About
AstraZeneca’s Medical Affairs teams worked with a growing volume of complex, unstructured medical data — from clinical trial protocols and patient observations to meeting notes with healthcare professionals. Processing this information manually slowed down their daily work and delayed access to critical insights. We were brought in to design an AI solution that could turn raw, fragmented data into structured, compliant, and ready-to-use reports.
Challenge, approach, and impact
How to automate CRM record analysis without compromising accuracy or compliance?
The system needed to process three main types of data: Clinical trial documentation — protocols, treatment results, and patient safety observations. Real-world evidence — feedback from doctors, patient stories, and healthcare reports. Field notes — unstructured comments and insights recorded by Medical Science Liaisons (MSLs) during conversations with healthcare providers.
How we built
Testimonials
AI Engineer @ Relevant Software
Relevant Software
“Working on AstraZeneca’s AI analytics platform was an exciting opportunity to apply advanced NLP and machine learning to real-world pharmaceutical data. Automating insight extraction from clinical records and field notes helped significantly reduce manual processing time while ensuring compliance — it was fulfilling to see how AI directly enhanced the Medical Affairs team’s efficiency and decision quality.“
Team structure
Client team
Serhii
Director, Bio-pharma medical evidence
The client stakeholders at AstraZeneca were working closely with the team at Relevant Software
Agency team
AI Consultant
Production
