TY - GEN

T1 - Detection of arousals in Parkinson's disease patients

AU - Sørensen, Gertrud Laura

AU - Kempfner, Jacob

AU - Jennum, Poul

AU - Sorensen, Helge B D

PY - 2011

Y1 - 2011

N2 - Arousal from sleep are short awakenings, which can be identified in the EEG as an abrupt change in frequency. Arousals can occur in all sleep stages and the number and frequency increase with age. Frequent arousals during sleep results in sleep fragmentation and is associated with daytime sleepiness. Manual scoring of arousals is time-consuming and the inter-score agreement is highly varying especially for patients with sleep related disorders. The aim of this study was to design an arousal detection algorithm capable of detecting arousals from sleep, in both non-REM and REM sleep in patients suffering from Parkinson's disease (PD). The proposed algorithm uses features from EEG, EMG and the manual sleep stage scoring as input to a feed-forward artificial neural network (ANN). The performance of the algorithm has been assessed using polysomnographic (PSG) recordings from a total of 8 patients diagnosed with PD. The performance of the algorithm was validated using the leave-one-out method resulting in a sensitivity of 89.8 % and a positive predictive value (PPV) of 88.8 %. This result is high compared to previous presented arousal detection algorithms.

AB - Arousal from sleep are short awakenings, which can be identified in the EEG as an abrupt change in frequency. Arousals can occur in all sleep stages and the number and frequency increase with age. Frequent arousals during sleep results in sleep fragmentation and is associated with daytime sleepiness. Manual scoring of arousals is time-consuming and the inter-score agreement is highly varying especially for patients with sleep related disorders. The aim of this study was to design an arousal detection algorithm capable of detecting arousals from sleep, in both non-REM and REM sleep in patients suffering from Parkinson's disease (PD). The proposed algorithm uses features from EEG, EMG and the manual sleep stage scoring as input to a feed-forward artificial neural network (ANN). The performance of the algorithm has been assessed using polysomnographic (PSG) recordings from a total of 8 patients diagnosed with PD. The performance of the algorithm was validated using the leave-one-out method resulting in a sensitivity of 89.8 % and a positive predictive value (PPV) of 88.8 %. This result is high compared to previous presented arousal detection algorithms.

U2 - 10.1109/IEMBS.2011.6090757

DO - 10.1109/IEMBS.2011.6090757

M3 - Conference article

VL - 2011

SP - 2764

EP - 2767

JO - Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings

JF - Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings

SN - 0589-1019

ER -