Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
November 21, 2025Ultrasonic Imaging

Novel Clinical Hybrid Deep Framework for Denoising and Anatomical Segmentation in Challenging Ultrasound Conditions

View Full Paper
Ask AI
Bookmark
Share

Authors

TSTaher SlimiAKAnouar Ben Khalifa

Discussion

Loading...

Member takes

Overview

This framework improves noise suppression in ultrasound imaging, indicating enhanced anatomical segmentation reliability and clinical interpretation.

Key Points

  • The research aims to develop a hybrid framework to enhance denoising and anatomical segmentation in ultrasound images.
  • Introduced a hybrid framework combining SAFIE for denoising with superpixel-based hypergraph modeling for segmentation.
  • Utilized gradient-based refinement and neural ordinary differential equations for improved performance.
  • Conducted qualitative evaluations by four radiologists to assess image quality and inter-observer agreement.
  • Achieved significant noise suppression and enhanced visibility of anatomical structures in ultrasound imaging.
  • Demonstrated robust performance compared to state-of-the-art methods as per quantitative analyses.
  • Inter-observer agreement was measured using Fleiss’ kappa, indicating strong reliability among assessments.

Cite This Study

Slimi et al. (2025) studied this question.

synapsesocial.com/papers/6924e3e6c0ce034ddc34ec3ahttps://doi.org/10.1177/01617346251384596
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Denoising of Medical Ultrasound Images Using Variational Autoencoders Guided by Attention Mechanisms2026
  2. 2Unsupervised speckle noise reduction technique for clinical ultrasound imaging2024 · 1 citations
  3. 3An Efficient Multi-Scale Wavelet Approach for Dehazing and Denoising Ultrasound Images Using Fractional-Order Filtering2024 · 3 citations
  4. 4A Complementary Global and Local Knowledge Network for Ultrasound Denoising with Fine-grained Refinement2024
  5. 5Hybrid Physics-Driven Deep Learning for Enhanced Ultrasound Image Quality and Speckle Noise Suppression2026