Applying Natural Language Processing to Automate Pharmacy Processes

A major US-based pharmacy chain needed to reliably automate the process of converting doctors’ prescriptions to standardized printed labels that are attached to customers’ prescription medicines. The doctors’ notes, written in expert English abundant with idioms and conventions used in medical practice, amounted to more than 100 million unique strings of text to be processed as a data stream with minimal latency at tens of transactions per second.

The requirements were to accurately determine the true intent of each prescription and translate its often hastily written and confusing language into clear textual and visual instructions for the consumer. To guarantee absolute patient safety, a custom-built knowledge system had to be deployed for validating dosage recommendations, checking for potential cross-interactions as well as minimizing potential risks arising from common misunderstandings. A warning and feedback workflow with human experts in the loop was implemented to resolve ambiguous cases, resulting in 100% accuracy for all accepted prescriptions.

The solution Wolfram Consulting Group delivered to the pharmacy chain was architected on the basis of Wolfram|Alpha, which provided a domain-specific knowledgebase and a natural language processing (NLP) framework that could be extended to master this unusual and complex task. Wolfram Consulting also curated additional proprietary data and incorporated it into the business process. This solution was delivered through an API that could be easily integrated into the client’s existing infrastructure. The solution greatly reduced the cost and time needed to generate prescription labels while ensuring the client’s highest standards for patient safety.