IntensityClip brick¶
Clip the intensity range as prescribed by the percentiles¶
Remove outliers at both ends of the intensity distribution and fit into a given dtype. To more robustly determine the clipping thresholds, data are removed of spikes with a median filter. Once the thresholds are calculated, the denoised data are thrown away and the thresholds are applied on the original image.
Adapted from niworkflows.
Mandatory inputs parameters:
- in_file (a string representing an existing file)
Input image which intensity will be clipped (valid extensions: [.nii, .nii.gz]).
ex. '/home/username/data/raw_data/T1w.nii'
Optional inputs with default value parameters:
- dtype (one of int16, float32, uint8, optional, default value is int16)
Output datatype.
ex. int16
- invert (a boolean, optional, default value is False)
Finalize by inverting contrast.
ex. False
- nonnegative (a boolean, optional, default value is True)
Whether input intensities must be positive.
ex. True
- p_max (a float, optional, default value is 99.9)
Percentile for the upper bound.
ex. 99.9
- p_min (a float, optional, default value is 10.0)
Percentile for the lower bound.
ex. 10.0
Outputs parameters:
- out_file (a strings representing a file)
Path of the clipped image (xtensions: [.nii, .nii.gz]).
ex. '/home/username/data/derived_data/T1w_clipped.nii'
Usefull links: