Automatic classification of Alzheimer's Disease vs. Frontotemporal dementia: A spatial decision tree approach with FDG-PET
- Resource Type
- Conference
- Authors
- Sadeghi, N.; Foster, N. L.; Wang, A. Y.; Minoshima, S.; Lieberman, A. P.; Tasdizen, T.
- Source
- 2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro Biomedical Imaging: From Nano to Macro, 2008. ISBI 2008. 5th IEEE International Symposium on. :408-411 May, 2008
- Subject
- Bioengineering
Computing and Processing
Signal Processing and Analysis
Alzheimer's disease
Dementia
Decision trees
Principal component analysis
Positron emission tomography
Classification tree analysis
Cities and towns
Brain
Pixel
Clinical diagnosis
Brain imaging
decision tree
FDG-PET
Alzheimer’s Disease
Frontotemporal dementia
- Language
- ISSN
- 1945-7928
1945-8452
We introduce a novel approach for the automatic classification of FDG-PET scans of subjects with Alzheimers Disease (AD) and Frontotemporal dementia (FTD). Unlike previous work in the literature which focuses on principal component analysis and predefined regions of interest, we propose the combined use of information gain and spatial proximity to group cortical pixels into empirically determined regions that can best separate the two diseases. These regions are then used as attributes in a decision tree learning framework. We demonstrate that the proposed method provides better classification accuracy compared to other methods on a group of 48 autopsy confirmed AD and FTD patients.