Excelra logo.
MakerTech green 3D crop.

AI-powered clinical data extraction.

Problem

Excelra faced a significant challenge: extracting structured data from complex clinical trial documents. Their highly skilled team was spending countless hours manually processing documents, particularly struggling with unlabeled charts and graphs across diverse disease areas like Type 2 Diabetes, Rheumatoid Arthritis, and Non-Small Cell Lung Cancer. They needed a solution that could maintain accuracy while dramatically improving efficiency.

Solution

We developed an AI-powered data extraction system that could handle the complexity of clinical trial documents. The solution combines advanced AI models with optical character recognition to identify and extract data from complex charts and tables, even when data points lack explicit labels. Built on Azure App Service for scalability and using Azure Functions for workflow automation, the system processes documents across multiple disease areas while maintaining strict accuracy standards.

MakerTech purple 3D crop 1.

Results

The AI solution transformed Excelra's document processing workflow. The system achieved over 75% accuracy for targeted variables and 100% accuracy for mandatory variables – critical benchmarks in clinical data processing. Built to handle diverse disease areas, it freed skilled employees from routine data extraction while maintaining the precision needed for medical research. Azure's managed services kept operational costs predictable, creating a sustainable foundation for future growth.

MakerTech purple 3D crop 3.

Technologies

OpenAI logo.
Microsoft Azure logo.
Python logo.
Swagger logo.