This script runs HCP-like minimal preprocessing pipeline for diffusion weighted images.

• FSL
• nibabel
• python 2.7

The input requirement is based on NPRL group at Emory acquisition protocol. For each subject following files must be available:

• b0 image right to left acquisition
• b0 image left to right acquisition
• Diffusion image right to left phase-encoded acquisition
• Diffusion image left to right phase-encoded acquisition
• bvector (.bvec) for each diffusion image
• b-value (.bval) for each diffusion image
  

The naming scheme for the files must be set in the config.json file placed in the root of the stydy folder. Sample config.json file is included in this repository.

{
  "filenames": {
    "b0_rl": [ #should contain two strings identifying b0 right to left images
      "b0",  
      "RL"
    ],
    "b0_lr": [ #should contain two strings identifying b0 left to right images
      "b0",
      "LR"
    ],
    "dwi_rl": [  #should contain two strings identifying weighted right to left images
      "D",
      "RL"
    ],
    "dwi_lr": [ #should contain two strings identifying weighted left to right images
      "D",
      "LR"
    ]
  }
}

Put all subjects' data in the folder <study>/Inputs . <study> is an optional name for the study and the parent folder of Inputs. In the Inputs directory each subject have her own subfolder. Picture below is a sample directory structure for a study with 3 participants.

1- open your Terminal

2 - type (option -t is a readout time of the acquisition)

python preprocess.py -i path_to_Study -t <fsl_readout_time>

After the code finishes, you can find all preprocessed data in <study>/Outputs directory