Deterministic ECG codec with Python + Rust core; staged proof on 4 standard clinical corpora (MIT-BIH, NSTDB, EDB, PTB-XL)
Project description
ZPE-Bio
Deterministic ECG codec repository with a Python CLI, a Rust core crate, and committed validation artifacts. This README is intentionally limited to claims that are backed by proof artifacts already on disk and by tests that run in CI. ZPE-Bio is one of 17 independent encoding products in the Zer0pa portfolio, each developed for its own domain.
This repo's claimed surface is ECG staged proof only. It does not use the wearable runbooks as release evidence, and it does not make a generalized biosignal or regulatory claim.
Anchored Proof Surface
| Claim | Proof Artifact On Disk | CI Test Anchor |
|---|---|---|
| MIT-BIH benchmark writer emits summary + aggregate artifacts | validation/results/BENCHMARK_SUMMARY.md, validation/results/mitdb_python_only/mitdb_aggregate.json |
tests/test_benchmark_mitdb.py |
| PTB-XL benchmark writer emits committed-style summary artifacts | validation/results/ptbxl/summary.json |
tests/test_benchmark_physionet.py |
| Clinical ECG round-trip remains deterministic and high-fidelity in the Python codec | validation/results/BENCHMARK_SUMMARY.md |
tests/test_codec.py, tests/test_deterministic_replay.py |
| Python and Rust codec implementations remain parity-gated in CI | core/rust/, tests/test_parity.py |
tests/test_parity.py |
Not Claimed
- No public release-readiness verdict
- No generalized biosignal victory claim
- No Bio Wearable closure claim
- No regulatory or FDA claim
Benchmark Results (committed artifacts, beta)
Results below are from committed benchmark artifacts. All runs are CPU-only Python codec; no hardware path is implied.
MIT-BIH Arrhythmia Database (48 records, full corpus)
| Metric | Value |
|---|---|
| Records processed | 48/48 |
| Integrity pass rate | 48/48 (100%) |
| Mean compression ratio | 1.323 |
| Mean SNR | 43.3 dB |
| Mean RMSE | 3.24 uV |
| Mean PRD | 1.12% |
| Max PRD | 2.32% |
Source: validation/results/BENCHMARK_SUMMARY.md, validation/results/mitdb_python_only/mitdb_aggregate.json | CI: tests/test_benchmark_mitdb.py
MIT-BIH Noise Stress Test Database (nstdb, 15 entries)
| Metric | Value |
|---|---|
| Records processed | 15/15 |
| Integrity pass rate | 15/15 (100%) |
| Mean compression ratio | 1.310 |
| Mean SNR | 60.5 dB |
| Max PRD | 1.96% |
Source: validation/results/nstdb/summary.json | CI: tests/test_benchmark_physionet.py
European ST-T Database (edb, 90 entries)
| Metric | Value |
|---|---|
| Records processed | 90/90 |
| Integrity pass rate | 90/90 (100%) |
| Mean compression ratio | 1.376 |
| Mean SNR | 52.5 dB |
| Max PRD | 4.34% |
Source: validation/results/edb/summary.json | CI: tests/test_benchmark_physionet.py
PTB-XL ECG Database (100 records sample)
| Metric | Value |
|---|---|
| Records processed | 100/100 |
| Integrity pass rate | 100/100 (100%) |
| Mean compression ratio | 1.576 |
| Mean SNR | 32.0 dB |
| Max PRD | 5.29% |
Source: validation/results/ptbxl/summary.json | CI: tests/test_benchmark_physionet.py
Note on PTB-XL SNR/PRD: PTB-XL records are 12-lead 500 Hz clinical studies; the lower SNR relative to MIT-BIH reflects higher signal diversity and more channels compressed per segment, not codec regression. Integrity passes 100/100.
Local Validation
python -m venv .venv
source .venv/bin/activate
python -m pip install --upgrade pip
python -m pip install -e ".[dev,validation]"
pytest -q
python -m build
Quick ECG smoke check:
python -m zpe_bio roundtrip --mode clinical --samples 250
Repository Map
python/zpe_bio/: Python package and CLIcore/rust/: Rust codec crateembedded/: embedded reference firmware treetests/: repo-local pytest suitescripts/: benchmark and operator scriptsvalidation/results/: committed benchmark outputsvalidation/runbooks/: execution and boundary runbooksdocs/: repo documentation and regulatory/reference material
Read Next
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file zpe_bio-0.2.1.tar.gz.
File metadata
- Download URL: zpe_bio-0.2.1.tar.gz
- Upload date:
- Size: 102.9 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.1.0 CPython/3.13.12
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
abc4f3203ee4a6649bd7be9014ad8798bfff631753c5d8dff87885dd3d24dbfd
|
|
| MD5 |
3411cc076499bf53881eff95722e6fc7
|
|
| BLAKE2b-256 |
e758a9e19d26b1f6070539c6e18331c8fbe77902f164cdca37a3bb836fd6bd2e
|
File details
Details for the file zpe_bio-0.2.1-py3-none-any.whl.
File metadata
- Download URL: zpe_bio-0.2.1-py3-none-any.whl
- Upload date:
- Size: 108.7 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.1.0 CPython/3.13.12
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
569a52eae1a440dac6563938ba4244ab795e745dc3c0299f44a6f6efa3b39ad0
|
|
| MD5 |
aeed9fd4217cb68254fe3b294012b980
|
|
| BLAKE2b-256 |
38e99d0c10de99df4f6f20658309aade073bec9b3eb2153cbe4de048444f6452
|