Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-1923
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-10-19T18:14:38.244090Z
status: published
lens: g_space
sector: health
headline: Human vs Artificial Intelligence in Detecting Coronary Occlusions in ST-Elevation Myocardial Infarction: Retrospective Study
dek: Background Occlusion myocardial infarction (OMI) on the electrocardiogram (ECG) can present as ST-elevation myocardial infarction (STEMI), non-STEMI, ST depression, or other patterns (eg, de Winter T waves and hyperacute T waves), posing a diagnostic challenge. Even experienced physicians may overlook OMI on ECG, leading to delays in catheterization laboratory activation. Recently, smartphone apps using AI have been shown to exhibit promising accuracy in ECG interpretation in large datasets. Objective The object…
gain_title: On 33 diagnostically challenging ECGs, the AI app PMCardio detected every occlusion myocardial infarction (100% sensitivity and NPV), while three cardiologists reached 62% to 85% sensitivity.
problem_title: (none)
trace_subject: (none)
gain_reading: On 33 diagnostically challenging ECGs, the AI app PMCardio detected every occlusion myocardial infarction (100% sensitivity and NPV), while three cardiologists reached 62% to 85% sensitivity.
gain_evidence: AI analysis demonstrated a sensitivity and NPV of 100%
problem_reading: (none)
problem_evidence: (none)
quick_read: Researchers retrospectively compared an AI smartphone application, PMCardio, with three interventional cardiologists on 33 emergency department ECGs that had been hard to interpret at admission. The clinicians, blinded to final diagnosis, judged whether each ECG signaled an occlusion requiring immediate coronary angiography, and angiography served as the reference standard.

The result matters because overlooked occlusion myocardial infarction can delay catheterization laboratory activation. The AI missed none of the occlusions in this set, but the sample was small and retrospective, so the finding signals potential rather than proven benefit for real-time clinical decisions.
limitation: The study is retrospective and uses only 33 selected, diagnostically challenging ECGs, so the results may not generalize to routine emergency care.
tag: Evidence-backed gain
key_points: The study included 33 diagnostically challenging ECGs from patients admitted to the emergency department. | Of the 33 patients, 23 had acute coronary syndromes, and 13 of those (57%) were occlusion myocardial infarction. | The cardiologists' sensitivities were 62%, 85%, and 62%, with misclassification rates of 33%, 36%, and 42%. | The AI had 100% sensitivity and NPV, 60% specificity, 62% PPV, and a 24% misclassification rate.
rundown: Occlusion myocardial infarction can appear as STEMI, non-STEMI, ST depression, or patterns such as de Winter T waves and hyperacute T waves, which makes it a diagnostic challenge even for experienced physicians.

Three interventional cardiologists independently assessed each ECG while blinded to the final diagnosis. Their performance was benchmarked against coronary angiography, with any lesion of at least 90% stenosis counted as positive.

Cardiologists had specificities of 70%, 50%, and 55% and PPVs of 57%, 52%, and 47%. The AI's specificity of 60% and PPV of 62% were moderate, so its advantage lay mainly in sensitivity, NPV, and overall misclassification.
sources:
- peer_reviewed | JMIR Cardio | https://doi.org/10.2196/88399 | 2026-10-14
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

bb448756a6f5c33b303102979a2eecf0c6ecb591a2cf417d0cdc188bcda79ca9

previous

0000000000000000000000000000000000000000000000000000000000000000