The prediction of cardiac abnormality and enhancement in minority class accuracy from imbalanced ECG signals using modified deep neural network models.
Hari Mohan RaiKalyan ChatterjeeSerhii DashkevychPublished in: Comput. Biol. Medicine (2022)
Keyphrases
- ecg signals
- neural network model
- minority class
- class imbalanced
- class imbalance
- prediction accuracy
- mit bih arrhythmia database
- highly skewed
- imbalanced data
- imbalanced datasets
- class distribution
- imbalanced data sets
- neural network
- nearest neighbour
- ensemble methods
- cost sensitive learning
- rare events
- imbalanced class distribution
- artificial neural networks
- multilayer perceptron
- classification error
- single class
- original data
- decision boundary
- multi class
- sampling methods
- data sets
- training data
- classification accuracy
- training dataset
- prediction model
- cost sensitive
- active learning
- feature vectors
- high dimensionality
- bp neural network
- feature selection