Comparison of Blind Source Separation Preprocessings Applied to Magnetoencephalogram Recordings to Improve the Classification of Alzheimer’s Disease Patients
نویسندگان
چکیده
This study compares diverse Blind Source Separation (BSS) techniques applied to magnetoencephalogram (MEG) background activity in order to improve the classification of Alzheimer’s Disease (AD) patients. MEG recordings from 18 AD patients and 13 control subjects were decomposed with the following BSS algorithms: AMUSE, SOBI, FastICA, and extended Infomax. Whereas AMUSE ranked the extracted BSS components naturally, the SOBI, FastICA, and Infomax sources were ordered considering their spectral content by increasing Median Frequency (MF). For each BSS algorithm, a one-way analysis of variance with age as a covariate was applied to define a subset of components with the most significant differences between subject groups. Then, these relevant subsets of components were used to partially reconstruct the MEG signals. ROC curves and linear discriminant analysis were used to assess the classification of the subjects’ MF values with and without the BSS preprocessing. The results indicated that the SOBI preprocessing increased the classification accuracy from 77.4% to 80.6% and that the algorithms AMUSE, SOBI, and extended Infomax improved the area under the ROC curve.
منابع مشابه
Assessment of classification improvement in patients with Alzheimer's disease based on magnetoencephalogram blind source separation
OBJECTIVES In this pilot study, we intended to assess whether a procedure based on blind source separation (BSS) and subsequent partial reconstruction of magnetoencephalogram (MEG) recordings might enhance the differences between MEGs from Alzheimer's disease (AD) patients and elderly control subjects. MATERIALS AND METHODS We analysed MEG background activity recordings acquired with a 148-ch...
متن کاملExplorer Blind source separation to enhance spectral and non - linear features of magnetoencephalogram recordings
متن کامل
Early Detection of Alzheimer’s Disease by Blind Source Separation and Bump Modeling of EEG Signals
The early detection Alzheimer’s disease is an important challenge. Using blind source separation, wavelet time-frequency transforms and “bump modeling” of electro-encephalographic (EEG) recordings, a set of features describing the recordings of mildly impaired patients and of controls subject is built. Feature selection by the random probe method leads to the selection of a few reliable feature...
متن کاملBlind source separation to enhance spectral and non-linear features of magnetoencephalogram recordings. Application to Alzheimer's disease.
This work studied whether a blind source separation (BSS) and component selection procedure could increase the differences between Alzheimer's disease (AD) patients and control subjects' spectral and non-linear features of magnetoencephalogram (MEG) recordings. MEGs were acquired with a 148-channel whole-head magnetometer from 62 subjects (36 AD patients and 26 controls), who were divided rando...
متن کاملEarly Detection of Alzheimer's Disease by Blind Source Separation, Time Frequency Representation, and Bump Modeling of EEG Signals
The early detection Alzheimer’s disease (AD) is an important challenge. In this paper, we propose a novel method for early detection of AD using electroencephalographic (EEG) recordings: first a blind source separation algorithm is applied to extract the most significant spatio-temporal components; these components are subsequently wavelet transformed; the resulting timefrequency representation...
متن کامل