Individual subject classification of mixed dementia from pure subcortical vascular dementia based on subcortical

Hee Jin Kim1, Jeonghun Kim, Hanna Cho

  • 1Department of Neurology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea.

Plos One
|October 17, 2013
PubMed

Insights

This study shows that analyzing hippocampal and amygdalar shape with MRI can accurately distinguish mixed dementia from subcortical vascular dementia, aiding treatment decisions for Alzheimer's disease pathology.

Area of Science:

  • Neuroimaging
  • Neurology
  • Dementia Research

Background:

  • Subcortical vascular dementia (SVaD) often co-occurs with Alzheimer's disease (AD) pathology, forming mixed dementia.
  • Accurate identification of mixed dementia is crucial for effective amyloid-targeted therapies.

Purpose of the Study:

  • To differentiate mixed dementia from pure SVaD using MRI.
  • To investigate the utility of hippocampal and amygdalar shape analysis in this discrimination.

Main Methods:

  • Utilized 11C-Pittsburgh compound B positron emission tomography (PiB-PET) to assess brain amyloid deposition in 68 SVaD patients.
  • Employed incremental learning with hippocampal and amygdalar shape analysis via MRI.
  • Defined PiB positivity with a retention ratio > 1.5.

Main Results:

  • 23 out of 68 SVaD patients (33.8%) were PiB(+), indicating mixed dementia.
  • Hippocampal shape analysis achieved 77.9% accuracy; amygdalar shape analysis achieved 75.0% accuracy.
  • Combined hippocampal and amygdalar shape analysis yielded the highest accuracy of 82.4%.

Conclusions:

  • Incremental learning using hippocampal and amygdalar shape effectively distinguishes mixed dementia from pure SVaD.
  • Amyloid and vascular pathologies differentially impact hippocampal and amygdalar shape.
  • This MRI-based approach offers a promising method for diagnosing mixed dementia.