ULECGNet: An Ultra-Lightweight End-to-End ECG Classification Neural Network

计算机科学 人工神经网络 支持向量机 人工智能 现场可编程门阵列 端到端原则 专用集成电路 心跳 模式识别(心理学) 灵活性(工程) 微控制器 统计分类 机器学习 特征提取 计算机硬件 计算机安全 数学 统计
作者
Jianbiao Xiao,Jiahao Liu,Huanqi Yang,Qingsong Liu,Ning Wang,Zhen Zhu,Yulong Chen,Long Yu,Liang Chang,Liang Zhou,Jun Zhou
出处
期刊:IEEE Journal of Biomedical and Health Informatics [Institute of Electrical and Electronics Engineers]
卷期号:26 (1): 206-217 被引量:44
标识
DOI:10.1109/jbhi.2021.3090421
摘要

ECG classification is a key technology in intelligent electrocardiogram (ECG) monitoring. In the past, traditional machine learning methods such as support vector machine (SVM) and K-nearest neighbor (KNN) have been used for ECG classification, but with limited classification accuracy. Recently, the end-to-end neural network has been used for ECG classification and shows high classification accuracy. However, the end-to-end neural network has large computational complexity including a large number of parameters and operations. Although dedicated hardware such as field-programmable gate array (FPGA) and application-specific integrated circuit (ASIC) can be developed to accelerate the neural network, they result in large power consumption, large design cost, or limited flexibility. In this work, we have proposed an ultra-lightweight end-to-end ECG classification neural network that has extremely low computational complexity (∼8.2k parameters & ∼227k multiplication/addition operations) and can be squeezed into a low-cost microcontroller (MCU) such as MSP432 while achieving 99.1% overall classification accuracy. This outperforms the state-of-the-art ECG classification neural network. Implemented on MSP432, the proposed design consumes only 0.4 mJ and 3.1 mJ per heartbeat classification for normal and abnormal heartbeats respectively for real-time ECG classification.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
Lucas应助烤冷面采纳,获得10
刚刚
WYM完成签到,获得积分10
1秒前
爱学习的叭叭完成签到,获得积分10
2秒前
tetrisxzs完成签到,获得积分10
2秒前
SciGPT应助guishan采纳,获得10
2秒前
2秒前
3秒前
liu完成签到,获得积分10
3秒前
停停停发布了新的文献求助10
3秒前
5秒前
5秒前
顾矜应助语收采纳,获得30
6秒前
7秒前
科研通AI6.4应助小盒采纳,获得10
7秒前
7秒前
Akim应助OAIX采纳,获得10
7秒前
852应助边伯贤采纳,获得10
7秒前
7秒前
8秒前
科研通AI6.2应助cc采纳,获得10
8秒前
豆沙饭团完成签到 ,获得积分10
10秒前
10秒前
量子星尘发布了新的文献求助10
10秒前
11秒前
科研顺路发布了新的文献求助10
11秒前
ftinscience应助酵母君采纳,获得10
12秒前
石子完成签到 ,获得积分10
12秒前
zrkkk发布了新的文献求助10
13秒前
可爱的函函应助三木采纳,获得10
13秒前
韩韩发布了新的文献求助10
13秒前
激昂的问玉完成签到,获得积分10
13秒前
医学发布了新的文献求助10
15秒前
16秒前
16秒前
16秒前
17秒前
17秒前
18秒前
18秒前
爆米花应助子凯采纳,获得10
18秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
Relation between chemical structure and local anesthetic action: tertiary alkylamine derivatives of diphenylhydantoin 1000
Signals, Systems, and Signal Processing 610
Discrete-Time Signals and Systems 610
Principles of town planning : translating concepts to applications 500
Iron‐Sulfur Clusters: Biogenesis and Biochemistry 400
Healable Polymer Systems: Fundamentals, Synthesis and Applications 400
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 纳米技术 有机化学 物理 生物化学 化学工程 计算机科学 复合材料 内科学 催化作用 光电子学 物理化学 电极 冶金 遗传学 细胞生物学
热门帖子
关注 科研通微信公众号,转发送积分 6071474
求助须知:如何正确求助?哪些是违规求助? 7902985
关于积分的说明 16340155
捐赠科研通 5211752
什么是DOI,文献DOI怎么找? 2787572
邀请新用户注册赠送积分活动 1770300
关于科研通互助平台的介绍 1648148