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 12, 2025

Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

9.6K

Random Fourier Features-Based Deep Learning Improvement with Class Activation Interpretability for Nerve Structure

Cristian Alfonso Jimenez-Castaño1, Andrés Marino Álvarez-Meza2, Oscar David Aguirre-Ospina3

  • 1Automatic Research Group, Universidad Tecnológica de Pereira, Pereira 660003, Colombia.

Sensors (Basel, Switzerland)
|November 27, 2021
PubMed
Summary

Related Concept Videos

Neural Circuits01:25

Neural Circuits

1.8K
Neural circuits and neuronal pools are two of the main structures found in the nervous system. Neural circuits are networks of neurons that work together to carry out a specific task or process. They consist of interconnected neurons and glial cells, which provide structural and metabolic support.
Neuronal pools are collections of nerve cells with similar functions and interact through chemical and electrical signals. These pools include both interneurons (the central neural circuit nodes that...
1.8K

You might also read

Related Articles

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

Sort by
Same author

Gaussian Connectivity-Driven EEG Imaging for Deep Learning-Based Motor Imagery Classification.

Sensors (Basel, Switzerland)·2026
Same author

Takens-Based Kernel Transfer Entropy Connectivity Network for Motor Imagery Classification.

Sensors (Basel, Switzerland)·2025
Same author

CrowdAttention: An Attention Based Framework to Classify Crowdsourced Data in Medical Scenarios.

Sensors (Basel, Switzerland)·2025
Same author

Integrating Cacao Physicochemical-Sensory Profiles via Gaussian Processes Crowd Learning and Localized Annotator Trustworthiness.

Foods (Basel, Switzerland)·2025
Same author

EEG Signal Prediction for Motor Imagery Classification in Brain-Computer Interfaces.

Sensors (Basel, Switzerland)·2025
Same author

EEG-Based Music Emotion Prediction Using Supervised Feature Extraction for MIDI Generation.

Sensors (Basel, Switzerland)·2025

This study introduces a kernel-based deep learning method to improve ultrasound nerve segmentation for peripheral nerve blocking (PNB). The approach enhances accuracy and interpretability, aiding regional anesthesia procedures.

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Anesthesiology

Background:

  • Peripheral nerve blocking (PNB) relies on accurate nerve localization via ultrasound.
  • Deep learning methods for ultrasound nerve segmentation face challenges with noise and data interpretability.
  • Existing complex architectures lack transparency in feature learning for nerve structures.

Purpose of the Study:

  • To enhance ultrasound nerve structure segmentation using a kernel-based deep learning approach.
  • To improve the interpretability of deep learning models in medical imaging.
  • To provide a more generalized and accurate segmentation for peripheral nerve blocks.

Main Methods:

  • A kernel-based deep learning enhancement using random Fourier features was applied to semantic segmentation architectures (FCN, U-net, ResUnet).
Keywords:
class activation mappingdeep learningnerve structure segmentationrandom Fourier featuresultrasound images

Related Experiment Videos

Last Updated: Oct 12, 2025

Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

9.6K
  • The method was tested on two ultrasound image datasets for peripheral nerve blocking.
  • GradCam++ was extended for class-activation mapping to interpret learned features.

    • Main Results:

    • The kernel-based approach demonstrated improved generalization capabilities in segmenting various nerve structures.
    • Enhanced data interpretability was achieved, revealing key features distinguishing nerves from the background.
    • The method proved effective for both shallow and deep neural network architectures.

    Conclusions:

    • The proposed kernel-based deep learning enhancement improves ultrasound nerve segmentation accuracy and interpretability for PNB.
    • This method offers a valuable tool for regional anesthesia, potentially reducing adverse effects.
    • The approach provides a transparent and generalized solution for automated nerve segmentation in medical ultrasound.