论文部分内容阅读
Background: Loud snoring is one of the principle symptoms of obstructive sleep apnea-hypopnea syndrome (OSAHS).Snoring sounds analyzing has been a potentially cost-effective and reliable alternative for the diagnosis of OSAHS.However, no investigation of the accuracy of snoring signal analyzing for the diagnosis of OSAHS in Chinese Han population has been reported.Objective: To investigate whether whole-night snoring detection and analysis aids diagnosis of OSAHS in the Han population bya new snore-analyzing technique.Materials and Methods: One hundred and thirty-nine snorers were consecutively recruited and one hundred and twenty patients were finally enrolled in this study.Snoring sounds were recorded using a non-contact microphone and polysomnography (PSG) were performedsimultaneously throughout the night.Snoring episodes were automatically recorded and analyzed by an engineer afterwards.Snoring signals rhythm and frequency domainwas analyzed based on frequency energy endpoint detection (FEP) and Earth moves distance (EMD), for each subject in order to harvest EMD-calculated AHI (AHIEMD).Lastly, we compared the AHIEMD with the PSG-monitored AHI (AHIpsG).Results: The accurate rate of AHIEMD compared with AHIpsG was 96.7%, 90.0%, 86.7% and 96.7% in non-, mild, moderate, severe patients with OSAHS respectively.The AHIEMD was correlated with the AHIPSG (r2=0.954, p<0.001).An area under the receiver operating characteristic (ROC) curve of 99.5%,97.6% and 99.8% for AHIEMD thresholds of 5, 15 and 30 events/h, respectively, was obtained for OSAHS detection.Altman-Bland analysis revealed 93.3% agreement of AHIEMD with AHIPSG.Conclusions: This new methodology for identifying OSAHS by analyzing snores is feasible and reliable in the Han population.The snoring sound based technology could be a promising tool for OSAHS screening and diagnosis.Further multicenter studies with larger sample of subjects are warranted to confirm this.