Skip to content

harmonization_clinicalcombat

ComBAT harmonization of clinical MRI data. It’s a ComBAT implementations for adapting sites to a reference site along with ready-to-run scripts to prepare datasets, fit a model, apply the harmonization and analyze the outputs. Two harmonization methods are available pairwise (linear) and clinic (non-linear).

Keywords : Harmonization, Pairwise, Clinical, ComBAT


Format : tuple path(ref_site), path(move_site)

TypeDescriptionMandatoryPattern
ref_sitefileCSV for Reference site data for one metric, all bundles and subjects. Must include sid,site,bundle,metric,mean,age,sex,handedness,disease. The Disease column must include at least the label HC.True*.{csv,csv.gz}
move_sitefileCSV for Moving site data for one metric, all bundles and subjects. Must include sid,site,bundle,metric,mean,age,sex,handedness,disease. The Disease column must include at least the label HC.True*.{csv,csv.gz}

Format : path(*.model.csv)

TypeDescriptionMandatoryPattern
*.model.csvfileHarmonization fitted model.True*.model.csv

Format : path(*.harmonized.csv.gz)

TypeDescriptionMandatoryPattern
*.harmonized.csv.gzfileHarmonized moving site data.True*.harmonized.csv.gz

Format : path(qc_reports)

TypeDescriptionMandatoryPattern
qc_reportsdirectoryBDQC reports for harmonized data.Trueqc_reports/

Format : path(figures)

TypeDescriptionMandatoryPattern
figuresdirectoryFigures generated to visualize harmonization results.Truefigures/

Format : path(versions.yml)

TypeDescriptionMandatoryPattern
versions.ymlfileFile containing software versionsTrueversions.yml

TypeDescriptionDefaultChoices
methodstringHarmonization strategy to use clinic (non-linear) or pairwise (linear)clinic
bundleslistList of bundles subset used to plot. By default all bundles in the csv input file were used to compute model harmonization.all
limit_age_rangebooleanDrop reference subjects outside the moving age range.disabled
ignore_sexbooleanRemove sex variable when estimating the model. If all subjects have the same value, it will be automatically applied.disabled
ignore_handednessbooleanRemove handedness variable when estimating the model. If all subjects have the same value, it will be automatically applied.disabled
no_empiral_bayesbooleanIgnore the empirical Bayesian estimatedisabled
regul_refintegerRidge penalty applied to reference regression. Parameter use for clinic method only.0
regul_movstringMoving site penalty or auto-tuning. Parameter use for clinic method only.-1
degreeintegerPolynomial degree used for age. It depend on method used (1 for parwise, 2 for clinic)None- 1
- 2
nuintegerVariance hyperparameter for the moving site. Parameter use for clinic method only.5
tauintegerCovariate hyperparameter for the moving site. Parameter use for clinic method only.2
degree_qcintegerQC model degree override (0 reuses the harmonization degree).0

DescriptionDOI
clinical-ComBATMethod for harmonizing MRI data across different sites.10.48550/arXiv.2511.04871


Last updated : 2025-12-22