🏥 WHO Health Intelligence Gateway

All-in-one intelligence tool for Global Health Monitoring

This Actor unifies the three pillars of health data into a single, automated feed. It does the work of a data analyst in minutes:

1. 📊 The Past: Fetches official statistics (GHO API).
2. 📚 The Present: Aggregates WHO Publications & Reports via API (with optional PDF text extraction).
3. 🔬 The Future: Tracks ongoing Clinical Trials for cures (R&D).

---

🎯 Perfect For

• 🏛️ NGOs & Humanitarian Aid – Map active outbreaks to deploy resources efficiently
• 💊 Pharma & Biotech – Track disease trends + competitor R&D pipelines
• 📰 Data Journalists – Scan 100+ PDFs for keywords instantly
• 🔬 Health Researchers – Access structured WHO data without API complexity
• 🤖 AI Developers – RAG-ready health data for LLM systems
• 📊 Risk Analysis: Monitor stability and health risks in specific regions for corporate intelligence

---

⚙️ Key Features

📚 WHO Publications API + PDF Text Extraction

Access 18,000+ WHO publications via official API.
Optionally downloads PDFs and extracts full text content.

• Fetches publications by keyword and date range
• Downloads and parses PDF documents
• Extracts text from complex PDF layouts

🧠 Free NLP Location Detection (No API Key Required)

Uses embedded Natural Language Processing (compromise.js) to analyze PDF text at zero extra cost.

• Auto-Location Detection: Finds cities and countries (e.g., "Yemen", "Sudan", "Kenya") inside extracted PDF text automatically
• Smart Cleaning: Removes duplicates, filters generic regions, and strips quotes/formatting noise

🔔 Watchlist Alerts

Define keywords (e.g., Cholera, Level 3, Outbreak). The Actor scans report titles and full PDF text.

• If a match is found, the ⚠️ ALERTS column in your CSV is flagged
• Optional: Send instant notifications to Slack or Discord

📊 Organized Multi-Dataset Output (Limited-permissions friendly)

This Actor keeps permissions limited by producing:

• Combined dataset (default): All normalized items are pushed to the run’s default dataset for immediate viewing in the Output tab (linked via {{links.apiDefaultDatasetUrl}}).
• 3 separate per-run datasets: GHO / Publications / Trials each go into their own dataset created uniquely per run (so old runs don’t mix with new ones).
• Stable discovery of those per-run datasets: The Actor writes a DATASET_LINKS JSON record to the run’s default key-value store, and the output schema links to it (key-value store is created per run).[1]
• HTML dashboard: Stored as a KV record and linked from output schema.[1]
• CSV export (optional): Stored as a KV record when outputFormat=csv.[1]

---

🎛️ Input Configuration

Section	Setting	Description
📚 Publications	Keyword Filter	Search term (e.g., "cholera", "mpox")
📚 Publications	Extract PDF Text	Enable this to download PDFs and extract text (includes NLP location detection)
📊 GHO Stats	Indicators	Codes like WHOSIS_000001 (Life Expectancy), MORT_100 (Mortality)
📊 GHO Stats	Years/Date Range	Filter by specific years or year range (startYear/endYear)
🔔 Alerts	Alert Keywords	Keywords to flag documents (e.g., "outbreak", "emergency")
🔬 Trials	Target Disease	Disease to search for (e.g., "Malaria", "Cancer")
🔬 Trials	Status/Phase	Filter by recruitment status and trial phase
📤 Output	Output format	json (default) or csv

---

📦 Outputs (Where to find what)

1) Combined results (default dataset)

• In Apify Console → Run → Output, open Combined dataset (default).
• This points to {{links.apiDefaultDatasetUrl}}/items as defined by Apify output schema variables.

2) Separate per-run datasets (GHO / Publications / Trials)

Because these datasets are created per run with run-scoped names, their IDs/URLs are published in:

• Key-value store record: DATASET_LINKS (JSON) in the run’s default KV store.[1]
• In Apify Console → Run → Output, open Separate datasets (links) (it links to {{links.apiDefaultKeyValueStoreUrl}}/records/DATASET_LINKS).

Expected DATASET_LINKS format:

{
  "ghoItems": "https://api.apify.com/v2/datasets/<id>/items",
  "publicationsItems": "https://api.apify.com/v2/datasets/<id>/items",
  "trialsItems": "https://api.apify.com/v2/datasets/<id>/items"
}

3) Raw source JSON (KV records)

• DATA_GHO (raw GHO items)
• DATA_PUBLICATIONS (raw publications)
• DATA_TRIALS (raw trials)

These are stored in the run’s default key-value store (one per run).[1]

4) HTML dashboard + CSV

• Dashboard: OUTPUT_DASHBOARD (HTML record linked from output schema using the /records/<key> form).
• CSV: OUTPUT_CSV (generated only if outputFormat=csv).[1]

---

📤 Output Example

The Actor produces a comprehensive dataset. Here is a simplified preview:

[
  {
    "type": "Publication",
    "title": "Multi-country outbreak of cholera, Report #32",
    "date": "11/26/2024",
    "url": "https://www.who.int/publications/m/item/...",
    "extractedLocations": ["Zimbabwe", "Mozambique", "Ethiopia", "Kenya"],
    "alertTriggered": true,
    "matchedKeywords": ["cholera", "outbreak"]
  },
  {
    "type": "GHO Indicator",
    "title": "WHOSIS_000001 - IND",
    "date": "2020",
    "value": 69.17521966
  },
  {
    "type": "Clinical Trial",
    "title": "Phase 3 Malaria Vaccine Study in Children",
    "nctId": "NCT05790889",
    "status": "RECRUITING",
    "phase": "PHASE2",
    "conditions": "Malaria",
    "url": "https://clinicaltrials.gov/study/NCT05790889"
  }
]

---

⚖️ Legal & Compliance

This Actor respects WHO and ClinicalTrials.gov terms of service:

• ✅ Uses official WHO Publications API (https://www.who.int/api/hubs/meetingreports)
• ✅ Uses official WHO GHO OData API (https://ghoapi.azureedge.net/api)
• ✅ Uses official ClinicalTrials.gov API v2 (https://clinicaltrials.gov/api/v2/studies)
• ✅ No web scraping - all data accessed via public APIs
• ✅ PDF downloads respect standard HTTP protocols

---

Built with ❤️ and </> by The_Rook - Try it now 🚀

1