NeuroImage. Clinical
-
NeuroImage. Clinical · Jan 2014
Randomized Controlled Trial Comparative StudyRandom Forest ensembles for detection and prediction of Alzheimer's disease with a good between-cohort robustness.
Computer-aided diagnosis of Alzheimer's disease (AD) is a rapidly developing field of neuroimaging with strong potential to be used in practice. In this context, assessment of models' robustness to noise and imaging protocol differences together with post-processing and tuning strategies are key tasks to be addressed in order to move towards successful clinical applications. In this study, we investigated the efficacy of Random Forest classifiers trained using different structural MRI measures, with and without neuroanatomical constraints in the detection and prediction of AD in terms of accuracy and between-cohort robustness. ⋯ The models trained using parcelled and high-dimensional (HD) input demonstrated equivalent performance, but the former was more effective in terms of computation/memory and time costs. The sensitivity/specificity for detecting MCI-to-AD conversion (but not AD/HC classification performance) was further improved from 79.5%/75%-83.3%/81.3% by a combination of morphometric measurements with ApoE-genotype and demographics (age, sex, education). When applied to the independent AddNeuroMed cohort, the best ADNI models produced equivalent performance without substantial accuracy drop, suggesting good robustness sufficient for future clinical implementation.
-
NeuroImage. Clinical · Jan 2014
Comparative StudyMeasurement of brain perfusion in newborns: pulsed arterial spin labeling (PASL) versus pseudo-continuous arterial spin labeling (pCASL).
Arterial spin labeling (ASL) perfusion-weighted imaging (PWI) by magnetic resonance imaging (MRI) has been shown to be useful for identifying asphyxiated newborns at risk of developing brain injury, whether or not therapeutic hypothermia was administered. However, this technique has been only rarely used in newborns until now, because of the challenges to obtain sufficient signal-to-noise ratio (SNR) and spatial resolution in newborns. ⋯ This study demonstrates that both ASL methods are feasible to assess brain perfusion in healthy and sick newborns. However, pCASL might be a better choice over PASL in newborns, as pCASL perfusion maps had a superior image quality that allowed a more detailed identification of the different brain structures.
-
NeuroImage. Clinical · Jan 2014
Alterations in the optic radiations of very preterm children-Perinatal predictors and relationships with visual outcomes.
Children born very preterm (VPT) are at risk for visual impairments, the main risk factors being retinopathy of prematurity and cerebral white matter injury, however these only partially account for visual impairments in VPT children. This study aimed to compare optic radiation microstructure and volume between VPT and term-born children, and to investigate associations between 1) perinatal variables and optic radiations; 2) optic radiations and visual function in VPT children. We hypothesized that optic radiation microstructure would be altered in VPT children, predicted by neonatal cerebral white matter abnormality and retinopathy of prematurity, and associated with visual impairments. 142 VPT children and 32 controls underwent diffusion-weighted magnetic resonance imaging at 7 years of age. ⋯ Neonatal white matter abnormalities and retinopathy of prematurity were associated with optic radiation diffusion values. Lower fractional anisotropy in the anterior sub-regions was associated with poor visual acuity and increased likelihood of other visual defects. This study presents evidence for microstructural alterations in the optic radiations of VPT children, which are largely predicted by white matter abnormality or severe retinopathy of prematurity, and may partially explain the higher rate of visual impairments in VPT children.
-
NeuroImage. Clinical · Jan 2014
Neonatal physiological correlates of near-term brain development on MRI and DTI in very-low-birth-weight preterm infants.
Structural brain abnormalities identified at near-term age have been recognized as potential predictors of neurodevelopment in children born preterm. The aim of this study was to examine the relationship between neonatal physiological risk factors and early brain structure in very-low-birth-weight (VLBW) preterm infants using structural MRI and diffusion tensor imaging (DTI) at near-term age. Structural brain MRI, diffusion-weighted scans, and neonatal physiological risk factors were analyzed in a cross-sectional sample of 102 VLBW preterm infants (BW ≤ 1500 g, gestational age (GA) ≤ 32 weeks), who were admitted to the Lucile Packard Children's Hospital, Stanford NICU and recruited to participate prior to routine near-term brain MRI conducted at 36.6 ± 1.8 weeks postmenstrual age (PMA) from 2010 to 2011; 66/102 also underwent a diffusion-weighted scan. ⋯ Results suggest that at near-term age, thalamus WM microstructure may be particularly vulnerable to certain neonatal risk factors. Interactions between albumin, bilirubin, phototherapy, and brain development warrant further investigation. Identification of physiological risk factors associated with selective vulnerability of certain brain regions at near-term age may clarify the etiology of neurodevelopmental impairment and inform neuroprotective treatment for VLBW preterm infants.
-
NeuroImage. Clinical · Jan 2014
Voxel-based clustered imaging by multiparameter diffusion tensor images for glioma grading.
Gliomas are the most common intra-axial primary brain tumour; therefore, predicting glioma grade would influence therapeutic strategies. Although several methods based on single or multiple parameters from diagnostic images exist, a definitive method for pre-operatively determining glioma grade remains unknown. We aimed to develop an unsupervised method using multiple parameters from pre-operative diffusion tensor images for obtaining a clustered image that could enable visual grading of gliomas. ⋯ Furthermore, the log-ratio value of each class of the 16-class diffusion tensor-based clustered images was compared between low- and high-grade gliomas, and the log-ratio values of classes 14, 15 and 16 in the high-grade gliomas were significantly higher than those in the low-grade gliomas (p < 0.005, p < 0.001 and p < 0.001, respectively). These classes comprised different patterns of the seven diffusion tensor imaging-based parameters. The results suggest that the multiple diffusion tensor imaging-based parameters from the voxel-based diffusion tensor-based clustered images can help differentiate between low- and high-grade gliomas.