Key result
CNN-based arrhythmia detection on GAP9 cuts energy consumption ~67% versus state-of-the-art transformers.
Why the study?
Wearable battery-powered ECG monitors require high energy efficiency, but state-of-the-art machine learning algorithms are extremely demanding in energy consumption, requiring solutions that balance accuracy and energy use to extend battery lifetime.
Absolute Event Rate: 95% vs 98%
An energy-efficient CNN-based methodology for ECG analysis on ultra-low-power platforms reduces energy consumption by 3x with only a marginal 3% drop in accuracy compared to state-of-the-art transformer models.
The widespread diffusion of long-term cardiac monitoring using wearable devices is a key opportunity for analyzing the health conditions of chronic patients. The continuous analysis of the heartbeat, even reduced to minimal configuration (i.e., two leads), can diagnose and keep track of many severe cardiac conditions, such as abnormal atrial and ventricular contractions. Since wearable devices are battery-powered, it is essential to design solutions that can improve the power efficiency of this monitoring, leveraging HW/SW optimization on low-power platforms. State-of-the-art algorithms based on advanced machine learning (ML) approaches achieve high accuracy but are extremely demanding in terms of energy consumption. In the context of battery-powered devices, determining a trade-off between accuracy and energy consumption is paramount to extending battery lifetime. This work presents a system design for analyzing the Electrocardiogram (ECG) signal to detect pathological conditions using an energy-efficient methodology based on Convolutional Neural Networks (CNNs). We assessed our solution on GAP9, a parallel microcontroller-class platform based on the RISC-V architecture. We achieved a 95.0% accuracy on the MIT-BIH Arrhythmia dataset, which includes five classes of pathological conditions. This value is marginally lower (3%) than the current state-of-the-art based on transformers. However, we identified the best energy-accuracy trade-off configuration, reducing the energy consumption of 3 x (0.03 mJ vs. 0.09 mJ) which guarantees a longer battery lifetime for critical applications.
No takes yet. Share an insight, caveat, or question.
Mazzoni et al. (2024) studied Arrhythmia. CNN-based methodology on GAP9 platform vs. State-of-the-art transformer-based algorithms was evaluated on Accuracy and energy consumption on the MIT-BIH Arrhythmia dataset. A CNN-based methodology on the GAP9 platform achieved 95.0% accuracy for arrhythmia detection, reducing energy consumption by 3x (0.03 mJ vs. 0.09 mJ) compared to state-of-the-art transformers.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: