Combining advanced MRI and artificial intelligence techniques could help doctors determine which patients with mild cognitive impairment (MCI) will suffer further cognitive decline, according to new research.

Roughly half of all patients with mild cognitive impairment eventually progress to Alzheimer’s disease, according to researchers at University Hospitals of Geneva, Switzerland. The key to treating Alzheimer’s in its early stages is determining which of those patients will progress to the disease. For their study, researchers analyzed 104 patients—35 were control subject, 69 had been diagnosed with MCI—using the MRI and artificial intelligence techniques.

The susceptibility-weighted MRI showed blood vessels in the brain with great detail, and also identified tiny microhemorrhages, or microbleeds. These microbleeds, along with increased levels of iron in the brain, were significantly more common among those with MCI. Researchers also analyzed the MRI data using support vector machines (SVMs), an artificial intelligence technique that uses algorithms to identify patterns. The SVM analysis picked out those patients that would progress to Alzheimer’s with 85% accuracy, according to the study. Results are scheduled to appear in the December issue of Radiology.