Poster Presentation

Search Abstracts | Symposia | Slide Sessions | Poster Sessions

EEG-based neural assessment of post-stroke aphasia in the acute phase: the potential application of neural tracking of natural speech

Poster Session E, Friday, October 2, 11:00 am - 1:00 pm, Wangari Maathai

Andi Smet¹, Ella Eycken¹, Robin Lemmens¹, Céline Gillebert¹, Tom Francart¹, Maaike Vandermosten¹; ¹KU Leuven

Globally, approximately 13 million people suffer from a stroke each year, with about one-third developing aphasia, an acquired language disorder. Standard behavioral tests used to diagnose aphasia focus on isolated stimuli, which are often confounded by comorbid cognitive impairments. There is therefore a need for objective assessment methods that more closely reflect natural language processing while minimizing task demands on patients. Neural envelope tracking is a promising technique in which brain activity measured with EEG is correlated to the speech envelope during natural continuous speech listening. The speech envelope contains temporal cues that are essential for speech comprehension, making neural tracking a relevant measure of language processing. Recent studies in chronic stroke populations have demonstrated reduced neural tracking in patients with aphasia compared to healthy controls, and have shown promising accuracy for individual classification. Similar group differences between stroke patients with and without aphasia have also been observed in our recent unpublished study (in preparation). However, these studies have focused on the chronic phase, when patients are neurologically more stable. In contrast, the clinical need for efficient and objective language assessment tools is greatest in the acute phase after stroke. This study investigates whether neural tracking of natural speech can differentiate between stroke patients with and without aphasia in the acute phase (<2 weeks post-stroke onset). In addition, a group of age-matched neurologically healthy controls is included as a third reference group. Patients with aphasia will be followed longitudinally into the chronic phase (>6 months post-stroke onset) to assess whether changes in behavioral recovery are reflected in neural tracking measures. We aim to recruit 100 stroke patients (50 with aphasia and 50 without aphasia) in the acute phase (bedside testing) and 50 healthy controls. Neural envelope tracking values will be quantified using a mutual information model and group differences will be analyzed via non-parametric cluster-based permutation tests. A classification model will be trained once a larger sample is available. The first longitudinal data will be available by the time of the conference. Preliminary results on the currently available data (n = 28) suggest decreased neural tracking in patients with aphasia compared to those without aphasia, particularly in the delta (0.5–4 Hz) and broadband (0.5–40 Hz) frequency ranges. These patterns appear consistent with previous findings reported in chronic-phase stroke. However, the current sample remains underpowered for definitive statistical inference, and formal statistical analyses will be performed on the final sample. In conclusion, neural envelope tracking presents a promising and efficient method for assessing language impairments in stroke patients. The current study investigates its potential in the acute phase of stroke, where objective bedside assessment tools are particularly needed. Preliminary findings suggest neural tracking differences between patients with and without aphasia that align with previous chronic-phase research, supporting its potential as a clinically relevant marker of language impairment after stroke.

Topic Areas: Disorders: Acquired,