Key result
CNN detects cardiac arrhythmia with ~99% accuracy, outperforming traditional machine learning models.
Why the study?
Choosing among various state-of-the-art models for heart arrhythmia prediction is difficult, especially for hospitals with less experienced doctors and limited resources.
Absolute Event Rate: 99% vs 97%
A Convolutional Neural Network can classify cardiac arrhythmias from ECG signals with 99% accuracy, offering a highly accurate automated detection method compared to traditional machine learning models.
CNN arrhythmia detection shows promise; leaves open prospective validation before clinical use.
Machine Learning is an integral part of Artificial Intelligence, is a science of statistical models and their algorithms. It is used to train a model to perform a real-world task without explicit commands. Heart Arrhythmia is a life-threatening disease dealing with an irregular heartbeat. ECG signals are the most accurate measure to find out the functionality of the cardiovascular-system. We put forward a solution to solve the difficulty of choosing among the various state-of-art models for heart arrhythmia prediction. This helps the hospitals with less experienced doctors to more accurately predict the disease at a lower cost. The Convolutional Neural Network model we built is more accurate than the existing models. We have also analyzed with other machine learning models such as Logistic Regression, Random Forest and XGBoost for the hospitals with lesser computation power.
No takes yet. Share an insight, caveat, or question.
Mathew et al. (2020) studied Heart Arrhythmia (n=47). Convolutional Neural Network vs. Logistic Regression, Random Forest, XGBoost was evaluated on Classification accuracy. The Convolutional Neural Network model achieved 99% accuracy in predicting cardiac arrhythmia, outperforming Random Forest, XGBoost, and Logistic Regression models.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: