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 Concept Videos

You might also read

Related Articles

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

Sort by
Same author

A retrospective study of coarse calcification as an independent risk factor for poor prognosis and early metastasis in paediatric neuroblastoma.

European journal of radiology·2026
Same author

A Synergistic DNA Walker-Track System with Tailored Spatial Compatibility for Rapid and Ultrasensitive Electrochemical Detection of MUC1.

Analytical chemistry·2026
Same author

Phase behavior of Tween 80/SDBS mixed microemulsions and mechanistic insights into enhanced oil separation from aged oily sludge.

Journal of hazardous materials·2026
Same author

Reported drug spectrum and disproportionality signals for malignant neoplasm progression in FAERS: a real-world pharmacovigilance study.

Frontiers in pharmacology·2026
Same author

CliPME: The clinical pathogenic bacteria mutation and expression database.

Journal of genetics and genomics = Yi chuan xue bao·2026
Same author

Mycobacterial non-homologous end joining is required for antiphage defense.

Nucleic acids research·2026

Related Experiment Video

Updated: Jul 9, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.8K

DM-CNN: Dynamic Multi-scale Convolutional Neural Network with uncertainty quantification for medical image

Qi Han1, Xin Qian1, Hongxiang Xu1

  • 1School of Intelligent Technology and Engineering, Chongqing University of Science and Technology, Chongqing 401331, PR China.

Computers in Biology and Medicine
|December 2, 2023
PubMed
Summary

This study introduces DM-CNN, a novel deep learning model for medical image analysis. DM-CNN enhances feature extraction and quantifies prediction uncertainty, achieving state-of-the-art results across multiple medical domains.

Keywords:
Deep learningDynamic convolutionMedical image classificationMulti-scale fusionUncertainty quantification

More Related Videos

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

406
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

42.6K

Related Experiment Videos

Last Updated: Jul 9, 2025

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.8K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

406
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

42.6K

Area of Science:

  • Medical image analysis
  • Artificial intelligence in healthcare
  • Deep learning for diagnostics

Background:

  • Convolutional Neural Networks (CNNs) have advanced medical image diagnosis.
  • Existing CNNs often suffer from limited feature information, inaccurate attention weights, and lack of prediction uncertainty quantification.
  • Addressing these limitations is crucial for reliable medical AI.

Purpose of the Study:

  • To propose a novel Convolutional Neural Network (CNN) model, DM-CNN, designed to overcome limitations in medical image analysis.
  • To enhance feature extraction capabilities and accurately quantify prediction uncertainty in deep learning models.
  • To achieve state-of-the-art performance in diverse medical imaging tasks.

Main Methods:

  • Developed DM-CNN incorporating four novel modules: Dynamic Multi-scale Feature Fusion (DMFF), Hierarchical Dynamic Uncertainty quantifies Attention (HDUQ-Attention), Multi-scale Fusion Pooling (MF Pooling), and Multi-objective Loss (MO loss).
  • DMFF fuses multi-scale features using adaptive convolution kernels.
  • HDUQ-Attention integrates attention tuning and Monte-Carlo (MC) dropout for uncertainty quantification.
  • MF Pooling optimizes multi-scale information processing and prevents overfitting.
  • MO loss addresses parameter discrepancies for efficient optimization.

Main Results:

  • DM-CNN achieved state-of-the-art classification performance on publicly available datasets in Dermatology, Histopathology, Respirology, and Ophthalmology.
  • The model demonstrated superior feature extraction and effective uncertainty quantification.
  • Experimental results validate the efficacy of the proposed modules and the overall DM-CNN architecture.

Conclusions:

  • DM-CNN offers a significant advancement in medical image analysis by improving feature representation and providing crucial uncertainty quantification.
  • The model's ability to maintain excellent performance while addressing uncertainty makes it highly valuable for clinical applications.
  • DM-CNN represents a robust solution for challenging medical deep learning tasks, enhancing diagnostic reliability.