Bioingeniería
Permanent URI for this collection
Browse
Browsing Bioingeniería by Subject "Apnea obstructiva del sueño"
Now showing 1 - 1 of 1
Results Per Page
Sort Options
Item Detección de la apnea obstructiva del sueño mediante el análisis de la variabilidad de frecuencia cardíaca de muy corto plazo utilizando señales electrocardiográficas(Universidad Santiago de Cali, 2022) Figueroa Peláez, Diana Carolina; Argüello Prada, Erick JavierThe obstructive sleep apnea syndrome (OSAS) is a currently prevalent respiratory disorder, which can cause short, medium and long- term alterations in the health of those who suffer from it. This is due to the cessation or interruption of gas exchange produced during each episode, so is very important it’s identification and diagnosis. However, only a small part of this population is diagnosed and treated, since the costs and inconveniences associated with the standard test used by physicians and specialized sleep centers are high. Based on the aforementioned issues, the present study proposes a method for detecting obstructive sleep apnea (OSAS) by analyzing very short-term heart rate variability (ultra-short-term HRV) of ECG signals. For this purpose, we used the free distribution database The Apnea-ECG Database, provided by PhysioNet/Computers in Cardiology Challenge 2000, from which we calculated 1-min series of RR intervals byusing algorithms that make it possible to extract parameters, both in the time and frequency domains, and non-linear parameters (Poincaré plots), which could suggest the presence of an episode of apnea. According to the results obtained, the parameters showing greater reliability in terms of sensitivity and specificity are the standard deviation of the NN intervals (SDNN), the standard deviation of the successive differences between adjacent NN intervals (SDSD), the square root of the mean of the squares of the successive differences between adjacent NN intervals (RMSSD), and the cardiac vagal index (CVI), being the one with the highest sensitivity (Se) and negative predictive value (VPN) the CVI with Se= 99.73% and VPN= 99.36%, and the one with the highest specificity (Sp) and positive predictive value (VPP) was SDNN with values of Sp= 82.90% and VPP= 67.86%, suggesting that these parameters might be used for the future development of system or devise that can detect OSAS episodes in a short time.