Artificial Intelligence Can Detect Stroke and Dementia Risk

Artificial intelligence, i.e., machine learning has identified one of the most frequent causes of dementia and stroke more accurately than current methods with the help of brain scan (CT).// A team of researchers at Imperial College London and the University of Edinburgh has developed new software to identify and measure the severity of small vessel disease which is one of the commonest causes of stroke and dementia. The findings of the study are published in Radiology.

The research team point out that the newly developed technology can help doctors to

• Provide the best treatment to patients more quickly in emergency settings
• Predict whether there are any chances of developing dementia
• Develop more personalized medicine

"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," said Dr. Paul Bentley, lead author and Clinical Lecturer at Imperial College London.

The new software developed is a first scan reading tool that could be useful in a comprehensive routine assessment of scan datasets and after several testing, this could be effectively used to assess patients with stroke at hospital admission, said Professor. Joanna Wardlaw, Head of Neuroimaging Sciences at the University of Edinburgh.

Small vessel disease (SVD)

Small vessel disease is a common neurological disease affecting older people which decreases the blood flow to the deep white matter connections of the brain, thereby damaging and destroying the brain cells.

SVD may cause

• Stroke
• Dementia
• Mood Disturbances

SVD increases with

• Age
• Hypertension
• Diabetes

Doctors can diagnose SVD by looking for changes to white matter in the brain during MRI or CT scans. However, to identify how far the disease has spread depends on doctor's assessment from the scan. Using CT scans, it is quite difficult to decide where the edges of the SVD are, and to estimate the severity of the disease, said Dr. Bentley.

Advertisement

However, MRI can detect and measure SVD more skillfully; It is not a frequently used method due to difficulty in the availability of scanner and suitability for emergency or older patients.

Details of the Study

Advertisement

"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. 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," said Dr. Bentley.

Dr. Bentley suggests that this software could help influence doctors decision-making in emergency neurological conditions and lead to more personalized medicine. For instance, in stroke, treatments such as 'clot-busting medications' can be quickly administered to unblock an artery. But, these treatments can be unsafe by causing bleeding, which becomes more common with an increase in the amount of SVD. The software could be utilized in future to determine the possible risk of hemorrhage in patients and doctors can decide on a personal basis, along with other factors, whether treatments like clot busters can be used or not.

Dr. Bentley also adds that the new software can help estimate the possibility of patients developing dementia or immobility, due to slowly progressive SVD. This would alert doctors to identify the potentially reversible causes such as high blood pressure or diabetes.

The study mainly used historical data of 1082 CT scans of stroke patients across 70 hospitals in the UK between 2000-2014, which includes cases from the Third International Stroke Trial.

Findings of the Study

The new software detected and measured a marker of SVD, and gave scores indicating the severity of disease ranging from mild to severe. The research team compared the results to a panel of expert doctors who analyzed the same scans for SVD severity. The level of agreement of the software with the expert doctors was as good as agreements between one expert doctor with another.

In 60 cases, the research team obtained MRI and CT in the same subjects and used the MRI to measure the exact amount of SVD. They observed that the software was 85 percent accurate at predicting the severity of SVD.

The research team is currently using similar approaches to measure the amount of brain shrinkage and other types of conditions commonly diagnosed on brain CT.

Source-Medindia