No abstract

PurposeTo determine the precision and accuracy of an automated method for segmenting white matter hyperintensities (WMH) on fast fluid-attenuated inversion-recovery (FLAIR) images in elderly brains at 3T.Materials and MethodsFLAIR images from 18 individuals (60–82 years, 9 females) with WMH burdens ranging from 1–80 cm3 were used. The protocol included the removal of clearly hyperintense voxels; two-class fuzzy C-means clustering (FCM); and thresholding to segment probable WMH. Two false-positive minimization (FPM) methods using white matter templates were tested. Precision was assessed by adding synthetic hyperintense voxels to brain slices. Accuracy was validated by comparing automatic and manual segmentations. Whole-brain, voxel-wise metrics of similarity, under- and overestimation were used to evaluate both precision and accuracy.ResultsPrecision was high, as the lowest accuracy in the synthetic datasets was 93%. Both FPM strategies successfully improved overall accuracy. Whole-brain accuracy for the FCM segmentation alone ranged from 45%–81%, which improved to 75%–85% using the FPM strategies.ConclusionThe method was accurate across the range of WMH burden typically seen in the elderly. Accuracy levels achieved or exceeded those of other approaches using multispectral and/or more sophisticated pattern recognition methods. J. Magn. Reson. Imaging 2010;31:1311–1322. © 2010 Wiley-Liss, Inc.

Background and Purpose Post-stroke cognitive impairment (PSCI) is typified by prominent deficits in processing speed and executive function. However, the underlying neuroanatomical substrates of executive deficits are not well understood and further elucidation is needed. There may be utility in fractionating executive functions to delineate neural substrates. Methods One test amenable to fine delineation is the Trail Making Test (TMT), which emphasizes processing speed (TMT-A) and set-shifting (TMT-B-A difference, proportion, quotient scores and TMT-B set-shifting errors). The TMT was administered to two overt ischemic stroke cohorts from a multinational study: (i) a chronic stroke cohort (N=61) and (ii) an acute-sub-acute stroke cohort (N=45). Volumetric quantification of ischemic stroke and White Matter HyperIntensities (WMH) was done on MRI, along with ratings of involvement of cholinergic projections, using the previously published Cholinergic Hyperintensities Projections Scale (CHIPS). Damage to the superior longitudinal fasciculus (SLF), which co-localizes with some cholinergic projections, was also documented. Results Multiple linear regression analyses were completed. While larger infarcts (β=0.37, p<0.0001) were associated with slower processing speed, CHIPS severity (β=0.39, p<0.0001) was associated with all metrics of set shifting. Left SLF damage, however, was only associated with the difference score (β=0.17, p=0.03). These findings were replicated in both cohorts. Patients with ≥2 TMT-B set shifting errors also had greater CHIPS severity. Conclusions In this multinational stroke cohort study, damage to lateral cholinergic pathways and the SLF emerged as significant neuroanatomical correlates for executive deficits in set shifting.

The ability to properly distinguish facial emotions has a protracted development, not maturing until well into adolescence. Emotional faces activate emotion-specific neural networks in adults; whether these networks are operational in children is not known. Using an implicit face-processing task in 10-year-old children, we determined that the emotions of fear, disgust and sadness recruited distinct neural systems. These systems included a number of regions typically associated with processing emotions in adults, namely the amygdala and parahippocampal gyrus, insula and cingulate gyrus, as well as the fusiform and superior temporal gyri. Thus, in spite of immature behavioral responses to emotional faces in explicit tasks, neural networks for emotion-specific processing are present in young children.

Hippocampal volumetry derived from structural MRI is increasingly used to delineate regions of interest for functional measurements, assess efficacy in therapeutic trials of Alzheimer’s disease (AD) and has been endorsed by the new AD diagnostic guidelines as a radiological marker of disease progression. Unfortunately, morphological heterogeneity in AD can prevent accurate demarcation of the hippocampus. Recent developments in automated volumetry commonly use multitemplate fusion driven by expert manual labels, enabling highly accurate and reproducible segmentation in disease and healthy subjects. However, there are several protocols to define the hippocampus anatomically in vivo, and the method used to generate atlases may impact automatic accuracy and sensitivity – particularly in pathologically heterogeneous samples. Here we report a fully automated segmentation technique that provides a robust platform to directly evaluate both technical and biomarker performance in AD among anatomically unique labeling protocols. For the first time we test head-to-head the performance of five common hippocampal labeling protocols for multi-atlas based segmentation, using both the Sunnybrook Longitudinal Dementia Study and the entire Alzheimer’s Disease Neuroimaging Initiative 1 (ADNI-1) baseline and 24-month dataset. We based these atlas libraries on the protocols of (Haller et al., 1997; Killiany et al., 1993; Malykhin et al., 2007; Pantel et al., 2000; Pruessner et al., 2000), and a single operator performed all manual tracings to generate de facto “ground truth” labels. All methods distinguished between normal elders, mild cognitive impairment (MCI), and AD in the expected directions, and showed comparable correlations with measures of episodic memory performance. Only more inclusive protocols distinguished between stable MCI and MCI-to-AD converters, and had slightly better associations with episodic memory. Moreover, we demonstrate that protocols including more posterior anatomy and dorsal white matter compartments furnish the best voxel-overlap accuracies (Dice Similarity Coefficient = 0.87–0.89), compared to expert manual tracings, and achieve the smallest sample sizes required to power clinical trials in MCI and AD. The greatest distribution of errors was localized to the caudal hippocampus and alveus-fimbria compartment when these regions were excluded. The definition of the medial body did not significantly alter accuracy among more comprehensive protocols. Voxel-overlap accuracies between automatic and manual labels were lower for the more pathologically heterogeneous Sunnybrook study in comparison to the ADNI-1 sample. Finally, accuracy among protocols appears to significantly differ the most in AD subjects compared to MCI and normal elders. Together, these results suggest that selection of a candidate protocol for fully automatic multi-template based segmentation in AD can influence both segmentation accuracy when compared to expert manual labels and performance as a biomarker in MCI and AD.

No abstract