Myriad Genetics adopts AWS GenAI Intelligent Document Processing Accelerator tool to fix costly, slow medical document workflows.

Myriad Genetics has rebuilt its medical document pipeline using AWS’s GenAI Intelligent Document Processing Accelerator, cutting costs and accelerating turnaround across its clinical divisions. The company needed a faster and more accurate system as rising document volumes pushed prior authorization workflows to a breaking point. Their older pipeline relied on pattern-based classification and manual extraction. As a result, it is high latency, rising operational costs, and heavy staffing requirements.

With AWS Bedrock and the GenAI IDP Accelerator, Myriad replaced rigid rules with models that understand medical context. Amazon Nova Pro now classifies documents through semantic reasoning instead of format matching. Engineers refined prompts using negative guidance and format cues. In which this enables the model to distinguish between similar document types. This shift raised accuracy from 94% to 98%, cut classification costs by 77%, and reduced processing time from 8.5 minutes to 1.5 minutes per document. The accelerator’s no-code configuration layer also let Myriad tune prompts. As well as attributes, and logic without rebuilding the system.

Information extraction, once fully manual, was automated using a multimodal approach that combined Textract’s forms and tables with Amazon Nova Premier’s reasoning capabilities. Few-shot visual examples, chain-of-thought instructions, and prompt caching helped the model resolve ambiguous checkboxes. As well as overlapping marks, and nuanced clinical phrases. The new extraction system reached 90% accuracy. This matches human specialists while processing documents in about 1.3 minutes each.

These improvements removed key bottlenecks in Myriad’s prior authorization pipeline. Faster classification shaved two minutes off each case. As a result, saves 300 hours monthly in Women’s Health alone. The company expects over $132,000 in annual savings from classification efficiencies and major productivity gains across Oncology and Mental Health. With scalable serverless infrastructure and AI-driven understanding, Myriad now moves documents through clinical workflows with speed, precision, and far lower cost.