Baptist Health Taps Eon AI to Catch the Findings 70% of Patients Miss

Baptist Health Herbert Wertheim Cancer Institute has begun using Eon, an artificial intelligence platform that reads radiology reports alongside the electronic medical record, to catch incidental findings before they slip past follow-up care. The tool is live for incidental lung and pancreas findings today, with breast, thyroid, kidney and liver on the roadmap, and it slots the work into a triage flow that flags faster-moving cases while it monitors slower ones through the software.

Dr. Leonard Kalman, acting system chief executive of Baptist Health Cancer Care and acting executive medical director of the Cancer Institute, framed the move as a way to identify patients who may benefit from follow-up care sooner and more consistently. It supports our commitment to early detection, providing coordinated, timely care and helping patients receive the right care at the right time, Kalman said.

How Eon Reads a Radiology Report

Eon’s software is built around a computational linguistics model that scans the free-text body of radiology reports rather than the images themselves. As each radiologist-dictated report is signed, it is delivered to Eon over a health data feed and analyzed for findings that look incidental, meaning the patient was scanned for one reason and something else turned up.

The platform then pulls from the electronic medical record through a FHIR integration to add smoking history, prior cancer history, primary care provider and preferred language, giving the case a complete clinical profile before it surfaces to a human reviewer.

Eon says its platform supports lung, pancreas, breast, thyroid, kidney, liver and cardiovascular conditions, and it pairs the AI with a team of expert clinical navigators who handle lower-risk cases so the in-house team can focus on complex ones. The tool organizes findings by level of concern inside the software, so care teams can decide what to escalate and what to monitor.

The 70% Follow-Up Gap

Incidental findings are not rare. Eon, citing published research, says up to 30% of imaging studies include one, and the Mid-South Baptist Memorial system estimated a much lower 0.362% rate of incidental pulmonary nodules before it ran its own prevalence study against Eon. When Eon analyzed 2,000 of Baptist Memorial’s reports, the rate came back at 2.28%.

The cost of missing them is the part that doesn’t show up on a billing sheet. Published research cited by Eon puts the share of patients who never receive appropriate follow-up at roughly 70%. Health systems using Eon, the company says, report follow-up completion rates averaging over 75% for patients with incidental pulmonary nodules, against a national average of approximately 30%.

That gap matters most where time is the variable that decides outcomes. A pancreatic cyst found early is a monitored patient; a Stage IV lung cancer diagnosed late is the case that prompted Eon’s founder, Dr. Akrum Al-Zubaidi, to build the company in the first place.

For a system that bills itself as a community-based cancer center serving patients across multiple states, the math behind that gap gets loud fast.

Why the Manual Process Buckles

Baptist Memorial CEO Parker Harris described a workflow before Eon in which staff were pulling information from multiple systems, validating and re-validating patients, and working through lists again and again just to make sure no one was missed. Staff were using three, four, sometimes five different systems; however, none of them worked well enough for the mission the organization was trying to accomplish, Harris said.

Every hospital had its own process, and because of that, there was no consistent way to ensure patients were identified and moved into the right pathways, he said. Our teams were doing double work, sometimes triple work, trying to keep up with the volume, and we knew it wasn’t sustainable.

The strain scales with the footprint. As a community-based cancer center, our mission is to ensure that a patient in Booneville, Mississippi, has the same opportunity for timely care as a patient in Memphis, Tennessee, or Jonesboro, Arkansas, Harris said. A single patient missed is a downstream diagnosis that arrives at a later, more expensive stage.

How a Finding Gets Triaged Inside the Tool

Baptist Health’s deployment is not a firehose of AI alerts dumped on clinicians. The tool organizes findings by level of concern so care teams can determine the most appropriate next steps, and findings that may require faster attention are flagged through established processes at the Institute. Other findings can be monitored and managed with appropriate follow-up through the software platform.

Patients and their ordering physician are informed of the findings when the AI surfaces them, and the ordering physician may choose to follow up directly based on the patient’s individual needs. Eon’s platform pairs the AI identification with evidence-based follow-up recommendations tailored to the patient’s imaging findings and risk factors, and the care plans adapt over time as new imaging or clinical events occur. Guideline anchors include the Fleischner Society criteria for lung nodules and ACC/AHA guidance for aortic aneurysms and coronary artery calcification.

What the Track Record Looks Like Elsewhere

Eon says it has surpassed half a billion radiology reports analyzed across more than 75 health systems and over 1,000 facilities, with the platform identifying over 2 million patients with incidental findings documented in radiology reports. The headline figure inside that footprint is the follow-up completion rate.

Baptist Memorial, a separate Baptist-affiliated system in the Mid-South, published results in December 2025 that read like a case for the deployment model Baptist Health Herbert Wertheim is now adopting. The system identified over twice as many incidental pulmonary nodules after Eon went live, half of which were in a higher-risk group needing timely clinical review, and it booked $3.44 million in incremental direct margin from downstream services.

Other published numbers from Eon’s customer base:

• RWJBarnabas Health identified 90 pancreatic cancers, with 55% classified as earlier-stage and non-metastatic, higher than historical controls, per a quote from Associate Chief Surgical Officer Russell Langan on Eon’s site.
• Our Lady of the Lake went from managing 200 lung screening patients per year to over 5,000 cases annually inside its screening and incidental system, per thoracic surgeon Dr. Emily Cassidy on Eon’s site.
• Lifepoint Health, which started its Eon program in 2018 focused on incidental pulmonary nodules, has scaled to 10 disease states and conditions including most recently breast, per Vice President of Clinical Technology and Systems Bartholomew Daugherty.

Health systems using Eon report follow-up completion rates on average of over 75% for patients with incidental pulmonary nodules, compared to the national average of approximately 30%.

Where the AI Stops and the Clinician Starts

Baptist Health is careful to frame Eon as support, not a substitute. The technology supports physicians and care teams by helping them review information more efficiently; it does not replace clinical judgement or decision making, the announcement states, and the use of AI-enabled technologies reflects Baptist Health’s commitment to bringing innovative tools into patient care while ensuring physicians and care teams remain at the center of medical decision-making.

The same guardrail is built into Eon’s product page. Unlike EHR modules or image-analysis tools that focus primarily on detection, Eon is purpose-built to manage the full lifecycle of an incidental finding, combining AI-driven identification, smart validation, evidence-based care plans and long-term follow-up tracking.

Kalman echoed that line in the announcement:

This technology is set to advance our capabilities by helping us identify patients who may benefit from follow-up care sooner and more consistently. It supports our commitment to early detection, providing coordinated, timely care and helping patients receive the right care at the right time.

What’s Next at the Cancer Institute

Baptist Health is starting with incidental lung and pancreas findings, where the published completion-rate gap is widest and the stage-shift case is strongest, and plans to expand into breast, thyroid, kidney and liver findings in future phases. The platform’s next-generation release, Eon 5.0, is rolling out to existing clients and preparing for broader commercial availability.

That expansion will put more disease states on the same lifecycle platform that Eon describes as evidence-based follow-up recommendations, instantly delivered, with care plans that adapt to new imaging or clinical events rather than closing when the radiology report is signed. For a cancer institute whose mission starts at early detection, the closing of the follow-up loop is the part that turns an AI pilot into a measurable shift in who gets treated at Stage I instead of Stage IV. Baptist Health Herbert Wertheim Cancer Institute plans to expand its incidental findings program beyond lung and pancreas into breast, thyroid, kidney and liver in future phases, according to the announcement.

Frequently Asked Questions

What does Eon’s AI actually do in a radiology report?

It scans the free-text body of the radiology report using a computational linguistics model, then pulls risk-factor data such as smoking history, prior cancer history, primary care provider and preferred language from the electronic medical record through a FHIR integration. The platform is described as purpose-built to manage the full lifecycle of an incidental finding, from identification through long-term follow-up tracking.

What counts as an incidental finding?

An incidental finding is something unexpected that turns up on an imaging exam performed for another reason, according to Baptist Health’s announcement. Common examples include lung nodules, pancreatic lesions, thyroid nodules and coronary artery calcifications.

Does the AI notify the patient directly?

No. Patients and their ordering physician are informed of the findings, and the ordering physician may choose to follow up directly based on the patient’s individual needs. Eon pairs the AI with a team of expert clinical navigators who handle lower-risk cases so in-house teams can focus on complex ones.

Why is the follow-up gap so large in the first place?

According to published research cited by Eon, up to 30% of imaging studies include an incidental finding, yet roughly 70% of those findings are not followed up on appropriately. Eon attributes the gap to manual processes, spreadsheets and EHR inboxes that make it difficult to consistently identify findings across radiology reports, apply guideline-based follow-up recommendations and ensure patients return for care.

How is this different from AI that reads the images themselves?

Eon is built to read the report text and manage follow-up, not to flag image pixels the way computer-vision tools do. Baptist Health’s announcement frames the tool as a way to help clinicians review information more efficiently, with physicians and care teams remaining at the center of medical decision-making.

Disclaimer: This article is for informational purposes only and does not constitute medical advice. Consult a qualified healthcare professional for medical decisions. Figures cited are accurate as of publication and may be updated by the organizations named.