Structural MRI in frontotemporal dementia: comparisons between hippocampal volumetry, tensor-based morphometry and...

Area of Science:

• Neuroimaging
• Neurology
• Radiology

Background:

• Magnetic Resonance Imaging (MRI) is crucial for diagnosing dementia-like diseases, including Frontotemporal Dementia (FTD).
• Standardized and accurate MRI analysis methods are needed for improved diagnosis and differentiation of neurodegenerative conditions.

Purpose of the Study:

• To compare the diagnostic accuracy of three automated MRI analysis techniques: Hippocampal Volumetry (HV), Tensor-based Morphometry (TBM), and Voxel-based Morphometry (VBM).
• To differentiate Frontotemporal Dementia (FTD) from Alzheimer's Disease (AD) and Mild Cognitive Impairment (MCI) using these methods in specific brain regions.

Main Methods:

• Thirty-seven FTD, 46 AD, 26 control, 16 progressive MCI (PMCI), and 48 stable MCI (SMCI) subjects underwent MRI scans.
• Analysis included HV for hippocampal volume, TBM for shape changes, and VBM for gray matter density.
• Correct Classification Rate (CCR), sensitivity (SS), and specificity (SP) were calculated to assess diagnostic performance.

Main Results:

• Voxel-based morphometry (VBM) demonstrated the highest accuracy in differentiating FTD from AD (CCR = 0.72), particularly in the lateral ventricles.
• VBM also showed high accuracy in distinguishing controls from FTD (CCR = 0.87), especially in the frontal and occipital horns of the lateral ventricles.
• Tensor-based morphometry (TBM) showed good performance in differentiating controls from FTD in the hippocampus and amygdala (CCR = 0.80).

Conclusions:

• Hippocampal atrophy is a feature observed in both Alzheimer's Disease (AD) and Frontotemporal Dementia (FTD).
• Voxel-based morphometry (VBM) emerged as the most accurate method for differentiating between FTD and AD among the techniques evaluated.