DIAN Data Resource Requests

Title:Assessing the association between perivascular dysfunction and beta-amyloid burden in subjects with or at risk for dementia

Date of Request:03/21/2018

Status:Approved

ID:DIAN-D1807

Aim 1:To assess group differences in perivascular dysfunction, under the hypothesis that the number of MRI-evident ePVS is: symptomatic carriers > asymptomatic carriers > mutation non-carriers.

Aim 2:To estimate the association between MRI-evident ePVS and PET-derived measures of beta-amyloid load in cross-sectional and longitudinal analyses, under the hypothesis that perivascular dysfunction contributes to the accrual of beta-amyloid.

Title:DIAN Elecsys CSF and Amyloid Imaging Concordance

Date of Request:02/21/2018

Status:Approved

ID:DIAN-D1806

Aim 1:Determine the concordance between CSF measures obtained via the Elecsys platform

Aim 2:Assess Elecsys CSF measure relative to other clinical and cognitive variables

Aim 3:Assess how well Elecsys measures predict clinical and cognitive progression

Title:DIAN data analysis in a tranSMART platform

Date of Request:02/16/2018

Status:Pending

ID:DIAN-D1805

Aim 1:Curate the DIAN data and prepare for tranSMART loading

Aim 2:Perform investigation of data in tranSMART platform

Title:Comparing Multifactorial Pathologic Trajectories in Asymptomatic Autosomal Dominant Alzheimer’s Disease Mutation Carriers and Late-Onset Alzheimer’s Disease patients

Date of Request:02/04/2018

Status:Approved

ID:DIAN-D1804

Aim 1:Identify disease triggering events in each DIAN subject, using all the available multimodal neuroimaging data

Aim 2:Subdivide and stage the DIAN subjects based on the identified individual causal pathological mechanisms (from Aim 1)

Aim 3:Compare both the obtained potential triggering events and the subjects’ subdivision/staging (from Aims 1 and 2 respectively) with the equivalent results previously obtained for LOAD patients from ADNI database.

Title:Quantification of Longitudinal Volume Change in Familial FTD (fFTD)

Date of Request:01/28/2018

Status:Approved

ID:DIAN-D1803

Aim 1:To quantify rates of change in structural (T1w) imaging in symptomatic fFTD

Aim 2:To quantify rates of change in structural (T1w) imaging in presymptomatic fFTD

Aim 3:To derive subject-specific maps of volumetric change in fFTD

Title:Linking brain connectivity changes to development of symptoms in autosomal dominant AD

Date of Request:01/27/2018

Status:Approved

ID:DIAN-D1802

Aim 1:To study the association of grey matter connectivity in autosomal dominant AD with estimated year of onset and cognitive decline.

Aim 2:To study associations between grey matter connectivity alterations and AD and injury markers in CSF (aβ42, aβ40, p-tau, t-tau, neurogranin, VLIP-1, SNAP-25, NRGN, YKL-40) and PET (aβ, FDG, and tau) cross-sectionally and longitudinally.

Aim 3:Compare cross-sectional findings in grey matter connectivity with longitudinal changes.

Title:Updated analyses of Dominantly Inherited Alzheimer Disease (AD) neuroimaging data using new biomarker, clinical, genetic, and psychometric data

Date of Request:01/26/2018

Status:Approved

ID:DIAN-D1801

Aim 1:Cross-sectional evaluation of the temporal ordering of imaging biomarkers in asymptomatic ADAD with current biomarker, clinical, genetic, and psychometric data.

Aim 2:Harmonization of refined ADAD imaging processed values with the ADNI sporadic AD studies (ADNI)

Title:Clinical and Molecular Characterization of PSEN1 and PSEN2 Variants of Unknown Pathogenicity

Date of Request:12/21/2017

Status:Approved

ID:DIAN-D1729

Aim 1:To describe imaging and fluid biomarkers for PSEN1 and PSEN2 variants of unknown pathogenicity

Title:Developing a machine-learning based biomarker model to predict Alzheimer’s disease progression

Date of Request:11/30/2017

Status:Approved

ID:DIAN-D1728

Aim 1:1. To train a machine learning algorithm that sensitively predicts ADAD stage (i.e. EYO) based on cross-sectional biomarker and imaging data in the DIAN cohort. The trained algorithm will be cross-validated in sporadic AD (ADNI) as a predictor of a. baseline cognition b. longitudinal disease progression

Aim 2:1. To train a machine learning algorithm that discriminates between mutation carrier (MC) and non-mutation carriers (NC) based on cross-sectional biomarker and imaging data in the DIAN cohort. Cross-validation of the trained model for diagnostic classification will be applied to cases with sporadic AD and controls from the ADNI data set.

Title:Time course of imaging markers of neurodegeneration in autosomal dominantly inherited Alzheimer’s disease assessed by Linear mixed effects models

Date of Request:11/29/2017

Status:Approved

ID:DIAN-D1727

Aim 1:The specific aim for the proposed work with the DIAN dataset is to make use of the machine learning paradigm for classification in combination with feature extraction methods such as partial least squares (PLS) algorithm [4] on a combined set of different imaging modalities. PLS is a very well-known algorithm in the neuroimaging field.

Aim 2:Additionally we aim for the examination of some other covariates which are known to affect neurodegeneration such as gender or genetic status by a LME approach, which represents a model of a response variable with fixed and random effects. These models comprise fitted coefficients, covariance parameters, design matrices, residuals and other diagnostic information.

Aim 3:Furthermore we are planning to follow an enhanced approach by querying the evolution of the selected features and by proposing predictive models based on supervised learning applied to the reference controls.