template_checker

check a header against best template

Module Attributes

ErrorCompare	Dictionary for mismatches in input (have key) and template (expect key)
CheckResult

Functions

arraystr_to_float(val)	Parse array values from different dicom types
clean_rt(errors)	Clean up errors that are not actually errors in "realtime" dicom headers.
find_errors(template, current_hdr[, allow_null])	Given a template and hdr, find any mismatches (non-conforming errors).
float_or_0(val)	float or zero
fuzzy_arr_check(have, expect)	Compare only 3 decimals of a float or array of floats.

Classes

CheckResult
ErrorCompare	Dictionary for mismatches in input (have key) and template (expect key)
TemplateChecker([db, context])	cache db connection and list of tags read a dicom file and report if it conforms to the expected template

class template_checker.CheckResult

• conforms: false when a acq2sqlite.DBQuery.CONSTS

  template-parameter between input and template mismatch
• errors: nested dict of {mismatched_param: {'have':...,'expect':...}} (parameter keyed dictionary with ErrorCompare values)
• input: dict of all parameters of an input dicom header

  (dcmmeta2tsv.TagValues)
• template: all the parameters of a template (matching Study, SeriesName)

  Also a dcmmeta2tsv.TagValues

Here’s an example of CheckResult datastructure in html/javascript on the static debug-enabled page

class template_checker.ErrorCompare

Dictionary for mismatches in input (have key) and template (expect key)

class template_checker.TemplateChecker(db=None, context='DB')

cache db connection and list of tags read a dicom file and report if it conforms to the expected template

check_file(dcm_path)

File disbatch for TemplateChecker.check_header()

Parameters:

dcm_path – path to dicom file with header/parameters to read.

Returns:

output of check_header

Return type:

CheckResult

check_header(hdr)

Check acquisition parameters against it’s template.

Parameters:

hdr – DB row or file dictionary desc. acq. to check against template

Returns:

Conforming status, errors, and comparison information

Return type:

CheckResult

template_checker.arraystr_to_float(val)

Parse array values from different dicom types

Parameters:

val (str) – array of floats stored as string

Returns:

converted to python list

Return type:

list[float]

>>> arraystr_to_float("[1.0, 2.0]") # saved dicomes
[1.0, 2.0]
>>> arraystr_to_float("1.0,2.0") # realtime dicoms
[1.0, 2.0]

template_checker.clean_rt(errors)

Clean up errors that are not actually errors in “realtime” dicom headers.

Currently (2025-02-26) only checks PixelResol using fuzzy_arr_check() :param errors: have/expect dict of errors :returns: errors with close ones removed

Parameters:

errors (dict[str, ErrorCompare])

Return type:

dict[str, ErrorCompare]

template_checker.find_errors(template, current_hdr, allow_null=[])

Given a template and hdr, find any mismatches (non-conforming errors).

Parameters:

• template (dict[str, str]) – expected values

• current_hdr (dict[str, str]) – values we currently have

• allow_null (list[str]) – current keys that can be null. see py:class:TemplateChecker.context

Returns:

dictionary of tag key names and the have/expect values

Return type:

dict[str, ErrorCompare]

>>> find_errors({"TR": "1300"}, {"TR": "1300"})
{}

>>> find_errors({"TR": "1300"}, {"TR": "2000"})
{'TR': {'expect': '1300', 'have': '2000'}}

template_checker.float_or_0(val)

float or zero

Parameters:

val (str)

Return type:

float

template_checker.fuzzy_arr_check(have, expect)

Compare only 3 decimals of a float or array of floats. Used by clean_rt() when comparing realtime dicom headers to DB

Parameters:

• have – string of floats on one side

• expect – string of floats on another

Returns:

true if they are roughly the same

Return type:

bool

>>> fuzzy_arr_check('2.00001', '2')
True
>>> fuzzy_arr_check('[2.00001, 0.0]', '2.0,0.0000001')
True
>>> fuzzy_arr_check('[2,0]', '2,1')
False