Fuzzy-entropy threshold based on a complex wavelet denoising technique to diagnose Alzheimer disease

Healthcare Technology Letters

article

Prinza Lazar¹  Rajeesh Jayapathy¹  Jordina Torrents-Barrena²  Beena Mol³  Mohanalin⁴  Domenec Puig² 

[1] Department of Electronics and Communication Engineering, Anna University;Department of Computer Engineering and Mathematics, University Rovira i Virgili;Department of Civil Engineering;Department of Electrical and Electronics Engineering

关键词: fuzzy systems;    diseases;    electroencephalography;    medical signal processing;    signal classification;    signal denoising;    wavelet transforms;    wavelet neural nets;    entropy;    mean square error methods;    fuzzy-entropy threshold;    complex wavelet denoising technique;    Alzheimer disease diagnosis;    irregularities;    electroencephalographic signals;    uncertainty;    classification rate;    multiresolution analysis;    optimum threshold;    AD EEG signals;    multiresolution wavelet;    Gaussian membership function;    signal-to-noise ratio;    lower root-mean-square error;    neural network scheme;   

DOI  :  10.1049/htl.2016.0022

学科分类：肠胃与肝脏病学

来源: Wiley

【 摘 要 】

The presence of irregularities in electroencephalographic (EEG) signals entails complexities during the Alzheimer's disease (AD) diagnosis. In addition, the uncertainty presented on EEG raises major issues in the improvement of the classification rate. The multi-resolution analysis through an optimum threshold will likely achieve better results in distinguishing AD and normal EEG signals. Hence, a fuzzy-entropy concept defined in a complex multi-resolution wavelet has been proposed to obtain the most appropriate threshold. First, the complex coefficients are fuzzified using a Gaussian membership function. Afterwards, the ability of the proposed fuzzy-entropy threshold has been compared with traditional thresholds in complex wavelet domain. Experimental results show that the authors’ methodology produces a higher signal-to-noise ratio and a lower root-mean-square error than traditional approaches. Moreover, a neural network scheme is performed along several features to classify AD from normal EEG signals obtaining a specificity of 87.5%.

【 授权许可】

CC BY|CC BY-ND|CC BY-NC|CC BY-NC-ND