基于声学信息检测阻塞性睡眠呼吸暂停的研究进展

打开文本图片集

中图分类号：R563.8 文献标志码：A DOI：10.11958/20250966

Newadvances on detecting obstructive sleep apnea based on acoustic information

YUHui1，LIU Hao12，CAIFengli3，ZHAO Jing1，BAI Xiangsen1，TIAN Guoliang1， ZHANG Hanyuel， ZHANG Liyuan4 1 Departmentof Biomedical Enginering，Tianjin UniversityTianjin3Oo2，China; 2 DepartmentofNetwork Iforation Tianjin 4th Center Hospital; 3 Department ofObstetrics and Gynecology， Juye County Hospitalof Traditional Chinese Medicine，4DepartmentofEquipmentandMaterial，Tianjin4th CentreHospital （2 Δ Corresponding Author E -mail： 13752631906@163.com

Abstract：Obstructive sleep apnea （OSA） is a common sleep disorder characterized byrepeated episodes of upper airwaycolapseandobstructionduring slep.PolysomnographyisthegoldstandardfordiagnosingOSA，butitisexpensive， time-consuming，and cancause discomfort forpatients.Inrecent years，acoustic-based approaches fordetecting OSA have emerged as aresearch focus.This review summarizes recent advances in OSA automatic detection techniques based on snoringandspeechsignals，andsystematicallyexaminestheirappicationsindagnosis，severityassessment，andlocaliation of obstructionsites.Findings indicate that theacousticfeatures ofsnoringand speechsignals hold significantvalueforOSA screning，and whencombined with machine learning anddeep learning models，itcanachieve high diagnosticacuracy. Futureresearchshould focuson elucidating therelationship betweenacousticfeaturesand thepathophysiological mechanismsofOSA，integratingmultimodal information，andadvancingtheclinicalapplicationofwearabledevices，withthe aim of promoting intelligent， non-invasive，and cost-effective screening technologies for OSA.

Key Words： sleep apnea，obstructive; snoring; speech; detection; obstruction location; severity

阻塞性睡眠呼吸暂停（OSA）是一种常见的睡眠障碍疾病，主要特征是睡眠期间上气道的反复塌陷和阻塞，导致呼吸暂停或低通气事件。（剩余18941字）