2019
DOI: 10.1109/access.2019.2956179
|View full text |Cite
|
Sign up to set email alerts
|

A Neural Network-Based ECG Classification Processor With Exploitation of Heartbeat Similarity

Abstract: This paper presents a neural network based processor with improved computation efficiency, which aims at multiclass heartbeat recognition in wearable devices. A lightweight classification algorithm that integrates both bi-directional long short-term memory (BLSTM) and convolutional neural networks (CNN) is proposed to deliver high accuracy with minimal network scale. To reduce energy consumption of the classification algorithm, the similarity between consecutive heartbeats is exploited to achieve a high degree… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
12
0

Year Published

2021
2021
2024
2024

Publication Types

Select...
6
1

Relationship

0
7

Authors

Journals

citations
Cited by 17 publications
(12 citation statements)
references
References 21 publications
0
12
0
Order By: Relevance
“…The hardware implementation of automatic ECG analysis systems is essential for ambulant monitorization of patients, and there are several examples in the literature for both ASIC [23,24] and FPGA [25,26] implementations. However, to the best of our knowledge, there are no hardware implementations of ECG signal processors that apply the Hermite fit for beat compression or classifications.…”
Section: Discussionmentioning
confidence: 99%
See 1 more Smart Citation
“…The hardware implementation of automatic ECG analysis systems is essential for ambulant monitorization of patients, and there are several examples in the literature for both ASIC [23,24] and FPGA [25,26] implementations. However, to the best of our knowledge, there are no hardware implementations of ECG signal processors that apply the Hermite fit for beat compression or classifications.…”
Section: Discussionmentioning
confidence: 99%
“…This module can be used as the input to systems to compress the ECG data as well as to classifiers. Despite the interest in producing hardware systems for real-time processing of ECG signals [23][24][25][26], to the best of our knowledge, this is the first time that Hermite function fitting with a complete preprocessing chain is implemented in hardware for ECG processing. The main contributions of this paper are as follows:…”
Section: Introductionmentioning
confidence: 99%
“…Although sequential models were specifically designed for sequence or time series, these types of models were used much less. Some studies had combined sequential and nonsequential neural network architectures [17,19,32,42,46,53]. After the neural networks, most of the models were classical machine learning methods, including linear models: support vector machines; decision trees; and similarity-based models, such as k-nearest neighbor classifiers.…”
Section: Model Construction Methods (Levels 7 and 8)mentioning
confidence: 99%
“…Smartphones were used in both benchmark [38][39][40] and nonbenchmark [21,30,31,35] studies. Embedded devices, however, had only been demonstrated in benchmark studies [41][42][43][44].…”
Section: Processing Device (Level 6)mentioning
confidence: 99%
“…It provided a significant increase in the power efficiency of the overall system due to a decrease in radio utilization, emphasizing the need to maximize local processing. Generic hardware accelerators discussed in [8,10,11], target common signal processing techniques predominantly used in biomedical applications. This ensures the use of accelerators in multiple varied algorithms and provides an energy-efficient solution with the flexibility to map a large set of biomedical applications.…”
Section: Introductionmentioning
confidence: 99%