A Method for Harmonizing Test-retest Multi-scanner Imaging Data

[Remi-Gau]

Added as an issue for book keeping

Source:
https://brainhack.live/

Leader:
Yinshan Wang

Harmonizing MRI data that acquired from multi-scanners may help reduce non-biological bias. ComBat (named for \combining batches") is an empirical Bayesian method for data harmonization that was originallydesigned for genomics (Johnson et al., 2007). It has recently been adapted to neuroimaging studies and applied to diverse data types including diffusion tensor imaging (DTI) (Fortin et al., 2017), cortical thicknesses (Fortinet al., 2018), functional connectivity measurements (Yu et al., 2018), and radiomic features derived from positron emission tomography (PET) imaging (Orlhac et al., 2018).

Details

• Programming languages:
  Matlab, R

• A list of requirements for taking part in the project (education level / English level /all the informat on you consider important for participants)

1. Chinese: Good 2. Programming languages: R 3. Knowledge related Freesurfer thickness analysis will be good

What participants gain/learn from this project

1. Hands on MRI morphimical analysis 2. Using R to handle repeated measurement data.

• Preferred maximal number of participants:
  4

• How many non-programming participants can join you during your project, what profession can they be?
  2