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

Computed Tomography01:10

Computed Tomography

Tomography refers to imaging by sections. Computed tomography (CT) is a non-invasive imaging technique that uses computers to analyze several cross-sectional X-rays to reveal minute details about structures in the body.
The technique was invented in the 1970s and is based on the principle that as X-rays pass through the body, they are absorbed or reflected at different levels. In the technique, a patient lies on a motorized platform while a computerized axial tomography (CAT) scanner rotates...

You might also read

Related Articles

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

Sort by
Same author

Enhanced Deep Neural Network for Prostate Segmentation in Micro-Ultrasound Images.

Sensors (Basel, Switzerland)·2025
Same author

Improved UNet with Attention for Medical Image Segmentation.

Sensors (Basel, Switzerland)·2023
Same author

Wave Equation Modeling via Physics-Informed Neural Networks: Models of Soft and Hard Constraints for Initial and Boundary Conditions.

Sensors (Basel, Switzerland)·2023
Same author

DeepUCT: Complex cascaded deep learning network for improved ultrasound tomography.

Physics in medicine and biology·2022

Related Experiment Video

Updated: May 14, 2026

An Experimental Protocol for Assessing the Performance of New Ultrasound Probes Based on CMUT Technology in Application to Brain Imaging
16:01

An Experimental Protocol for Assessing the Performance of New Ultrasound Probes Based on CMUT Technology in Application to Brain Imaging

Published on: September 24, 2017

CUCT-Net: End-to-End Signal-to-Image Learning for Quantized Speed-of-Sound Estimation and Tissue Segmentation in

Qinhan Gao1, Mohamed Khaled Almekkawy1

  • 1School of Electrical Engineering and Computer Science, Penn State University, University Park, PA 16802, USA.

Sensors (Basel, Switzerland)
|May 13, 2026
PubMed
Summary

A new deep learning framework, CUCT-Net, offers efficient Ultrasound Computed Tomography (UCT) by directly mapping ultrasound signals to speed-of-sound (SoS) and tissue images. This method reduces computational costs and improves robustness for medical imaging applications.

Keywords:
convolutional neural networkdeep learningsignal processingultrasound computed tomography

More Related Videos

Hybrid µCT-FMT imaging and image analysis
13:45

Hybrid µCT-FMT imaging and image analysis

Published on: June 4, 2015

Pipeline for Planning and Execution of Transcranial Ultrasound Neuromodulation Experiments in Humans
07:52

Pipeline for Planning and Execution of Transcranial Ultrasound Neuromodulation Experiments in Humans

Published on: June 28, 2024

Related Experiment Videos

Last Updated: May 14, 2026

An Experimental Protocol for Assessing the Performance of New Ultrasound Probes Based on CMUT Technology in Application to Brain Imaging
16:01

An Experimental Protocol for Assessing the Performance of New Ultrasound Probes Based on CMUT Technology in Application to Brain Imaging

Published on: September 24, 2017

Hybrid µCT-FMT imaging and image analysis
13:45

Hybrid µCT-FMT imaging and image analysis

Published on: June 4, 2015

Pipeline for Planning and Execution of Transcranial Ultrasound Neuromodulation Experiments in Humans
07:52

Pipeline for Planning and Execution of Transcranial Ultrasound Neuromodulation Experiments in Humans

Published on: June 28, 2024

Area of Science:

  • Medical Imaging
  • Computational Ultrasound
  • Deep Learning

Background:

  • Traditional Ultrasound Computed Tomography (UCT) methods, like Full Waveform Inversion (FWI), are computationally intensive and sensitive to acoustic contrasts.
  • Accurate speed-of-sound (SoS) estimation and tissue segmentation are crucial for UCT but challenging with conventional techniques.

Purpose of the Study:

  • To introduce the Multi-Channel Transducer Network (CUCT-Net), a deep learning framework for end-to-end UCT.
  • To enable direct quantized SoS estimation and tissue-level segmentation from raw ultrasound measurements, overcoming limitations of traditional FWI.

Main Methods:

  • CUCT-Net employs a multi-input encoder-decoder architecture for direct mapping from multi-static UCT measurements to image-space outputs.
  • The network utilizes Shift Units (SU) for enhanced fine-scale feature modeling, particularly under sparse sensing conditions.
  • Experiments were conducted using k-Wave simulations on various phantoms, including anatomical brain models, under both clean and noisy conditions.

Main Results:

  • The CUCT-Net achieved accurate quantized SoS estimation and direct tissue segmentation in simulations.
  • The framework demonstrated strong robustness to noise, enhanced by transfer learning.
  • CUCT-Net significantly reduced computational cost compared to FWI, maintaining stable performance with fewer sensors and complex tissue variations.

Conclusions:

  • CUCT-Net effectively reformulates UCT as a direct signal-to-image learning problem.
  • The end-to-end deep learning approach bypasses iterative inversion, offering efficient and robust performance for UCT.
  • The multi-input architecture facilitates information integration from multiple transducers, highlighting the potential of data-driven UCT for SoS estimation and tissue segmentation.