摘要
肺音听诊是医生诊断各种呼吸系统疾病的一种简单、无创的方法,可为检测和鉴别呼吸病理提供及时而有用的信息。然而,肺音中的噪声污染是采集肺音时不可避免的问题,在医生诊断患者时,噪声的存在会对肺音分析产生严重阻碍,因此有效去除多余噪声更有助于准确、客观地评价肺音信息。论文对不同去噪技术进行了讨论,包括小波去噪、最小均方自适应滤波、谱减法,评估了三种去噪性能指标,得出了相关比较性结论。
Lung sound auscultation is a simple and non-invasive method for doctors to diagnose various respiratory diseases,which can provide timely and useful information for the detection and differentiation of respiratory pathology.However,noise pollution in lung sounds is an inevitable problem when collecting lung sounds.When doctors diagnose patients,the presence of noise will produce serious obstacles to the analysis of lung sound,so the effective removal of excess noise is more conducive to accurate and objective evaluation of lung sound information.In this paper,different denoising techniques are discussed,including wavelet denoising,least-mean square adaptive filtering and spectral subtraction.The three denoising performance indexes are evaluated,and the relative comparative conclusions are obtained.
作者
郭涛
陈梦凡
李进
石帅
GUO Tao;CHEN Mengfan;LI jin;SHI Shuai(Key Laboratory of Instrumentation Science&Dynamic Measurement(North University of China),Ministry of Education,Taiyuan 030051;China Academy of Launch Vehicle Technology,China Aerospace Science and Technology Corporation,Beijing 100000)
出处
《计算机与数字工程》
2024年第4期1098-1102,共5页
Computer & Digital Engineering
基金
国家自然科学基金项目(编号:51975541)资助。
关键词
肺音
小波分析
最小均方自适应滤波
谱减法
lung sound
wavelet analysis
least mean square adaptive filtering
spectral subtraction
