ECG Arrhythmia Classification Using Transfer Learning from 2- Dimensional Deep CNN Features.
Milad SalemShayan TaheriJiann-Shiun YuanPublished in: BioCAS (2018)
Keyphrases
- transfer learning
- classification accuracy
- ecg signals
- feature extraction
- feature vectors
- feature set
- feature space
- text classification
- machine learning
- machine learning algorithms
- extracted features
- class labels
- heart rate variability
- mit bih arrhythmia database
- cross domain learning
- learning tasks
- knowledge transfer
- support vector
- cross domain
- beat classification
- support vector machine
- labeled data
- active learning
- convolutional neural network
- domain adaptation
- semi supervised learning
- cross domain text classification
- pairwise
- training set
- feature selection
- multi task
- training examples
- unsupervised learning
- knn
- learning process
- reinforcement learning