Formale Wissensrepräsentation klinischer Verläufe – §3–§22 (FM-3)
Project description
AION Clinical 2.0.0
Formale Wissensrepräsentation klinischer Verläufe – Referenzimplementierung der Abschnitte §3–§22 des Papers FM-3 (10.5281/zenodo.19548857).
Copyright 2026 Friedhelm Matten / ISCaD GmbH – EUPL-1.2
Installation
# Kern (stdlib-only, sehr leicht)
pip install aion-clinical
# Mit REST-API + PostgreSQL + Monitoring
pip install "aion-clinical[api,postgres,monitoring]"
# Komplettes Produktionspaket
pip install "aion-clinical[full]"
Paper-Implementierung
| § | Modul | Beschreibung |
|---|---|---|
| §11 | aion.query.language |
Formale Abfragesprache (Kohortenformeln) |
| §15 | aion.causal.pc_algorithm |
PC-Algorithmus + Bootstrap |
| §18 | aion.schema_evolution, aion.schema_registry |
Schemaevolution |
| §20 | aion.privacy.dp, aion.federation |
Differential Privacy + Föderierung |
| §21 | aion.ai.components, aion.ai.trajectories |
KI-Komponenten (7 Schichten) |
| §22 | aion.explain.shapley |
Erklärbarkeit (Shapley, kontrafaktisch, suffizient, beschränkt) |
Schnellstart
from aion.privacy.dp import LocalDP
# Lokale DP-Abfrage
dp = LocalDP(epsilon=1.0)
noisy_count = dp.release_local_count(true_count=42)
print(f"Noisy: {noisy_count:.1f}")
from aion.causal.pc_algorithm import PCAlgorithm
# Kausalstruktur lernen
pc = PCAlgorithm(alpha=0.05)
graph = pc.fit(data)
print(graph.adjacency())
from aion.explain.shapley import bounded_explanation
# Beschränkte Erklärung (§22.5)
result = bounded_explanation(events, feature_fn, model_fn, k_max=5)
print(result.selected_events) # garantiert |S| ≤ 5
REST-API starten
pip install "aion-clinical[full]"
aion # → http://localhost:8000
Zitation
@article{matten2026aion,
title = {Formale Wissensrepräsentation klinischer Verläufe},
author = {Matten, Friedhelm},
year = {2026},
doi = {10.5281/zenodo.19548857},
url = {https://zenodo.org/records/19548857},
}
Lizenz
EUPL-1.2 (European Union Public Licence). Siehe LICENSE.
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 aion_clinical-2.0.0.tar.gz.
File metadata
- Download URL: aion_clinical-2.0.0.tar.gz
- Upload date:
- Size: 62.7 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.1.0 CPython/3.13.5
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
1f623777dc039429be9c48cd1389883ede59f4c937177340b5ce470ae3229e0a
|
|
| MD5 |
c2eca3480f26fb0260bd4a8c6335378e
|
|
| BLAKE2b-256 |
21cb81541baddb84b4412979accb3704cb18a356f1bfca3ab9c28bfba92a1f1f
|
File details
Details for the file aion_clinical-2.0.0-py3-none-any.whl.
File metadata
- Download URL: aion_clinical-2.0.0-py3-none-any.whl
- Upload date:
- Size: 76.4 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.1.0 CPython/3.13.5
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
2fe37e87113ad055f62f79646f0d2ef1727481de2a06f2e5e8e052fedc2d97d3
|
|
| MD5 |
4d6f7b2797782e01a64cbeea10a78dbb
|
|
| BLAKE2b-256 |
b75cb8977e1dc21cc6e891cbac8fb797db543e7482346b0a8deccb86bf1ca5f1
|