As a drug moves through its life cycle from the research lab to commercialization and distribution, it picks up a lot of data along the way. Some of this data is structured, but close to 80% may be unstructured. Also, data collected outside the clinical trial ecosystem – such as from claims, wearables, and social media – is growing very fast. Such data is proving critical for managing the drug lifecycle effectively. Traditionally, we have to manually extract data to convert it to digital form first, before we can leverage the data for anything from analytics and reporting to signals and safety. However, AI and machine learning algorithms offer us new options to automate the digitalization of various forms of data and to perform predictive analytics and clustering on it. We are architecting an enterprise-class, scaleable, microservices based, serverless, cloud platform. This platform can ingest, process, transform, and report out valuable data-driven insights. As we implement the system, we have to make significant architectural decisions upfront, due to the complexity of the ecosystem in which the platform will operate. However, this is not only a technical challenge. There are big business challenges to go with it. In our approach to get the platform working with existing and often legacy technology platforms, we find ourselves addressing many business challenges as well. In this talk, I’ll share both the technology architecture and the business challenges we typically have to overcome. Here are some of the questions I’ll answer:
•How does the business perceive this new technology and what it takes to accept change?
•How to integrate the platform with existing business processes and systems?
•How do we manage structured and unstructured data?
•How do we architect for scalability and continuous adaptability?
The audience will walk away with great ideas on how to leverage AI to extract value from their data, as well as what they should not attempt to do.