Feat/eit parser complete#30
Merged
Merged
Conversation
- parse() reads preamble via np.memmap, resolves spec by frame-size
divisibility, memory-maps binary frames with FRAME_EIT_DTYPE
- calibration pipeline (EIDORS read_draeger_file.m, Adler 2016):
vv = FT_A * trans_A - FT_B * trans_B [calibrated transimpedance]
I_real = injection_current / FC_CURRENT [A]
V_diff = (voltage_A - voltage_B) / FV_VOLTAGE [V]
- aux_signals: timestamp, trans_A/B raw, I_real, V_diff, frame_counter, medibus
- warns if Framerate [Hz] missing from header before defaulting to 50 Hz
- RawImpedanceData: add aux_signals field; clarify measurements docstring
- README: DragerEitParser status Scaffold -> Implemented; add .eit quick-start example; update RawImpedanceData table entry - docs/data_model.md: document aux_signals field in RawImpedanceData with all keys, shapes, and units
- Parse patient01.eit and .bin, verify measurements shape and sanity - Check timestamps monotonic, print calibration constants - Compare frame count between .eit and .bin
- test_header_formats.py: registry contents, get_eit_specs(), frozen spec - test_eit_dtypes.py: itemsize 5495, field offsets trans_A=16 trans_B=2592, frombuffer round-trip - test_eit_utils.py: parse_eit_header() valid/error cases, field map keys, calibration constants, Time colon edge case - test_eit_parser.py: validate(), parse() shape/metadata/calibration/ aux_signals, fs warning, truncated file error, 5 real-file tests
tests/models/ — base_data, reconstructed_data tests/parsers/ — base_parser, routing, header_formats tests/parsers/draeger/ — bin dtypes/parser/utils, asc parser, eit dtypes/parser/utils tests/parsers/timpel/ — timpel dtypes, parser All 242 tests pass. git mv used to preserve history.
- Add eit_greit.py: reconstruct_greit() and build_greit() wrapping pyEIT GREIT solver for RawImpedanceData (208 measurements → 32×32 images). Handles external reference frames for cross-recording EELI comparison. Ref: Adler et al., Physiol. Meas. 30 (2009) S35–S55, DOI 10.1088/0967-3334/30/6/S03 - Update parsers/__init__.py: document DragerEitParser as implemented; add reconstruction utilities section - Update README: add GREIT quickstart snippet with pyeit optional dep - Update docs/data_model.md: document RawImpedanceData.measurements protocol note (208 = 16 elec × 13 pairs, adjacent drive) - Fix ruff: sort imports, remove unused variable (walrus operator), remove unused numpy import across 4 test files
…fmmu and §10 cross-correlation - Translate all Italian markdown cells to English throughout - §9: protocol diagnostic — std vs fmmu measurement ordering with side-by-side reconstruction at peak inspiration; explains why fmmu is expected to match Dräger hardware sweep order - §10: frame alignment check via cross-correlation of .eit global signal vs .bin global_signal; confirms lag=0 prerequisite for paired ML dataset construction - Summary table at end covering all validation checks - Narrative reframed: bridge .eit → pyEIT is functional; spatial quality gap vs vendor .bin motivates data-driven reconstruction (documented in summary, will be addressed in feat/ml-reconstruction)
- docs/data_model.md: now covers only data container specs (BaseData, ReconstructedFrameData, ContinuousSignalData, RawImpedanceData); overview table moved to parsers.md - docs/parsers.md (new): parser reference for all implemented parsers (DragerBinParser, DragerEitParser, DragerAscParser, TimpelTabularParser); includes binary frame layouts, calibration constants, measurement protocol note, GREIT bridge API, and loader utilities - README: update Documentation section with three-file structure
- tasks.csv: mark 2.5.2 done (measurement protocol documented); add 7.0.1 done (GREIT bridge), 7.0.2/7.0.3 in_progress (std/fmmu + cross-correlation — cells written, need Riccardo to run); add ML.1/ML.2/ML.3 todo (data-driven reconstruction branch) - learning_journal: full session entry — GREIT spatial diagnosis, std vs fmmu explanation, ML reconstruction rationale, eitprocessing finding, EELI cross-recording strategy, branch status
…r convention Measurement ordering: - Change parser_meas from 'std' to 'fmmu' in build_greit(). The PulmoVista 500 sweeps voltage measurements starting from the electrode adjacent to the current sink for each drive pattern. This matches pyEIT's fmmu (rotating) protocol, not std (absolute). Source: Dräger PulmoVista 500 IFU SW 1.3n p. 144 (measurement diagram). Image orientation — third correction added (fliplr): - Existing: negate (impedance convention) + rot90(k=1) (electrode 1 at 12 o'clock = anterior at top). - New: np.fliplr() after rot90 to match caudal-cranial CT convention. Dräger displays images viewed from feet: left of image = right of patient. pyEIT produces the anatomical mirror; fliplr corrects this. Source: IFU p. 146 — "il lato sinistro dell'immagine visualizza il lato destro del paziente". docs/parsers.md: update §3 (fmmu confirmed) and §6 (three-step orientation pipeline with IFU page citations).
…ot parallel The PulmoVista 500 records only the .eit file. The .bin is generated by Dräger's software applying Newton-Raphson reconstruction to each .eit frame in sequence. Frame-level alignment is therefore guaranteed by construction, not empirically — a temporal offset is architecturally impossible. Cross-correlation lag=0 is a verification, not a prerequisite. Also updates ML rationale: the paired dataset has zero misalignment noise because .bin[k] is the deterministic transform of .eit[k].
|
Codecov Report❌ Patch coverage is
📢 Thoughts on this report? Let us know! |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.