I co-founded FormIQ. From a fax queue to a hospital pilot.
Healthcare AI for document intake at health systems. We ran a two-month orthopedics pilot at Yale New Haven Health with 10 to 12 administrative staff: about 2,000 documents and roughly 300 staff hours saved, measured.
- Documents
- About 2,000
- Staff hours saved
- ~300, measured
- Users
- 10 to 12 staff
- Reached
- Final-stage procurement
Following the document through the work
A document moved from an incoming fax to the staff who routed it, to an administrator, to a physician, and back out by fax. Arrival to return took one to two weeks. We mapped those steps with the staff doing them rather than treating extraction as the whole problem.
What I owned
I led product and technical development, mapped intake, routing and human review with the administrative staff, and led most of the executive presentation. My cofounder brought the idea and the relationships, including the introduction to the executive team.
Designing the pilot around the constraint
We could not get access to the Epic instance on our timeline, so we designed a pilot that worked without it: templates for the most common documents, patient data still entered by hand, and every field editable. Similar documents were grouped so physicians could review them together.
How far it went
We cleared security review and clinical workflow review and reached final-stage procurement. The health system chose an established vendor on maturity, then invited us to continue through its Center for Health Care Innovation.
The document-processing method is covered by an NDA with Yale. This page describes the work and the workflow, not how the system worked.
The story behind it · 6 minWe lost on vendor maturityWe built for the workflow. The buyer also needed to know whether we could support it for years.