Brain MR Radiomics to Differentiate Cognitive Disorders

Area of Science:

• Neuroimaging
• Radiomics
• Cognitive Neuroscience

Background:

• Neurodegenerative disorders cause subtle brain changes years before cognitive decline.
• Early detection of cognitive impairment is crucial for timely intervention.

Purpose of the Study:

• To evaluate hippocampal texture and volumetric features from MRI for differentiating cognitive groups.
• To assess the utility of these features in classifying normal cognition, mild cognitive impairment, and Alzheimer's disease.
• To correlate imaging features with Clinical Dementia Rating (CDR) scores.

Main Methods:

• Analysis of 3-T T1-weighted MRI data from 173 patients (Alzheimer's Disease Neuroimaging Initiative).
• Extraction of texture and volumetric features from bilateral hippocampi.
• Binary classification using diagonal quadratic discriminant analysis with leave-one-out cross-validation.

Main Results:

• Hippocampal texture analysis effectively distinguished between normal cognition and early-stage impairment.
• Volumetric features were more effective in differentiating normal cognition from advanced impairment.
• The models demonstrated promising performance in classifying cognitive states.

Conclusions:

• Magnetic resonance (MR) radiomics, particularly hippocampal texture and volumetric analysis, shows potential for classifying cognitive groups.
• This approach may aid in the early detection and staging of cognitive decline associated with neurodegenerative disorders.