CardioNexus: Mapping Every Heartbeat with CMR and ECG.

Cross-chamber phenotypes combine ventricular and atrial measurements to describe chamber coupling, longitudinal pump function, and diastolic filling relationships. This page is source-audited for the current non-valvular ventricular_atrial_feature.csv rows. The same aggregate also contains mitral and tricuspid annular diameter rows, but those are documented on Valvular and flow phenotypes.

Modality: Cine short-axis CMR; cine long-axis CMR; ECG timing when available
UKB source: Data Fields 20208 and 20209, with ECG timing support
Pipeline step: Combined ventricular-atrial feature extraction
Outputs: ventricular_atrial_feature.csv, visualization/combined/AV_plane.png
Maturity: Source-audited phenotype page

Clinical question

Cross-chamber rows ask how atrial emptying, ventricular stroke volume, and atrioventricular plane motion relate within the same subject. CardiacNexus reports quantitative coupling features; it does not diagnose pulmonary hypertension, myocardial infarction, diastolic heart failure, dilated cardiomyopathy, or heart failure from a single row.

Anatomical and physiological definition

Atrioventricular plane displacement (AVPD) measures systolic descent of the atrioventricular plane. Published CMR protocols often use multi-view LV and RV annular landmarks, whereas the current CardiacNexus row is a simplified 4-chamber implementation [1] [3].

Isovolumetric pulmonary vein transit (IPVT) is the LV stroke-volume component not matched by LA total stroke volume in the same cycle. The IPVT ratio divides this difference by LA total stroke volume, so it is sensitive to both ventricular and atrial measurement quality [6].

Atrial contribution to LV filling estimates the LV filling increment between the ECG-derived LA pre-atrial-contraction frame and LV end diastole, normalized either to LV stroke volume or LV end-diastolic volume [4].

Source acquisition and UKB fields

The current route combines UK Biobank short-axis cine field 20209, long-axis cine field 20208, and ECG-derived atrial timing when available. It requires per-subject images and segmentations, aggregated ventricular and atrial feature CSVs, plus timeseries/ventricle.npz and timeseries/atrium.npz.

What exactly CardiacNexus measures

Atrioventricular plane displacement

eval_ventricular_atrial_feature.py calls evaluate_AVPD on the 4-chamber segmentation at the LV ED and ES frames from the ventricular time series. It writes a single AV: AVPD [cm] row and stores visualization/combined/AV_plane.png when visualization succeeds.

AVPD measurement panel from the CardiacNexus manuscript image cache citing Lindholm et al. 2022 — AVPD source panel from the manuscript image cache, cited there to Lindholm et al. 2022, shown to orient multi-view AV plane displacement measurement. CardiacNexus currently emits a simplified 4-chamber AVPD row, so this figure is method context rather than a one-to-one implementation diagram.Source: manuscript image cache citing Lindholm et al. 2022; permission and exact source-panel review pending for draft use.

Longitudinal contribution to stroke volume context from the manuscript image cache — Longitudinal contribution context figure from the manuscript image cache, cited there to Lindholm et al. 2022, shown to orient AV-plane contribution concepts. It is adjacent physiology context and does not add extra current CardiacNexus output columns.Source: manuscript image cache citing Lindholm et al. 2022; permission and exact source-panel review pending for draft use.

Septal and lateral contribution context from the manuscript image cache — Septal/lateral contribution context figure from the manuscript image cache, cited there to Ostenfeld et al. 2016, shown to orient regional contribution language. It is not a diagnostic pulmonary-hypertension classifier and does not replace the current AVPD/IPVT output mapping.Source: manuscript image cache citing Ostenfeld et al. 2016; permission and exact source-panel review pending for draft use.

Isovolumetric pulmonary vein transit

The same script computes IPVT as LV: SV [mL] - LA: Total SV (bip) [mL]. If IPVT is negative, the IPVT rows are skipped for that subject. The ratio is IPVT / LA total stroke volume.

Atrial contribution to LV filling

When LA: T_pre_a is available in timeseries/atrium.npz, the script samples LV: Volume [mL] at that frame and compares it with LVEDV and LVESV. If the pre-atrial-contraction LV volume is greater than LVEDV, the atrial-contribution rows are skipped.

Regional contribution boundary

Regional longitudinal, septal, and lateral stroke-volume contributions are important in the CMR literature [2], but they are not emitted by the current CardiacNexus combined-feature CSV. They are therefore documented as a planned/literature boundary, not as current output rows.

Output columns and units

Display family	Exact output column	Unit	Status	Schema note
AV plane displacement	`AV: AVPD [cm]`	cm	current	simplified 4-chamber implementation
IPVT	`AV: IPVT [mL]`	mL	conditional current	skipped when IPVT is negative
IPVT	`AV: IPVT Ratio`	ratio	conditional current	IPVT divided by LA total stroke volume
Atrial contribution	`AV: % AC to LVSV [%]`	%	conditional current	requires `LA: T_pre_a` and valid LV volume curve
Atrial contribution	`AV: % AC to LVEDV [%]`	%	conditional current	requires `LA: T_pre_a` and valid LV volume curve

Output reconciliation

Evidence layer	Result
Implementation source	5 current non-valvular combined rows checked against `eval_ventricular_atrial_feature.py`
Output inventory	`docs/data/output_column_inventory.yml` records `ventricular_atrial_feature` rows, including the valve rows documented on the valvular page
Phenotype dictionary	`docs/data/phenotype_dictionary.yml` links the 5 cross-chamber rows to this page
Page output table	all current cross-chamber rows are listed above

Required upstream inputs

la_4ch.nii.gz and seg4_la_4ch.nii.gz;
timeseries/ventricle.npz with LV ED/ES frames and LV volume curve;
timeseries/atrium.npz with LA: T_pre_a when atrial-contribution rows are expected;
aggregated ventricular_volume.csv;
aggregated atrial_volume.csv;
consistent subject eid membership in both aggregate tables.

Reference ranges with cohort and method context

Feature	Source	Cohort	Reference value	Status	Note
AVPD and regional contribution	Ostenfeld et al. [1]	pulmonary hypertension and controls	source-specific values	Verified context source	Multi-point protocol; current row is simplified
AVPD prognosis	Lindholm et al. [2]	PAH cohort and controls	source-specific cutoffs	Verified context source	Outcome context, not a CardiacNexus classifier
LV pumping contribution	Carlsson et al. [3]	healthy adults, athletes, DCM	method-specific values	Verified context source	Describes AVPD contribution to LV pumping
Atrial transport	Rahimtoola et al. [4]	myocardial infarction context	source-specific values	Verified context source	Historical atrial transport-function context
Automated CMR workflow	Ruijsink et al. [5]	UKB-scale CMR automation	quality-control context	Verified context source	Supports automation/QC caveat, not direct thresholds
IPVT	Aquaro et al. [6]	diastolic dysfunction CMR cohort	source-specific values	Verified context source	Method-compatible context; table-row extraction deferred

Source-located registry status: reference_range_sources.yml maps AVPD/regional contribution context to Ostenfeld 2016 and Lindholm 2022, LV pumping contribution to Carlsson 2007/2012 context, and IPVT to Aquaro 2019. Because the current CardiacNexus AVPD row is simplified and IPVT depends on paired atrial/ventricular volume curves, these rows remain source-located context rather than diagnostic thresholds.

Disease interpretation

AVPD and regional stroke-volume contribution have been studied in pulmonary arterial hypertension, dilated cardiomyopathy, and heart-failure prognosis [2] [3]. Atrial transport and IPVT relate to myocardial infarction and diastolic filling physiology [4] [6]. The disease badges on this page are interpretation aids only, not validated phenotype-to-ICD classifiers.

QC caveats and maturity boundary

Cross-chamber rows inherit missingness and QC failures from multiple source measurements. IPVT depends on coherent LV stroke volume and LA total stroke volume. Atrial contribution depends on ECG-derived pre-atrial-contraction timing and the LV volume curve. AVPD depends on 4-chamber segmentation quality and ED/ES frame consistency.

The current AVPD output is intentionally labeled as simplified because it does not reproduce the full multi-view LV/RV annular protocol used in several literature sources. Regional longitudinal, septal, and lateral contribution rows should not be shown as current CardiacNexus outputs until the pipeline emits them.

Implementation provenance

Feature family	Formula or computational route	Exact output columns	Source code file and function	Upstream dependencies	Conditional behavior	QC artifacts	Schema debt
AVPD	`evaluate_AVPD(seg4_la_4ch, nim_la_4ch, T_ED, T_ES)`	`AV: AVPD [cm]`	`eval_ventricular_atrial_feature.py`; `evaluate_AVPD`	4-chamber cine, 4-chamber segmentation, LV ED/ES frame indices	skipped if helper raises `ValueError`	`visualization/combined/AV_plane.png`	simplified 4-chamber AVPD, not full multi-view AVPD
IPVT	`LV: SV [mL] - LA: Total SV (bip) [mL]`; ratio to LA total stroke volume	`AV: IPVT [mL]`, `AV: IPVT Ratio`	`eval_ventricular_atrial_feature.py`	ventricular and atrial aggregate CSVs	skipped if IPVT is negative	no dedicated new figure	mixes ventricular and atrial upstream measurement assumptions
Atrial contribution	`(LVEDV - LV_pre_a) / LVSV` and `(LVEDV - LV_pre_a) / LVEDV`	`AV: % AC to LVSV [%]`, `AV: % AC to LVEDV [%]`	`eval_ventricular_atrial_feature.py`	`LA: T_pre_a`, LV volume time series, LVEDV, LVESV	skipped if timing is missing or pre-A volume exceeds LVEDV	no dedicated new figure	ECG timing availability controls row presence
Regional contribution boundary	not emitted by current pipeline	no current output column	not applicable	multi-view long-axis/short-axis regional contribution workflow would be required	literature context only	none	do not add regional rows unless pipeline emits them

Source audit

Current AVPD, IPVT, IPVT ratio, and atrial-contribution labels were checked against eval_ventricular_atrial_feature.py.
Valve diameter rows in the same aggregate remain documented on the valvular page and are not duplicated in this page's feature summary.
docs/data/reference_sources.yml is present and used as the curated reference-source registry for this page.
Textbook context boundary: implementation, AVPD, atrial transport, automated CMR, and IPVT sources are sufficient for draft rollout; Braunwald/Hurst background was not needed for the phenotype-specific output contract.
One AVPD measurement panel 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

Ostenfeld E, Stephensen SS, Steding-Ehrenborg K, Heiberg E, Arheden H, Radegran G, Holm J, Carlsson M. Regional contribution to ventricular stroke volume is affected on the left side, but not on the right in patients with pulmonary hypertension. The International Journal of Cardiovascular Imaging. 2016;32(8):1243-1253. doi:10.1007/s10554-016-0898-9.
Lindholm A, Kjellstrom B, Seemann F, Carlsson M, Hesselstrand R, Radegran G, Arheden H, Ostenfeld E. Atrioventricular plane displacement and regional function to predict outcome in pulmonary arterial hypertension. The International Journal of Cardiovascular Imaging. 2022;38(10):2235-2248. doi:10.1007/s10554-022-02616-w.
Carlsson M, Ugander M, Mosen H, Buhre T, Arheden H. Atrioventricular plane displacement is the major contributor to left ventricular pumping in healthy adults, athletes, and patients with dilated cardiomyopathy. American Journal of Physiology-Heart and Circulatory Physiology. 2007;292(3):H1452-H1459.
Rahimtoola SH, Ehsani A, Sinno MZ, Loeb HS, Rosen KM, Gunnar RM. Left atrial transport function in myocardial infarction. The American Journal of Medicine. 1975;59(5):686-694.
Ruijsink B, Puyol-Anton E, Oksuz I, Sinclair M, Bai W, Schnabel JA, Razavi R, King AP. Fully automated, quality-controlled cardiac analysis from CMR: validation and large-scale application to characterize cardiac function. JACC: Cardiovascular Imaging. 2020;13(3):684-695. doi:10.1016/j.jcmg.2019.05.030.
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.

Cross-chamber phenotypes