| Status |  |
| License |  |
| Python |  |
| Docs |  |
| Coverage |  |
| CI |    |
Python library for Electrical Impedance Tomography (EIT) recordings of the Dräger PulmoVista 500 and of Timpel: it reads the files, finds and checks each breath, computes the clinical indices of tidal recordings and PEEP trials, and writes them as tables for R, SPSS and databases. For research in mechanical ventilation and lung monitoring.
Documentation: kernel236.github.io/fastEIT
pip install git+https://github.com/Kernel236/fastEIT.gitfrom fasteit.export import recording_tables, write_csv
from fasteit.features import tidal_features
from fasteit.models import RecordingContext
from fasteit.parsers.loader import load_data
from fasteit.preprocessing import preprocess
raw = load_data("recording.bin") # 1. read the file
recording = preprocess(raw, context=RecordingContext(patient_id="P01", label="T0")) # 2. breaths
features = tidal_features(recording) # 3. clinical indices
write_csv(recording_tables(features), "results/") # 4. tables for R and SPSSA PEEP trial adds segment_peep_trial between steps 2 and 3. The
user guide covers both protocols, cutting a
recording, the review figures and the export.
| Step | |
|---|---|
| Files | Dräger .bin and Timpel exports (reconstructed images) are analysed; Dräger .eit (raw measurements) and .asc (signals) are read |
| Breaths | cardiac filter, breath detection with quality flags, exclusions, lung mask, regions; PEEP, plateau, driving pressure and tidal volume of each breath, in cmH2O and mL |
| PEEP trials | steps from the airway pressure or placed by hand, at the PEEP levels of the protocol; breaths analysed, driving pressure and tidal volume of each step |
| Clinical indices | EELI, TIV, share by region, global inhomogeneity, centre of ventilation, compliance; Costa collapse and overdistension, crossing PEEP, best PEEP, recruitment-to-inflation ratio |
| Tables | one row per recording, breath and PEEP step, keyed by patient, recording and phase; CSV for R, SPSS and Excel |
| Figures | the segments on the file, and the breaths used for the indices |
Every method states its source (paper and DOI, or an evidence notebook in notebooks/),
and the steps reimplemented from eitprocessing
are checked against it.
Alpha: the pipeline from a file to the tables is complete and tested; names and tables may still change before 1.0.
.eit files can be reconstructed with pyEIT through reconstruct_greit() (optional:
pip install "fasteit[pyeit] @ git+https://github.com/Kernel236/fastEIT.git"; GREIT,
Adler et al., Physiol. Meas. 2009, DOI: 10.1088/0967-3334/30/6/S03). This bridge is
experimental: it uses a generic circular mesh, and its images do not match the Dräger
reconstruction stored in .bin files.
Contributions are welcome: see CONTRIBUTING.md. This project follows the Contributor Covenant code of conduct.
This software is for research purposes only. It is not a medical device and has not been validated for clinical decision-making.
Apache License 2.0 - see LICENSE.