PubMed Scraper | Medical Literature Data Extraction API | Apify Actor

PubMed scraper and medical literature data extraction API. Extract papers, abstracts, and citations from PubMed with this Apify actor. Free tier included.

Built for biomedical researchers, pharma analysts, and AI developers who need clean, structured clinical literature without XML parsing.

Quick Start · API Reference · Pricing · Support

---

Quick Start

import { ApifyClient } from 'apify-client';
import 'dotenv/config';

const client = new ApifyClient({ token: process.env.APIFY_TOKEN });

const run = await client.actor('getascraper/pubmed-rag-extractor').call({
  searchTerm: 'SARS-CoV-2 vaccines',
  articleTypes: ['review', 'meta_analysis'],
  dateFrom: '2024-01-01',
  dateTo: '2024-01-31',
  maxPapers: 10,
});

const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items);

Output:

{
  "pmid": "38212345",
  "pmcid": "PMC10812345",
  "doi": "10.1000/example",
  "title": "mRNA Vaccines: A New Era in Immunology",
  "abstract": "Messenger RNA vaccines represent a breakthrough...",
  "authors": ["Smith J", "Doe A"],
  "journal": "Nature Medicine",
  "mesh_terms": ["mRNA Vaccines", "COVID-19"],
  "article_types": ["review", "meta_analysis"],
  "publication_date": "2024-01-15",
  "pubmed_url": "https://pubmed.ncbi.nlm.nih.gov/38212345",
  "pdf_url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10812345/pdf",
  "source": "full_text",
  "chunks": [
    {
      "idx": 0,
      "text": "The development of mRNA vaccines...",
      "tokens": 512
    }
  ]
}

---

Features

• PMC full-text extraction when Open Access is available
• Abstract fallback for non-OA papers
• MeSH term indexing for structured biomedical tagging
• cl100k_base tokenization with 512 tokens per chunk
• Drop-in for vector databases ready for LangChain, LlamaIndex, Qdrant

---

What this actor does

This Actor extracts PubMed papers by search term, MeSH terms, article types, and date range. It prioritizes full-text extraction from PMC Open Access when available, falling back to abstracts for non-OA papers.

It supports structured biomedical queries with MeSH descriptors, article type filtering, and date range selection. Each paper includes bibliographic metadata and token-aware chunks.

---

Installation

npm install

Copy the environment file and add your Apify API token:

cp .env.example .env

Open .env and replace your_apify_token_here with your actual Apify API token. Get one free at console.apify.com.

---

Input

Field	Type	Description	Default
searchTerm	string	Search query (e.g. SARS-CoV-2 vaccines)	none
meshTerms	array	MeSH Descriptor headings	none
articleTypes	array	review, meta_analysis, clinical_trial	none
dateFrom	string	Start date (YYYY-MM-DD)	none
dateTo	string	End date (YYYY-MM-DD)	none
maxPapers	integer	Max papers to extract	100

---

Output

Each paper is a structured JSON record with token-aware chunks. Download as JSON, CSV, Excel, or HTML.

Field	Description
pmid	PubMed ID (primary identifier)
pmcid	PubMed Central ID (present for PMC OA papers)
doi	DOI (when provided)
title	Paper title
abstract	Abstract as returned by E-utilities
authors	Author display names, order preserved
journal	Journal title
mesh_terms	MeSH Descriptor headings assigned by NLM
article_types	PubMed publication types
publication_date	YYYY-MM-DD
pubmed_url	Link to the PubMed citation
pdf_url	PMC PDF link, when available
source	full_text or abstract (text origin)
chunks	Token-aware chunks for RAG
chunks[].idx	0-indexed position
chunks[].text	Chunk text
chunks[].tokens	Token count (≤ 512)

See sample-output.json for a full example.

---

Pricing

$0.02 per paper.

A run of 100 papers typically completes in 1 to 2 minutes. Pay only for what you extract.

---

Use Cases

• Biomedical research: Build RAG corpora from PubMed without XML parsing
• Clinical evidence: Extract systematic reviews and meta-analyses for evidence synthesis
• Pharma intelligence: Monitor drug-related publications for competitive analysis
• Embedding pipelines: Feed pre-chunked text into vector databases for biomedical Q&A

---

FAQ

What article types can I filter by? review, meta_analysis, clinical_trial, randomized_controlled_trial, observational_study, and more.

How does full-text vs abstract work? The Actor first attempts to extract full text from PMC Open Access. If the paper is not OA, it falls back to the abstract.

Do I need MeSH terms? No. MeSH terms are optional. You can search by free-text query alone.

---

Support

Open an issue in the Apify Console.

---

Related Resources

• PubMed E-utilities documentation
• Apify Client for JavaScript

---

Ready to start extracting?

Open the PubMed Scraper for RAG on Apify