Artificial intelligence has successfully improved the accuracy of detection of cerebral small vessel disease from CT scans, British researchers have reported.
The machine learning technique delivered results almost as accurate as those obtained from MRI scans, according to a study conducted at Charing Cross Hospital, London, and reported in Radiology.
The software was developed at Imperial College in London and Edinburgh University.
The developers say it could help in making decisions about stroke care and in monitoring the development of dementia.
During the study, the AI system analysed more than 1,000 CT scans from stroke patients at 70 British hospitals. In 60 cases, the researchers were able to compare the findings with MRI results.
Professor Joanna Wardlaw, from Edinburgh University, said: “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.”
Fellow researcher Dr Paul Bentley, from Imperial College, 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.”
Rapid Automated Quantification of Cerebral Leukoaraiosis on CT Images: A Multicenter Validation Study. Radiology in press
Leave a Reply