Prediction Model for Brain Age May Help Detect Cognitive Decline

THURSDAY, July 1, 2021 (HealthDay News) — Patients with amnestic mild cognitive impairment (aMCI) have higher predicted age difference (PAD) than healthy controls, according to a study published online June 23 in Radiology: Artificial Intelligence.

Weijie Huang, Ph.D., from Beijing Normal University, and colleagues trained a machine learning approach to determine brain age based on T1-weighted MRI. Model training and testing was carried out on two datasets (616 healthy controls and 80 patients with aMCI from the Beijing Ageing Brain Rejuvenation Initiative [BABRI] and 589 healthy controls and 144 patients with aMCI from the Alzheimer’s Disease Neuroimaging Initiative [ADNI]).

The researchers found that patients with aMCI had higher PADs than those of healthy controls (BABRI: 2.65 ± 4.91 versus 0.18 ± 4.79; ADNI: 1.68 ± 5.28 versus 0.05 ± 4.41), based on the prediction model. In patients with aMCI, PAD was significantly associated with individual cognitive impairment in several cognitive domains. Higher PADs were also seen for APOE ε4 allele carriers versus noncarriers (3.76 ± 4.82 versus 0.10 ± 5.05) and patients with amyloid-positive versus amyloid-negative status (2.40 ± 5.25 versus 0.93 ± 5.20). The combination of PAD with other markers of Alzheimer disease at baseline differentiated between progressive and stable aMCI with an area under the curve of 0.87.

“This work suggests that PAD has the potential to be developed into a computerized marker for early diagnosis of cognitive impairment and monitoring response to treatment,” the authors write.

One author disclosed financial ties to the pharmaceutical industry.

Abstract/Full Text (subscription or payment may be required)