Background. Primary progressive aphasia (PPA) is a neurodegenerative syndrome in which diagnosis is usually challenging. Biomarkers are needed for diagnosis and monitoring. In this study, we aimed to evaluate Electroencephalography (EEG) as a biomarker for the diagnosis of PPA. Methods. We conducted a cross-sectional study with 40 PPA patients categorized as non-fluent, semantic, and logopenic variants, and 20 controls. Resting-state EEG with 32 channels was acquired and preprocessed using several procedures (quantitative EEG, wavelet transformation, autoencoders, and graph theory analysis). Seven machine learning algorithms were evaluated (Decision Tree, Elastic Net, Support Vector Machines, Random Forest, K-Nearest Neighbors, Gaussian Naive Bayes, and Multinomial Naive Bayes). Results. Diagnostic capacity to distinguish between PPA and controls was high (accuracy 75%, F1-score 83% for kNN algorithm). The most important features in the classification were derived from network analysis based on graph theory. Conversely, discrimination between PPA variants was lower (Accuracy 58% and F1-score 60% for kNN). Conclusions. The application of ML to resting-state EEG may have a role in the diagnosis of PPA, especially in the differentiation from controls. Future studies with high-density EEG should explore the capacity to distinguish between PPA variants.
24 February 2025 in Journal of Alzheimer’s Disease
Background: Non-invasive brain stimulation has shown positive results in maximizing the effects of language therapy in primary progressive aphasia (PPA). Due to the different patterns of brain damage in each aphasia variant, we hypothesized that patients with non-fluent and semantic variants would show a differential response to transcranial magnetic stimulation (TMS). Objective: We aimed to compare the clinical responses after a single session of repetitive TMS in the left inferior frontal gyrus (IFG) and the left dorsolateral prefrontal cortex (DLPC). Methodology: Twenty patients with PPA (14 with non-fluent and 6 with semantic variants) were assessed before and after repetitive TMS over the IFG, DLPC, and vertex with several language tasks, connected speech, and a subjective impression of change scale. Results: IFG stimulation was associated with an improvement in words per minute and the subjective assessment in the non-fluent variant, but no effects were found in the semantic variant. DLPC stimulation was associated with an improvement in words per minute, repetition, and naming latency in the non-fluent variant, and in naming and subjective impression of change in the semantic variant. Conclusions: Our study showed a differential effect of one session of brain stimulation over the IFG and DLPC in patients with non-fluent and semantic PPA variants. These findings suggest that the selection of the target of stimulation may be relevant for the success of brain stimulation and favor the use of DLPC over the IFG.