New software can predict stroke and dementia more accurately

Pioneering technology could help detect early signs of small vessel disease (SVD), which can lead to dementia and stroke.
The new software, which has been shown to be highly accurate, could make a big difference to people with diabetes because diabetes can accelerate SVD, as can age and hypertension (high blood pressure).
Imperial College London, which developed the software, has trialled it in a study using historical data taken from 1,082 brain (CT) scans of people who had a stroke between 2000-2014.
The technology identified SVD and gave a score based on the reading, indicating how severe the vessel damage was. The results showed the software was 85% accurate.
Dr Paul Bentley, lead author and Clinical Lecturer at Imperial College London, said: “This is the first time that machine learning methods have been able to accurately measure a marker of small vessel disease in patients presenting with stroke or memory impairment who undergo CT scanning.
“Our technique is consistent and achieves high accuracy relative to an MRI scan – the current gold standard technique for diagnosis. This could lead to better treatments and care for patients in everyday practice.”
The current method of determining SVD relies on doctors looking for changes in the brain’s white matter during an MRI or CT scan. However, it can be challenging to get an accurate idea of how far the vessel damage has spread.
“Current methods to diagnose the disease through CT or MRI scans can be effective, but it can be difficult for doctors to diagnose the severity of the disease by the human eye,” added Dr Bentley.
“The importance of our new method is that it allows for precise and automated measurement of the disease. This also has applications for widespread diagnosis and monitoring of dementia, as well as for emergency decision-making in stroke.”
Professor Joanna Wardlaw, head of Neuroimaging Sciences at the University of Edinburgh, added: “This is a first step in making a scan reading tool that could be useful in mining large routine scan datasets and, after more testing, might aid patient assessment at hospital admission with stroke.”
The study findings have been published in the journal Radiology.