In Pharma, a new drug goes through rigorous testing before it is released commercially. Even so, adverse drug reactions often are found after it’s released.

Such adverse reactions are reported through different channels and different formats mostly unstructured. Currently humans manually document and categorize the severity of the reaction so that appropriate action can be taken.

Some of these adverse reactions could be fatal and require immediate action. However, the volume of information to filter through makes scaling with human labor difficult. Artificial intelligence offers an alternate.

We worked with one of the top five Pharma companies in the world to automate the process by using intelligent document processing techniques.

In the presentation, we will share the architecture of the platform and show how we were able to achieve over 80% accuracy. After the system was rolled out, the human experts just have to validate cases that could not be handled by the system, freeing up much of their time to focus on value-added work.

Learning Objectives:

• How AI is used in health for efficacy?
• What are the key considerations to implement an AI system
• |How to experiment with AI in the context of your business?