Home › Cardiac arrest › AI ECG model excludes ACS after cardiac arrest

Cardiac arrest

AI ECG model excludes ACS after cardiac arrest

Artificial Intelligence-Assisted, Electrocardiogram-Derived, Sequential Rule-Based Decision Tree for Exclusion of Acute Coronary Syndrome as the Etiology of Out-of-Hospital Cardiac Arrest: A Single-Center Retrospective Cohort Study
Resuscitation · 2026-10 · Added 31 October 2026

A South Korean single-centre study built an AI-assisted ECG decision tree that classified about half of post-OHCA patients as 'ACS not likely' with no adjudicated ACS case misclassified in development or temporal validation.

Study type

Cohort

Design

Single-centre retrospective cohort with temporal validation

Participants

400 adults with post-ROSC OHCA (329 development / 71 validation)

Main finding

The tree labelled 52.0% of development and 45.1% of validation patients 'ACS not likely' with zero missed ACS cases; yield concentrated in non-shockable rhythms (64.8% of non-ACS cases excluded, AUROC 0.824)

Caveat

Single centre and retrospective, using a deep-learning ECG platform nowhere near ambulance deployment

Bottom line

Hypothesis-generating

Learning goal

Weigh the role of AI-assisted ECG interpretation in guiding post-ROSC destination and cath lab decisions.

More

• Topic: Cardiac arrest
• Daily scan: 31 October 2026

Related papers on Cardiac arrest

• Women still get less bystander CPR in public

  Resuscitation Plus · 31 Oct 2026 · Registry · Supports current practice

  A retrospective cohort of 10,445 public out-of-hospital cardiac arrests in North Carolina and Singapore found women were less likely to receive bystander CPR in both regions (aOR 0.87 and 0.78), with survival favouring women in North Carolina but men in Singapore.

• Real-time BVM feedback linked to better arrest ventilation

  Resuscitation · 31 Oct 2026 · Cohort · Hypothesis-generating

  A US prospective before/after study in Milwaukee found real-time BVM feedback (Real BVM Help) was associated with the share of OHCA ventilations meeting both rate and tidal-volume targets rising from 7.5% to 39.0%.

• ECG signs appear minutes before post-ROSC rearrest

  Resuscitation · 31 Oct 2026 · Cohort · Hypothesis-generating

  A US retrospective cohort of 1,298 post-ROSC OHCA patients found routine ECG features — wider QRS, lower heart rate, higher heart-rate variability — were associated with rearrest, with prediction best one minute before it (AUROC 0.74).

• AED coverage depends on how responders travel

  Frontiers in Public Health · 30 Oct 2026 · Modelling · Context only

  A Chinese spatial optimisation study of 17,842 OHCA dispatch records in Wuhan modelled AED coverage under different responder travel modes and found demand-weighted coverage ranged from 85% (driving) to 32% (walking), with universal walking coverage needing 521 extra AEDs versus 68 by car.