2×advanced liver disease diagnoses (5 more per 1,000)

Because of AI, primary care teams at Mayo Clinic diagnosed advanced liver disease in about 5 more patients per 1,000 adults who had a routine ECG.

Arandomized isolated pre-windowunreviewed

The Advance trial (Nature Medicine, Dec 2025) cluster-randomized 98 primary care teams; when an ECG-reading model flagged likely advanced chronic liver disease, clinicians were notified, and new diagnoses within 180 days were 1.0% vs 0.5% among 15,596 adults with ECGs (78 of 8,034 vs 35 of 7,562).

Sources

primaryNature Medicine: Detection of undiagnosed liver cirrhosis via AI-enabled electrocardiogram — a pragmatic, cluster-randomized clinical trialnature.com
primaryClinicalTrials.gov NCT05782283 (results posted)clinicaltrials.gov
secondarynpj Digital Medicine: Clinician engagement shapes the impact of AI-based ECG screening for chronic liver diseasedoi.org

Editor's notes

Grades and numbers from the PubMed abstract and the registry results record; the Nature Medicine full text is paywalled and was not read. verify: achieved dates are the registry's study start (2023-04-18) and completion (2025-09-01), not ECG enrollment dates from the methods section.

History

Oct 22, 2026Poster number and label set from the entry's own figures. Still unreviewed.
Oct 22, 2026The editor has not yet read the Nature Medicine paper or the registry record; published unreviewed.
Oct 22, 2026Added by weekly research run from the scout's registry lead; window pre (trial completed Sept 2025).

Cite this entry

Permalink https://becauseofai.org/entry/advance-ai-ecg-liver

“Because of AI, primary care teams at Mayo Clinic diagnosed advanced liver disease in about 5 more patients per 1,000 adults who had a routine ECG.” Because of AI, updated Oct 22, 2026. https://becauseofai.org/entry/advance-ai-ecg-liver

Caveat: The outcome is a diagnosis, not better health: whether earlier diagnosis changed treatment or survival was not measured, clinician follow-through was uneven, and the trial ran 2023–2025.
Achieved: Apr 2023 – Sep 2025
Published: Dec 17, 2025
Compared with: Care teams notified of AI-ECG results vs usual care, cluster-randomized 1:1 across 45 primary care practices; primary endpoint new advanced chronic liver disease diagnosis within 180 days
Effect: OR 2.09 (95% CI 1.22–3.55) for a new diagnosis (0.97% vs 0.46% (registry counts 78/8,034 vs 35/7,562): about 5 more diagnoses per 1,000 patients)
Reach: 15,596 adults, 98 care teams, 45 Mayo Clinic primary care practices
Conflicts: Mayo Clinic licensed the ECG model to Anumana; several authors may benefit through a related patent, and two sit on Anumana's scientific advisory board.
What would move the grade: Follow-up showing that the extra diagnoses led to surveillance, treatment, or fewer decompensations would turn a detection gain into a health gain.
Updated: Oct 22, 2026 unreviewed