Brain MRI Volumetry for Cognitive Prediction in Early Alzheimer’s: A Head-to-Head Software Comparison

In a retrospective study of 255 early Alzheimer’s patients, brain volumetry software-based models predicted MMSE score with r=0.56, but domain-specific cognitive scores were modestly correlated (r=0.21-0.40) and visuospatial function not significantly predicted.

• Retrospective analysis of 255 amyloid PET-confirmed early Alzheimer’s disease patients; volumetric features from four software (FreeSurfer, AQUA, DeepBrain, A-finder) were combined with seven clinical variables to train a multi-layer perceptron model.
• FreeSurfer achieved the highest correlation for MMSE (r=0.56) and most domains; AQUA led for attention (r=0.40); DeepBrain for memory (r=0.38). Visuospatial function showed no significant prediction across all models.
• Limitations include retrospective, single-center design and lack of external or prospective validation, limiting generalizability of software-specific comparisons.

RadPigeon summaries are original and for information only. They are not clinical advice.