Search

Alzheimer's disease dementia (ADD) is the most diffuse neurodegenerative disorder belonging to mild cognitive impairment (MCI) and dementia in old persons. This disease is provoked by an abnormal accumulation of amyloid-beta and tauopathy proteins in the brain. Very recently, the first disease-modifying drug has been licensed with reserve (i.e., Aducanumab). Therefore, there is a need to identify and use biomarkers probing the neurophysiological underpinnings of human cognitive functions to test the clinical efficacy of that drug. In this regard, event-related electroencephalographic potentials (ERPs) and oscillations (EROs) are promising candidates. Here, an Expert Panel from the Electrophysiology Professional Interest Area of the Alzheimer's Association and Global Brain Consortium reviewed the field literature on the effects of the most used symptomatic drug against ADD (i.e., Acetylcholinesterase inhibitors) on ERPs and EROs in ADD patients with MCI and dementia at the group level. The most convincing results were found in ADD patients. In those patients, Acetylcholinesterase inhibitors partially normalized ERP P300 peak latency and amplitude in oddball paradigms using visual stimuli. In these same paradigms, those drugs partially normalize ERO phase-locking at the theta band (4-7 Hz) and spectral coherence between electrode pairs at the gamma (around 40 Hz) band. These results are of great interest and may motivate multicentric, double-blind, randomized, and placebo-controlled clinical trials in MCI and ADD patients for final cross-validation.

[EN] This paper presents two applications of Independent Component Analysis Mixture Modeling (ICAMM) for the classification and prediction of data. The first one of these extensions is Sequential ICAMM (SICAMM), an ICAMM structure that takes into account the sequential dependence in the feature record. This algorithm can be used to classify input observations in a given set of mutually-exclusive classes. The performance of SICAMM is tested with simulations and compared against that of the base ICAMM algorithm and of a Dynamic Bayesian Network (DBN). All three methods are also used to classify real electroencephalographic (EEG) signals to compute hypnograms, a clinical tool used to help in the diagnosis of sleep disorders. The second extension of ICAMM is PREDICAMM, an estimation algorithm that makes use of the ICAMM parameters in order to reconstruct missing samples from a set of data. This predictor is used to reconstruct real EEG data from a working memory experiment, and its performance is compared to that of a classical predictor for EEG signals: sphere splines. Prediction performance is measured with four error indicators: signal-to-interference ratio, KullbackLeibler divergence, correlation, and mean structural similarity index. Both extensions of the base ICAMM algorithm have achieved a higher performance than other methods

[EN] This paper presents two applications of Independent Component Analysis Mixture Modeling (ICAMM) for the classification and prediction of data. The first one of these extensions is Sequential ICAMM (SICAMM), an ICAMM structure that takes into account the sequential dependence in the feature record. This algorithm can be used to classify input observations in a given set of mutually-exclusive classes. The performance of SICAMM is tested with simulations and compared against that of the base ICAMM algorithm and of a Dynamic Bayesian Network (DBN). All three methods are also used to classify real electroencephalographic (EEG) signals to compute hypnograms, a clinical tool used to help in the diagnosis of sleep disorders. The second extension of ICAMM is PREDICAMM, an estimation algorithm that makes use of the ICAMM parameters in order to reconstruct missing samples from a set of data. This predictor is used to reconstruct real EEG data from a working memory experiment, and its performance is compared to that of a classical predictor for EEG signals: sphere splines. Prediction performance is measured with four error indicators: signal-to-interference ratio, KullbackLeibler divergence, correlation, and mean structural similarity index. Both extensions of the base ICAMM algorithm have achieved a higher performance than other methods

This paper presents a new application of independent component analysis mixture modeling (ICAMM) for prediction of electroencephalographic (EEG) signals. Demonstrations in prediction of missing EEG data in a working memory task using classic methods and an ICAMM-based algorithm are included. The performance of the methods is measured by using four error indicators: signal-to-interference (SIR) ratio, Kullback-Leibler divergence, correlation at lag zero and mean structural similarity index. The results show that the ICAMM-based algorithm outperforms the classical spherical splines method which is commonly used in EEG signal processing. Hence, the potential of using mixtures of independent component analyzers (ICAs) to improve prediction, as opposed on estimating only one ICA is demonstrated.