Biomarker discovery for sparse classification of brain images in Alzheimer's disease
| Authors | |
|---|---|
| Year of publication | 2012 |
| Type | Article in Periodical |
| Magazine / Source | Annals of the BMVA |
| MU Faculty or unit | |
| Citation | |
| Web | http://www.bmva.org/annals/2012/2012-0002.pdf |
| Field | Neurology, neurosurgery, neurosciences |
| Description | We describe a computationally efficient biomarker discovery approach, based on a combination of penalised regression and a resampling method, for the identification of localised brain regions that are highly discriminative between two groups of brain images. The proposed procedure has been applied for classification of brain images in subjects with Alzheimer’s disease and mild cognitive impairment using baseline FDG-PET data and both baseline and longitudinal MRI data. Results of nine independent classification experiments show that the selected biomarkers are consistent with well-known patterns of atrophy, hypometabolism and progression of Alzheimer’s disease that have been reported in previous studies. The overall classification performance, which has been assessed for statistical significance, is comparable to related state-of-the-art findings. |
| Related projects: |