🔬 Clinical Evidence Research Assistant

AI-powered evidence research tool for Medical Affairs and HEOR teams. Enter any drug and indication — get an instant structured evidence summary pulling live data from ClinicalTrials.gov and PubMed, synthesised by Claude AI.

---

🎯 What It Does

A Medical Affairs professional manually searching for clinical evidence spends 2-3 hours per drug. This tool does it in 30 seconds.

Enter a drug + indication → get:

• Active clinical trials with endpoints, sponsors, and status
• Key published literature from PubMed
• AI-synthesised evidence summary with Medical Affairs implications
• Downloadable report

---

🛠️ Built With

Tool	Purpose
Python	Core language
ClinicalTrials.gov API v2	Live trial data
PubMed / Biopython Entrez	Published literature
Claude API (Anthropic)	AI evidence synthesis
Streamlit	Web interface

---

📁 Project Structure

├── clinical_trials.py    # Fetches live data from ClinicalTrials.gov
├── pubmed_search.py      # Searches PubMed for published literature
├── llm_synthesis.py      # Claude AI evidence synthesis engine
├── report_generator.py   # Assembles structured evidence report
├── streamlit_app.py      # Streamlit web interface
├── requirements.txt      # Python dependencies
└── .env.example          # API key template

---

🚀 Quick Start

1. Clone the repo

git clone https://github.com/LifeSciForge/Clinical_Evidence_Assistant.git
cd Clinical_Evidence_Assistant

2. Create virtual environment

python3 -m venv venv
source venv/bin/activate

3. Install dependencies

pip install -r requirements.txt

4. Add your API key

cp .env.example .env
# Edit .env and add your Anthropic API key

5. Run the app

streamlit run streamlit_app.py

Open browser at http://localhost:8501

---

💡 Example Searches

Drug	Indication	What You Get
pembrolizumab	NSCLC	15 trials, 10 papers, full synthesis
semaglutide	Type 2 Diabetes	Active trials, GLP-1 literature
lecanemab	Alzheimers	Phase 3 data, amyloid evidence
trastuzumab	breast cancer	HER2 trial landscape

---

🎯 Target Users

• Medical Affairs — evidence gap analysis, MSL briefings
• HEOR — payer-relevant endpoint identification
• Clinical Development — trial landscape mapping
• Pharma Strategy — rapid evidence scanning

---

🔑 API Key Setup

This app uses Claude AI for evidence synthesis.

1. Get your free API key at console.anthropic.com
2. Copy .env.example to .env
3. Add your key: ANTHROPIC_API_KEY=your_key_here

The app runs in placeholder mode without an API key — all trial and literature data still loads from live sources.

---

👤 Author

Pranjal Das
AI & Automation for Life Sciences
github.com/LifeSciForge

---

Data sources: ClinicalTrials.gov · PubMed · Claude AI (Anthropic)