There is evidence in electroencephalography that alpha, theta and delta band oscillations reflect cognitive and memory performances and that quantitative techniques can improve the electroencephalogram (EEG) sensitivity. This paper presents the results of comparative analysis of qEEG variables as reliable markers for Alzheimer's disease (AD). We compared the sensitivity and specificity between spectral analysis (spectA) and coherence (Coh) within the same group of AD patients. SpectA and Coh were calculated from EEGs of 40 patients with mild to moderate AD and 40 healthy elderly controls. The peak of spectA was smaller in the AD group than in controls. AD group showed predominance of slow spectA in theta and delta bands and a significant reduction of inter-hemispheric Coh for occipital alpha 2 and beta 1 and for frontal delta sub-band. ROC curve supported that alpha band spectA was more sensitive than coherence to differentiate controls from AD. Key words: electroencephalography, Alzheimer's disease, dementia, EEG, spectral analysis, coherence.EEGq na doença de Alzheimer: análise espectral versus coerência. O que é melhor? RESUMO Há evidências de que as oscilações das bandas teta, alfa e delta no eletroencefalograma podem refletir diferenças na cognição e memória; a sensibilidade deste método diagnóstico pode ser melhorada por técnicas de quantificação. Comparamos a sensibilidade e especificidade entre a análise espectral (spectA) e coerência (Coh) dentro do mesmo grupo de pacientes com doença de Alzheimer (DA) e contra um grupo controle. SpectA e Coh foram calculadas a partir de EEGs de 40 pacientes com DA leve a moderada e 40 idosos saudáveis. O pico do espectro foi menor no grupo DA que nos controles. O grupo DA também apresentou um espectro mais lento nas bandas teta e delta e menor coerência inter-hemisférica para as sub-bandas alfa 2 e beta 1 posterior e delta frontal. A curva ROC suporta que a análise espectral da banda alfa foi mais sensível que a coerência para diferenciar controles de DA. Palavras-Chave: eletroencefalografia, doença de Alzheimer, demência, EEG, análise espectral, coerência. CorrespondenceRenato Anghinah Rua Itacolomi 333 / cj 83

No abstract

The visual analysis of EEG has shown useful in helping the diagnosis of Alzheimer disease (AD) when the diagnosis remains uncertain, being used in some clinical protocols. However, such analysis is subject to the inherent equipment imprecision, patient movement, electrical records, and physician interpretation of the visual analysis variation. The Artificial Neural Network (ANN) could be a helpful tool, appropriate to address problems such as prediction and pattern recognition. In this work, it has use a new class of ANN, the Paraconsistent Artificial Neural Network (PANN), which is capable of handling uncertain, inconsistent, and paracomplete information, for recognizing predetermined patterns of EEG and to assess its value as a possible auxiliary method for AD diagnosis. Thirty three patients with Alzheimer's disease and 34 controls patients of EEG records were obtained during relaxed wakefulness. It was considered as normal patient pattern, the background EEG activity between 8.0 and 12.0 Hz (with an average frequency of 10 Hz), allowing a range of 0.5 Hz. The PANN was able to recognize waves that belonging to their respective bands of clinical use (theta, delta, alpha, and beta), leading to an agreement with the clinical diagnosis at 82% of sensitivity and at 61% of specificity. Supported with these results, the PANN could be a promising tool to manipulate EEG analysis, bearing in mind the following considerations: the growing interest of specialists in EEG analysis visual and the ability of the PANN to deal directly imprecise, inconsistent and paracomplete data, providing an interesting quantitative and qualitative analysis.

In this work we summarize some of our studies on Paraconsistent Artificial Neural Networks applied to electroencephalography. In particular we've applied in the study of the Alzheimer Disease and Attention-Deficit/Hyperactivity Disorder (ADHD).

No abstract