Markiewicz, Pawel J.
Optimized data preprocessing for multivariate analysis applied to 99mTc-ECD SPECT data sets of Alzheimer's patients and asymptomatic controls
2011-01, Merhof, Dorit, Markiewicz, Pawel J., Platsch, Günther, Declerck, Jerome, Weih, Markus, Kornhuber, Johannes, Kuwert, Torsten, Matthews, Julian C., Herholz, Karl
Multivariate image analysis has shown potential for classification between Alzheimer's disease (AD) patients and healthy controls with a high-diagnostic performance. As image analysis of positron emission tomography (PET) and single photon emission computed tomography (SPECT) data critically depends on appropriate data preprocessing, the focus of this work is to investigate the impact of data preprocessing on the outcome of the analysis, and to identify an optimal data preprocessing method. In this work, technetium-99methylcysteinatedimer (99mTc-ECD) SPECT data sets of 28 AD patients and 28 asymptomatic controls were used for the analysis. For a series of different data preprocessing methods, which includes methods for spatial normalization, smoothing, and intensity normalization, multivariate image analysis based on principal component analysis (PCA) and Fisher discriminant analysis (FDA) was applied. Bootstrap resampling was used to investigate the robustness of the analysis and the classification accuracy, depending on the data preprocessing method. Depending on the combination of preprocessing methods, significant differences regarding the classification accuracy were observed. For 99mTc-ECD SPECT data, the optimal data preprocessing method in terms of robustness and classification accuracy is based on affine registration, smoothing with a Gaussian of 12 mm full width half maximum, and intensity normalization based on the 25% brightest voxels within the whole-brain region.
Classification accuracy of multivariate analysis applied to 99mTc-ECD SPECT data in Alzheimer s disease patients and asymptomatic controls
2009-10, Merhof, Dorit, Markiewicz, Pawel J., Declerck, Jérôme, Platsch, Günther, Matthews, Julian C., Herholz, Karl
With increasing life expectancy in developed countries, there is a corresponding increase in the frequency of diseases typically associated with old age, in particular dementia. In recent research, multivariate analysis of Positron Emission Tomography (PET) datasets has shown potential for classification between Alzheimer s disease (AD) patients and asymptomatic controls. In this work, the feasibility of multivariate analysis using Principal Component Analysis (PCA) and Fisher Discriminant Analysis (FDA) of Single Photon Emission Computed Tomography (SPECT) data is investigated. In order to obtain robust and reliable results, bootstrap resampling is applied and the robustness and classification accuracy of PCA/FDA are investigated. The robustness of the analysis is assessed by estimating the distribution of the angle between PCA/FDA discriminative vectors generated by bootstrap resampling, and the classification predictive accuracy is assessed using the .632 bootstrap estimator. The results indicate that PCA/FDA on SPECT data enables a robust differentiation between AD patients and asymptomatic controls based on three principal components, with a classification accuracy of 89%.