Mount Sinai researchers show that deep learning applied to ECGs can detect COPD early and accurately across diverse populations.
Key Details
- 1Study analyzed over 208,000 ECGs from more than 18,000 COPD cases and 49,000 matched controls spanning 2006-2023.
- 2A convolutional neural network achieved AUC 0.80 (internal), 0.82 (external validation), and 0.75 (UK Biobank).
- 3Model explainability linked predictions to clinically relevant ECG features (notably P-wave changes).
- 4Validation included distinct hospitals in New York City and patients from the UK Biobank.
- 5ECG-based AI tool uses standard 10-second, 12-lead data, enabling scalability and low cost for broad COPD screening.
Why It Matters
This work demonstrates that AI-driven ECG interpretation can provide scalable, non-invasive early detection for COPD, a major global health burden. Such AI models could expand access to screening in resource-limited settings, improving patient outcomes through earlier intervention.
Source
EurekAlert
Related News
•EurekAlert
AI-Enhanced Imaging Breaks Barriers in Complex Media for Biomedical Use
Researchers developed a physics-based machine learning imaging system that enhances visibility through complex media, holding promise for biomedical imaging.
•EurekAlert
AI Model Accelerates Accurate Dose Prediction for CyberKnife Radiotherapy
A physics- and spatially informed diffusion model accelerates Monte Carlo–level dose estimation for CyberKnife radiotherapy using planning CT and anatomical data.
•EurekAlert
New G-AUDIT Tool Detects Hidden Bias in Medical AI Training Data
Johns Hopkins and FDA researchers unveil G-AUDIT, a tool to identify hidden biases in medical AI datasets before model training.