WP3 Data Access & Disease Biomarkers from Large-scale Population Studies

Rationale:
Numerous natural history and epidemiological studies in Alzheimer’s disease and related dementias have been performed in order to identify the most promising biomarkers for early detection of the disease, monitoring disease progression, and measuring the effectiveness of potential treatments during clinical trials. These initiatives have collected substantial amounts of imaging, cognitive, fluid, lifestyle, and genetic data on the healthy ageing, those who have memory problems and likely to develop dementia, and patients affected by Alzheimer’s and other dementias. While many of these studies have scanned hundreds of patients numerous times, very little has been done to leverage these global efforts and pool all the generated data together. In this package, we will combine all of this information and extract relevant image-based biomarkers using existing, well-validated tools in order to provide the most comprehensive set of data on the disease process. This biomarker data will then serve as inputs for the phenomenological and biomechanical models that will be constructed in other work packages.
Objectives:
  • To provide a complete data management solution, based on the current VPH-Share infrastructure, that allows access to data, derived data, and resulting biomarkers from multiple cohorts to the researchers involved in the modelling activities that form the core of WP5 and WP6.
  • To establish data processing pipelines for extracting biomarkers from several large data cohorts.
  • To extract these imaging biomarkers from all imaging data available to the consortium
  • To identify the biomarkers and other characteristics (both from single imaging modalities and multi modalities) that provide the most accurate inputs for phenomenological and biomechanical models.
Activities:
  • Data repositories for derived data from retrospective studies
  • Analysis of retrospective data cohorts
  • Establishing connectivity with retrospective data sources
  • Storage of derived data and biomarkers
  • Data management infrastructure for prospective data
  • Specification, optimisation of methods and construction of image analysis pipelines
  • Extraction of personalised data for phenomenological modelling
  • Extraction of personalised data for mechanistic modelling