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

Imaging Studies II: Ultrasonography01:24

Imaging Studies II: Ultrasonography

IntroductionUltrasonography, or renal ultrasound, is a noninvasive medical imaging technique that uses high-frequency sound waves to visualize the kidneys, ureters, bladder, and surrounding tissues.Indications for Urinary System UltrasonographyUrinary system ultrasonography is indicated in various clinical scenarios, such as:Kidney Stones (Urolithiasis): To detect and monitor the size and presence of kidney or urinary tract stones.Hydronephrosis: To assess the dilation of the renal pelvis and...

You might also read

Related Articles

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

Sort by
Same author

Mapping Microvascular Flow via Radon Transform Ultrasound: Technical Advances and Pilot Application.

BME frontiers·2026
Same author

Thermal and imaging effects of Feraheme in MR-guided focused ultrasound: a phantom study.

Physics in medicine and biology·2026
Same author

Kidney Stone Research in the Rochester Epidemiology Project.

Mayo Clinic proceedings·2026
Same author

Early Experience With Beta-Lactam Therapeutic Drug Monitoring in an Outpatient Parenteral Antimicrobial Therapy (OPAT) Program.

Open forum infectious diseases·2026
Same author

Accuracy and precision of automated kidney stone detection on CT.

Abdominal radiology (New York)·2026
Same author

Fast 3D Ultrasound Localization Microscopy via Projection-based Processing Framework.

IEEE transactions on medical imaging·2026
Same journal

PF-DAformer: Proximal Femur Segmentation via Domain Adaptive Transformer for Dual-Center QCT.

Biomedical signal processing and control·2026
Same journal

Reconstructing 12-lead ECG from reduced lead sets using an encoder-decoder convolutional neural network.

Biomedical signal processing and control·2026
Same journal

Explainable artificial intelligence in electrocardiography: A systematic review.

Biomedical signal processing and control·2026
Same journal

SeRL: Style-embedding representation learning for unsupervised CT images synthesis from unpaired MR images.

Biomedical signal processing and control·2026
Same journal

CNN-Autoformer: Automated EEG-Based Seizure Detection and Localization Using Hybrid Deep Learning.

Biomedical signal processing and control·2025
Same journal

Generalizable Multimodal Retinal Image Registration via Label-free Vessel Segmentation.

Biomedical signal processing and control·2025
See all related articles
  1. Home
  3. Self-supervised Deep Learning For Denoising In Ultrasound Microvascular Imaging.
  1. Home
  3. Self-supervised Deep Learning For Denoising In Ultrasound Microvascular Imaging.

Related Experiment Video

High-Speed Ultraviolet Photoacoustic Microscopy for Histological Imaging with Virtual-Staining assisted by Deep Learning
09:31

High-Speed Ultraviolet Photoacoustic Microscopy for Histological Imaging with Virtual-Staining assisted by Deep Learning

Published on: April 28, 2022

Self-supervised Deep Learning for Denoising in Ultrasound Microvascular Imaging.

Lijie Huang1, Jingyi Yin1, Jingke Zhang1

  • 1Department of Radiology, Mayo Clinic College of Medicine and Science, Mayo Clinic, Rochester, MN, USA.

Biomedical Signal Processing and Control
|May 18, 2026

View abstract on PubMed

Summary
This summary is machine-generated.

This article introduces a new artificial intelligence method called Half-Angle-to-Half-Angle (HA2HA) to improve the clarity of ultrasound images of tiny blood vessels. By training the system to recognize consistent blood flow patterns while ignoring random background noise, the researchers successfully enhanced image quality in both animal and human organ scans without needing manual labels.

Keywords:
DopplerNoise2NoiseSelf-supervised deep learningdenoisingradio-frequencyultrasound microvascular imagingradio-frequency signal processingcontrast-to-noise ratiovascular quantificationcolor Doppler imagingimage denoising

Frequently Asked Questions

More Related Videos

Two-Dimensional Super-Resolution Visualization of Rat Brain Microvasculature Using Ultrasound Localization Microscopy
07:26

Two-Dimensional Super-Resolution Visualization of Rat Brain Microvasculature Using Ultrasound Localization Microscopy

Published on: March 28, 2025

Related Experiment Videos

High-Speed Ultraviolet Photoacoustic Microscopy for Histological Imaging with Virtual-Staining assisted by Deep Learning
09:31

High-Speed Ultraviolet Photoacoustic Microscopy for Histological Imaging with Virtual-Staining assisted by Deep Learning

Published on: April 28, 2022

Two-Dimensional Super-Resolution Visualization of Rat Brain Microvasculature Using Ultrasound Localization Microscopy
07:26

Two-Dimensional Super-Resolution Visualization of Rat Brain Microvasculature Using Ultrasound Localization Microscopy

Published on: March 28, 2025

Area of Science:

  • Biomedical engineering research within Self-supervised Deep Learning for medical imaging
  • Diagnostic radiology and vascular physiology studies

Background:

Ultrasound microvascular imaging provides noninvasive insights into physiological processes but faces significant limitations regarding image clarity. That uncertainty drove researchers to seek better ways to handle low signal-to-noise ratios in deep tissue. Prior research has shown that unfocused plane wave imaging often produces noisy data that complicates clinical diagnosis. No prior work had resolved the difficulty of maintaining vascular detail while removing background interference in contrast-free scenarios. This gap motivated the development of automated frameworks capable of processing radio-frequency signals directly. Traditional approaches frequently struggle to balance noise reduction with the preservation of delicate microvascular structures. It was already known that deep learning models usually require large labeled datasets for effective training. This study addresses the need for a label-free technique that functions across diverse clinical imaging environments.

Purpose Of The Study:

The researchers propose the Half-Angle-to-Half-Angle framework, which constructs training pairs from complementary angular subsets of radio-frequency data. This mechanism exploits the consistency of vascular signals across angles while treating noise as a variable component to be filtered out during the self-supervised training process.

The study utilizes radio-frequency blood flow data as the primary input for the neural network. This raw signal format allows the model to perform denoising before the final image reconstruction, which benefits secondary processes like color Doppler imaging.

The authors state that beamformed radio-frequency data is necessary because it preserves the raw signal characteristics required for the model to distinguish between consistent vascular flow and random noise. This domain-specific input enables the network to learn features that are lost in standard image-based processing.

The researchers use complementary angular subsets of the beamformed data to create training pairs. This component acts as the supervisory signal, allowing the network to learn the underlying structure of blood flow without needing human-provided labels or ground-truth images.

The team measured a maximum contrast-to-noise ratio improvement of approximately 10 decibels in contrast-enhanced pig kidney data. This specific metric quantifies the significant enhancement in image quality achieved by the model compared to conventional imaging techniques.

The authors propose that their label-free approach provides a generalizable solution for robust vascular imaging. They suggest that this method could facilitate more reliable disease diagnosis by overcoming the limitations of low signal-to-noise ratios in deep tissue scenarios.

The study aims to introduce the Half-Angle-to-Half-Angle framework as a novel solution for denoising in ultrasound microvascular imaging. Researchers sought to overcome the persistent challenge of low signal-to-noise ratios in unfocused plane wave imaging. This limitation often hinders the accurate quantification of vascular structures and impairs reliable disease diagnosis. The authors were motivated by the need for a label-free method that functions effectively in deep tissue environments. They aimed to demonstrate that self-supervised learning could extract high-quality vascular signals from raw radio-frequency data. The team specifically addressed the difficulty of maintaining consistent flow information while removing random noise artifacts. This research was driven by the goal of providing a generalizable tool for both contrast-free and contrast-enhanced clinical scenarios. The investigators intended to validate their framework across diverse biological datasets to ensure broad applicability.

Main Methods:

The researchers developed a self-supervised training strategy to process beamformed radio-frequency blood flow signals. Their review approach involved constructing training pairs from complementary angular subsets of the acquired data. This design ensures that vascular signals remain consistent across the subsets while noise varies independently. The team trained the model using in-vivo data obtained from porcine kidney samples. They evaluated the performance of the framework on independent test datasets collected from unseen subjects. These test sets included both contrast-free and contrast-enhanced samples from human liver and kidney tissues. The authors compared the output of their model against conventional imaging methods to validate the enhancement. This systematic evaluation confirms the ability of the framework to handle diverse clinical scenarios without manual supervision.

Main Results:

The HA2HA framework achieved a maximum contrast-to-noise ratio improvement of approximately 10 decibels in contrast-enhanced pig kidney data. This finding indicates a substantial enhancement in the overall quality of the generated images. The researchers observed that the model effectively suppresses noisy backgrounds in human liver color Doppler imaging results. These improved images exhibited clearer visualization of microvascular flow compared to standard techniques. The model demonstrated consistent performance across both contrast-free and contrast-enhanced imaging scenarios. The authors noted that denoising directly in the radio-frequency domain benefits various downstream vascular quantification processes. These results highlight the versatility of the proposed method for different types of ultrasound data. The study confirms that the approach provides a robust and generalizable solution for clinical vascular imaging.

Conclusions:

The researchers propose that the HA2HA framework provides a robust solution for enhancing vascular visualization in various clinical settings. Their findings suggest that denoising directly within the radio-frequency domain supports multiple downstream diagnostic processes. The authors indicate that this approach successfully improves microvascular flow visibility while suppressing background artifacts. They highlight that the method remains effective for both contrast-free and contrast-enhanced imaging modalities. The study demonstrates that the model generalizes well across different subjects, including both porcine and human tissues. The authors conclude that their technique offers a practical alternative to conventional methods requiring extensive manual annotation. This work implies that self-supervised learning can significantly elevate the standard of ultrasound-based diagnostic imaging. The team maintains that their strategy facilitates more reliable vascular quantification for future medical applications.