Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Oct 10, 2025

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
14:08

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

42.9K

Multistage Pruning of CNN Based ECG Classifiers for Edge Devices.

Li Xiaolin, Rajesh C Panicker, Barry Cardiff

    Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference
    |December 11, 2021
    PubMed
    Summary

    This study introduces a novel multistage pruning technique for convolutional neural network (CNN) models used in electrocardiogram (ECG) analysis. The method significantly reduces model complexity for wearable devices without sacrificing diagnostic accuracy.

    Related Concept Videos

    Instrumentation Amplifier01:25

    Instrumentation Amplifier

    758
    An electrocardiography (ECG) machine is an essential piece of medical equipment used to monitor the electrical activity of the heart. It operates by detecting small electrical changes on the skin that result from the depolarization of the heart muscle during each heartbeat. However, these signals are in the microvolt range and can be easily overwhelmed by noise or interference.
    To overcome this challenge, an ECG machine utilizes an instrumentation amplifier. This specialized amplifier is...
    758

    You might also read

    Related Articles

    Articles linked to this work by shared authors, journal, and citation graph.

    Sort by
    Same author

    Exogenous additives for acrylamide mitigation in thermally processed foods: A review of safety risks and quality impacts.

    Food chemistry·2026
    Same author

    Novel insights into the protective role of crocetin from gardenia jasminoides against mitochondrial hypoxia.

    Scientific reports·2025
    Same author

    MLFusion: Multilevel Data Fusion using CNNs for atrial fibrillation detection.

    Computers in biology and medicine·2025
    Same author

    Signal-quality-aware multisensor fusion for atrial fibrillation detection.

    Healthcare technology letters·2025
    Same author

    ECG Biometric Authentication Using Self-Supervised Learning for IoT Edge Sensors.

    IEEE journal of biomedical and health informatics·2024
    Same author

    Sensor-Location-Specific Joint Acquisition of Peripheral Artery Bioimpedance and Photoplethysmogram for Wearable Applications.

    Sensors (Basel, Switzerland)·2023

    Area of Science:

    • Biomedical Engineering
    • Artificial Intelligence
    • Cardiology

    Background:

    • Smart wearable devices enable real-time electrocardiogram (ECG) monitoring for arrhythmia detection, improving patient outcomes.
    • Deep learning, specifically Convolutional Neural Networks (CNNs), effectively detects anomalous ECG beats but faces computational complexity challenges for edge devices.
    • Existing CNN models require substantial computational resources, limiting their deployment on low-powered wearable devices.

    Purpose of the Study:

    • To develop a novel multistage pruning technique for reducing the complexity of CNN models used in ECG analysis.
    • To enable the implementation of accurate ECG classification models on resource-constrained edge devices.
    • To minimize performance degradation typically associated with network pruning methods.

    Related Experiment Videos

    Last Updated: Oct 10, 2025

    Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
    14:08

    Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

    Published on: April 13, 2013

    42.9K

    Main Methods:

    • A novel multistage network pruning technique was developed and applied to an existing CNN model for ECG classification.
    • The technique focused on reducing model parameters and computational complexity while maintaining high performance.
    • Performance was evaluated against traditional pruning methods and a baseline model at 60% sparsity.

    Main Results:

    • The proposed multistage pruning technique achieved 97.7% accuracy and an F1 score of 93.59% for ECG classification at 60% sparsity.
    • This represents a significant improvement over traditional pruning with fine-tuning, with accuracy increased by 3.3% and F1 score by 9%.
    • The method resulted in a 60.4% decrease in run-time complexity compared to the baseline CNN model.

    Conclusions:

    • The novel multistage pruning technique effectively reduces CNN model complexity for ECG analysis, making it suitable for low-powered edge devices.
    • This approach offers a substantial improvement in accuracy and F1 score compared to existing pruning methods, with negligible performance loss.
    • The reduced computational and run-time complexity facilitates the real-time monitoring of arrhythmias using wearable devices.