发表机构
Computer Architecture for Embedded Systems, University of Twente, The Netherlands; Biomedical Signals and Systems, University of Twente, The Netherlands; Faculty of Science, Computer Science, Vrije Universiteit Amsterdam, The Netherlands（嵌入式系统计算机架构，特文特大学，荷兰; 生物医学信号与系统，特文特大学，荷兰; 科学学院，计算机科学，阿姆斯特丹自由大学，荷兰）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究针对可穿戴设备心律失常检测中深度学习算法功耗高的问题，通过数据精度降低和近似乘法等近似技术，在保持分类性能的同时显著降低功耗，延长了可穿戴设备电池寿命。
AI 中文摘要
心血管疾病是全球主要死因，心律失常等病症常需长期监测以有效检测和诊断。当前可穿戴监测设备笨重、不舒适且依赖临床医生手动评估心电图。深度学习算法在心律失常检测和分类中性能优越，但计算复杂度和高功耗限制了在可穿戴设备中的应用。本文研究使用近似技术降低深度学习架构的功耗，同时保持可接受的分类性能。在一个先进的深度学习模型及其硬件架构中研究了数据精度降低和近似乘法等技术。使用MIT - BIH心律失常数据库对模型进行训练和验证，并对采用各种近似乘法器的硬件实现进行综合和评估。与最先进的8.75微瓦（和2.08微焦）参考架构相比，我们提出的架构在12千赫兹时功耗为3.07微瓦（和2.17微焦），功耗降低64.9%，同时提供可接受的输出质量，即93.7%的分类准确率和92.1%的灵敏度。在100兆赫兹时，我们提出的架构功耗为9.45毫瓦（和0.8微焦），与最先进架构相比能耗降低61.5%。这些结果表明，我们提出的近似方法在保持所需心律失常分类性能的同时，显著延长了可穿戴设备的电池寿命。
英文摘要
Cardiovascular diseases are the leading cause of death worldwide, and conditions such as arrhythmia often require long-term monitoring for effective detection and diagnosis. However, current wearable monitoring devices are bulky, uncomfortable, and typically rely on clinicians to manually evaluate electrocardiograms (ECGs). While Deep Learning (DL) algorithms have shown superior performance in arrhythmia detection and classification, their computational complexity coupled with high power consumption limit deployment in wearable devices. To address this challenge, this paper investigates the use of approximation techniques to reduce the power and energy consumption of DL architectures while maintaining acceptable classification performance. Specifically, techniques such as data precision reduction and approximate multiplication are investigated in a state-of-the-art DL model and its corresponding hardware architecture. The model is trained and validated using the MIT-BIH Arrhythmia Database, and hardware implementations employing various approximate multipliers are synthesized and evaluated. Compared with the state-of-the-art 8.75 μW (and 2.08 μJ) reference architecture, our proposed architecture consumes 3.07 μW (and 2.17 μJ) at 12 kHz, showing 64.9% reduction in power consumption while providing an acceptable output quality, i.e., 93.7% classification accuracy and 92.1% sensitivity. At 100 MHz, our proposed architecture consumes 9.45 mW (and 0.8 μJ), showing 61.5% reduction in energy consumption as compared to the state-of-the-art architecture. These results demonstrate that our proposed approximations significantly extend wearable device battery life while preserving the required arrhythmia classification performance.
