Predicting language outcome and recovery after stroke.

A stroke patient's ability to speak or understand language in the months after their brain injury could be predicted by a new web-based tool, rather than left to guesswork. Around a third of stroke survivors experience aphasia—loss of language abilities—yet doctors currently have no reliable way to forecast who will recover and how quickly. This project tackles that gap by combining three parallel approaches: identifying which brain lesions and patient characteristics lead to long-term communication problems; mapping the brain networks that compensate for damaged language areas using functional imaging; and building a probabilistic prediction tool from that data. If successful, the research will produce an easy-to-use online system that gives patients, carers, and clinicians a personalised recovery timeline based on how similar cases have progressed. This could transform post-stroke care from vague reassurance to concrete planning—helping families prepare for care needs, therapists target their efforts, and patients set realistic expectations. The tool would not replace clinical judgment but would add a data-driven layer to decisions about speech therapy intensity and discharge timing. The work is applied from the outset, designed specifically for clinical deployment rather than as a fundamental science investigation.