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: Jun 28, 2026

Artificial Intelligence-Based System for Detecting Attention Levels in Students
06:37

Artificial Intelligence-Based System for Detecting Attention Levels in Students

Published on: December 15, 2023

An Energy-Efficient ECG Classifier With On-Chip Learning Using Binarized Convolutional Neural Network.

Rui Zhang, Ranran Zhou, Xinyi Han

    IEEE Transactions on Biomedical Circuits and Systems
    |September 17, 2025
    PubMed
    Summary

    This study introduces an energy-efficient ECG classifier using binarized convolutional neural networks (bCNNs) and on-chip learning. The novel approach enhances accuracy and reduces power consumption for wearable applications.

    Related Concept Videos

    Classification of Signals01:30

    Classification of Signals

    1.3K
    In signal processing, signals are classified based on various characteristics: continuous-time versus discrete-time, periodic versus aperiodic, analog versus digital, and causal versus noncausal. Each category highlights distinct properties crucial for understanding and manipulating signals.
    A continuous-time signal holds a value at every instant in time, representing information seamlessly. In contrast, a discrete-time signal holds values only at specific moments, often denoted as x(n), where...
    1.3K

    You might also read

    Related Articles

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

    Sort by
    Same author

    Investigating the Shared Mechanisms of Endocrine-Disrupting Chemicals in Urogenital Tumors.

    Biology·2026
    Same author

    A High-Performance Bimetallic Ru<sub>1</sub>Mo<sub>6</sub> Active Site for Thermal Ammonia Synthesis under Mild Conditions.

    Journal of the American Chemical Society·2026
    Same author

    Boosting SERS Performance of TiO<sub>2</sub> Microspheres via Phytic Acid Modification: Application to Bisphenol A Detection in Consumer Products.

    ACS applied materials & interfaces·2026
    Same author

    Nitrogen fertilization increases soil organic carbon through distinct pathways in contrasting cropland soils.

    Science China. Life sciences·2026
    Same author

    Realizing efficient ammonia electrosynthesis enabled by metal-organic framework-derived CuCo-LDH.

    Physical chemistry chemical physics : PCCP·2026
    Same author

    Sustained-Release Reducing Microenvironment Tailors Highly Crystalline FeS<sub><i>x</i></sub> Shells on Zero-Valent Iron for Enhanced Electron Transfer.

    Environmental science & technology·2026

    Area of Science:

    • Biomedical Engineering
    • Artificial Intelligence
    • Hardware Acceleration

    Background:

    • Binarized convolutional neural networks (bCNNs) offer low power for ECG classification via 1-bit quantization.
    • Existing bCNN methods process full ECG images, ignoring sparsity and increasing computational overhead.
    • Inter-patient variability and binarization-induced accuracy loss hinder performance in current bCNN ECG classifiers.

    Purpose of the Study:

    • To propose an energy-efficient ECG classifier using bCNN with on-chip learning.
    • To reduce power consumption and memory usage through patch-by-patch computation.
    • To enhance classification accuracy across patients by integrating R-peak interval data for model updates.

    Main Methods:

    • Implemented a patch-by-patch computation approach, processing only relevant ECG data patches.

    Related Experiment Videos

    Last Updated: Jun 28, 2026

    Artificial Intelligence-Based System for Detecting Attention Levels in Students
    06:37

    Artificial Intelligence-Based System for Detecting Attention Levels in Students

    Published on: December 15, 2023

  • Employed an on-chip learning mechanism to update bCNN model weights using ECG features and R-peak intervals.
  • Designed a reconfigurable convolutional processing element array and a base-2 softmax structure for hardware efficiency.

    • Main Results:

    • Achieved 97.55% classification accuracy and 89.15% specificity on an FPGA verification.
    • The classifier occupies a small area (0.43 mm²) using a 55 nm CMOS process.
    • Demonstrated low average energy consumption: 0.12 μJ per classification and 0.09 μJ per on-chip learning.

    Conclusions:

    • The proposed bCNN with on-chip learning offers an energy-efficient solution for ECG classification.
    • The patch-by-patch approach and optimized hardware structures significantly reduce resource utilization.
    • The classifier's performance and low power profile make it ideal for wearable ECG monitoring systems.