The common, classic fax is a fascinating case study in the difficulties of automation. Watch this video as Jess Czelusniak, Associate Vice President of Applications and CNIO at VHC Health, and Tim Hoskins, Vice President of Solutions Architecture at Vyne Medical, discuss the challenges of classifying and sorting incoming fax documents. Through machine learning, automation, and staff training, they reduced the process from 10-15 minutes (described in a 29-page document) to less than 3 minutes per fax, and are reducing the burden even further.

The hospital in this case study has 537 beds and is located in Arlington, VA. It is a level 2 trauma center that sees 60,000-65,000 ER patients per year.

Czelusniak describes how the completely manual filing system they had was overwhelming as the volume of faxes burgeoned. Government regulations requiring more information sharing, in the absence of digital connectivity, led to the flood of faxes.

Two tasks must be accomplished for each incoming fax: inserting it in the right place in the right patient record, and notifying the right doctors. Both tasks were threatened under the old system.

For each fax, a staff person had to extract the information needed to file the fax properly. The manual process used a document management system before the fax went into the Epic EHR, and during a switch between systems, the staff person had to wait for several minutes in the middle of the transfer. Naturally, a staff person would often forget to finish the process, or forget which patient the fax covered.

Delays in getting information to doctors was common—sometimes up to three weeks. Often the patient saw the document in the MyChart portal before the doctor received it.

When Vyne Medical was brought in to apply their machine learning and automation technology, a thorough experimentation period ensued. A hundred different document types had to be recognized. A wide range of staff were brought into the effort, and that 29-page document turned out to be invaluable to explain the workflow needing automation.

Some documents still appear in handwriting, but Hoskins says they have improved their system to point where it recognizes handwritten text with 82-85% accuracy.

Since people have a greater skill for discernment, human in the loop verification might be necessary in the process to ensure accuracy and provide quicker patient care. The experience at Vyne Medical is that the filing of some simple types of documents, such IDs, can be fully automated, but a clinical document with serious impact on a patient calls for human review. But their system makes it easy to view the document on the screen, and automatically extracts key information such as patient name and date of birth for filing.

When Vyne Medical took on the surgical department, they discovered that the department’s biggest challenge was prioritizing surgeries. Each surgical document contained a surgical request date, and it took the development team only ten minutes to create the tool that extracted it automatically and allowed them to sort documents by that date. This was a game changer for the surgical department.

Even though the benefits of the new system were obvious, there was angst among users about the change. Training is still required, and turnover in those staff positions is high. Czelusniak was patient and kept reminding the staff how far they had come.

During this automation effort, Czelusniak found a department still receiving paper faxes, but even that process was fixed by Vyne Medical.

In general, Hoskins says, clerical positions are ripe for automation. Hiring and retention are hard in data entry jobs. Automation is a complex technical project, but saves money in the end.

The next steps at VHC Health are to automate more documents, add more departments, and allow doctors to sign faxed forms without having to print and rescan them. Vyne Medical plans to implement generative AI in 2026 to allow more advanced automation.  Check out our interview with VHC Health and Vyne Medical to learn more.

Learn more about VHC Health: https://www.vhchealth.org/

Learn more about Vyne Medical: https://vynemedical.com/

Listen and subscribe to the Healthcare IT Today Interviews Podcast to hear all the latest insights from experts in healthcare IT.

And for an exclusive look at our top stories, subscribe to our newsletter and YouTube.

Tell us what you think. Contact us here or on Twitter at @hcitoday. And if you’re interested in advertising with us, check out our various advertising packages and request our Media Kit.

Vyne Medical is a proud sponsor of Healthcare Scene.