ways of automating the process. The research results are encouraging, suggesting that the automatic classification of sleep arousals is an achievable task.
The published paper can be found here.
References
- American Academy of Sleep Medicine. International Classification of Sleep Disorders. American Academy of Sleep Medicine, 2014.
- Bonnet MH, Doghramji K, Roehrs T, Stepanski EJ, Sheldon SH,Walters AS,Wise M, Chesson AL. The scoring of arousal in sleep: Reliability, validity, and alternatives. Journal of Clinical Sleep Medicine 2007;03(2):133–145.
- Magalang UJ, Chen NH, Cistulli PA, Fedson AC, Gíslason T, Hillman D, Penzel T, Tamisier R, Tufik S, Phillips G, Pack AI, for the SAGIC Investigators. Agreement in the
scoring of respiratory events and sleep among international sleep centers. Sleep 2013;36(4):591–596.
Photo by Nox Medical Copyright ©2019
Topic: Research & Publications
