An OSAHS evaluation method based on multi-features acoustic analysis of snoring sounds

Jiang, Yanmei; Peng, Jianxin; Song, Lijuan

doi:10.1016/j.sleep.2021.06.012

Cited by 10 publications

(7 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…PR800 800 Hz power ratio represents the ratio of snoring spectral energy above and below 800 Hz. Previous studies (Jiang et al 2021) have shown that this feature has a good effect on distinguishing simple snorers from OSAHS patients. PR800 in frame k is defined as:…”

Section: Formantsmentioning

confidence: 87%

“…However, few studies clearly address how to identify OSAHS patients based on these detected snoring sounds. In the work of Jiang et al (2021) and Sola et al (2007), a subject is considered as OSASH patient if the number of OSAHS related snoring sounds is equal to or greater than simple snoring sounds during the whole night. Based on this definition of OSAHS patients, their works both achieve relatively high accuracy for OSAHS patients' detection.…”

Section: Discussionmentioning

confidence: 99%

“…It is important to distinguish OSAHS patients and simple snorers by analysis of some biological signals rather than PSG to avoid the waste of medical resources. It is has been demonstrated by previous studies that there is a significantly different distribution of acoustic features between OSAHS patients and simple snorers (Fiz et al 1996, Jané et al 2000, Koo et al 2017, Jiang et al 2021. The work of Ng et al (2008) showed that snoring sounds of OSAHS patients had higher formants.…”

Section: Introductionmentioning

confidence: 88%

“…OSAHS is characterized by multiple partial or complete collapses of the upper airway structure during sleep that results in hypopnea and apnea events (Berry et al 2012). Apnea-Hypopnea Index (AHI) is a recognized indicator to evaluate the severity of OSAHS patients, which represents the average number of hypopnea and apnea events per hour throughout the whole night (Hafezi et al 2020, Jiang et al 2021, Malhotra et al 2021. According to the American Sleep Medicine Society (AASM) (2012), snorers are recognized as simple snorers, mild, moderate, and severe OSAHS patients with AHI 5, 5<AHI 15, 15 < AHI 30, and AHI > 30 respectively.…”

Section: Introductionmentioning

confidence: 99%

“…The model achieved high sensitivity with 94.1%, but low specificity with 73.7%. Jiang et al (2021) applied Top-6 acoustic features selected by random forest algorithm and classified the two kinds of snoring sounds using logistic regression model. Their work could exactly distinguish simple snorers and OSAHS patients.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Automatic identifying OSAHS patients and simple snorers based on Gaussian mixture models

et al. 2023

Self Cite

View full text Add to dashboard Cite

Snoring is a typical symptom of obstructive sleep apnea-hypopnea syndrome (OSAHS). In this study, an effective OSAHS patient detection system based on subject independence using snoring sounds is presented. A series of acoustic features including Mel cepstral coefficient, bark sub-band feature, perceptual linear prediction coefficient, pitch frequency, formant, 800Hz power ratio, spectral entropy, and gammatone cepstral coefficient are extracted. The feature selection based on the Fisher ratio is conducted and the Gaussian mixture models (GMM) of OSAHS patients and simple snorers were established respectively. For the test subject, the difference in log-likelihood probabilities of the two GMMs is calculated to identify the OSAHS patients. The proposed model achieves 90% accuracy and 95.65% precision based on leave-one-subject-out cross-validation using selected features with a dimension of 100. The average prediction time of the proposed model is 0.134 ± 0.005s. The promising results demonstrate the effectiveness and low computational cost of diagnosing OSAHS patients using snoring sounds at home.

show abstract

Section: Formantsmentioning

confidence: 87%

Section: Discussionmentioning

confidence: 99%