Real Time Data Processing for Detection of Apnea using Android Phone
S. R. Patil1, Prachi Shewale2, Aditi Agrawal3, Vandana Choudhari4, Balika Doke5
1Prof. Dr.S.R.Patil, Computer Department, University of Pune ,Bharati Vidyapeeth College of Engineering For women, Pune, India.
2Miss. Prachi Shewale, Computer Department, University of Pune ,Bharati Vidyapeeth College of Engineering for Women, Pune, India.
3Miss. Aditi Agrawal, Computer Department, University of Pune ,Bharati Vidyapeeth College of Engineering for Women, Pune, India.
4Miss. Vandana Choudhari, Computer Department, University of Pune ,Bharati Vidyapeeth College of Engineering for Women, Pune, India.
5Miss. Balika Doke, Computer Department, University of Pune ,Bharati Vidyapeeth College of Engineering for Women, Pune, India.
Manuscript received on January 05, 2014. | Revised Manuscript received on January 11, 2014. | Manuscript published on January 15, 2014. | PP: 1-5 | Volume-2 Issue-2, January 2014. | Retrieval Number: A0556122113/2014©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Sleep apnea (or sleep apnoea in British English) is a type of sleep disorder characterized by pauses in breathing or instances of shallow or infrequent breathing during sleep. Each pause in breathing, called an apnea, can last from at least ten seconds to several minutes, and may occur 5 to 30 times or more an hour. Similarly, each abnormally shallow breathing event is called a hypopnea. Sleep apnea is often diagnosed with an overnight sleep test called a polysomnogram, or “sleep study”. The final diagnosis of sleep apnea is established by an overnight polysomnography (PSG) that involves the recording and the studying of several neurologic and cardio-respiratory signals. Those PSGs are carried out in sleep laboratories with attending systems and specialized staff. Because these studies are expensive, it is very relevant to find reliable diagnostic alternatives using fewer biological signals and providing a high level of usability. Identifying the presence of sleep apneas from blood oxygen saturation signal fragments taken from pulsioximetry systems (SPO2). In order to build the classifier, all the methods with which we worked were trained and tested with annotated SpO2 signals available in the Apnea-ECG Database. Another additional requirement we considered was that the classifier should run in real time using, at each particular moment, past information in the SpO2 signal and not information contained in the whole signal. Moreover, we implemented a monitoring system that detects apneic events in real time while the patient is sleeping, which can be sometimes used as a valid alternative to PSGs. This monitoring system constitutes of a desktop application consisting historical database and a mobile device in which our apnea classifier runs performing a local real-time analysis that allows the system to take an active role in the monitoring process. This system can also record patients’ nocturnal pulsioximetry and send data to a specific health center to be evaluated by qualified medical staff.
Keywords: Data mining, real-time monitoring, sleep apnea and hypopnea syndrome (SAHS) detection, SpO2 signal analysis.