Researchers at Dublin Institute of Technology (DIT) are applying the power of machine learning to prevent and treat stroke, one of the most severe medical problems worldwide.
Researchers at Dublin Institute of Technology (DIT) are applying the power of machine learning to prevent and treat stroke, one of the most severe medical problems worldwide.
Almost 1 million people suffer a stroke each year in Europe and millions more are impacted by it. Imagine if a computer application could analyse reams of medical data and make recommendations about a stroke patient’s treatment across the course of their entire illness – a one-stop-shop to help doctors assess whether a patient is at risk of stroke or what kind of rehabilitation they need.
DIT is a partner on the PRECISE4Q project, led by Charité Universität Medizin in Berlin, one of the largest university hospitals in Europe, in collaboration with nine academic and industry partners across Europe. The team is creating the Digital Stroke Patient Platform enabling – for the first time – personalised treatment which addresses an individual patients’ needs in all stages of the stroke lifecycle: prevention, rehabilitation and reintegration into society.
This platform is at the cutting edge of a revolutionary shift in healthcare towards what is known as ‘personalised medicine’. A key EU priority for the future, personalised medicine is grounded on the idea that it is possible to produce predictive computational models that can target the best treatment options for each individual patient’s needs. This application of artificial intelligence in medicine is relatively new, transforming how diseases are discovered and treated.
The Digital Stroke Patient Platform marks the first time that machine learning will be applied across a stroke patient’s entire illness. The goal is to improve the prevention and treatment of the disease, and to enhance each individual’s quality of life after stroke.
The toll of stroke on patients and loved ones is often immense. Stroke occurs when there is a sudden lack of blood supply to the brain, which can cause brain tissue to die. Many survivors are left with a range of challenging outcomes, including problems with mobility, vision, speech, memory and depression.
In Europe, the impact of stroke has a significant financial cost of over €40 billion each year, putting a major strain on healthcare systems and rehabilitation resources.
The project lead for DIT, Professor John D Kelleher, Academic Leader, Information, Communications & Entertainment Institute, has been awarded €1.3 million to build the predictive models that underpin the platform.
Professor Kelleher explains, “We are working to develop a set of models for each stage of stroke treatment - prevention, therapy, rehabilitation and reintegration. These systems will be capable of analysing large data sets from a wide variety of sources, integrating them into self-learning computer models. On some tasks, artificial Intelligence can process and analyse data much faster and more precisely than humans. It can also generate meaningful insights due to the sheer volume and variation of data that can be analysed.”
The information feeding into the models includes clinical research data from hospitals, for example genetic, biochemical, and brain imaging data, alongside big data sets such as health registries and electronic health records.
“The patient’s own medical data and images will then be inputted into the system and analysed against the population and research data. The platform, which will be easily accessed via an application on a computer or desktop, will enable doctors to make more accurate and faster diagnoses for those at risk of stroke, and it will also help them create the personalised prevention or treatment plans. For example, the reintegration model will predict a long-term quality of life profile for a patient, with the goal of enabling long-term resilience to stroke and maximising each patient’s quality of life.’
The project, which is funded under the EU horizon 2020 programme, aims to have a major impact on the lives of those who suffer a stroke, as well as those who are at risk.
If the platform can accurately pinpoint those who are most at risk, early interventions such as lifestyle improvements and healthy living measures could increase life expectancy and save lives. It could also reduce the costs to healthcare systems by keeping people healthier and focusing resources where they are most needed.
The platform also has the potential to improve quality of life through the personalised reintegration plans. “Early intervention will increase the number of stroke patients who can live independently or return to work. This is especially important given that statistics are showing that an increasing number of young people are suffering from stroke.”
If successful, the platform could minimise the burden of stroke for the individual and for society, and ultimately lead to people leading longer and healthier lives.
