dcmmeta2tsv

Give a tab separated metadata value line per dicom file.

Module Attributes

TagValues

keys are names from taglist.txt, also has dcm_path key for file

Functions

csa_fetch(csa_tr, itemname)	safely look into csa_tr dicom dictionary.
read_csa(csa)	extract parameters from siemens CSA :param csa: content of siemens private tag (0x0029, 0x1010) :return: nibabel's csareader dictionary or None if cannot read
read_known_tags([tagfile])	read in tsv file like with header name,tag,desc.
read_shims(csa_s)
read_tags(dcm_path, tags)	Read dicom header and isolate tags
tagpair_to_hex(csv_str)	move our text files has tags like "0051,1017" to pydicom indexe like (0x51,0x1017)

Classes

DicomTagReader()	Class to cache read_known_tags() output
NULLVAL()	Container to imitate pydicom.dcmread.

class dcmmeta2tsv.DicomTagReader

Class to cache read_known_tags() output

read_dicom_tags(dcm_path)

return values of dicom header priority fields ordered as defined in taglist.txt :param dcm_path: path to dciom :return: dict[taglist.txt tagname, tag value]

>>> dtr = DicomTagReader()
>>> hdr = dtr.read_dicom_tags('example_dicoms/RewardedAnti_good.dcm')
>>> list(hdr.values()) 
[1, 'p2', '154833.265000', '20220913', ...

>>> list(hdr.values())[-1]
'example_dicoms/RewardedAnti_good.dcm'

Parameters:

dcm_path (PathLike)

Return type:

dict[str, str]

class dcmmeta2tsv.NULLVAL

Container to imitate pydicom.dcmread. object that has obj.value for when a dicom tag does not exist. Using “null” to match AFNI’s dicom_hinfo missing text

dcmmeta2tsv.TagValues

keys are names from taglist.txt, also has dcm_path key for file

dcmmeta2tsv.csa_fetch(csa_tr, itemname)

safely look into csa_tr dicom dictionary. Expect nested structure like 'tags'->itemname->'items'->[0].

In future, might want to check itemname and pull out more than the first array item.

>>> csa_fetch({'notags':'badinput'}, 'PhaseEncodingDirectionPositive')
'null'
>>> csa_fetch({'tags':{'ImaPATModeText': {'items': [1]}}}, 'ImaPATModeText')
1

Parameters:

• csa_tr (dict)

• itemname (str)

Return type:

str

dcmmeta2tsv.read_csa(csa)

extract parameters from siemens CSA :param csa: content of siemens private tag (0x0029, 0x1010) :return: nibabel’s csareader dictionary or None if cannot read

>>> read_csa(None) is None
True

Return type:

dict | None

dcmmeta2tsv.read_known_tags(tagfile='taglist.txt')

read in tsv file like with header name,tag,desc. skip comments and header

Parameters:

tagfile – text tsv file to get name,tag(hex pair),desc from

Returns:

file parsed into a list of dictonaires

Return type:

list[Tag]

dcmmeta2tsv.read_shims(csa_s)

Param:

csa_s 0x0029,0x1020 CSA Series Header Info:: csa_s = dcmmeta2tsv.read_csa(dcm.get(())

Returns:

list of shim values in order of CHM matlab code

Parameters:

csa_s (dict | None)

Return type:

list

CHM maltab code concats

sAdjData.uiAdjShimMode sGRADSPEC.asGPAData[0].lOffset{X,Y,Z} sGRADSPEC.alShimCurrent[0:4] sTXSPEC.asNucleusInfo[0].lFrequency

>>> csa_s = pydicom.dcmread('example_dicoms/RewardedAnti_good.dcm').get((0x0029, 0x1020))
>>> ",".join(read_shims(read_csa(csa_s)))
'1174,-2475,4575,531,-20,59,54,-8,123160323,4'
>>> read_shims(None)  
['null', ...'null']

dcmmeta2tsv.read_tags(dcm_path, tags)

Read dicom header and isolate tags

Parameters:

• dcm_path (PathLike) – dicom file with headers to extract

• tags (list[Tag]) – ordered dictionary with ‘tag’ key as hex pair, see tagpair_to_hex()

Returns:

dict[tag,value] values in same order as tags

Return type:

dict[str, str]

>>> tr = {'name': 'TR', 'tag': (0x0018,0x0080), 'loc': 'header'}
>>> ipat = {'name': 'iPAT', 'tag': 'ImaPATModeText', 'loc': 'csa'}
>>> list(read_tags('example_dicoms/RewardedAnti_good.dcm', [ipat, tr]).values())
['p2', '1300.0', 'example_dicoms/RewardedAnti_good.dcm']

>>> list(read_tags('example_dicoms/DNE.dcm', [ipat,tr]).values())
['null', 'null', 'example_dicoms/DNE.dcm']

dcmmeta2tsv.tagpair_to_hex(csv_str)

move our text files has tags like “0051,1017” to pydicom indexe like (0x51,0x1017)

Parameters:

csv_str – comma separated string to convert

Returns:

dicom header tag hex pair

Return type:

tuple[int, int]

>>> tagpair_to_hex("0051,1017")
('0x51', '0x1017')