Biometric contrastive learning for data-efficient deep learning from electrocardiographic images.
Veer SanghaAkshay KhunteGregory HolsteBobak J MortazaviZhangyang WangEvangelos K OikonomouRohan KheraPublished in: Journal of the American Medical Informatics Association : JAMIA (2024)
A pretraining strategy that leverages biometric signatures of different ECGs from the same patient enhances the efficiency of developing AI models for ECG images. This represents a major advance in detecting disorders from ECG images with limited labeled data.
Keyphrases
- deep learning
- artificial intelligence
- convolutional neural network
- big data
- electronic health record
- machine learning
- optical coherence tomography
- heart rate variability
- heart rate
- case report
- left ventricular
- blood pressure
- gene expression
- heart failure
- dna methylation
- mitral valve
- left atrial
- data analysis
- positron emission tomography