Automated sleep scoring and sleep apnea detection in children David P. Baraglia a , Matthew J. Berryman a , Scott W. Coussens b , Yvonne Pamula c , Declan Kennedy c , A. James Martin c and Derek Abbott a a Centre for Biomedical Engineering and School of Electrical and Electronic Engineering, The University of Adelaide, SA 5005, Australia. b Department of Physiology, School of Moleculaar and Biomedical Sciences, The University of Adelaide, SA 5005, Australia. c Department of Pulmonary Medicine, Women’s and Children’s Hospital, 72 King William Road, North Adelaide, SA 5006, Australia. ABSTRACT This paper investigates the automated detection of a patient’s breathing rate and heart rate from their skin conductivity as well as sleep stage scoring and breathing event detection from their EEG. The software developed for these tasks is tested on data sets obtained from the sleep disorders unit at the Adelaide Women’s and Children’s Hospital. The sleep scoring and breathing event detection tasks used neural networks to achieve signal classification. The Fourier transform and the Higuchi fractal dimension were used to extract features for input to the neural network. The filtered skin conductivity appeared visually to bear a similarity to the breathing and heart rate signal, but a more detailed evaluation showed the relation was not consistent. Sleep stage classification was achieved with and accuracy of around 65% with some stages being accurately scored and others poorly scored. The two breathing events hypopnea and apnea were scored with varying degrees of accuracy with the highest scores being around 75% and 30%. Keywords: Sleep, apnea, EEG, neural network 1. INTRODUCTION 1.1 Objectives This paper investigates the development of software capable of automatically detecting sleeping disorders in children. Three tasks have been proposed for the software: 1. the measurement of depth of sleep of a patient, 2. the detection of breathing irregularities during sleep, and 3. the extraction of heart and breathing signals of a patient. The first task was to investigate the use filtering techniques on the skin conductivity to extract the heart rate and breathing rate from only the skin conductivity. The second task was to develop software capable of accurately scoring a patient’s sleep. This task investigated the effectiveness of the two signal processing techniques of Fourier transform and fractal dimension in the classification of sleep stages. The third task investigated the detection of breathing events using only the EEG as input data. The same two signal processing techniques as in the second task were used in this task as well. Data sets have been provided from the Sleep Disorders Unit at the Adelaide Women’s and Children’s Hospital. The data includes a set of electroencephalograph (EEG), thoracic and abdominal breathing data and skin conductivity data. By using various signal processing techniques on the data it is hoped that the depth of sleep of a patient can be accurately estimated as well as the detection of occurrences of events during sleep where breathing is obstructed. It is hoped that the results of this project will contribute towards the development of automated systems for the diagnosis of sleep apnea in children. Complex Systems, edited by Axel Bender, Proc. of SPIE Vol. 6039, 60390T, (2006) · 0277-786X/06/$15 · doi: 10.1117/12.638867 Proc. of SPIE Vol. 6039 60390T-1