Dataset Information

Multi-contrast multi-atlas parcellation of diffusion tensor imaging of the human brain.

ABSTRACT: In this paper, we propose a novel method for parcellating the human brain into 193 anatomical structures based on diffusion tensor images (DTIs). This was accomplished in the setting of multi-contrast diffeomorphic likelihood fusion using multiple DTI atlases. DTI images are modeled as high dimensional fields, with each voxel exhibiting a vector valued feature comprising of mean diffusivity (MD), fractional anisotropy (FA), and fiber angle. For each structure, the probability distribution of each element in the feature vector is modeled as a mixture of Gaussians, the parameters of which are estimated from the labeled atlases. The structure-specific feature vector is then used to parcellate the test image. For each atlas, a likelihood is iteratively computed based on the structure-specific vector feature. The likelihoods from multiple atlases are then fused. The updating and fusing of the likelihoods is achieved based on the expectation-maximization (EM) algorithm for maximum a posteriori (MAP) estimation problems. We first demonstrate the performance of the algorithm by examining the parcellation accuracy of 18 structures from 25 subjects with a varying degree of structural abnormality. Dice values ranging 0.8-0.9 were obtained. In addition, strong correlation was found between the volume size of the automated and the manual parcellation. Then, we present scan-rescan reproducibility based on another dataset of 16 DTI images - an average of 3.73%, 1.91%, and 1.79% for volume, mean FA, and mean MD respectively. Finally, the range of anatomical variability in the normal population was quantified for each structure.

SUBMITTER: Tang X

PROVIDER: S-EPMC4014574 | biostudies-other | 2014

REPOSITORIES: biostudies-other

ACCESS DATA

Publications

Multi-contrast multi-atlas parcellation of diffusion tensor imaging of the human brain.

Tang Xiaoying X Yoshida Shoko S Hsu John J Huisman Thierry A G M TA Faria Andreia V AV Oishi Kenichi K Kutten Kwame K Poretti Andrea A Li Yue Y Miller Michael I MI Mori Susumu S

PloS one 20140508 5

In this paper, we propose a novel method for parcellating the human brain into 193 anatomical structures based on diffusion tensor images (DTIs). This was accomplished in the setting of multi-contrast diffeomorphic likelihood fusion using multiple DTI atlases. DTI images are modeled as high dimensional fields, with each voxel exhibiting a vector valued feature comprising of mean diffusivity (MD), fractional anisotropy (FA), and fiber angle. For each structure, the probability distribution of eac ...[more]

PMID: 24809486

Similar Datasets

Project description:We investigated the potential of machine learning for diagnostic classification in late-life major depression based on an advanced whole brain white matter segmentation framework. Twenty-six late-life depression and 12 never depressed individuals aged > 55 years, matched for age, MMSE, and education underwent brain diffusion tensor imaging and a multi-contrast, multi-atlas segmentation in MRIcloud. Fractional anisotropy volume, mean fractional anisotropy, trace, axial and radial diffusivity (RD) extracted from 146 white matter parcels for each subject were used to train and test the AdaBoost classifier using stratified 12-fold cross validation. Performance was evaluated using various measures. The statistical power of the classifier was assessed using label permutation test. Statistical analysis did not yield significant differences in DTI measures between the groups. The classifier achieved a balanced accuracy of 71% and an Area Under the Receiver Operator Characteristic Curve (ROC-AUC) of 0.81 by trace, and a balanced accuracy of 70% and a ROC-AUC of 0.80 by RD, in limbic, cortico-basal ganglia-thalamo-cortical loop, brainstem, external and internal capsules, callosal and cerebellar structures. Both indices shared important structures for classification, while fornix was the most important structure for classification by both indices. The classifier proved statistically significant, as trace and RD ROC-AUC scores after permutation were lower than those obtained with the actual data (P = 0.022 and P = 0.024, respectively). Similar results were obtained with the Gradient Boosting classifier, whereas the RBF-kernel Support Vector Machine with k-best feature selection did not exceed the chance level. Finally, AdaBoost significantly predicted the class using all features together. Limitations are discussed. The results encourage further investigation of the implemented methods for computer aided diagnostics and anatomically informed therapeutics.

Project description:BACKGROUND:Brain microstructural maturation progresses rapidly in the third trimester of gestation and first weeks of life, but typical microstructural development may be influenced by the presence of critical congenital heart disease (CHD). OBJECTIVE:The aim of this study was to investigate the pattern of white matter (WM) microstructural development in neonates with different types of critical CHD. The secondary aim was to examine whether there is an association between WM microstructural maturity and neonatal ischemic brain injury. METHODS:For this prospective, longitudinal cohort study, 74 term born neonates underwent diffusion tensor imaging (DTI) before (N?=?56) and after (N?=?71) cardiac surgery performed <30?days of life for transposition of the great arteries (TGA), single ventricle physiology with aortic arch obstruction (SVP-AO), left- (LVOTO) or right ventricle outflow tract obstruction (RVOTO). Microstructural integrity was investigated by fractional anisotropy (FA) and by mean diffusivity (MD) in 16 white matter (WM) structures in three WM regions with correction for postmenstrual age. Ischemic brain injury was defined as moderate-severe white matter injury or stroke. RESULTS:Before cardiac surgery, the posterior parts of the corona radiata and internal capsule showed significantly higher FA and lower MD compared to the anterior parts. Centrally-located WM structures demonstrated higher FA compared to peripherally-located structures. Neonates with TGA had higher FA in projection-, association- and commissural WM before surgery, when compared to other CHD groups. Neonates with LVOTO showed lower preoperative MD in these regions, and neonates with SVP-AO higher MD. Differences in FA/MD between CHD groups were most clear in centrally located WM structures. Between CHD groups, no differences in postoperative FA/MD or in change from pre- to postoperative FA/MD were seen. Neonatal ischemic brain injury was not associated with pre- or postoperative FA/MD. CONCLUSIONS:Collectively, these findings revealed brain microstructural WM development to follow the same organized pattern in critical CHD as reported in healthy and preterm neonates, from posterior-to-anterior and central-to-peripheral. Neonates with TGA and LVOTO showed the most mature WM microstructure before surgery and SVP-AO the least mature. Degree of WM microstructural immaturity was not associated with ischemic brain injury.

Dataset Information

Multi-contrast multi-atlas parcellation of diffusion tensor imaging of the human brain.

Publications

Multi-contrast multi-atlas parcellation of diffusion tensor imaging of the human brain.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets