No abstract

No abstract

Background and Objective: The interpretation of the electrocardiogram (ECG) gives clinical information and helps in the assessing of the heart function. There are distinct ECG patterns associated with a specific class of arrythmia. The convolutional neural network is actually one of the most applied deep learning algorithms in ECG processing. However, with deep learning models there are many more hyperparameters to tune. Selecting an optimum or best hyperparameter for the convolutional neural network algorithm is challenging. Often, we end up tuning the model manually with different possible range of values until a best fit model is obtained. Automatic hyperparameters tuning using Bayesian optimization (BO) and evolutionary algorithms brings a solution to the harbor manual configuration. Methods: In this paper, we propose to optimize the Recurrent one Dimensional Convolutional Neural Network model (R-1D-CNN) with two levels. At the first level, a residual convolutional layer and one-dimensional convolutional neural layers are trained to learn patientspecific ECG features over which the multilayer perceptron layers can learn to produce the final class vectors of each input. This level is manual and aims to lower the search space. The second level is automatic and based on proposed algorithm based BO. Results: Our proposed optimized R-1D-CNN architecture is evaluated on two publicly available ECG Datasets. The experimental results display that the proposed algorithm based BO achieves an optimum rate of 99.95%, while the baseline model achieves 99.70% for the MIT-BIH database. Moreover, experiments demonstrate that the proposed architecture fine-tuned with BO achieves a higher accuracy than the other proposed architectures. Conclusions: Our architecture achieves a good result compared to previous works and based on different experiments.

No abstract

No abstract