How Inova Modernized Its Data Architecture in 6 Months to Deploy 70+ AI Applications
Inova, a nonprofit health system serves Northern Virginia
and Washington, D.C. Every year they handle over 4 million patient visits
across hospitals, primary care practices and emergency centers. The
organization has checked over 300 health applications for AI and has put in
place more than 70 different AI apps.
The challenge was not finding AI tools. The challenge was
getting the data foundation ready for them.
The 4-Year Roadmap Compressed to 6 Months
To support its AI goals Inova recently worked with data
integration company Fivetran on a data modernization effort. The health system
sped up a four-year data modernization plan to six months to become AI-ready
and move from testing to live production.
By centralizing and standardizing data movements with
Fivetran Inova has created a foundation that supports AI agents, copilots and
intelligent applications built on trusted governed data.
The Tech Stack
Inova combined technology platforms:
1.
Fivetran. Data integration
and ingestion
2.
DBT Labs. Shared data
infrastructure layer for quality and governance
3.
Astronomer. Orchestration
4.
Microsoft Fabric. Analytics
platform
5.
Databricks. Analytics and
AI workloads
The health system manages over 1,000 technology contracts
spanning clinical, imaging, laboratory, financial and SaaS solutions. Each with
its own data models, APIs and limitations.
What Inova Is Building
1. Ambient AI Scribes
than 1,000 providers at Inova use Abridge as an ambient AI
scribe. The health system is also testing AI scribes in specialty inpatient
areas with Abridge.
2. Agentic AI for Patient Referrals
Inova uses AI to refer patients to providers outside the
Inova network. These AI voice agents send data back to Inova so clinicians can
assist community partners if patients develop complications.
3. AI-Driven Scheduling
AI tools search provider schedules for slots and deliver
data to access teams for proactive patient booking.
4. Administrative Burden Reduction
Inova is working on a project with Elevance Health and
Abridge to reduce burden in precharting workflows.
5. Inovas AI Distribution Portal
The health system developed its branded digital AI
distribution portal, where staff can access AI agents from multiple vendor
platforms.
The Philosophy Behind It
Jon McManus, Inovas Chief Data and AI Officer said the
health system is not trying to compete with third-party app developers. It is
simply incorporating AI capabilities into workflows with AI.
Sometimes they want to control the UI, the look and feel.
Inova uses the models from OpenAI and Anthropic.
AI Literacy as a Priority
McManus compares navigating AI features to using a
wayfinding app.
You need a map; you need to know where the bathrooms are;
you need to know there's an app to get reservations on rides. In healthcare
they have to think in a way.
He added that it is hard for healthcare professionals to
become AI literate if they must navigate different expressions of AI.
Avoiding Vendor Lock-In
Inova carefully selects partners to avoid being locked into
vendor contracts.
They pick their partners carefully. They. Ensure that the
underlying techniques and technologies they serve don't make them too
single-threaded.
Having a diversity of products in the tech stack gives the
health system flexibility.
Governance and Guardrails
Every AI application at Inova is held to the validation bar.
Every pipeline, every agent every model is judged against
whether it makes care safer equitable or more accessible.
When patients enter an exam room providers discuss that the
visit will be recorded using AI.
Governance doesn't have to be an obstacle, to AI.
They've deeply woven those frameworks together so that
they're implementing solutions and governing in a way that maintains
reliability, performance and safety.
