Sleep Apnea Syndrome Screening by Tri-axial Accelerometer, Oximeter and Phenotype Information
| Autor: | Wu, Jhao-Cheng., 吳朝成 |
|---|---|
| Rok vydání: | 2017 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 106 Sleep apnea syndrome (SAS) is a well-known sleep disorder nowadays. People suffering SAS cease breathing during sleeping because airway are partially or completely blocked caused by upper respiratory collapse repeatedly. SAS reduces quality of life by daytime sleepiness, memory declination, sexual dysfunction, and even myocardial infarction or cerebral vascular accident during sleeping. In general, people hardly notice SAS and need to do overnight sleep test, which is called polysomnography (PSG) in the sleep center of the hospital. PSG test is a limited environment, uncomfortable and expensive examination. In order to simplify the complex diagnosis, this thesis uses two tri-axial accelerometers (TAA), which are low-cost and small sensors, stick on the left side of thorax and abdomen separately to sense the thoracic movement (THO) and abdominal movement (ABD) signals. This thesis proposed a sleep apnea / hypopnea event detection algorithm by THO, ABD movement signals and blood oxygen saturation levels (SpO2) signal during sleep. The proposed algorithm first combines two three-dimensional signals for two tri-axial accelerometers (TAA) into two one-dimensional signals, TAA-ABD and TAA-THO, respectively, and segments the overnight recorded TAA-ABD and TAA-THO signals into 10-second windows to extract two features from both TAA-ABD and TAA-THO, fundamental frequency ratio and 95% quantile amplitude ratio. The SpO2 signal is segmented into 20-second windows with six features, minimum, maximum, median, mean, variance of the first derivative, and difference between median and minimum. The proposed algorithm uses a SpO2 desaturation detector to detect SpO2 desaturation, and a support vector machine (SVM) with ten features to construct classifiers and a SVM state machine to identify the apnea and hypopnea events. In this study, inspired by the physiological knowledge, we propose a phenotype-based approach. It is well known that the gender, body-mass index (BMI), and age are intimately related to the sleep apnea pattern and severity. With the phenotypical information we can reduce the training subject from 63 to 15 to improve our training speed. And achieve 82:26% accuracy of AHI classification. The results indicates that the proposed algorithm has great potential to classify the severity of patients in clinical examinations for both the screening and the homecare purposes. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|