Key result
A distributed multioutput CNN architecture for continuous ECG monitoring achieved 99.2% accuracy on the MIT-BIH Arrhythmia Data Set and 7x more energy efficiency compared to state-of-the-art works.
Why the study?
Continuous monitoring of ECG signals is challenging in low-power wearable devices due to energy and memory constraints.
Does a distributed multioutput CNN architecture improve accuracy and energy efficiency for continuous ECG monitoring in low-power wearable devices?
Does a distributed multioutput CNN architecture improve accuracy and energy efficiency for continuous ECG monitoring in low-power wearable devices?
Effect estimate: 7x more energy efficiency
A novel distributed multioutput CNN architecture for wearable devices achieves high accuracy (99.2%) for arrhythmia detection while significantly improving energy efficiency.
May enable continuous ECG monitoring on low-power wearables; leaves open clinical validation before practice adoption.
The recent developments in wearable devices and the Internet of Medical Things (IoMT) allow real-time monitoring and recording of electrocardiogram (ECG) signals. However, continuous monitoring of ECG signals is challenging in low-power wearable devices due to energy and memory constraints. Therefore, in this article, we present a novel and energy-efficient methodology for continuously monitoring the heart for low-power wearable devices. The proposed methodology is composed of three different layers: 1) a noise/artifact detection layer to grade the quality of the ECG signals; 2) a normal/abnormal beat classification layer to detect the anomalies in the ECG signals; and 3) an abnormal beat classification layer to detect diseases from ECG signals. Moreover, a distributed multioutput convolutional neural network (CNN) architecture is used to decrease the energy consumption and latency between the edge–fog/cloud. Our methodology reaches an accuracy of 99.2% on the well-known MIT-BIH Arrhythmia Data Set. Evaluation on real hardware shows that our methodology is suitable for devices having a minimum RAM of 32 kb. Moreover, the proposed methodology achieves7×more energy efficiency compared to state-of-the-art works.
No takes yet. Share an insight, caveat, or question.
Demirel et al. (2021) studied Arrhythmia. Distributed multioutput convolutional neural network (CNN) architecture vs. State-of-the-art works was evaluated on Accuracy on MIT-BIH Arrhythmia Data Set (7x more energy efficiency). A distributed multioutput CNN architecture for continuous ECG monitoring achieved 99.2% accuracy on the MIT-BIH Arrhythmia Data Set and 7x more energy efficiency compared to state-of-the-art works.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: