Prague Brains Group

Statistical Analyses

We perform various levels of statistical analyses in the extent of individual needs, using many statistical software packages to suit specific properties of your research:

  • Descriptive statistics
  • Parametric and non-parametric inferential methods
  • Predictive statistical correlation models
  • Predictive statistical regression models

Types of software we work with, including graphical softwares:

  • R (R Foundation)
  • Python (Python Software Foundation)
  • STATISTICA (StatSoft)
  • MATLAB (MathWorks)
  • OriginPro (OriginLab Corporation)
  • Excel (Microsoft Corporation)
  • MetaXL (EpiGear)

AI Methods

 

Methodology of Systematic Reviews and Meta-Analyses

We provide guidance through the process of systematic review/meta-analysis construction based on many years of experience. Our work is based on extended cooperation for creating high-quality systematic reviews via efficient and evidence-based solutions.

Our work for you includes assistance during the construction of PICO statement, establishing inclusion and exclusion criteria, literature search, literature selection, data extraction, and we perform final both qualitative and quantitative data synthesis with all kinds of graphical illustrations. Final assistance during the manuscript preparation is included as well.

Guidelines, risk of bias assessments and software we work with:

Guidelines:

  • Working with the latest versions of Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA), as a globally accepted evidence-based minimum set of items for reporting in systematic reviews and meta-analyses.
  • Working with the latest versions of Meta-analysis of Observational Studies in Epidemiology (MOOSE) statements (less frequently used).
  • Performing advisable preregistration of your systematic review/meta-analysis at the International Prospective Register of Systematic Reviews (PROSPERO; https://www.crd.york.ac.uk/Prospero).

Risk of Bias Assessment:

  • Newcastle-Ottawa Scale for nonrandomized studies
  • Jadad Scale for randomized clinical trials
  • Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2)
  • Cochrane Risk of Bias Assessments (Rob 2, ROBINS-I, ROB-ME)

Assessments of heterogeneity and statistical analysis:

  • Modelling of forest-plots via MetaXL (EpiGear) Software or within stand-alone packages, such as Cochrane RevMan:

Stand-alone packages:

We routinely used software packages used for statistical analyses, more in detail introduced in an individual page.

We create adequate modifications of abovementioned scales in order to provide bias assessment of the best fit for individual needs.

MRI data analyses