Dataset Information

A Bayesian Spatial Model to Predict Disease Status Using Imaging Data From Various Modalities.

ABSTRACT: Relating disease status to imaging data stands to increase the clinical significance of neuroimaging studies. Many neurological and psychiatric disorders involve complex, systems-level alterations that manifest in functional and structural properties of the brain and possibly other clinical and biologic measures. We propose a Bayesian hierarchical model to predict disease status, which is able to incorporate information from both functional and structural brain imaging scans. We consider a two-stage whole brain parcellation, partitioning the brain into 282 subregions, and our model accounts for correlations between voxels from different brain regions defined by the parcellations. Our approach models the imaging data and uses posterior predictive probabilities to perform prediction. The estimates of our model parameters are based on samples drawn from the joint posterior distribution using Markov Chain Monte Carlo (MCMC) methods. We evaluate our method by examining the prediction accuracy rates based on leave-one-out cross validation, and we employ an importance sampling strategy to reduce the computation time. We conduct both whole-brain and voxel-level prediction and identify the brain regions that are highly associated with the disease based on the voxel-level prediction results. We apply our model to multimodal brain imaging data from a study of Parkinson's disease. We achieve extremely high accuracy, in general, and our model identifies key regions contributing to accurate prediction including caudate, putamen, and fusiform gyrus as well as several sensory system regions.

SUBMITTER: Xue W

PROVIDER: S-EPMC5879954 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:The dominant approach to neuroimaging data analysis employs the voxel as the unit of computation. While convenient, voxels lack biological meaning and their size is arbitrarily determined by the resolution of the image. Here, we propose a multivariate spatial model in which neuroimaging data are characterised as a linearly weighted combination of multiscale basis functions which map onto underlying brain nuclei or networks or nuclei. In this model, the elementary building blocks are derived to reflect the functional anatomy of the brain during the resting state. This model is estimated using a Bayesian framework which accurately quantifies uncertainty and automatically finds the most accurate and parsimonious combination of basis functions describing the data. We demonstrate the utility of this framework by predicting quantitative SPECT images of striatal dopamine function and we compare a variety of basis sets including generic isotropic functions, anatomical representations of the striatum derived from structural MRI, and two different soft functional parcellations of the striatum derived from resting-state fMRI (rfMRI). We found that a combination of ?50 multiscale functional basis functions accurately represented the striatal dopamine activity, and that functional basis functions derived from an advanced parcellation technique known as Instantaneous Connectivity Parcellation (ICP) provided the most parsimonious models of dopamine function. Importantly, functional basis functions derived from resting fMRI were more accurate than both structural and generic basis sets in representing dopamine function in the striatum for a fixed model order. We demonstrate the translational validity of our framework by constructing classification models for discriminating parkinsonian disorders and their subtypes. Here, we show that ICP approach is the only basis set that performs well across all comparisons and performs better overall than the classical voxel-based approach. This spatial model constitutes an elegant alternative to voxel-based approaches in neuroimaging studies; not only are their atoms biologically informed, they are also adaptive to high resolutions, represent high dimensions efficiently, and capture long-range spatial dependencies, which are important and challenging objectives for neuroimaging data.

Project description:ContextBecause ectopic ACTH-secreting (EAS) tumors are often occult, improved imaging is needed.ObjectiveOur objective was to evaluate the utility of [(111)In-DTPA-d-Phe]pentetreotide scintigraphy [octreotide (OCT)] imaging at 6 mCi [low OCT (LOCT)] and 18 mCi [high OCT (HOCT)], [(18)F]fluorodeoxyglucose (FDG)-positron emission tomography (PET) and [(18)F]l-3,4-dihydroxyphenylalanine (F-DOPA)-PET scans, computed tomography (CT), and magnetic resonance imaging (MRI).Design and settingThe study was a prospective evaluation at a clinical research center.PatientsForty-one subjects participated, 30 (17 female) with resected EAS tumors and 11 (three female) with occult EAS, based on inferior petrosal sinus sampling results and imaging studies.InterventionINTERVENTION included CT and MRI of neck, chest, abdomen, LOCT (with or without HOCT) and FDG- or F-DOPA-PET without CT every 6-12 months.Main outcome measureTumor identification was the main outcome measure.ResultsMost recent results were analyzed. Eighteen patients had tumor resected on the first visit; otherwise, surgery occurred 33 +/- 25 (9-99) months later. Tumor size was 1.9 +/- 1.7 (0.8-8.0) cm; 83% were intrathoracic. CT, MRI, LOCT, HOCT, FDG-PET, and F-DOPA-PET had sensitivities per patient of 93% [95% confidence interval (CI) = 79-98%], 90% (95% CI = 74-96%), 57% (95% CI = 39-73%), 50% (95% CI = 25-75%), 64% (95% CI = 35-85%), and 55% (95% CI = 28-79%) and positive predictive values (PPV) per lesion of 66, 74, 79, 89, 53, and 100%, respectively. LOCT and PET detected only lesions seen by CT/MRI; abnormal LOCT or F-DOPA-PET improved PPV of CT/MRI. By modality, the fraction of patients with one or more false-positive findings was 50% by CT, 31% by MRI, 18% by L/HOCT, and 18% by FDG-PET. Eight occult EAS patients had 64 +/- 58 (9-198) months follow-up; others had none.ConclusionsHigh sensitivity and PPV suggest thoracic CT/MRI plus LOCT scans for initial imaging, with lesion confirmation by two modalities.

Dataset Information

A Bayesian Spatial Model to Predict Disease Status Using Imaging Data From Various Modalities.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets