Dataset Information

Personalized microstructural evaluation using a Mahalanobis-distance based outlier detection strategy on epilepsy patients' DTI data - Theory, simulations and example cases.

ABSTRACT: Quantitative MRI methods have recently gained extensive interest and are seeing substantial developments; however, their application in single patient vs control group comparisons is often limited by inherent statistical difficulties. One such application is detecting malformations of cortical development (MCDs) behind drug resistant epilepsies, a task that, especially when based solely on conventional MR images, may represent a serious challenge. We aimed to develop a novel straightforward voxel-wise evaluation method based on the Mahalanobis-distance, combining quantitative MRI data into a multidimensional parameter space and detecting lesion voxels as outliers. Simulations with standard multivariate Gaussian distribution and resampled DTI-eigenvalue data of 45 healthy control subjects determined the optimal critical value, cluster size threshold, and the expectable lesion detection performance through ROC-analyses. To reduce the effect of false positives emanating from registration artefacts and gyrification differences, an automatic classification method was applied, fine-tuned using a leave-one-out strategy based on diffusion and T1-weighted data of the controls. DWI processing, including thorough corrections and robust tensor fitting was performed with ExploreDTI, spatial coregistration was achieved with the DARTEL tools of SPM12. Additional to simulations, clusters of outlying diffusion profile, concordant with neuroradiological evaluation and independent calculations with the MAP07 toolbox were identified in 12 cases of a 13 patient example population with various types of MCDs. The multidimensional approach proved sufficiently sensitive in pinpointing regions of abnormal tissue microstructure using DTI data both in simulations and in the heterogeneous example population. Inherent limitations posed by registration artefacts, age-related differences, and the different or mixed pathologies limit the generalization of specificity estimation. Nevertheless, the proposed statistical method may aid the everyday examination of individual subjects, ever so more upon extending the framework with quantitative information from other modalities, e.g. susceptibility mapping, relaxometry, or perfusion.

SUBMITTER: Gyebnar G

PROVIDER: S-EPMC6756533 | biostudies-literature | 2019

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Personalized microstructural evaluation using a Mahalanobis-distance based outlier detection strategy on epilepsy patients' DTI data - Theory, simulations and example cases.

Gyebnár Gyula G Klimaj Zoltán Z Entz László L Fabó Dániel D Rudas Gábor G Barsi Péter P Kozák Lajos R LR

PloS one 20190923 9

Quantitative MRI methods have recently gained extensive interest and are seeing substantial developments; however, their application in single patient vs control group comparisons is often limited by inherent statistical difficulties. One such application is detecting malformations of cortical development (MCDs) behind drug resistant epilepsies, a task that, especially when based solely on conventional MR images, may represent a serious challenge. We aimed to develop a novel straightforward voxe ...[more]

PMID: 31545838

Similar Datasets

Project description:Binary phenotypes commonly arise due to multiple underlying quantitative precursors and genetic variants may impact multiple traits in a pleiotropic manner. Hence, simultaneously analyzing such correlated traits may be more powerful than analyzing individual traits. Various genotype-level methods, e.g., MultiPhen (O'Reilly et al. []), have been developed to identify genetic factors underlying a multivariate phenotype. For univariate phenotypes, the usefulness and applicability of allele-level tests have been investigated. The test of allele frequency difference among cases and controls is commonly used for mapping case-control association. However, allelic methods for multivariate association mapping have not been studied much. In this article, we explore two allelic tests of multivariate association: one using a Binomial regression model based on inverted regression of genotype on phenotype (Binomial regression-based Association of Multivariate Phenotypes [BAMP]), and the other employing the Mahalanobis distance between two sample means of the multivariate phenotype vector for two alleles at a single-nucleotide polymorphism (Distance-based Association of Multivariate Phenotypes [DAMP]). These methods can incorporate both discrete and continuous phenotypes. Some theoretical properties for BAMP are studied. Using simulations, the power of the methods for detecting multivariate association is compared with the genotype-level test MultiPhen's. The allelic tests yield marginally higher power than MultiPhen for multivariate phenotypes. For one/two binary traits under recessive mode of inheritance, allelic tests are found to be substantially more powerful. All three tests are applied to two different real data and the results offer some support for the simulation study. We propose a hybrid approach for testing multivariate association that implements MultiPhen when Hardy-Weinberg Equilibrium (HWE) is violated and BAMP otherwise, because the allelic approaches assume HWE.

Project description:Background and purposeThe ability of DTI to track the progression of microstructural damage in patients with inherited ataxias has not been explored so far. We performed a longitudinal DTI study in patients with spinocerebellar ataxia type 2.Materials and methodsTen patients with spinocerebellar ataxia type 2 and 16 healthy age-matched controls were examined twice with DTI (mean time between scans, 3.6 years [patients] and 3.3 years [controls]) on the same 1.5T MR scanner. Using tract-based spatial statistics, we analyzed changes in DTI-derived indices: mean diffusivity, axial diffusivity, radial diffusivity, fractional anisotropy, and mode of anisotropy.ResultsAt baseline, the patients with spinocerebellar ataxia type 2, as compared with controls, showed numerous WM tracts with significantly increased mean diffusivity, axial diffusivity, and radial diffusivity and decreased fractional anisotropy and mode of anisotropy in the brain stem, cerebellar peduncles, cerebellum, cerebral hemisphere WM, corpus callosum, and thalami. Longitudinal analysis revealed changes in axial diffusivity and mode of anisotropy in patients with spinocerebellar ataxia type 2 that were significantly different than those in the controls. In patients with spinocerebellar ataxia type 2, axial diffusivity was increased in WM tracts of the right cerebral hemisphere and the corpus callosum, and the mode of anisotropy was extensively decreased in hemispheric cerebral WM, corpus callosum, internal capsules, cerebral peduncles, pons and left cerebellar peduncles, and WM of the left paramedian vermis. There was no correlation between the progression of changes in DTI-derived indices and clinical deterioration.ConclusionsDTI can reveal the progression of microstructural damage of WM fibers in the brains of patients with spinocerebellar ataxia type 2, and mode of anisotropy seems particularly sensitive to such changes. These results support the potential of DTI-derived indices as biomarkers of disease progression.

Dataset Information

Personalized microstructural evaluation using a Mahalanobis-distance based outlier detection strategy on epilepsy patients' DTI data - Theory, simulations and example cases.

Publications

Personalized microstructural evaluation using a Mahalanobis-distance based outlier detection strategy on epilepsy patients' DTI data - Theory, simulations and example cases.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets