Skip to main content

Systematic sanity checks on imaging datasets within an XNAT environment

Project description


pipeline status coverage report python pypi

Main ConceptsCommandsExamplesInstallContributing

bbrc-validator is a Python-based software package that performs automatic quality assessment of neuroimaging datasets and their processing derivatives, through collections of "checkpoints". bbrc-validator is built on two core concepts: Tests and Validators.

  • A Test checks a specific trait from a given resource (either an imaging session or a single scan). It asks a specific question whose answer can be either True or False (eg. "Does this MRI scan have a conversion to NIfTI available?"). As such, Tests may be seen as unit tests . A Test class is defined by two attributes (passing and failing) that refer to two "real-life" cases (one expected to pass the Test and another expected to fail it). In addition, these attributes are systematically used by the CI testing.

  • A Validator is a collection of Test objects that may be executed against any XNAT imaging resource (by referring to their experiment identifiers). Running a Validator on a given experiment takes its associated set of tests, runs them sequentially and collects their results in a JSON object. A human-readable report can be generated (as a PDF document) with the results of the whole procedure.

Main Concepts

  • Test:

     class MyTest():
         """ Test functionality description """
         passing = 'PASSING_CASE_ID'
         failing = 'FAILING_CASE_ID'
         def run():     # executes the Test logic and returns some Results
             return Results(has_passed=test_outcome, data=some_data)  
         def report():  # provides a human-readable version of Results data
  • Validator:

     class MyValidator():
         def __init__():
             self.tests = [MyTest, ...]        
         def run():     # runs all Tests sequentially             
         def dump():    # compiles all Test results in a single JSON object    
         def report():  # generates a human-readable PDF report based on the results
  • Result: Represents the outcome from the execution of a Test. It includes a boolean attribute has_passed (representing the outcome of Test execution) and some additional data object (optionally used for storing contextual information from the execution).


Executes the specified Validator against a given image resource (a.k.a XNAT experiment) and generates (a) a JSON object with the results of all the Tests and (b) a human-readable PDF report.

usage: [-h] --config CONFIG --experiment EXPERIMENT
                        [--validator VALIDATOR] --output OUTPUT [--verbose]

Run a validator against an experiment

optional arguments:
  -h, --help                             show this help message and exit
  --config CONFIG, -c CONFIG             XNAT configuration file
  --experiment EXPERIMENT, -e EXPERIMENT XNAT experiment unique identifier
  --validator VALIDATOR, -v VALIDATOR    Validator name (default:ArchivingValidator)
  --output OUTPUT, -o OUTPUT             PDF file to store the report
  --verbose, -V                          Display verbosal information (optional)

Given a specific type of Validator, collects all results available in an XNAT instance and compiles them in a CSV file.

usage: [-h] --config CONFIG --version VERSION
                            [--validator VALIDATOR] --output OUTPUT 
                            [--project PROJECT] [--verbose]

Compile validation scores

optional arguments:
  -h, --help                    show this help message and exit
  --config CONFIG               XNAT configuration file
  --version VERSION, -v VERSION Filter specific version
  --validator VALIDATOR         Validator name (default:ArchivingValidator)                      
  --output OUTPUT, -o OUTPUT    CSV output file
  --verbose, -V                 Display verbosal information (optional)

Enables the creation of tables such as the following example obtained from ArchivingValidator (table trimmed to fit the dimensions of the page).

Tests included:

  1. HasUncompressedPixelData
  2. HasCorrectSequences
  3. HasBvecBval
  4. IsClassicDICOM
  5. HasDuplicatedSequences
  6. HasNifti
  7. HasPhilipsPrivateTags
  8. IsStudyDescriptionCorrect
Sums 11 11 0 11 11 6 0 11
BBRCDEV_E00211 ![#c5f015]( ![#c5f015]( ![#f03c15]( ![#c5f015]( ![#c5f015]( ![#f03c15]( ![#f03c15]( ![#c5f015](
BBRCDEV_E00210 ![#c5f015]( ![#c5f015]( ![#f03c15]( ![#c5f015]( ![#c5f015]( ![#c5f015]( ![#f03c15]( ![#c5f015](
BBRCDEV_E00213 ![#c5f015]( ![#c5f015]( ![#f03c15]( ![#c5f015]( ![#c5f015]( ![#c5f015]( ![#f03c15]( ![#c5f015](
BBRCDEV_E00212 ![#c5f015]( ![#c5f015]( ![#f03c15]( ![#c5f015]( ![#c5f015]( ![#c5f015]( ![#f03c15]( ![#c5f015](
BBRCDEV_E00196 ![#c5f015]( ![#c5f015]( ![#f03c15]( ![#c5f015]( ![#c5f015]( ![#c5f015]( ![#f03c15]( ![#c5f015](
BBRCDEV_E00214 ![#c5f015]( ![#c5f015]( ![#f03c15]( ![#c5f015]( ![#c5f015]( ![#f03c15]( ![#f03c15]( ![#c5f015](
BBRCDEV_E00217 ![#c5f015]( ![#c5f015]( ![#f03c15]( ![#c5f015]( ![#c5f015]( ![#f03c15]( ![#f03c15]( ![#c5f015](
BBRCDEV_E00216 ![#c5f015]( ![#c5f015]( ![#f03c15]( ![#c5f015]( ![#c5f015]( ![#c5f015]( ![#f03c15]( ![#c5f015](
BBRCDEV_E00219 ![#c5f015]( ![#c5f015]( ![#f03c15]( ![#c5f015]( ![#c5f015]( ![#f03c15]( ![#f03c15]( ![#c5f015](
BBRCDEV_E00218 ![#c5f015]( ![#c5f015]( ![#f03c15]( ![#c5f015]( ![#c5f015]( ![#f03c15]( ![#f03c15]( ![#c5f015](
BBRCDEV_E00198 ![#c5f015]( ![#c5f015]( ![#f03c15]( ![#c5f015]( ![#c5f015]( ![#c5f015]( ![#f03c15]( ![#c5f015](


Create a Validator and review the list of its Tests

  1. Set a pyxnat connection to the XNAT instance hosting the data requiring validation.
  2. Create an instance of SPM12SegmentValidator, a Validator for segmentations produced using SPM12 Segment.
  3. Print out a list of included tests.
import pyxnat
intf = pyxnat.Interface(config='.xnat.cfg')

from bbrc.validation import SPM12SegmentValidator
spmv = SPM12SegmentValidator(lut={}, xnat_instance=intf)

print('{} tests (`{}`):'.format(spmv.__class__.__name__, spmv.version))
SPM12SegmentValidator tests (`d6ca22c1`):

[<bbrc.validation.processing.spm.HasCorrectNumberOfItems at 0x273dee24e88>,
 <bbrc.validation.processing.spm.HasCorrectItems at 0x273dee247c8>,
 <bbrc.validation.processing.spm.HasCorrectSPMVersion at 0x273dda8f4c8>,
 <bbrc.validation.processing.spm.HasCorrectMatlabVersion at 0x273dee28848>,
 <bbrc.validation.processing.spm.HasCorrectOSVersion at 0x273dee287c8>,
 <bbrc.validation.processing.spm.SPM12SegmentSnapshot at 0x273dee249c8>,
 <bbrc.validation.processing.spm.HasNormalSPM12Volumes at 0x273dee28788>,
 <bbrc.validation.processing.spm.SPM12SegmentExecutionTime at 0x273dee28bc8>]

Run SPM12SegmentValidator against an MRI session,'XNAT_E00001')
2021-02-04 12:12:54,635 - root - INFO - Running <bbrc.validation.processing.spm.HasCorrectNumberOfItems object at 0x00000273DEE24E88>
2021-02-04 12:12:54,964 - root - ERROR - XNAT_E00001 has 15 items (different from 16)
2021-02-04 12:12:55,572 - root - INFO - Running <bbrc.validation.processing.spm.HasCorrectItems object at 0x00000273DEE247C8>
2021-02-04 12:12:56,120 - root - INFO - Running <bbrc.validation.processing.spm.HasCorrectSPMVersion object at 0x00000273DDA8F4C8>
2021-02-04 12:12:56,592 - root - INFO - Running <bbrc.validation.processing.spm.HasCorrectMatlabVersion object at 0x00000273DEE28848>
2021-02-04 12:12:56,782 - root - INFO - Running <bbrc.validation.processing.spm.HasCorrectOSVersion object at 0x00000273DEE287C8>
2021-02-04 12:12:57,001 - root - INFO - Running <bbrc.validation.processing.spm.SPM12SegmentSnapshot object at 0x00000273DEE249C8>
2021-02-04 12:13:04,997 - root - INFO - * Creating snapshots...
2021-02-04 12:13:46,472 - root - INFO - Saved in /tmp/tmp3j664u27.png
2021-02-04 12:13:46,515 - root - INFO - Running <bbrc.validation.processing.spm.HasNormalSPM12Volumes object at 0x00000273DEE28788>
2021-02-04 12:13:50,552 - root - INFO - Running <bbrc.validation.processing.spm.SPM12SegmentExecutionTime object at 0x00000273DEE28BC8>

Collect results from SPM12SegmentValidator execution,

import json 
result = spmv.dump()
{'experiment_id': 'XNAT_E00001',
 'version': 'd6ca22c1',
 'generated': '2021-02-04, 12:13',
 'HasCorrectItems': {'has_passed': False,
  'data': ["Missing SPM12_SEGMENT items: ['pyscript_setorigin.m']."]},
 'HasCorrectSPMVersion': {'has_passed': True, 'data': []},
 'HasCorrectMatlabVersion': {'has_passed': True, 'data': []},
 'HasCorrectOSVersion': {'has_passed': True, 'data': []},
 'SPM12SegmentSnapshot': {'has_passed': True, 
  'data': ['/tmp/tmp3j664u27.png']},
 'HasNormalSPM12Volumes': {'has_passed': True,
  'data': ['Volumes: 773592.1702940931 524339.7480925963']},
 'SPM12SegmentExecutionTime': {'has_passed': True, 'data': ['0:07:15']}}

Generate a human-readable PDF report from the results,

import tempfile
_,fp = tempfile.mkstemp(suffix='.pdf')
print('Report created: {}'.format(fp))
Loading pages (1/6)
Counting pages (2/6)                                               
Resolving links (4/6)                                                       
Loading headers and footers (5/6)                                           
Printing pages (6/6)

Report created: '/home/jhuguet/notebooks/bbrc-validator/tmpcexwvwj5.pdf'


bbrc-validator can be installed via pip,

pip install bbrc-validator

bbrc-validator requires wkhtmltopdf for PDF generation. A static build release (with QT patches) is recommended, see available releases here by OS/distribution.

On Ubuntu 18.04:

dpkg -i wkhtmltox_0.12.6-1.bionic_amd64.deb
apt --fix-broken -y install

On CentOS 7:

yum -y  localinstall wkhtmltox-0.12.6-1.centos7.x86_64.rpm


bbrc-validator is still under active development. The currently included Tests and Validators have been tailored to the particular needs and context of the Barcelonaβeta Brain Research Center and as such might differ with the needs from other organizations.
However, the software was designed with an aim towards genericity, modularity and reusability. Since all Tests are based upon the same template (eg. each of them being linked to XNAT data resources as test cases), this makes them virtually shareable across groups and makes bbrc-validator open to public contributions.

Contact us for details on how to contribute or open an issue to start a discussion.


Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

bbrc-validator-0.5.17.tar.gz (135.3 kB view hashes)

Uploaded source

Built Distribution

bbrc_validator-0.5.17-py3-none-any.whl (169.7 kB view hashes)

Uploaded py3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page