1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465 |
- """
- Read hippocampal subfield volumes computed by Freesurfer
- https://surfer.nmr.mgh.harvard.edu/fswiki/HippocampalSubfields
- https://github.com/freesurfer/freesurfer/tree/release_6_0_0/HippoSF
- >>> from freesurfer_volume_reader.freesurfer import HippocampalSubfieldsVolumeFile
- >>>
- >>> for volume_file in HippocampalSubfieldsVolumeFile.find('/my/freesurfer/subjects'):
- >>> print(volume_file.read_volumes_mm3())
- >>> print(volume_file.read_volumes_dataframe())
- """
- import re
- import typing
- import pandas
- import freesurfer_volume_reader
- class HippocampalSubfieldsVolumeFile(freesurfer_volume_reader.SubfieldVolumeFile):
- # https://surfer.nmr.mgh.harvard.edu/fswiki/HippocampalSubfields
- FILENAME_PATTERN = (
- r"^(?P<h>[lr])h\.hippoSfVolumes"
- r"(?P<T1>-T1)?(-(?P<analysis_id>.+?))?\.v10.txt$"
- )
- FILENAME_REGEX = re.compile(FILENAME_PATTERN)
- FILENAME_HEMISPHERE_PREFIX_MAP = {"l": "left", "r": "right"}
- def __init__(self, path: str):
- super().__init__(path=path)
- subject_dir_path = self._absolute_path.parent.parent
- self.subject = subject_dir_path.name
- filename_match = self.FILENAME_REGEX.match(self._absolute_path.name)
- assert filename_match, self._absolute_path
- filename_groups = filename_match.groupdict()
- assert (
- filename_groups["T1"] or filename_groups["analysis_id"]
- ), self._absolute_path
- self.hemisphere = self.FILENAME_HEMISPHERE_PREFIX_MAP[filename_groups["h"]]
- self.t1_input = filename_groups["T1"] is not None
- self.analysis_id = filename_groups["analysis_id"]
- def read_volumes_mm3(self) -> typing.Dict[str, float]:
- subfield_volumes = {}
- for line in (
- self._absolute_path.read_text(encoding="ascii").rstrip().split("\n")
- ):
- # https://github.com/freesurfer/freesurfer/blob/release_6_0_0/HippoSF/src/segmentSubjectT1T2_autoEstimateAlveusML.m#L8
- # https://github.com/freesurfer/freesurfer/blob/release_6_0_0/HippoSF/src/segmentSubjectT1T2_autoEstimateAlveusML.m#L1946
- subfield_name, subfield_volume_mm3_str = line.split(" ")
- subfield_volumes[subfield_name] = float(subfield_volume_mm3_str)
- return subfield_volumes
- def read_volumes_dataframe(self) -> pandas.DataFrame:
- volumes_frame = self._read_volume_series().reset_index()
- volumes_frame["subject"] = self.subject
- volumes_frame["hemisphere"] = self.hemisphere
- # volumes_frame['hemisphere'] = volumes_frame['hemisphere'].astype('category')
- volumes_frame["T1_input"] = self.t1_input
- volumes_frame["analysis_id"] = self.analysis_id
- return volumes_frame
|