Screening Sleep Disordered Breathing using ECG Signals


M. Palaniswami;
Post Doctoral Research Fellow: Ahsan Khandoker, Jayavardhana Gubbi.

Student: Chandan Karmakar, Kris Nilsen.

Eugene Zilberg, Compumedics Ltd;
David Burton, Compumedics Ltd.

Introduction: Sleep apnoea hypopnea syndrome (SAHS) is a common sleep related breathing disorder that is usually diagnosed through expensive studies in sleep laboratories. Undiagnosed SAHS is associated with cardiovascular morbidity as well as excessive daytime sleepiness and poor quality of life.
Significance: If SAHS could be diagnosed using only ECG recordings, it could be possible to diagnose SAHS inexpensively from ECG recordings acquired in the patient?s home.
Applications: This in turn may help prioritize patients, so that those in greatest need of treatment will undergo full PSG recordings in a timely manner, while those without apnoea will be able to avoid this tedious procedure.
Challenges: In this collaborative research project, we are working on developing algorithms to detect sleep disordered breathing based on ECGs and nonlinear modelling of heart rate variability. We investigated into different machine learning techniques [support vector machines (SVM), Neural Network (NN), Quadratic discriminant (QD) model ] with an aim to find an appropriate model for the automatic detection of SAHS types from their respective overnight ECG recordings and estimation of relative degree of sleep disordered breathing. This investigation may provide essential information for introducing a novel screening device that can aid sleep specialist or other physicians in the initial assessment of patients with suspected SAHS and estimate the relative risk of sleep related breathing disorder, thereby indicating need for referral for overnight sleep studies [i.e. polysomnogram (PSG) recording].
top of page