Dataset Information

Predicting cortical ROIs via joint modeling of anatomical and connectional profiles.

ABSTRACT: Localization of cortical regions of interests (ROIs) in structural neuroimaging data such as diffusion tensor imaging (DTI) and T1-weighted MRI images has significant importance in basic and clinical neurosciences. However, this problem is considerably challenging due to the lack of quantitative mapping between brain structure and function, which relies on the availability of multimodal training data including benchmark task-based functional MRI (fMRI) images and effective machine learning algorithms. This paper presents a novel joint modeling approach that learns predictive models of ROIs from concurrent task-based fMRI, DTI, and T1-weighted MRI datasets. In particular, the effective generalized multiple kernel learning (GMKL) algorithm and ROI coordinate principal component analysis (PCA) model are employed to infer the intrinsic relationships between anatomical T1-weighted MRI/connectional DTI features and task-based fMRI-derived functional ROIs. Then, these predictive models of cortical ROIs are evaluated by cross-validation studies, independent datasets, and reproducibility studies. Experimental results are promising. We envision that these predictive models can be potentially applied in many scenarios that have only DTI and/or T1-weighted MRI data, but without task-based fMRI data.

SUBMITTER: Zhang T

PROVIDER: S-EPMC3927783 | biostudies-literature | 2013 Aug

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Predicting cortical ROIs via joint modeling of anatomical and connectional profiles.

Zhang Tuo T Zhu Dajiang D Jiang Xi X Ge Bao B Hu Xintao X Han Junwei J Guo Lei L Liu Tianming T

Medical image analysis 20130412 6

Localization of cortical regions of interests (ROIs) in structural neuroimaging data such as diffusion tensor imaging (DTI) and T1-weighted MRI images has significant importance in basic and clinical neurosciences. However, this problem is considerably challenging due to the lack of quantitative mapping between brain structure and function, which relies on the availability of multimodal training data including benchmark task-based functional MRI (fMRI) images and effective machine learning algor ...[more]

PMID: 23666264

Similar Datasets

Project description:Visual attentional bias forms for rewarding and punishing stimuli in the environment. While this attentional bias is adaptive in healthy situations, it is maladaptive in disorders such as drug addiction or PTSD. In both these disorders, the ability to exert control over this attentional bias is associated with drug abstinence rates or reduced PTSD symptoms, indicating the interaction of visual attention, cognitive control, and stimulus association. The inferior parietal lobule (IPL) is central to attention, while the prefrontal cortex (PFC) is critical for reward, cognitive control, and attention. Importantly, regions of the IPL and PFC commonly project to the rostral dorsal caudate (rdCaud) of the striatum. We propose an anatomical network architecture in which IPL projections converge with PFC projections in a connectional hub in the rdCaud, providing an anatomical substrate for the interaction of these projections and their competitive influence on striatal processing. To investigate this, we mapped the dense projections from the caudal IPL and prefrontal (dlPFC, vlPFC, OFC, dACC, and dmPFC) regions that project to the medial rdCaud with anatomical tract-tracing tracer injections in monkeys. These inputs converge in a precise site in the medial rdCaud, rostral to the anterior commissure. Small retrograde tracer injections confirmed these inputs to the medial rdCaud and showed that a proximal ventral striatal location has a very different pattern of cortical inputs. We next used human resting-state functional connectivity MRI (fcMRI) to examine whether a striatal hub exists in the human medial rdCaud. Seed regions in the human medial rdCaud revealed cortical correlation maps similar to the monkey retrograde injection results. A subsequent analysis of these correlated cortical regions showed that their peak correlation within the striatum is in the medial rdCaud, indicating that this is a connectional hub. In contrast, this peak striatal correlation was not found in the ventral striatal location, suggesting that this site is not a connectional hub of cortical regions. Taken together, this work uses the precision of monkey anatomy to identify a connectional hub of IPL and PFC projections in the medial rdCaud. It also translates this anatomical precision to humans, demonstrating that, guided by anatomy, connectional hubs can be identified in humans with fcMRI. These connectional hubs provide more specific treatment targets for drug addiction, PTSD, and other neurological and psychiatric disorders involving the striatum.

Project description:MOTIVATION:Alternative splicing generates multiple isoforms from a single gene, greatly increasing the functional diversity of a genome. Although gene functions have been well studied, little is known about the specific functions of isoforms, making accurate prediction of isoform functions highly desirable. However, the existing approaches to predicting isoform functions are far from satisfactory due to at least two reasons: (i) unlike genes, isoform-level functional annotations are scarce. (ii) The information of isoform functions is concealed in various types of data including isoform sequences, co-expression relationship among isoforms, etc. RESULTS:In this study, we present a novel approach, DIFFUSE (Deep learning-based prediction of IsoForm FUnctions from Sequences and Expression), to predict isoform functions. To integrate various types of data, our approach adopts a hybrid framework by first using a deep neural network (DNN) to predict the functions of isoforms from their genomic sequences and then refining the prediction using a conditional random field (CRF) based on co-expression relationship. To overcome the lack of isoform-level ground truth labels, we further propose an iterative semi-supervised learning algorithm to train both the DNN and CRF together. Our extensive computational experiments demonstrate that DIFFUSE could effectively predict the functions of isoforms and genes. It achieves an average area under the receiver operating characteristics curve of 0.840 and area under the precision-recall curve of 0.581 over 4184 GO functional categories, which are significantly higher than the state-of-the-art methods. We further validate the prediction results by analyzing the correlation between functional similarity, sequence similarity, expression similarity and structural similarity, as well as the consistency between the predicted functions and some well-studied functional features of isoform sequences. AVAILABILITY AND IMPLEMENTATION:https://github.com/haochenucr/DIFFUSE. SUPPLEMENTARY INFORMATION:Supplementary data are available at Bioinformatics online.

Dataset Information

Predicting cortical ROIs via joint modeling of anatomical and connectional profiles.

Publications

Predicting cortical ROIs via joint modeling of anatomical and connectional profiles.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets