omni.pipelines.func.align#

Functions

create_reference_and_moco(output_path, ...)

Create reference image and motion correction.

deoblique_func(output_path, func)

Deoblique functional image.

omni.pipelines.func.align.create_reference_and_moco(output_path: str, func_do: str, TR: float, slice_times: List[float], sed: int, loops: List[int] = [1, 1, 1], subsample: List[int] = [5, 3, 1], borders: List[int] = [1, 1, 1], use_allineate: bool = False)[source]#

Create reference image and motion correction.

This function computes a quick intensity mask to get rid of voxels that are way outside the skull. It then computes an initial DVARs estimate to exclude volumes. It finds the worst 5% of volumes by DVARs and excludes volumes adjacent to those worst volumes (a centered window of size 3). The DVARs time course is then smooothed to find long stretches (~100 frames) of minimal motion, excluding the first/last 10% of the scan.

We use these extracted frames to create a reference functional image for further processing.

Parameters:
output_pathstr

Output path to write out files to.

func_dostr

Deobliqued functional image.

TRfloat

Repetition time of scan.

slice_timesList[float]

List of slice times.

sedint

Axis of slice encoding direction.

loopsList[int]

Number of loops for SpaceTimeRealign.

subsampleList[int]

Subsampling for SpaceTimeRealign.

bordersList[int]

Borders for SpaceTimeRealign.

use_allineatebool

Sets whether to use 3dAllineate for framewise alignment instead.

Returns:
str

Path to motion corrected functional.

str

Path to reference functional.

omni.pipelines.func.align.deoblique_func(output_path: str, func: str)[source]#

Deoblique functional image.

Parameters:
output_pathstr

Output path to write out files to.

funcstr

Functional image.

Returns:
str

Functional image deobliqued.