Keynote Lecture: Fatemeh Geranmayeh

Fatemeh Geranmayeh

Fatemeh Geranmayeh

From models of language recovery after stroke to Large Language Models in stroke

Saturday, September 13, 2025, 2:00 – 3:00 pm, Elstad Auditorium

Speaker: Dr. Fatemeh Geranmayeh, Imperial College London

‘Language’ is a uniquely human faculty that not only distinguishes us from other species but also holds profound significance for individuals' identity, autonomy, and quality of life. Following a stroke, impairments in speech production and language comprehension are among the most devastating consequences, associated with poorer outcomes across virtually all post-stroke functional and quality of life measures. Understanding the organisation and recovery of speech and language has thus been a major research priority since the 19th-century foundational work of classical neuroanatomical figures.

Yet, studying language recovery remains challenging. Recovery trajectories are notoriously heterogeneous and unfold along multidimensional axes of language function. Moreover, because language function, and the broader neural architecture that supports it, is not localised but distributed, we must adopt a truly whole-brain approach to map recovery mechanisms effectively.

This calls for a reconceptualisation of how we measure and model speech production and language in the brain, particularly in the context of recovery. Whole-brain health approaches, coupled with scalable, high-resolution assessments of cognition and language, are essential. Digital cognitive tools now enable remote, longitudinal testing of language and cognition, and when combined with recent advances in automatic speech recognition, natural language processing, and large language models, promise to transform both our scientific understanding and research translation.

In this keynote, I will trace a scientific arc from traditional models of post-stroke language recovery to emerging applications of digital and computational technologies. I will discuss how integrating large-scale speech data with modern machine learning approaches may allow us not only to better characterise
language recovery but also to democratise research participation and increase equity in neuroscience.

About Fatemeh Geranmayeh

Dr. Fatemeh Geranmayeh is a Clinician Scientist at the Department of Brain Sciences, Imperial College London. She leads the Clinical Language & Cognition Group, focusing on post-stroke language and cognitive impairments, vascular dementia, and the deep cognitive and clinical phenotyping of patients with cerebrovascular disease. Dr. Geranmayeh's academic journey began in 2004 when she earned a First Class Honours BSc degree in Neuroscience, receiving the Goldberg-Schachmann and Freda Becker Award from the University of London. She completed her medical degree in 2006 and pursued a PhD under the mentorship of the late Professor Richard Wise, which she completed in 2015.

Her research employs advanced neuroimaging techniques to investigate the neural mechanisms underlying speech production recovery following acute stroke. In 2022, she initiated the IC3 study, a multimodal clinical study to track longitudinal cognitive and language recovery post-stroke, incorporating novel blood biomarkers, brain imaging, and automated speech recognition and cognitive assessments.

In addition to her research, Dr. Geranmayeh is a practicing neurology consultant, running a specialist vascular cognitive clinic, and co-Director of the Alzheimer's Society VIDA (Vascular and Immune contributors to DementiA) Doctoral Training Centre.

SNL Account Login


Forgot Password?
Create an Account