Researchers at the Radiological Society of North America have developed an artificial intelligence (AI)-based brain age prediction model that can quantify deviations from a healthy brain-aging trajectory in patients with mild cognitive impairment.
The study was published back in June in Radiology: Artificial Intelligence.
Early Detection of Cognitive Impairment
According to the researchers, the model could be used to aid in early detection of cognitive impairment.
Individuals who suffer from amnestic mild cognitive impairment (aMCI), a transition phase from normal aging to Alzheimer’s disease, have memory deficits that are more serious than the normal for their age and education. However, it is not serious enough to affect their daily function.
The study involved Ni Shu, Ph.D., from State Key Laboratory of Cognitive Neuroscience and Learning, Beijing Normal University, in Beijing, China, along with other colleagues.
The team utilized a machine learning approach to train a brain age prediction model, which was based on the T1-weighted MR images of 974 healthy adults between the ages of 49.3 and 95.4 years.
The trained model was then applied to estimate the predicted age difference of aMCI patients in datasets from the Beijing Aging Brain Rejuvenation Initiative, which included 616 healthy controls and 80 aMCI patients, and the Alzheimer’s Disease Neuroimaging Initiative, which included 589 healthy controls and 144 aMCI patients.
Besides this, the team also looked at the associations between the predicted age difference and cognitive impairment, genetic risk factors, pathological biomarkers of Alzheimer’s, and clinical progression in aMCI patients.
The Study’s Results
The study’s results demonstrated that aMCI patients had brain-aging trajectories distinct from the typical normal aging trajectory. The proposed brain age prediction models would be able to quantify individual deviations from this normal trajectory.
The team also found that the predicted age difference was strongly linked with individual cognitive impairment of aMCI patients in domains like memory, attention, and executive function.
“The predictive model we generated was highly accurate at estimating chronological age in healthy participants based on only the appearance of MRI scans,” the paper stated. “In contrast, for aMCI, the model estimated brain age to be greater than 2.7 years older on average than the patient’s chronological age.”
The model also showed that progressive aMCI patients suffer from more deviations from typical normal aging than stable aMCI patients. Through the use of tools like the predicted age difference score and biomarkers for Alzheimer’s, the progressions of aMCI can be predicted better.
By combining the predicted age difference with other biomarkers for Alzheimer’s, the best performance for accurately differentiating between progressive aMCI and stable aMCI can be achieved.
“This work indicates that predicted age difference has the potential to be a robust, reliable and computerized biomarker for early diagnosis of cognitive impairment and monitoring response to treatment,” the authors said.