Background A previous study suggests that resting-state EEG biomarkers measured at prefrontal region (Fp1, and Fp2) are moderately correlated with Mini-Mental State Examination (MMSE) scores of elderly people with Alzheimer’s disease. In this study, our objective was to investigate whether resting-state EEG biomarkers recorded from frontal region are correlated with each MMSE sub-scores. 20 elderly patients diagnosed as Alzheimer’s disease entered to the study. After completion of MMSE, subjects underwent EEG for 5 min with closed eyes condition. We measured median frequency, theta/alpha power ratio, and relative powers. To examine the relationship between these features and MMSE sub-scores first, Pearson correlation coefficients were computed for each feature and MMSE sub-scores. Then, p values were computed for each correlation. Finally, a Bonferroni correction was done. Results Nine correlations have been found for markers recorded from F3, F7, and Fz. Alpha and beta relative powers were the markers which shows correlations. We found that MMSE overall, attention, and calculation scores are significantly correlated with beta relative powers recorded from F3, and Fz, and alpha relative power from F7. Orientation to time scores were correlated with F3, and Fz beta relative powers. The only correlation found for orientation to place was beta relative power of F3. Conclusions Our results indicate that there are correlations between frontal EEG markers and MMSE sub-scores of patients with Alzheimer’s disease. The results show that alpha and beta relative powers are markers correlated with MMSE scores. It seems that if we want to develop predicting models for Alzheimer’s disease, using data recorded from other frontal electrodes, especially what we have introduced should be considered.

Purpose: The purpose of this study is to use linear and non-linear features extracted from Electroencephalography (EEG) signal to predict the Mini-Mental State Examination (MMSE) test score by machine learning algorithms. Materials and Methods: First, the MMSE test was taken from 20 subjects that were referred with the initial diagnosis of dementia. Then, the brain activity of subjects was recorded via EEG signal. After preprocessing this signal, various linear and non-linear features are extracted from it that are used as input to machine learning algorithms to predict MMSE test scores in three levels. Results: Based on the experiments, the best classification result is related to the Long Short-Term Memory (LSTM) network with 68% accuracy. Conclusion: Findings show that by using machine learning algorithms and features extracted from EEG signal the MMSE scores are predicted in three levels. Although deep neural networks require a lot of data for training, the LSTM network has been able to achieve the best performance. By increasing the number of subjects, it is expected that the classification results will also increase.

No abstract