Postmortem neuropathologies are associated with retrospective tensor-based morphometry findings in Alzheimer's...

Understanding the relationship between postmortem neuropathological findings and in vivo imaging biomarkers is crucial for advancing the early diagnosis and treatment of Alzheimer's disease (AD). This study investigates the association between postmortem neuropathologies and retrospective Tensor-Based Morphometry (TBM) findings across the Alzheimer's dementia continuum.

This study utilized tensor-based morphometry (TBM) to investigate structural brain changes associated with AD neuropathology. Data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) were analyzed, including MRI scans and neuropathological findings from 98 participants. Logistic regression was used to explore the relationship between TBM indices and neuropathological outcomes. The diagnostic performance of TBM was evaluated using accuracy analysis.

Significant differences in TBM indices (StatROI and TempROI) were observed between the dementia and non-dementia groups. Logistic regression indicated that higher StatROI and TempROI indices were associated with lower odds of AD pathology, such as amyloid plaques (OR = 0.939 and 0.927), neurofibrillary tangles (OR = 0.961 and 0.931), and neuritic plaques (OR = 0.943 and 0.916). Accuracy analysis demonstrated good discriminatory power of TBM indices in differentiating pathologically confirmed AD from non-AD dementia (sensitivity up to 61.29, specificity up to 93.33%) and cognitively normal individual (sensitivity up to 82.26, specificity up to 75%).

TBM is a promising tool for predicting the underlying neuropathology of AD and distinguishing between different causes of dementia. Integrating TBM into clinical assessments can enhance the accuracy of AD diagnosis and improve patient management.