What question did this study set out to answer?

To develop and validate a hybrid deep-learning architecture that removes speckle noise from medical ultrasound images while retaining critical fine anatomical details.

January 1, 2026Critical Reviews in Biomedical Engineering

Denoising of Medical Ultrasound Images Using Variational Autoencoders Guided by Attention Mechanisms

View Full Paper

Ask AI

Bookmark

Authors

TSTaher Slimi HAHouda AKREMI AKA. Said E. Khalifa

Discussion

Member takes

Overview

Algorithm evaluation study demonstrates superior speckle noise reduction across ultrasound datasets, indicating enhanced anatomical visualization for clinical diagnosis.

Key Points

To develop and validate a hybrid deep-learning architecture that removes speckle noise from medical ultrasound images while retaining critical fine anatomical details.
Integrated large kernel attention (LKA) to capture broad contextual features with a convolutional variational autoencoder (CVAE) for generative image reconstruction.
Benchmarked the proposed LKA-CVAE framework against state-of-the-art denoising approaches across six diverse ultrasound imaging datasets.
Achieved an average peak signal-to-noise ratio (PSNR) of 53.64 dB (SD 0.028), representing a 10 dB improvement over the top competing method.
Attained an average structural similarity index measure (SSIM) of 0.913 (SD 0.03), yielding a statistically significant improvement of 0.041 compared to leading alternatives.

Cite This Study

Slimi et al. (2026) studied this question.

synapsesocial.com/papers/6a70c8ba35aa2c282ce21b89 https://doi.org/10.1615/critrevbiomedeng.2026060806

View Full Paper

Ask AI

Bookmark

January 1, 2026Critical Reviews in Biomedical Engineering

Denoising of Medical Ultrasound Images Using Variational Autoencoders Guided by Attention Mechanisms

View Full Paper

Ask AI

Bookmark

Authors

TSTaher Slimi HAHouda AKREMI AKA. Said E. Khalifa

Discussion

Member takes

Overview

Algorithm evaluation study demonstrates superior speckle noise reduction across ultrasound datasets, indicating enhanced anatomical visualization for clinical diagnosis.

Key Points

To develop and validate a hybrid deep-learning architecture that removes speckle noise from medical ultrasound images while retaining critical fine anatomical details.
Integrated large kernel attention (LKA) to capture broad contextual features with a convolutional variational autoencoder (CVAE) for generative image reconstruction.
Benchmarked the proposed LKA-CVAE framework against state-of-the-art denoising approaches across six diverse ultrasound imaging datasets.
Achieved an average peak signal-to-noise ratio (PSNR) of 53.64 dB (SD 0.028), representing a 10 dB improvement over the top competing method.
Attained an average structural similarity index measure (SSIM) of 0.913 (SD 0.03), yielding a statistically significant improvement of 0.041 compared to leading alternatives.

Cite This Study

Slimi et al. (2026) studied this question.

synapsesocial.com/papers/6a70c8ba35aa2c282ce21b89 https://doi.org/10.1615/critrevbiomedeng.2026060806

Also Consider

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

View Full Paper

Ask AI

Bookmark