Deep Learning-based Classification of Dementia using Image Representation of Subcortical Signals

TL;DR

The study tackles noninvasive dementia classification with EEG by extracting scout time-series from deep brain regions (hippocampus, amygdala, and thalamus) via sLORETA, converting them into 128×128 continuous wavelet transform images, and classifying with convolutional neural networks. It evaluates multiple CNN backbones and fusion strategies on two high-density EEG datasets (BrainLat and IITD-AIIA), demonstrating that a DenseNet201 model paired with the product of posterior probabilities from left and right regions achieves the best performance. The forward model is expressed as $V = A \tilde{S} + Z$, with sLORETA estimating source currents to obtain discriminative subcortical signals; the approach yields 94.17% accuracy on BrainLat and 77.72% on IITD-AIIA, with ROC-AUC values near 1.0 and 0.92 respectively. Overall, the results indicate that subcortical EEG sources coupled with image-based DL can effectively differentiate AD, FTD, and MCI/HC, offering a promising pathway for early dementia diagnosis and differential prognosis.

Abstract

Dementia is a neurological syndrome marked by cognitive decline. Alzheimer's disease (AD) and Frontotemporal dementia (FTD) are the common forms of dementia, each with distinct progression patterns. EEG, a non-invasive tool for recording brain activity, has shown potential in distinguishing AD from FTD and mild cognitive impairment (MCI). Previous studies have utilized various EEG features, such as subband power and connectivity patterns to differentiate these conditions. However, artifacts in EEG signals can obscure crucial information, necessitating advanced signal processing techniques. This study aims to develop a deep learning-based classification system for dementia by analyzing scout time-series signals from deep brain regions, specifically the hippocampus, amygdala, and thalamus. The study utilizes scout time series extracted via the standardized low-resolution brain electromagnetic tomography (sLORETA) technique. The time series is converted to image representations using continuous wavelet transform (CWT) and fed as input to deep learning models. Two high-density EEG datasets are utilized to check for the efficacy of the proposed method: the online BrainLat dataset (comprising AD, FTD, and healthy controls (HC)) and the in-house IITD-AIIA dataset (including subjects with AD, MCI, and HC). Different classification strategies and classifier combinations have been utilized for the accurate mapping of classes on both datasets. The best results were achieved by using a product of probabilities from classifiers for left and right subcortical regions in conjunction with the DenseNet model architecture. It yields accuracies of 94.17$\%$ and 77.72$\%$ on the BrainLat and IITD-AIIA datasets, respectively. This highlights the potential of this approach for early and accurate differentiation of neurodegenerative disorders.

Figures (6)

• Figure 1: Block diagram depicting the proposed method. The processed EEG signals are utilized to extract scout time series from the hippocampus, amygdala and thalamus using sLORETA. The signals are segmented and divided into left and right regions. Subsequently, the CWT-based images are fed to separate classifiers for images corresponding to left and right regions. $z_L$ and $z_R$ represent the latent representation of the classifiers, while $\hat{y}_L$ and $\hat{y}_R$ denote classifier predictions. The latent embeddings are fused using Early and Tensor Fusion, while the individual classifier outputs are fused using probability sum and product.
• Figure 2: Grand Average EEG Source Localization plots (front view) for AD, FTD, and HC cases from the BrainLat dataset at timestamps 70s, 70.5s, and 71s. These plots are generated using the Brainstorm Toolbox. The activation maps were set to 20$\%$ amplitude, with the amplitude threshold parameter set to "Maximum: Global" for each case.
• Figure 3: Depiction of the Early Fusion and Tensor Fusion Network approaches. $z_L$ and $z_R$ represent latent embeddings from the left and right classifiers, respectively.
• Figure 4: Confusion matrix using a combination of DenseNet201 and $\hat{y}_{mul}$ for (a) BrainLat dataset and (b) IITD-AIIA dataset.
• Figure 5: Scatter plot depicting clusters corresponding to each of the classes obtained by applying dimensionality reduction using t-SNE on the latent embedding vector for (a) BrainLat dataset and (b) IITD-AIIA dataset.