RT info:eu-repo/semantics/conferenceObject
T1 Pulse rate variability analysis to enhance oximetry as at-home alternative for sleep apnea diagnosing
A1 Gutiérrez Tobal, Gonzalo César
A1 Álvarez González, Daniel
A1 Vaquerizo Villar, Fernando
A1 Barroso García, Verónica
A1 Martín Montero, Adrián
A1 Crespo Senado, Andrea
A1 Campo Matias, Félix del
A1 Hornero Sánchez, Roberto
AB This study focuses on the at-home Sleep apnea-hypopneasyndrome (SAHS) severity estimation. Three percentoxygen desaturation index ðODI3Þ from nocturnalpulse-oximetry has been commonly evaluated as simplifiedalternative to polysomnography (PSG), the standardin-hospital diagnostic test. However, ODI3 has shownlimited ability to detect SAHS as it only sums upinformation from desaturation events. Other physiologicalsigns of SAHS can be found in respiratory and cardiacsignals, providing additional helpful data to establishSAHS and its severity. Pulse rate variability time series(PRV), also derived from nocturnal oximetry, is considereda surrogate for heart rate variability, which providesboth cardiac and respiratory information. In this study,200 oximetric recordings obtained at patients home wereinvolved, divided into training (50%) and test (50%)groups. ODI3 and PRV were obtained from them, thelatter being characterized by the extraction of statisticalfeatures in time domain, as well as the spectral entropyfrom the commonly used very low (0–0.04 Hz.), low(0.04–0.15 Hz.), and high (0.15–0.4 Hz.) frequencybands. The ODI3 and PRV features were joined in amulti-layer perceptron artificial neural network (MLP),trained to estimate the apnea-hypopnea index (AHI),which is the PSG-derived parameter used to diagnoseSAHS. Our results showed that single ODI3 rightlyassigned 62.0% of the subjects from the test group intoone out the four SAHS severity degrees, reaching 0.470Cohens kappa, and 0.840 intra-class correlationcoefficient (ICC) with the actual AHI (accuracies of90.0, 88.0 and 82.0% in the increasing AHI cutoffs usedto define SAHS severity). By contrast, our MLP modelrightly assigned 75.0% of the subjects into their correspondingSAHS severity level, reaching 0.614 j and0.904 ICC (accuracies of 93.0, 88.0 and 90.0%). Theseresults suggest that SAHS diagnosis could be accuratelyconducted at-patients home by combining ODI3 and PRVfrom nocturnal oximetry
YR 2018
FD 2018
LK http://uvadoc.uva.es/handle/10324/31359
UL http://uvadoc.uva.es/handle/10324/31359
LA eng
NO Producción Científica
DS UVaDOC
RD 23-nov-2024