Pulmonary lymphatic transport

The lymphatics module implements the pulmonary fluid and lymphatic transport equations published by Ashworth et al. (2023). It was restored from the historical EdeTede implementation and integrated with the current develop data structures without including the later surfactant model.

Model inputs

The workflow combines two independently calculated inputs:

• perfusion supplies mean capillary pressure, capillary sheet surface area, and blood transit time; and

• ventilation supplies the minimum and maximum terminal elastic recoil pressure through export_terminal_lymphatic_inputs and import_terminal.

After perfusion geometry and results have been established, call define_problem_type('lymphatic_transport'). This preserves the existing perfusion unit fields while allocating the additional lymphatic fields. Then import any archived inputs and call lymphatic_transport.

Model parameters

The published human values remain the defaults, so existing workflows do not need to set any parameters. A value can be changed from Python before calling lymphatic_transport with:

from aether.parameter_types import update_lymphatics

update_lymphatics('lung_mass_g', 1.5)

Calling update_lymphatics('help', 0.0) prints the current values. Parameter names and human defaults are:

Name	Default	Meaning
lung_mass_g	639	Total lung mass in g
breathing_rate_bpm	15	Respiratory rate in breaths/min
capillary_hydraulic_conductivity	4.41335e-8	Historical capillary conductivity value used by the published model
interstitial_capacity_ml_per_100g	30	Maximum interstitial fluid volume per 100 g of lung
initial_interstitial_saturation	0.48	Initial interstitial volume as a fraction of capacity
interstitial_compartment_a_fraction	0.005	Fraction of interstitial capacity assigned to compartment A
interstitial_pressure_min_mmhg	-8	Minimum interstitial pressure in mmHg
interstitial_pressure_max_mmhg	-1	Maximum interstitial pressure in mmHg
lymphatic_pressure_min_mmhg	-8	Minimum initial-lymphatic pressure in mmHg
lymphatic_pressure_max_mmhg	1	Maximum initial-lymphatic pressure in mmHg
lymphatic_density	1	Lymphatic-to-capillary exchange-area ratio
lymphatic_saturation_threshold	0.3	Saturation at which the conductivity polynomial becomes active
lymphatic_baseline_conductivity_ratio	1.48	Baseline lymphatic-to-capillary conductivity ratio
lymphatic_conductivity_coefficient_1 to _6	845.87, -2416.7, 2388.5, -922.24, 125.85, -0.0067	Fifth-order saturation/conductivity polynomial coefficients
pressure_phase_offset_radians	1.570796326794895	Lymphatic pressure phase offset relative to breathing
integration_steps_per_transit	96	Number of integration steps per capillary transit time
convergence_tolerance	5e-6	Interstitial-saturation convergence tolerance

lymphatic_surface_area_ratio is accepted as a descriptive alias for lymphatic_density. The retained lymphatic_integrity, test_time, and reflection_coefficient names do not affect the active published hydrostatic model: integrity has no implemented valve/backflow equation, test_time is superseded by convergence control, and the osmotic pathway is inactive with its default reflection coefficient of zero.

Output fields

get_ne_lymph_flux and get_ne_lymph_intsat return the element-field indices for average flux and interstitial saturation. These are dedicated fields: lymphatic transport does not overwrite the unstrained vascular radii. Terminal results are available from export_terminal_lymphatic.

Reproducibility and provenance

The equations in src/lib/lymphatics.f90 retain the published implementation. Integration fixes are limited to elapsed-time reporting, safe unit-field resizing, explicit terminal-to-unit matching, and dedicated lymphatic export fields. Runtime can vary with processor load; numerical outputs, rather than the elapsed-time message, should be compared when checking reproducibility.

The complete Python workflow is maintained in the matching lymphatics_Ashworth2023 example in the lung-group-examples repository.

The source implementation contains a historical note that output units may need further verification for a possible factor-of-1000 conversion. That scientific calibration has not been altered by this restoration and should be resolved in a separately validated change.