TY  - CHAP
U1  - Konferenzveröffentlichung
A1  - Boiko, Andrei
A1  - Gaiduk, Maksym
A1  - Seepold, Ralf
A1  - Martínez Madrid, Natividad
T1  - Accelerometer based system for unobtrusive sleep apnea detection
T2  - 27th International Conference on Knowledge Based and Intelligent Information and Engineering Sytems, KES2023, 6 - 8 September, Athens, Greece (Procedia Computer Science, Vol. 225)
N2  - Sleep is an essential part of human existence, as we are in this state for approximately a third of our lives. Sleep disorders are common conditions that can affect many aspects of life. Sleep disorders are diagnosed in special laboratories with a polysomnography system, a costly procedure requiring much effort for the patient. Several systems have been proposed to address this situation, including performing the examination and analysis at the patient's home, using sensors to detect physiological signals automatically analysed by algorithms. This work aims to evaluate the use of a contactless respiratory recording system based on an accelerometer sensor in sleep apnea detection. For this purpose, an installation mounted under the bed mattress records the oscillations caused by the chest movements during the breathing process. The presented processing algorithm performs filtering of the obtained signals and determines the apnea events presence. The performance of the developed system and algorithm of apnea event detection (average values of accuracy, specificity and sensitivity are 94.6%, 95.3%, and 93.7% respectively) confirms the suitability of the proposed method and system for further ambulatory and in-home use.
KW  - Contactless measurement
KW  - Accelerometer
KW  - Vital signals
KW  - Health monitoring
KW  - Sleep apnea
Y1  - 2023
SN  - 1877-0509
SS  - 1877-0509
U6  - https://doi.org/10.1016/j.procs.2023.10.148
DO  - https://doi.org/10.1016/j.procs.2023.10.148
N1  - Corresponding author: Andrei Boiko
VL  - 225
SP  - 1592
EP  - 1600
PB  - Elsevier
CY  - Amsterdam
ER  -