Boosting Based Classification of Event Related Potentials for Early Diagnosis of Alzheimer´s Disease

Author

Stepenosky, Nicholas ; Topalis, Apostolos ; Syed, Hussein ; Green, Deborah ; Kounios, John ; Clark, Christopher ; Polikar, Robi

Author_Institution

Dept. of Electr. & Comput. Eng., Rowan Univ., Glassboro, NJ

fYear

2005

fDate

6/27/1905 12:00:00 AM

Firstpage

2494

Lastpage

2497

Abstract

With the number of the elderly population affected by Alzheimer´s disease (AD) rising, the need to find an accurate, inexpensive and non-intrusive procedure that can be made available to community healthcare providers for early diagnosis of Alzheimer´s disease is becoming more and more urgent as a major health concern. Several recent studies have looked at analyzing electroencephalogram signals through the use of wavelets and neural networks. In this study, multiresolution wavelet analysis, coupled with the ensemble of classifiers based boosting algorithm is used on the P300 component of the event related potentials (ERP) to determine the feasibility of the approach as a diagnostic tool for early diagnosis of AD. The technique and its promising initial results are presented

Keywords

bioelectric potentials; diseases; electroencephalography; geriatrics; medical signal processing; neural nets; patient diagnosis; signal classification; wavelet transforms; Alzheimer disease; P300 component; boosting-based classification; classifiers; early diagnosis; electroencephalogram signals; event related potentials; healthcare; multiresolution wavelet analysis; neural networks; Algorithm design and analysis; Alzheimer´s disease; Boosting; Enterprise resource planning; Medical services; Neural networks; Senior citizens; Signal analysis; Signal resolution; Wavelet analysis; Alzheimer´s disease; ERP; P300; artificial neural networks; boosting; oddballparadigm; wavelet analysis;

fLanguage

English

Publisher

ieee

Conference_Titel

Engineering in Medicine and Biology Society, 2005. IEEE-EMBS 2005. 27th Annual International Conference of the

Conference_Location

Shanghai

Print_ISBN

0-7803-8741-4

Type

conf

DOI

10.1109/IEMBS.2005.1616975

Filename

1616975