Clinical Decision Support Python Packages

haenv

HAEnv — synthetic longitudinal patients with code-derived gold, for evaluating health agents over time

pycyclops

A toolkit for evaluating and monitoring AI models in clinical settings

miscnn

A framework for Medical Image Segmentation with Convolutional Neural Networks and Deep Learning

conmedrl

Offline constrained reinforcement learning toolkit for critical-care research, with ICU data processing, FQI/FQE, and interoperability tools.

metaxu

An AI assurance and provenance layer for healthcare — model-agnostic, agent-agnostic, EHR-agnostic.

inflammation-prediction-risk-models

Early prediction of inflammation from ICU time-series data (XGBoost) — runs end-to-end on synthetic data; original MIMIC-III workflow preserved for credentialed users.

aucmedi

a framework for Automated Classification of Medical Images

bh-sentinel-ml

Open-source clinical safety signal detection for behavioral health systems. Text in. Flags out. Clinician decides.

fhir-mcp

The trustworthy FHIR bridge for AI agents — MCP server with reproducible clinical evals, reversible keyed de-identification, and CDS Hooks.

pgx-mcp

MCP server for pharmacogenomics — ClinVar, PharmGKB, gnomAD, OpenTargets, and clinical trials from Claude

bh-sentinel-core

Open-source clinical safety signal detection for behavioral health systems. Text in. Flags out. Clinician decides.

medmcp

Medical reasoning layer for AI agents — deterministic symptom risk signals, drug interactions, labs, ICD-11; HTTP API + MCP (stdio). Not a substitute for clinical care.

keystone-bench

A clinical decision-evidence benchmark: pull the one fact an answer rests on, and see whether the assistant's action moves with the evidence. Built on HealthBench.

dentrx

An open, guideline-cited dental prescribing reference (antibiotics & analgesics) — US + UK, adults & children. Educational reference, not a prescription.

moisscode

A domain-specific language for clinical decision support and biotech workflow automation

miscnn-tf-2-0

A framework for Medical Image Segmentation with Convolutional Neural Networks and Deep Learning

medeval-harness

Emergency-department triage assistant where the LLM is structurally barred from deciding urgency. 68 YAML rules in pure Python assign the ESI level; the model only extracts structured facts and rephrases the rationale in plain English. LangGraph, Langfuse-traced, with a 50-case eval harness gating CI.

miscnn-tf-1-14

A framework for Medical Image Segmentation with Convolutional Neural Networks and Deep Learning