Machine learning improves dementia, stroke diagnosis
dementia diagnosis machine learning

Machine learning improves dementia, stroke diagnosis

Scientists from Imperial College London and the University of Edinburgh have developed software capable of detecting dementia and stroke precursors using CT scans.

The development, according to the study’s lead author and clinical lecturer at Imperial College London, Dr Paul Bentley, “could lead to better treatments and care for patients in everyday practice”.

The research team’s machine learning software has been trained using 1,082 CT scans of stroke patients, taken from 70 hospitals across the UK between 2000 and 2014. The software is able to identify and measure a marker of Small Vessel Disease (SVD), a common precursor to strokes and dementia, which reduces blood flow to the brain’s deep white matter connections.

Doctors currently rely on CT and/or MRI scans to detect SVD. But observing changes in white matter over time can be difficult for the human eye, meaning that estimating the severity of the disease and, in turn, the likelihood of dementia or a stroke is a major challenge.

Dr Bentley 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.”

So far the software has proven to be 85 percent accurate at predicting the severity of SVD.

The team envisages a time in the not too distant future when precautionary scans are taken at scale, analysed by the software, and the results are delivered in a fraction of the time. “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 strokes,” said Dr Bentley.

Automating diagnosis in this way could, in theory, free up countless hours for doctors, and help improve patient care across the NHS. It would also reduce the burden on MRI procedures, which are often not suitable for elderly patients.

Quantifying the likelihood of dementia

The team believes that, eventually, the software will be able to quantify the likelihood that patients will develop dementia or immobility, as both correlate to some degree with progressive SVD. As a result, doctors could soon be able to alert patients and advise on preventative measures and treatable causes, such as diabetes and high blood pressure.

Professor Joanna Wardlaw, head of neuroimaging sciences at the University of Edinburgh, added that the research could speed up patient assessments during hospital visits. “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 patients’ assessment at hospital admission with a stroke.”

The study has been featured in the publication Radiology. The research took place at Charing Cross Hospital, which is part of the Imperial College Healthcare NHS Trust.

Internet of Business says

The potential of AI, machine learning, and data analytics to help diagnose, treat, or even prevent serious illnesses is becoming increasingly apparent, with advances in early cancer detection among the many other benefits. In the long run, this could help bring about a future in which more and more diseases become manageable conditions that patients live with, rather than suffer from, or can minimise the effects of for as long as possible.

Combined with advances in dedicated wearable technology, and applications that add specialist detection capabilities to consumer devices, such as the Fitbit and Apple Watch, the future of healthcare looks exciting, particularly if new technology can help minimise the cost of care at a time of falling investment in real terms, and soaring elderly populations.

Other new technology applications can help athletes and others recover from serious injuries, and manage their own recovery programmes.

Here are some of Internet of Business’ recent reports in the healthtech space: