How Operational AI Is Reducing Friction in Multi-Location Imaging Networks

Running 60 outpatient imaging centers across six states means competition is relentless. For Capitol Imaging Services, the center that contacts a patient first wins the appointment — and the revenue. That pressure drove CTO Tim Haley to stop evaluating operational AI and start deploying it in partnership with AbbaDox.

Their first-year results are a candid look at what real-world AI adoption actually feels like — including the parts that didn't go as planned.

Key Takeaways

• Modular deployment works better than a big-bang launch. AbbaDox built Capitol Imaging's AI stack one use case at a time, starting with scheduling in the markets where the need was sharpest. Once guardrails were set for the first region, subsequent markets moved faster.
• Voice AI requires cultural tuning. When AbbaDox deployed their voice scheduling agent (Abby) in southern locations using a voice profile built for a northern market, it flopped. Patients pushed back. Sentiment analysis flagged the issue and a voice adjustment produced an immediate improvement in success rates.
• The front-office numbers are significant. Across the network, 96% of inbound faxes are now processed automatically, with more than 17,000 imaging orders handled in a single week. AI-handled scheduling maintains a 94% patient satisfaction rate.
• Staff roles are evolving, not disappearing. As scheduling automation absorbs routine tasks, patient care coordinators are moving into clinical support roles rather than being displaced.

The next phase Haley is building toward takes the AI workflow from the first fax all the way to front-door arrival: insurance verification, patient forms via text, pre-arrival payment, and self-check-in.

🔗 Read the full article on Radiology Business