Forrest Health AI Flags 173 Lung Nodules in Epic in 6 Weeks

Why It Matters

Early detection of incidental lung nodules can dramatically improve outcomes, especially for rural patients who often lack consistent follow‑up. The AI‑navigator hybrid model demonstrates how EHR‑embedded intelligence can close care gaps and streamline physician workflows.

Key Takeaways

• AI flagged 173 lung nodules in six weeks, improving early detection
• Navigator validates alerts and forces physician acknowledgment via Epic pop‑up
• System routes patients to primary care or residency clinic for follow‑up
• Forrest Health among first adopters of Epic’s lung‑nodule workflow

Pulse Analysis

The integration of artificial intelligence directly into Epic’s electronic health record marks a pivotal shift in how hospitals surface hidden pathologies. By mining radiology reports for incidental lung nodules, the AI engine surfaces findings that would otherwise be buried in unrelated emergency‑room visits. This proactive approach aligns with broader industry moves toward predictive analytics, where algorithms sift through unstructured data to flag actionable insights before clinicians even see the report.

For rural health networks, the value proposition is especially compelling. Many patients in underserved areas lack a regular primary‑care relationship, leading to delayed diagnosis of potentially malignant nodules. Forrest Health’s navigator role bridges this gap: after the AI flags a lesion, a human reviewer confirms the finding, triggers an Epic pop‑up that obliges the treating physician to acknowledge the alert, and then coordinates referral to a primary‑care provider or the residency clinic. This hybrid model blends machine speed with human judgment, ensuring both accuracy and accountability while streamlining care pathways.

The broader implications extend beyond lung health. Successful deployment demonstrates that embedded AI can be scaled to other incidental findings—such as coronary calcium scores or renal masses—within the same EHR framework. As more health systems adopt similar workflows, the industry may see a new standard for automated, yet clinician‑validated, surveillance programs. However, challenges remain, including alert fatigue, data privacy, and the need for robust validation across diverse patient populations. Addressing these will be critical for realizing the full promise of AI‑enhanced EHRs in improving population health.