Note

This page is a reference documentation. It only explains the class signature, and not how to use it. Please refer to the user guide for the big picture.

brainprep.datasets.openms.OpenMSDataset

class brainprep.datasets.openms.OpenMSDataset(datadir)

Bases: object

Open Multiple Sclerosis (OpenMS) anatomical dataset.

This dataset [1] contains Magnetic Resonance (MR) images of Multiple Sclerosis (MS) patients with corresponding ground truth segmentations of white matter lesion changes.

Parameters:

datadirDirectory

Directory where data will be stored.

Attributes:

_urlstr

Internal URL used to fetch data.

References

[1]

Žiga Lesjak, Franjo Pernuš, Boštjan Likar, and Žiga Špiclin. Validation of white-matter lesion change detection methods on a novel publicly available MRI image database. Neuroinformatics, 14(4):403–420, October 2016.

Examples

>>> from brainprep.config import Config
>>> from pathlib import Path
>>> from brainprep.datasets import OpenMSDataset
>>>
>>> datadir = Path("/tmp/brainprep-data")
>>> datadir.mkdir(parents=True, exist_ok=True)
>>> dataset = OpenMSDataset(datadir)
>>> with Config(verbose=False):
...     data = dataset.fetch(
...         subject="01",
...         modality="T1w",
...         dtype="cross_sectional",
...     )
>>> data
Bunch(
  description: PosixPath('...')
  anat: PosixPath('...')
)

__init__(datadir)

fetch(subject, modality, dtype='cross_sectional')

Fetch data.

Parameters:

subjectstr

the subject identifier. This identifier must lie in [‘01’ - ‘30’], [‘01’ - ‘20’], for cross sectional or longitudinal data respectively.

modalitystr

the modality to be fetched: ‘T1w’, ‘T2w’ or ‘FLAIR’.

dtypestr

the type of data to download: ‘cross_sectional’ or ‘longitudinal’. Default ‘cross_sectional’.

Returns:

datasetBunch

the fetched data path. Keys are either ‘sub-{subject}’ or ‘sub-{subject}_ses-{timepoint}’ for cross sectional and longitudinal data, respectively. A ‘description’ entry is also available.

sanity_check(subject, modality, dtype='cross_sectional')

Check that the fetch parameters are correct.

Parameters:

subjectstr

the subject identifier. This identifier must lie in [‘01’ - ‘30’], [‘01’ - ‘20’], for cross sectional or longitudinal data respectively.

modalitystr

the modality to be fetched: ‘T1w’, ‘T2w’ or ‘FLAIR’.

dtypestr

the type of data to download: ‘cross_sectional’ or ‘longitudinal’. Default ‘cross_sectional’

Raises:

ValueError

If the fetch input parameters are not correct.