New AI technology to detect dementia

23 October 2018 (Last Updated October 23rd, 2018 14:28)

Researchers at University College London (UCL) have developed a new artificial intelligence (AI) technology, called SuStaIn, to identify and differentiate various dementias, including Alzheimer’s disease.

New AI technology to detect dementia
Credit: A Health Blog.

Researchers at University College London (UCL) have developed a new artificial intelligence (AI) technology, called the SuStaIn algorithm, to identify and differentiate various dementias, including Alzheimer’s disease.

The machine learning algorithm works by analysing patient MRI scans and can automatically disentangle varying patterns of progression in different patients.

The new technology is also expected to enable the detection of individuals who may respond best to different treatments.

"The machine learning algorithm works by analysing patient MRI scans and can automatically disentangle varying patterns of progression in different patients."

When tested, the SuStaIn algorithm has reportedly identified three separate subtypes of Alzheimer’s disease.

The researchers said that early detection of the subtypes and the use of non-invasive MRI scanning provides a better chance of determining the best treatment for patients.

UCL Centre for Medical Image Computing researcher Alexandra Young said: “Individuals might present with similar symptoms to each other, but using SuStaIn we can find that they belong to different subgroups.

“This allows us to predict more accurately how their disease will progress and diagnose it earlier.”

The team noted that drugs fail in clinical trials for Alzheimer’s disease because a variety of very different patients are recruited and tested for a single treatment.

Centre for Medical Image Computing professor Daniel Alexander added: “A treatment with a strong effect on a particular subgroup of patients may show no overall effect on the full population so fail the drug trial.

“SuStaIn provides a way to show treatment effects on distinct subgroups, potentially expediting treatments to market.”

Currently, the researchers are working on expanding the use of the machine learning SuStaIn algorithm to additional diseases.