An Intelligent Algorithm for Home Sleep Apnoea Test Device

An Intelligent Algorithm for Home Sleep Apnoea Test Device

Ahsan H. Khandoker (The University of Melbourne, Australia)
Copyright: © 2012 |Pages: 15
DOI: 10.4018/978-1-60960-818-7.ch516
OnDemand PDF Download:
No Current Special Offers


In this chapter authors try to develop a system for Sleep apnea with the help of machine learning algorithms using ECG signals.  The application of an intelligent machine learning technique (Support Vector Machines, SVM) to diagnose the patients with sleep apnea syndrome using Electrocardiogram (ECG) signal.  Sleep apnea syndrome is a medical condition caused by sleep apnea which is defined as the cessation of breathing for short periods during sleep.
Chapter Preview


Obstructive sleep apnoea syndrome (OSAS) is a common problem defined by frequent cessation of breathing due to the partial or complete obstruction of upper airway for short periods during sleep. This is typically accompanied by a reduction in blood oxygen saturation and leads to wakening from sleep in order to breathe. It is a common sleep related breathing disorder with a reported prevalence of 4% in adult men and 2% in adult women (Young, 1993).Excessive daytime sleepiness is the most common complaint. The fragmented sleep due to OSAS can result in poorer daytime cognitive performance, increased risk of motor vehicle and workplace accidents, depression, diminished sexual function, and memory loss (Coleman, 1999; Neito, 2000). Undiagnosed OSAS is now regarded as an important risk factor for the development of cardiovascular diseases (e.g. hypertension, stroke, congestive heart failure, left ventricular hypertrophy, acute coronary syndromes) (Young, 1997). OSAS can be treated by applying continuous positive airway pressure (CPAP) through the nose which prevents upper airway from collapsing. If patients are identified and then treated at an early stage of OSAS, the adverse health effects can be reduced (Dimsdale, 2000). Therefore, early recognition of subjects at risk of OSAS is essential.

The severity of OSAS is typically quantified by the number of apneas and hypopneas per hour of sleep, a quantity that has been termed Apnea-Hypopnea Index (AHI). Different populations have different AHI values. Specific cutoffs are typically used to establish the diagnosis of OSAS (Polysomnography task force, 1997; Flemons, 2003). For example, as of this writing, the Medicare criteria for reimbursement of continuous positive airway pressure (CPAP) therapy are AHI ≥15 events/hour, or AHI ≥5 events/hour associated with symptoms (e.g., daytime somnolence and fatigue). However, a variety of AHI thresholds ranging between 5 and 40 have been used as suggestive of OSAS in different studies (Khandoker, 2009a). Approximately two to four percent of middle-aged women and men, respectively, have been estimated to have an AHI≥15 events/hour and excessive daytime somnolence in the population-based Wisconsin Sleep Cohort Study (Young, 1993).Using an AHI cutoff of ≥5 events/hour without the symptoms of excessive daytime sleepiness puts the prevalence at 9% for women and 24% for men. The symptom of excessive daytime sleepiness is quite variable and not always present in patients with OSAS. Thus, most people suffering from OSAS remain undiagnosed and untreated (Khandoker, 2009a). More recent studies also suggest a high prevalence (i.e., prevalence of AHI ≥ 5 in adults age 30-69 of 17%), perhaps due to increasing obesity rates in later years (Young, 2005).

Complete Chapter List

Search this Book: