Dear ARG Members,
We hope that you all had good Easters, and do not have chocolate hang-overs!
Please find below a summary of Rachel Brown’s talk that she is kindly giving on the 1st May. Hopefully her teaser below about of the development of PLORAS individualised post stroke recovery prediction tool will spur you on to come back and see us!!
All the best,
Firle & Sharon
Project: Predicting Language Outcome and Recovery After Stroke
Summary: The acute onset of aphasia following a stroke is a traumatic and life-changing experience for patients and relatives. The inevitable question asked of many speech and language therapists, doctors and other health professionals at this time is, “Will I (or my relative) be able to speak normally again?” To date, the responses to this question tend to be aimed at providing encouragement whilst perhaps setting relatively low expectations to ‘prepare’ patients and relatives for the worst. It is currently extremely difficult to provide patients with anything other than general or anecdotal information about the likely duration and extent of language recovery after aphasic stroke.
Our project aims to create a clinical tool for patients and clinicians which will provide individualised predictions about the most likely course of recovery from aphasia after stroke, based on the patient’s MRI brain image. These predictions are based on the exact location, size and shape of the lesion at a very detailed level. This has become possible due to advances in neuroimaging and computational techniques. In practical terms, future patients with aphasia can be shown the rate and level of improvement that other individuals with the same area of brain damage experienced over time. We are constantly working towards understanding the factors that improve the predictions. We make the predictions based on what we know from the participants we have tested to date, including their scores on language tests (Comprehensive Aphasia Test; Swinburn, Porter and Howard, 2004) and an MRI scan. We have built a system that can make those predictions and it can also learn from patient data directly to keep improving the predictions it makes. Critically, each prediction indicates the level of confidence that we have in it. For example, we can be confident when we know that more than 90% of patients with the same lesion site recovered within a particular time frame, but we are less confident if only 50% of patients recovered, because this indicates that the effect of the lesion varies across patients.
In addition, we aim to explain why some patients respond better to treatment than others and why some patients recover more quickly than others. We have found some ‘critical lesion sites’ for language problems. Damage to these areas consistently results in persistent aphasia (aphasia lasting a duration of five years or more), whereas patients without damage to these areas are more likely to recover within a few years post stroke. We hope the findings about recovery of language will give information to guide therapy for aphasia in the future.
We will shortly be sending out a questionnaire to everyone attending the next ARG meeting, to find out about your experience of recovery conversations with people with aphasia and their families.