Reproducibility package for: "When does intravenous oxygen help? A physiological simulation of oxygen microbubble infusion across two regimes of critical oxygen economy: the difficult airway and cardiac arrest."

# Reproducibility package: intravenous oxygen microbubbles across two oxygen economiesPhysiological simulation code, generated data and figures accompanying the manuscript"When does intravenous oxygen help? A physiological simulation of oxygen microbubbleinfusion across two regimes of critical oxygen economy: the difficult airway and cardiacarrest."This deposit lets any reader reproduce every numerical result and figure in the paperfrom source.## Contents```code/ oxygen_model.py Core model: shared oxygen-transport core (Severinghaus dissociation with Bohr correction, Berggren shunt mixing, supply-limited consumption, dissociation-curve-limited microbubble offloading) plus the two regime simulators (difficult airway, asphyxial arrest) and the BVM extension. analysis_final.py Deterministic runs, Monte Carlo (2,000 iterations/regime), dose-response and metabolic-suppression sweeps; writes results_final.json and figures F1-F5. sobol_arrest.py Variance-based global sensitivity analysis (Sobol, SALib) sobol_airway.py of the oPMB benefit in each regime (1,024 base samples, 10,240 evaluations/regime); write sobol_*.json and F7. bvm_shunt_sweep.py Two-way beta x shunt sweep of oPMB benefit (Figure 9), showing the single effective-oxygenation threshold. swine_validation.py External plausibility check of the arrest model against the measured swine SaO2 trajectory (Garcia Mancebo et al. 2024, Fig 4b); produces F8.data/ results_final.json Deterministic + Monte Carlo summary outputs. sobol_arrest.json First- and total-order Sobol indices (arrest regime). sobol_airway.json First- and total-order Sobol indices (airway regime). swine_fig4b_values.csv Group-mean arterial SaO2 by minute, extracted from the published Source Data of Garcia Mancebo et al. 2024 (Fig 4b), used as the comparison target for swine_validation.py.figures/ F1-F8 *.png 300 dpi figures as used in the manuscript.```## RequirementsPython 3.10+ and the packages in `requirements.txt`:```numpyscipymatplotlibSALibopenpyxl```Install with: `pip install -r requirements.txt`## How to reproduceFrom the `code/` directory:```python analysis_final.py # deterministic, Monte Carlo, dose-response -> results_final.json, F1-F5python sobol_arrest.py # arrest global sensitivity -> sobol_arrest.jsonpython sobol_airway.py # airway global sensitivity -> sobol_airway.jsonpython swine_validation.py # swine plausibility check (prints model vs measured)```The Sobol scripts take a few minutes each (10,240 model evaluations per regime).## Notes on interpretation- The arrest "time above critical cerebral oxygen delivery" endpoint is a physiological surrogate, not survival or neurological outcome.- All arrest simulations use a fixed 20-minute horizon; curves reaching that horizon are horizon-censored, not biologically plateaued.- The swine analysis is an external plausibility check of the oxygen-delivery and offloading machinery: CPR-generated flow was selected within a physiologically plausible range to match the measured trajectory and was not independently reported in the source study. It is not independent validation of the human-scale prediction.- Monte Carlo distributions reflect parameter uncertainty (prior-predictive sweeps), not Bayesian updating from observed data.## Source data attribution`data/swine_fig4b_values.csv` is derived from the open-access Source Data of:Garcia Mancebo J, Sack K, Hartford J, et al. Systemically injected oxygen within rapidlydissolving microbubbles improves the outcomes of severe hypoxaemia in swine. NatureBiomedical Engineering 2024;8(11):1396-1411. https://doi.org/10.1038/s41551-024-01266-8(CC BY 4.0).