Artificial Intelligence-Enabled Electrocardiogram Improves the Diagnosis and Prediction of Mortality in Patients With Pulmonary Hypertension
Problem
Elevated pulmonary artery pressure in pulmonary hypertension requires early detection from ECG to enable prompt diagnosis and treatment.
Why it matters
Pulmonary hypertension is disabling and life-threatening, and late detection risks detrimental consequences.
What the AI did
A deep-learning neural network model analyzed ECGs to identify patients with elevated pulmonary artery pressure and related prognostic implications.
Results
AUC 0.88 (sensitivity 81.0%, specificity 79.6%) for detecting ePAP; diagnostic odds ratio >8 for most factors; external validation on a Japan cohort
Impact
AI-ECG detected ePAP consistently across age, sex, and comorbidities, with external cohort validation.
Record
| Venue | JACC Asia |
|---|---|
| Published | 2022 |
| Work type | article |
| Authors | Chih‐Min Liu, Edward S.C. Shih, Jhih-Yu Chen, Chih-Han Huang, I‐Chien Wu, Pei‐Fen Chen, Satoshi Higa, Nobumori Yagi et al. |
| Citations | 46 (OpenAlex, at import time) |
| References | 34 |
| Model or method | deep learning neural network |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1016/j.jacasi.2022.02.008 |
| Open access | gold |
| Keywords | electrocardiogram, pulmonary hypertension, deep learning, diagnosis, mortality prediction |
Importance rating
significant — thousands to millions affected
Abstract calls pulmonary hypertension disabling and life-threatening; AI-ECG detects ePAP (AUC 0.88) and predicts cardiovascular mortality (HR 3.69), validated in external Japan cohort.