A Cost-Effective and Intelligent Holter Device for Accurate Cardiac Arrhythmia Detection

Problem: Health risks from cardiac arrhythmias remain a major concern because current detection systems tend to be costly while also lacking accessibility especially in areas with limited resources. Aims: The purpose of this research is to create a portable Holter monitoring device which can detect cardiac arrhythmias accurately through machine learning algorithms while maintaining cost-effectiveness. Contribution: This research combines an inexpensive ESP32 microcontroller with conventional machine learning algorithms to optimize performance rather than relying on complex deep learning models like earlier studies. Method: The ESP32 microcontroller and the AD8232 ECG sensor were used together to build a data acquisition system. Research teams created and evaluated Support Vector Machines (SVM), K-Nearest Neighbors (KNN) and Multilayer Perceptron (MLP) models to classify arrhythmias. Results: The SVM model reached its peak performance with an accuracy of 78.53% achieved through the use of a linear kernel combined with feature selection from a random forest algorithm. The performance of both KNN and MLP models proved to be promising which underscored the importance of hyperparameter tuning and feature selection. Conclusion: The research successfully proved the possibility of designing an affordable intelligent Holter device with effective arrhythmia detection capabilities. Implication: The results indicate a possibility for widespread availability of cost-effective cardiac monitoring which could enhance patient care along with early arrhythmia detection in resource-scarce settings.