kwneuro.noddi module¶

class kwneuro.noddi.Noddi(volume, directions)¶

A Noddi result

Parameters:

directions: VolumeResource¶: The NODDI image volume. It is a 4D volume, with the first three dimensions being spatial and the final dimension indexing the directions.

static estimate_from_dwi(dwi, mask=None, dpar=0.0017, n_kernel_dirs=500)¶

Estimate Noddi from a DWI.

Parameters:

dwi (Dwi) – The source DWI
mask (VolumeResource | None) – Optionally, a boolean 3D volume that has indicates where the fit should take place, such as a brain mask.
dpar (float) – The parallel diffusivity to be used in the model fitting. If not provided, the default value of 1.7e-3 mm^2/s is used, which is suitable for white matter. For gray matter, a value of 1.3e-3 mm^2/s is recommended.
n_kernel_dirs (int) – The number of directions to use when generating the AMICO NODDI kernels. This value represents the total
Default (count of possible orientations for the response functions across the half-sphere.) –

Return type:

Noddi

Returns: A Noddi resource containing the estimated parameters.

property fwf: VolumeResource¶: Free Water Fraction (FWF) map as a 3D volume.

get_modulated_ndi_odi()¶

Compute the modulated maps, NDI*TF and ODI*TF, where TF = 1 - FWF.

Return type:: tuple[VolumeResource, VolumeResource]
Returns:: Returns 3D volumes for modulated NDI and ODI maps, as VolumeResources.

load()¶

Load any on-disk resources into memory and return a DTI with all loadable resources loaded.

property ndi: VolumeResource¶: Neurite Density Index (NDI) map as a 3D volume.

property odi: VolumeResource¶: Orientation Dispersion Index (ODI) map as a 3D volume.

save(path)¶

Save all resources to disk and return a Noddi with all resources being on-disk.

Parameters:: path (Path | str) – The desired file save location, a nifti file path.
Return type:: Noddi

Returns: A Noddi with its internal resources being on-disk.

volume: VolumeResource¶: The NODDI image volume. It is a 4D volume, with the first three dimensions being spatial and the final dimension indexing the noddi outputs.