Clinical trials generate vast amounts of complex, unstructured data. Cleaning, organizing, and managing this data always proves challenging to clinical trial organizations. Also, in clinical trials, it is very important to maintain a compliant record of data for regulatory and reporting purposes.
Our client, a global leader in contract research services, healthcare data science and technology solutions, wanted to design a fast, efficient, and compliant electronic clinical trial documentation
process. We were proud to partner with the trailblazer in global healthcare industry to infuse AI in clinical trial processes. Our Decision Sciences Lab DL Libraries and Text Analytics Framework were used to build taxonomies and Deep Learning (DL) Models to classify and digitize the various trial documents with least manual intervention.
Key Features of Solution
- Constructed AI as a black box to extract information from the documents (Digitized & Scanned)
- Built scalable microservices to accommodate different file types, 26 of them (pdf, doc, pptx, xlsx etc.)
- Pipelined all the microservices to run as an application
- The documents go through a pipeline of six microservices for text/image classification
Top Business Benefits
Leveraging AI and ML, we helped the client digitize highly complex clinical trial documents, helping the client achieve the below listed benefits.
- 20 minutes of classification time is reduced to less than a minute
- ML based document classification embedded inside the custom application, which can be leveraged for other BU.
- Less human intervention and more consistent classification lead to 70% cost reduction over time