Unlocking Fast and Accurate Healthcare Document Processing with AWS Generative AI

Revolutionizing Document Processing in Healthcare

As healthcare organizations face significant challenges in managing overwhelming volumes of documentation, Myriad Genetics has taken a major leap forward in streamlining their processes. By partnering with the AWS Generative AI Innovation Center, Myriad leveraged advanced technologies to transform the way they handle critical documents across Women’s Health, Oncology, and Mental Health sectors. The previous system, combining Amazon Textract and Comprehend, while effective, revealed operational bottlenecks and high expenses, prompting the need for a more robust solution.

Challenges with Manual Document Handling

The intricacies of medical documentation, combined with strict regulatory requirements, left Myriad struggling with a dependency on manual data extraction methods. Despite their existing solution achieving a commendable 94% classification accuracy, it incurred costs of $15,000 monthly for just one business unit, and bore an unacceptable classification latency of 8.5 minutes per document. To optimize processing efficiency, Myriad required a transformative approach to both classification and data extraction.

Introducing the GenAI IDP Accelerator

The launch of the open-source Generative AI Intelligent Document Processing (GenAI IDP) Accelerator presented a new avenue for Myriad. By utilizing Amazon Bedrock for processing and model optimization, this innovative approach could automate document classification, thereby cutting down processing time significantly and minimizing costs. The integration of generative AI allowed for enhanced context understanding, an essential capability for extracting nuanced medical information accurately.

Technical Implementation and Results

Through precise prompt engineering and model selection strategies, Myriad’s new document processing pipeline increased operational scalability significantly. With the GenAI IDP Accelerator, the company automated workflows that were once manually intensive, such as Key Information Extraction (KIE), leading to improved efficiencies and reduced operational burdens. The results have been substantial—not only in terms of cost savings but also in refining prior authorization workflows, ensuring swift and accurate patient care.

Future Perspectives on AI in Document Processing

As organizations continue to encounter massive document workloads, the insights gained from Myriad’s journey underscore the transformative potential of generative AI in document processing. The experience illustrates that with the right technology, healthcare providers can significantly enhance operational efficiencies and accuracy in patient care. The implications of this technology extend beyond just Myriad, hinting at a future where healthcare sectors globally can elevate their documentation handling processes with AI-driven innovations.