The classification of electrocardiogram (ECG) signals plays a crucial role in identifying heart-related diseases and ensuring early detection for timely treatment. However, manually detecting arrhythmia is a challenging task, highlighting the need for an automatic technique to detect abnormal heart signals. In this study, the researchers focused on classifying five classes of ECG arrhythmic signals from Physionet's MIT-BIH Arrhythmia Dataset using Artificial Neural Networks (ANN). Specifically, they proposed a Convolutional Neural Network (CNN) model customized for categorizing ECG signals. The results demonstrated that the planned CNN model achieved an overall accuracy of 95.2% in categorizing arrhythmia. The average precision and recall of the proposed model were found to be 95.2% and 95.4%, respectively. These findings indicate that the CNN model can effectively detect irregularities in heart rhythm at an early stage. In addition to the existing summary, it is worth noting that this research was presented at the 4th International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) in October 2020. The study was conducted by Mohammad Mahmudur Rahman Khan, Md. Abu Bakr Siddique, Shadman Sakib, Anas Aziz, Abyaz Kader Tanzeem, and Ziad Hossain. The authors also provided further context by discussing the importance of ECG signal classification in identifying heart-related diseases and selecting appropriate treatments for patients. They highlighted that deep learning techniques such as CNNs have shown significant success in ECG signal classification tasks. Overall, this study contributes to the development of automated techniques for detecting abnormal heart signals through accurate classification of ECG arrhythmic data using a customized CNN model which can effectively identify irregularities in heart rhythm at an early stage with high accuracy and precision rates..
Scientists have been studying how to tell if someone's heart is healthy or not by looking at their heartbeats. They found that using a special kind of computer program called Artificial Neural Networks can help them do this automatically. They made a new computer model called Convolutional Neural Network that can look at the heartbeats and tell if there is something wrong with them. This model was very good at finding problems, with an accuracy of 95.2%. The scientists presented their research at a conference in October 2020. The study was done by a group of researchers named Mohammad Mahmudur Rahman Khan, Md. Abu Bakr Siddique, Shadman Sakib, Anas Aziz, Abyaz Kader Tanzeem, and Ziad Hossain.
Definitions
- Classification: Sorting things into groups based on their similarities.
- ECG signals: The electrical signals produced by the heart that are recorded using an electrocardiogram (ECG) machine.
- Arrhythmia: A condition where the heartbeat is irregular or abnormal.
- Artificial Neural Networks (ANN): Computer programs designed to mimic the way the human brain works and learn from data.
- Convolutional Neural Network (CNN): A type of artificial neural network commonly used for image recognition tasks.
- Accuracy: How correct or accurate something is.
- Precision: How close measurements are to each other.
- Recall: The ability to remember or recognize something.
- Irregularities: Things that are not
Classifying Electrocardiogram (ECG) Signals with Artificial Neural Networks
The classification of electrocardiogram (ECG) signals plays a crucial role in identifying heart-related diseases and ensuring early detection for timely treatment. However, manually detecting arrhythmia is a challenging task, highlighting the need for an automatic technique to detect abnormal heart signals. In October 2020, researchers from Bangladesh presented their findings on classifying five classes of ECG arrhythmic signals from Physionet's MIT-BIH Arrhythmia Dataset using Artificial Neural Networks (ANNs) at the 4th International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud). The research team included Mohammad Mahmudur Rahman Khan, Md. Abu Bakr Siddique, Shadman Sakib, Anas Aziz, Abyaz Kader Tanzeem and Ziad Hossain.
Background
An electrocardiogram (ECG) is a medical test that records electrical activity in the heart over time. It can be used to diagnose various cardiac conditions such as arrhythmias or other abnormalities in the rhythm of the heartbeat. Manual detection of these irregularities is difficult due to its complexity and requires specialized training; thus making it necessary to develop automated techniques for accurate diagnosis. Deep learning techniques such as Convolutional Neural Networks (CNNs) have shown significant success in ECG signal classification tasks and this study aimed to further explore this potential by creating a customized CNN model for categorizing ECG signals from Physionet's MIT-BIH Arrhythmia Dataset into five classes: Normal Sinus Rhythm (N), Atrial Fibrillation/Flutter (A), Premature Ventricular Contraction/Beat/Extrasystole(V), Left Bundle Branch Block Beat/Premature Junctional Complexes(L), Right Bundle Branch Block Beat/Premature Junctional Complexes(R).
Methodology
The proposed CNN model was designed with two convolution layers followed by two fully connected layers which were optimized through backpropagation algorithm using Adam optimizer with learning rate 0.001 and batch size 32 samples per epochs. The model was trained on 80% of data set while 20% was used for testing purposes; resulting in an overall accuracy of 95.2%. The average precision and recall achieved by the proposed model were found to be 95.2% and 95.4%, respectively indicating its effectiveness at detecting irregularities in heart rhythm at an early stage accurately..
Conclusion
This study contributes to the development of automated techniques for detecting abnormal heart signals through accurate classification of ECG arrhythmic data using a customized CNN model which can effectively identify irregularities in heart rhythm at an early stage with high accuracy and precision rates.. This research highlights how deep learning models are increasingly being applied successfully across medical fields including cardiology where they can provide more reliable results than manual methods alone
The classification of the electrocardiogram (ECG) signal has a vital impact on identifying heart-related diseases. This can ensure the premature finding of heart disease and the proper selection of the patient's customized treatment. However, the detection of arrhythmia is a challenging task to perform manually. This justifies the necessity of a technique for automatic detection of abnormal heart signals. Therefore, our work is based on the classification of five classes of ECG arrhythmic signals from Physionet's MIT-BIH Arrhythmia Dataset. Artificial Neural Networks (ANN) have demonstrated significant success in ECG signal classification. Our proposed model is a Convolutional Neural Network (CNN) customized to categorize the ECG signals. Our result testifies that the planned CNN model can successfully categorize arrhythmia with an overall accuracy of 95.2%. The average precision and recall of the proposed model are 95.2% and 95.4%, respectively. This model can effectively be used to detect irregularities of heart rhythm at an early stage.
