Calculating AHI combine oximetery and snore sound

The paper proposes the diagnosis of Sleep Apnea Hypopnea Syndrome (SAHS) only using snore sound and SaO2 signals. The time-frequency characteristics of snore sound signal are extracted to locate the snore events, while SaO2 signal is used to determine whether SaO2 decreased concurrence. Therefore respiratory events can be detected definitely. Finally patient´s Apnea Hypopnea Index (AHI) is calculated by counting the number of detected respiratory events during nocturne. Experiment results show that the sensitivity can reach up to 96% and specificity to 96% as AHI≥5. The accuracy of SAHS with different severity can reach up to 90.8% compared with polysomnography (PSG). It suggests that diagnosis of SAHS from the snore sound and SaO2 signals is possible. The portable devices could be developed to screen SAHS and sleep monitor in home.