Diagnosis of Alzheimer’s Disease by Time-Dependent Power Spectrum Descriptors and Convolutional Neural Network Using EEG Signal

Figure 1. The block diagram of the TD-PSD feature extraction method for EEG signals.

Figure 2. A conventional sublayer in CNN.

Figure 3. The place of electrodes of EGG signals.

Figure 4. Plots of sampling from EEG signals for three categories of MCI, AD, and HC.

Figure 5. Time-frequency analysis of EEG signals for MCI patients.

Figure 6. Time-frequency analysis of EEG signals for AD patients.

Figure 7. Time-frequency analysis of EEG signals for HC samples.

Figure 8. Conceptual diagram of the process used in this paper.

Figure 9. Confusion matrix of classification methods: (a) KNN, (b) SVM, and (c) LDA.

Figure 10. The architecture of the CNN layer for classification Alzheimer's disease.

Figure 11. Accuracy and loss values for classification process of the presented CNN.

Figure 12. Confusion matrix of the presented CNN method for classification of patients suffering from Alzheimer's.