A Deep Learning Algorithm for Detecting Acute Pericarditis by Electrocardiogram
Problem
Acute pericarditis is easily confused with STEMI in emergency-department chest-pain patients, risking misdiagnosis from 12-lead ECGs.
Why it matters
Misdiagnosing pericarditis as heart attack could lead to harmful, unnecessary cardiac interventions.
What the AI did
A deep learning model was trained on raw 12-lead ECG signals to detect acute pericarditis, tested in a human-machine competition and against a commercial automatic algorithm.
Results
not stated in source (abstract truncated before performance figures)
Impact
Evidence supports development and internal validation of an ECG-based pericarditis detection model, not clinical deployment.
Record
| Venue | Journal of Personalized Medicine |
|---|---|
| Published | 2022 |
| Work type | article |
| Authors | Yu‐Lan Liu, Chin‐Sheng Lin, Cheng‐Chung Cheng, Chin S. Lin |
| Citations | 24 (OpenAlex, at import time) |
| References | 52 |
| Model or method | pericarditis-DLM (deep learning model); architecture not stated in source |
| Organization | not stated in source |
| Primary source | https://doi.org/10.3390/jpm12071150 |
| Open access | gold |
| Keywords | deep learning, ECG, pericarditis, STEMI, emergency medicine |
Importance rating
real but narrower advance
Pericarditis/STEMI confusion in ED chest-pain patients risks harmful unnecessary cardiac interventions; meaningful diagnostic safety benefit, but abstract states no scale of affected patients.