Diagnosis of heart disease using a convolutional neural network based on artificial intelligence

Heart disease is the most common disease with complications, mortality, and causes numerous complications and problems for people in society. Given the numerous advances in various fields, including disease diagnosis methods, early and accurate diagnosis is always needed by the research community. Diagnostic methods based on artificial intelligence (AI) have paved their way as a powerful and effective tool in various fields, and their use is increasing day by day. Systematic and fully automated diagnosis of cardiac arrhythmias with acceptable accuracy is always of interest to researchers, and finding innovative ways is needed. Intelligent systems based on convolutional neural networks can be considered an effective solution in making automated diagnostic systems intelligent, which pave the way for disease differentiation. In this study, cardiac arrhythmias were diagnosed using a convolutional neural network combined with artificial intelligence. Here, the signal spectrogram was used as the input to the convolutional neural network. Training and testing data based on artificial intelligence have been used. Accuracy and sensitivity have been calculated using a ۱۰-fold cross-validation method. The proposed method achieved an average accuracy of ۹۹.۷۱% and an average sensitivity of ۹۹.۶۴%. The proposed method based on artificial intelligence has reduced the computation time and system complexity and achieved acceptable accuracy. It can be said that the proposed method has the necessary ability to distinguish arrhythmia and is very suitable.

کلیدواژه ها:

Artificial intelligence ، deep learning ، convolutional neural network ، cardiac arrhythmias