A novel approach to analyzing brain structure that focuses on the shape, rather than the size, of particular features may allow identification of individuals who are in the early, pre-symptomatic stages of Alzheimer’s disease.
A team of investigators used advanced computational tools to analyze data from standard MRI scans. They found that people with Alzheimer’s disease, including those diagnosed partway through a multiyear study, had greater levels of asymmetry in key brain structures: differences in shape between the left and right sides of the brain. Their study has been published in the journal Brain.
The team developed a computer-aided system, called BrainPrint, for representing the whole brain based on the shape, rather than the size or volume, of structures. Originally described in a 2015 article in NeuroImage, BrainPrint appears to be as accurate as a fingerprint in distinguishing among individuals. In a recent paper in the same journal, the researchers demonstrated the use of BrainPrint for automated diagnosis of Alzheimer’s disease.
The current study used BrainPrint to analyze structural asymmetries in a series of MR images of almost 700 participants in the National Institutes of Health-sponsored Alzheimer’s Disease Neuroimaging Initiative. BrainPrint analysis of the data revealed that initial, between-hemisphere differences in the shapes of the hippocampus and amygdala—structures known to be sites of neurodegeneration in Alzheimer’s disease—were highest in individuals with dementia and lowest in healthy controls. Among those originally classified with mild cognitive impairment, baseline asymmetry was higher in those that progressed to Alzheimer’s dementia and became even greater as symptoms developed. Increased asymmetry was also associated with poorer cognitive test scores and with increased cortical atrophy.
Reprinted from materials provided by Mass General.