OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Data & Analytics

Openmed

Openmed is an Apache-2.0 healthcare AI SDK for extracting structured clinical entities and de-identifying sensitive information in medical text. Its Python runtime processes locally after required model artifacts are available, with documented implementation paths for Apple MLX, Android ONNX Runtime Mobile, browser Transformers.js and Swift applications through OpenMedKit.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Openmed — OpenViST project profile in AI Data & Analytics.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Openmed when building a software workflow that needs clinical entity extraction or de-identification and you can select, acquire and evaluate appropriate model artifacts for the target runtime.

Who it is for

  • Developers building clinical text-processing applications.
  • Healthcare technology teams evaluating local clinical NLP or de-identification workflows.
  • Researchers or engineers who need to extract entities such as diseases and drugs from medical text.

Documented uses

  • Extract clinical entities from text, including disease and drug mentions.
  • Redact sensitive identifiers from clinical notes.
  • Build local clinical text-processing applications for supported desktop, mobile or browser runtimes.
Before you choose

Requirements & limits

The SDK source is released under Apache-2.0. Model and dataset terms vary. Model downloads and some optional integrations may use a network.

  • The core runtime can process locally after required model artifacts are available, but acquiring models and using some optional integrations may require network access.
  • Supported models and operations vary by runtime and artifacts; the repository does not specify which operations are available through every mobile or browser path.
  • The SDK license does not determine model or dataset licensing; review each separately.
  • The repository describes a Safe Harbor-aligned configuration targeting 18 identifier categories, while stating that use of the SDK alone does not establish HIPAA compliance.
  • Deployment owners are responsible for evaluating privacy behavior and clinical fitness for their application.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install OpenMed with the Hugging Face runtime

Set up OpenMed’s core and Hugging Face runtime on Linux, macOS, or Windows for CPU or CUDA use.

Before you start

  • Python 3.10 or later.
  • Linux, macOS, or Windows; CPU or CUDA.

Steps

  1. Install the core package and Hugging Face runtime with the command below.

    Documented command
    pip install --upgrade "openmed[hf]"

  2. Use the Python API example with analyze_text, passing clinical text and a model name such as disease_detection_superclinical.

  3. Inspect the returned entities, for example by printing each entity’s label and text.

Expected outcome

The package is installed, and the documented API pattern returns structured entities after required model artifacts are available.

Guide limitations

  • The local runtime requires model artifacts; model downloads may use a network.
  • Model and dataset terms vary and should be reviewed. Clinical fitness, privacy behavior, and deployment suitability require owner validation.
  • The documented example does not establish clinical validation.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
7 October 2026
Original website title
maziyarpanahi/openmed · Official homepage title · retained in its source language
How we review sourcesSuggest a correction