Professor Alex Leff on an app for regaining speech after a stroke
23 October 2024
On World Stroke Day, Professor Alex Leff from UCL Queen Square Institute of Neurology talks about the impact of stroke on language processing and how a new app, iTalkBetter, can help with language deficits.
Professor Alex Leff is Professor of Cognitive Neurology and an Honorary Consultant Neurologist. His main focus is on cognitive rehabilitation, particularly the field of acquired language disorders and vision.
Imagine waking up one day unable to express your thoughts or understand the written or spoken words of those around you. This is the reality for about one-third of people who develop aphasia following a stroke. Aphasia is a complex language and communication disorder that can disrupt speech, comprehension, writing, and even numerical abilities. Professor Alex Leff, a Professor of Cognitive Neurology and an Honorary Consultant Neurologist has worked with stroke patients for the best part of three decades, he explains that the symptoms of aphasia can vary a lot between people depending on which part of the brain has been damaged.
“Language in the brain is like a network. You can think of it a bit like a bit like a train network. Knock out a peripheral station and the effect on the whole network won’t be too bad, but if you knock out a central station it will affect much of the network. That’s why some people have mild symptoms following an aphasic stroke, whilst other have severe and wide-ranging symptoms.”
Professor Leff goes on to say that although symptoms can be severe, most people do go on to make improvements in communication and language over time, especially when given focused therapy. Currently the most effective therapies offered are delivered by trained healthcare professionals, such as speech and language therapists and psychologists who will focus on certain aspects of communication. The majority of people who have aphasia following a stroke are offered around five or six hours of therapy while they are an inpatient in hospital and then roughly the same amount of therapy once they are discharged. Although the therapies are very effective, learning language again takes a lot of time, much longer than 12 hours.
“One way of thinking about the time it takes to re-learn language is thinking about how long it took you to first learn language, or take just one language skill, reading for example. You probably hardly remember how long it took, but it took you years, and hours and hours of practice. The same can be true for re-learning language. Another way of thinking about it is how long would it take you to learn a new language? And that’s with an undamaged brain.”
Professor Alex Leff has dedicated much of his academic and clinical career to cognitive rehabilitation and saw the discrepancy between what was currently offered to people with aphasia and what was desperately needed. Along with Professor Jenny Crinion and the Neurotherapeutics group at the Institute of Cognitive Neuroscience, this led to the development of a series of apps targeting specific language impairments such as reading, comprehensive and speaking. Crucially, the idea is that users can practice in their own time, without any limits on the amount of therapy they receive.
“These apps focus on individual language impairments, so if you want to get better at reading, you would use the reading app, if you want to get better at speaking, you would use a speaking app, and this is where iTalkBetter fits in. It focuses on speech production by showing people pictures and getting them to name the picture. The app provides prompts and feedback on each picture and eventually the aim is that they be able to access that word without having a prompt at all.”
Feedback is a fundamental aspect of aphasia treatment; however, speech recognition software has historically not been able to handle speech errors. iTalkBetter uses UCL-developed software, ‘Nuva’, which is able to identify and correct speech errors that people with aphasia make and offer prompts.
The team tested iTalkBetter within a standard randomised control trial with people who had aphasia following a stroke. They showed that practicing with the app significantly improved users’ ability to retrieve and name words. They also showed a knock-on effect into conversational speech, whereby users were able to produce more words per minute to describe a complicated picture following six-weeks’ practice with the app. Improvements were paired with brain changes all well; brain regions related to language perception, production and control increased in volume after practice with the iTalkBetter app.
Professor Leff is keen that iTalkBetter reaches those who need it. Together with Professor Jenny Crinion they are forming a spin-out company called NeuroSpeech AI which will focus on further developing the apps and making them accessible to a wider user-base.
Listen to Professor Alex Leff talk about 'iTalkBetter'.
Related
- Professor Alex Leff's academic profile
- Queen Square Institute of Neurology
- News story: iTalkBetter app significantly improves speech in stroke patients