Atrial function

Feature summary

Badge legend: Phenotype-family colors group feature types only. Status colors are reserved for evidence tables: muted green means verified, muted amber means method-dependent or schema debt, gray outline means needs verification, and red is reserved for blocked items. Disease badges use neutral navigation styling and are literature-context links only, not diagnoses, classifiers, or validated phenotype-to-ICD associations.

Atrial function phenotypes describe reservoir, conduit, and booster-pump behavior from the same long-axis atrial volume curves used by Atrial structure. This page documents CardiacNexus current output behavior; it does not define diagnostic thresholds for atrial fibrillation, heart failure, or diastolic dysfunction.

Modality
Cine long-axis CMR
UKB source
Data Field 20208 plus ECG timing when available
Pipeline step
Long-axis atrial volume curve extraction and derivative analysis
Outputs
atrial_volume.csv, timeseries/atrium.npz, atrial volume/rate QC plots
Maturity
Source-audited phenotype page

Clinical question

Clinicians use atrial function to interpret whether an enlarged or normal-sized atrium has impaired reservoir, conduit, or booster-pump performance. LA emptying fraction and phasic function are discussed in CMR and echocardiography literature as markers of atrial remodeling, diastolic dysfunction, and prognosis, but CardiacNexus outputs are quantitative research phenotypes rather than classifiers [1] [2].

Anatomical and physiological definition

Reservoir function reflects atrial filling while the ventricle contracts and the mitral/tricuspid annulus descends. Conduit function reflects passive emptying into the ventricle during early diastole. Booster-pump function reflects active atrial contraction late in diastole when sinus rhythm and timing information allow that phase to be identified [1].

Source acquisition and UKB field

The current extraction uses UK Biobank cine long-axis CMR, UK Biobank Data Field 20208, plus resting ECG-derived timing when the pre-atrial-contraction frame is needed. Required image inputs and segmentation boundary assumptions are inherited from the atrial structure pipeline.

What exactly CardiacNexus measures

The functional rows use the same biplane LA and 4ch RA volume curves documented on the atrial-structure page. Indexed values require query_BSA(subject). Conduit and booster rows additionally require an ECG-derived T_pre_a that passes implementation timing checks.

Total emptying fraction

CardiacNexus computes LA and RA total emptying fraction from maximum and minimum framewise atrial volume:

Copyable formulaEF_total = (V_max - V_min) / V_max * 100

Conduit and booster fractions

When ECG-derived pre-atrial-contraction timing passes checks, CardiacNexus records LA: V_pre_a [mL] and separates LA passive/conduit and active/booster components:

Copyable formulaEF_conduit = (V_max - V_pre_a) / V_max * 100
Copyable formulaEF_booster = (V_pre_a - V_min) / V_pre_a * 100

Atrial expansion index

The expansion index uses the same extrema but normalizes by minimum volume:

Copyable formulaEI = (V_max - V_min) / V_min * 100

Peak emptying rates

For LA peak emptying rates, the implementation smooths the LA biplane volume curve, differentiates it, finds two negative derivative extrema, and reports their absolute values as PER-E and PER-A. The ratio LA: PER-E/PER-A is skipped when the implemented outlier guard detects an extreme value.

Left atrial volume-time curve showing Vmax Vmin and pre-atrial contraction volume
LA volume-time curve from the manuscript image cache, cited there to Peters et al. 2021, shown to orient Vmax, Vmin, and Vpre-A timing. CardiacNexus output labels and conditional skips remain defined by the implementation table below.Source: manuscript image cache citing Peters et al. 2021; permission and exact source-panel review pending for draft use.
LA volume time and dV/dt curve from the CardiacNexus manuscript image cache citing Aquaro et al. 2019
LA volume/time and dV/dt curve from the manuscript image cache, cited there to Aquaro et al. 2019, shown to orient early and late peak emptying rate concepts. CardiacNexus derivative smoothing, output names, and conditional skips are defined in the page text and tables.Source: manuscript image cache citing Aquaro et al. 2019; permission and exact source-panel review pending for draft use.
Left atrial reservoir conduit and booster pump phasic function context
LA reservoir, conduit, and booster timing context from the manuscript image cache, shown as broad phasic-function orientation. The displayed phases are context for interpreting current CardiacNexus emptying and conduit/booster rows, not validation examples or diagnostic thresholds.Source: manuscript image cache; permission and exact source-panel review pending for draft use.

Output columns and units

Display familyExact output columnUnitStatusSchema note
LA total emptyingLA: Total SV (bip) [mL]mLcurrentLA maximum minus minimum biplane volume
LA total emptyingLA: EF_total [%]%currentReservoir summary from LA biplane extrema
RA total emptyingRA: EF_total [%]%currentUses the current RA 4ch volume curve despite the legacy NPZ label
Expansion indexLA: EI [%]%currentMaximum/minimum volume based
Expansion indexRA: EI [%]%currentMaximum/minimum volume based
LA peak emptyingLA: PER-E [mL/s]mL/scurrent when derivative QC passesReported as a positive magnitude
LA peak emptyingLA: PER-A [mL/s]mL/scurrent when derivative QC passesReported as a positive magnitude
LA peak emptying indexedLA: PER-E/BSA [mL/s/m^2]mL/s/m²current when BSA and PER passBSA-dependent
LA peak emptying indexedLA: PER-A/BSA [mL/s/m^2]mL/s/m²current when BSA and PER passBSA-dependent
LA peak emptying ratioLA: PER-E/PER-Aunitlessconditional current outputOutlier guard may skip the ratio
LA pre-atrial contractionLA: V_pre_a [mL]mLconditional current outputRequires ECG-derived timing
LA pre-atrial contraction indexedLA: V_pre_a/BSA [mL/m^2]mL/m²conditional current outputRequires ECG-derived timing and BSA
LA active functionLA: EF_booster [%]%conditional current outputAlso called active EF in implementation comments
LA passive functionLA: EF_conduit [%]%conditional current outputAlso called passive EF in implementation comments

Output reconciliation

Evidence layerResult
Implementation source14 atrial-function outputs documented here: total emptying, expansion index, LA peak emptying rates, ECG-timed pre-atrial volume, conduit fraction, and booster fraction
Output inventorydocs/data/output_column_inventory.yml includes the same 14 outputs under atrial_volume.functional_columns
Phenotype dictionarydocs/data/phenotype_dictionary.yml links the same 14 outputs to this page
Page output tableall 14 output labels are listed above

Required upstream inputs

  • la_2ch.nii.gz, la_4ch.nii.gz, seg_la_2ch.nii.gz, and seg_la_4ch.nii.gz.
  • The atrial structure volume curves and timeseries/atrium.npz.
  • BSA lookup for indexed values.
  • Resting ECG XML and successful timing checks for T_pre_a, conduit, and booster outputs.

Reference ranges with cohort and method context

FeatureSourceCohortReference valueStatusNote
LA emptying fraction and phasic functionHoit 2014 and CMR atrial function reviews [1] [2]adult CMR/clinical review contextstudy-specificVerified context sourceUseful for interpretation, not a CardiacNexus-specific normal limit
Biplane LA volume and phasic functionWandelt et al. CMR biplane/Simpson comparison [3]CMR method-comparison cohortmethod-dependentVerified context sourcePhase definitions and boundary conventions differ across studies
UKB CMR chamber contextPetersen et al. UK Biobank CMR reference ranges [4]UK Biobank CMR reference cohortstructural context onlyVerified context sourceFunctional atrial rows remain verification-queued

Source-located registry status: docs/data/reference_range_sources.yml records Hoit 2014, Wandelt 2017, and Petersen 2017 as verified context sources for atrial-function interpretation, but the functional atrial rows remain verification-queued rather than promoted numeric normal ranges. This page therefore treats LA/RA emptying fractions and PER/PFR as method-dependent physiology context until row-level source tables and CardiacNexus phase definitions are adjudicated.

Disease interpretation

Reduced atrial emptying fraction and altered phasic function are discussed as markers of atrial remodeling, AF substrate, HFpEF/diastolic dysfunction, and HCM-related loading abnormalities [1] [2]. These disease badges are navigation context only. They do not indicate that CardiacNexus diagnoses AF, HF, or HCM from atrial function alone.

QC caveats and maturity boundary

Atrial function inherits segmentation, landmark, and volume-curve errors from the atrial structure pipeline. PER values are sensitive to temporal resolution, smoothing, arrhythmia, and noisy masks. Conduit and booster outputs are conditional: missing or rejected ECG timing leads to missing conditional outputs rather than imputed values.

Implementation provenance

Current outputs are implemented in src/feature_extraction/Long_Axis_20208/eval_atrial_volume.py.

Feature familyFormula or computational routeExact output columnsSource code file and functionUpstream dependenciesConditional behaviorQC artifactsSchema debt
Total emptying and expansionextrema of LA biplane and RA 4ch volume curvesLA: Total SV (bip) [mL], LA: EF_total [%], RA: EF_total [%], LA: EI [%], RA: EI [%]eval_atrial_volume.py; base feature update and EI blocksla_2ch.nii.gz, la_4ch.nii.gz, seg_la_2ch.nii.gz, seg_la_4ch.nii.gzsubject skipped if long-axis inputs or segmentation failtimeseries/atrium.npz, atrial volume QC plotsRA volume uses 4ch route despite legacy (bip) NPZ label
LA peak emptying ratessmoothed LA biplane volume derivative; two negative extrema reported as positive magnitudesLA: PER-E [mL/s], LA: PER-A [mL/s], LA: PER-E/BSA [mL/s/m^2], LA: PER-A/BSA [mL/s/m^2], LA: PER-E/PER-Aeval_atrial_volume.py; PER derivative blockLA biplane volume curve, temporal resolution, BSA lookuprows skipped if derivative fitting, peak checks, BSA, or ratio guard failsatrial PER/time-series QC plotPER values are derivative-derived and smoothing-sensitive
ECG-timed pre-atrial contractionresting ECG timing converted to cine frame and checked against volume-curve landmarksLA: V_pre_a [mL], LA: V_pre_a/BSA [mL/m^2], LA: EF_booster [%], LA: EF_conduit [%]eval_atrial_volume.py; pre-atrial-contraction blockresting ECG XML timing, temporal resolution, LA biplane curve, BSA lookuprows skipped if ECG timing is missing, outside frame range, or inconsistent with curve landmarksLA: T_pre_a in timeseries/atrium.npz; ECG/time-series QC plotconditional outputs require sinus-timing assumptions and external ECG availability

Source audit

  • Formula claims were checked against the current implementation strings and update blocks in eval_atrial_volume.py.
  • Full atrial-function output coverage was reconciled across the implementation contract, page text, docs/data/output_column_inventory.yml, and docs/data/phenotype_dictionary.yml.
  • docs/data/reference_sources.yml exists and is used here as the current curated reference-source registry for scientific and implementation source metadata.
  • One LA peak emptying rate curve figure is displayed from a page-local public path and registered in docs/data/figure_provenance.yml; permission and exact source-panel review remain pending for draft use.

References

  1. Hoit BD. Left atrial size and function: role in prognosis. Journal of the American College of Cardiology. 2014;63(6):493-505. doi:10.1016/j.jacc.2013.10.055. PMID:24291276.
  2. Peters DC, Lamy J, Sinusas AJ, Baldassarre LA. Left atrial evaluation by cardiovascular magnetic resonance: sensitive and unique biomarkers. European Heart Journal - Cardiovascular Imaging. 2021;23(1):14-30. doi:10.1093/ehjci/jeab221. PMID:34718484; PMCID:PMC8685602.
  3. Wandelt LK, Kowallick JT, Schuster A, Wachter R, Stumpfig T, Unterberg-Buchwald C, Steinmetz M, Ritter CO, Lotz J, Staab W. Quantification of left atrial volume and phasic function using cardiovascular magnetic resonance imaging: comparison of biplane area-length method and Simpson's method. The International Journal of Cardiovascular Imaging. 2017;33(11):1761-1769. doi:10.1007/s10554-017-1160-9. PMID:28523471.
  4. Petersen SE, Aung N, Sanghvi MM, Zemrak F, Fung K, Paiva JM, Francis JM, Khanji MY, Lukaschuk E, Lee AM, Carapella V, Kim YJ, Leeson P, Piechnik SK, Neubauer S. Reference ranges for cardiac structure and function using cardiovascular magnetic resonance in Caucasians from the UK Biobank population cohort. Journal of Cardiovascular Magnetic Resonance. 2017;19(1):18. doi:10.1186/s12968-017-0327-9. PMID:28178995; PMCID:PMC5304550.
  5. Aquaro GD, Pizzino F, Terrizzi A, Carerj S, Khandheria BK, Di Bella G. Diastolic dysfunction evaluated by cardiac magnetic resonance: the value of the combined assessment of atrial and ventricular function. European Radiology. 2019;29(3):1555-1564.