Award-winning AI streamlines medical fax workflows
(Photo by UTHealth Houston)
McWilliams School of Biomedical Informatics at UTHealth Houston was named a 2026 Amazon Web Services Champion Sept. 15 at the WS IMAGINE conference in Chicago. The school received the honor for a collaboration that turned artificial intelligence research into a clinical solution that has processed more than 1.8 million medical faxes and saved more than 73,000 hours of staff time.
The honor recognizes the school’s partnership with the UTHealth Houston Center for Digital Healthcare Innovation to develop iDFAX, a secure, HIPAA-compliant generative AI solution that streamlines the time-intensive process of handling medical faxes.
The teams brought together McWilliams School of Biomedical Informatic’s expertise in AI research and natural language processing with the Center for Digital Healthcare Innovation’s clinical and information technology expertise to move the technology from research into everyday clinical operations.
Medical faxes remain a significant part of healthcare operations. Before iDFAX, clinical staff spent approximately 2.5 minutes on each incoming fax, manually reviewing unstructured records, identifying patient information, and entering documents into Epic and OnBase.
Turning AI research into clinical impact
The collaboration began in June 2023, when the two teams set out to find a better way to manage the time-intensive process of handling medical faxes.
Researchers at McWilliams School of Biomedical Informatics led AI research, natural language processing algorithm development, and model optimization, while the Center for Digital Healthcare Innovation led clinical workflow architecture, integration with Epic and OnBase, enterprise scalability, and implementation across clinics.
The resulting iDFAX platform used Amazon Bedrock and other AWS services to securely automate portions of the workflow and link incoming documents with the appropriate electronic health record.
Since its implementation, iDFAX has processed more than 1.8 million faxes, including more than 100,000 per month, and saved more than 73,000 hours of staff time.
“Being named a 2026 AWS Champion is a powerful testament to the synergetic partnership between McWilliams School of Biomedical Informatics and the Center for Digital Healthcare Innovation,” said Xiaoqian Jiang, PhD, chair of the Department of Health Data Science and Artificial Intelligence and the Christopher Sarofim Family Professor in Biomedical Informatics and Bioengineering at McWilliams School of Biomedical Informatics. “This collaboration proves how academic medical centers can rapidly translate AI innovation directly into clinical impact.”
A collaborative approach to innovation
The AWS Champion recognition exhibited a two-year effort to move an AI concept from research and development into a production environment capable of supporting healthcare operations at scale.
The teams developed an enterprise pipeline using Amazon Bedrock, Amazon S3, Amazon DynamoDB, Amazon Simple Queue Service, and AWS Direct Connect, incorporating privacy safeguards and automated electronic health record linking throughout the process.
The collaboration allowed each team to contribute its distinct expertise. McWilliams School of Biomedical Informatics researchers focused on advancing and optimizing the AI behind the platform, while the Center for Digital Healthcare Innovation translated those capabilities into clinical workflows and integrated the technology into the systems used by care teams.
That integration of research and clinical operations was critical to moving iDFAX beyond a promising AI concept and into everyday use, said Babatope Fatuyi, chief health innovation officer at UTHealth Houston.
“Academic medical centers are full of promising AI. What is rare is AI that survives contact with the clinic,” Fatuyi said. “iDFAX worked because the McWilliams School of Biomedical Informatics and the Center for Digital Healthcare Innovation built it together, research and operations in the same room, with privacy and safety designed in from the first line of code. More than 1.8 million faxes processed and over 73,000 staff hours returned to patient care is what that partnership looks like in practice.”
The work has also been highlighted by AWS as an example of generative AI implementation in healthcare, including AWS Public Sector Blog features published in March 2025 and July 2026.
