Subspace corrected relevance learning with application in neuroimaging.

In machine learning, data often comes from different sources, but combining them can introduce extraneous variation that affects both generalization and interpretability. For example, we investigate the classification of neurodegenerative diseases using FDG-PET data collected from multiple neuroimaging centers. However, data collected at different centers introduces unwanted variation due to differences in scanners, scanning protocols, and processing methods.

Comparison of univariate and multivariate analyses for brain [18F]FDG PET data in α-synucleinopathies.

Background: Brain imaging with [18F]FDG-PET can support the diagnostic work-up of patients with α-synucleinopathies. Validated data analysis approaches are necessary to evaluate disease-specific brain metabolism patterns in neurodegenerative disorders. This study compared the univariate Statistical Parametric Mapping (SPM) single-subject procedure and the multivariate Scaled Subprofile Model/Principal Component Analysis (SSM/PCA) in a cohort of patients with α-synucleinopathies.

Relation of 18-F-Dopa PET with hypokinesia-rigidity, tremor and freezing in Parkinson's disease.

Introduction: In this retrospective study concerning patients with Parkinson's disease (PD) scanned with 18-F-Dopa PET (N = 129), we looked for an association between reduced 18-F-Dopa uptake and the key PD symptoms tremor and hypokinesia-rigidity. We hypothesized to find a stronger correlation between dopaminergic depletion in the striatum and hypokinesia-rigidity compared to tremor.

Methods: The onset side of symptoms (documented for 102 patients) as well as the first registered UPDRS (available for 79 patients) was used to correlate with F-Dopa uptake values in the caudate nucleus and putamen in this large retrospective sample.

Neuroimaging methods applied in Parkinson's disease.

Radiotracer methods provide regional in vivo quantified information about specific biochemical activities in brain tissue. The understanding of the principles governing radiotracer uptake into brain tissue determines the potential value of these tracers in assessing pathophysiology of brain diseases. Too often a reductionist view of images is taken to directly point to clinical features or even diagnoses of brain diseases.

Significance of non-presynaptic SPECT tracer methods in Parkinson's disease.

Tropane-related tracers for SPECT and PET scanning have shown their value in assessing the in vivo striatal presynaptic status of the dopaminergic neurotransmitter system in man. Previously, there were a few other tracers suggested to be useful in the study of in vivo brain biochemistry in humans, particularly in patients afflicted with Parkinson's disease (PD) or parkinsonian syndromes. We summarize the work concerning the two main tracers applied in that context using single photon emission tomography (SPECT) scanning in patients with PD: IBZM and HMPAO.