Health AI Platform

The Health AI Platform is the part of Loop that turns patient context, workflow definitions, tool calls, research knowledge, and ML services into repeatable health operations.

This section is intentionally practical:

It maps directly to code that exists in this repository.
It shows where the platform is production-ready today.
It calls out where an interface exists but the full implementation is still an extension point.

What the platform includes

Workflow orchestration with @loop/workflow-engine
Clinical context through the Patient Graph API and typed client
Agent tools through Luna’s tool registry
Knowledge ingestion through the research paper pipeline and embeddings
ML service boundaries through the Python FastAPI service and @loop/ml-client

Start here

Read the Overview
Follow Getting Started
Learn the 4D Chess Architecture
Build your first workflow in Guides

Section map

Core pages

Workflows

Integrations

Platform Components

Guides

API Reference

Platform in one snippet


import { WorkflowEngine } from '@loop/workflow-engine';
import { PatientGraphClientImpl } from '@loop/patient-graph-client';
 
const patientGraph = new PatientGraphClientImpl({
  baseUrl: process.env.PATIENT_GRAPH_API_URL!,
  getAuthToken: async () => process.env.PATIENT_GRAPH_API_KEY!,
});
 
const workflow = WorkflowEngine.load(`
id: thyroid-review-v1
name: Thyroid Review
version: 1.0.0
steps:
  - id: check-tsh
    type: check_biomarker
    action:
      type: check_value
      params:
        biomarker: TSH
        threshold: 2.5
        operator: ">"
  - id: recommend-selenium
    type: recommend_supplement
    condition: "check-tsh.result === true"
    action:
      type: recommend
      params:
        supplement: Selenium
        dosage: 200mcg
`);
 
const profile = await patientGraph.getPatientContext('user_123');
 
if (profile.ok) {
  const result = await workflow.execute({
    patientId: profile.data.externalId,
    data: { biomarker: { TSH: 4.2 } },
  });
 
  console.log(result.recommendations);
}

Current-state notes

The workflow engine is real and test-backed.
The Patient Graph service and client are real and actively routed.
Luna’s tool registry is real and powers agent capabilities.
The research ingestion pipeline exists in @loop/ai, but the admin hook currently uses a placeholder wrapper.
The ML service currently exposes a health endpoint; the broader ML surface is intentionally small in this repository snapshot.