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

Smart Sensor-Driven Gait Rehabilitation Walker Using Machine Learning for Predictive Home-Based Therapy.

Sensors (Basel, Switzerland)·2026
Same author

Analyzing Orbital Angular Momentum (OAM) mode via localized beam sampling.

Scientific reports·2026
Same author

Roadmap on singular optics and its applications.

Applied physics. B, Lasers and optics·2026
Same author

Decrypting chaotic visual ciphers via quasi quantum neural networks (Q²NNs).

Scientific reports·2026
Same author

Sensing Through Tissues Using Diffuse Optical Imaging and Genetic Programming.

Sensors (Basel, Switzerland)·2026
Same author

Quantum neural network-based compensation of distorted orbital angular momentum beams in complex media.

Scientific reports·2025

Related Experiment Video

Updated: Jul 23, 2025

Digital Inline Holographic Microscopy DIHM of Weakly-scattering Subjects
10:16

Digital Inline Holographic Microscopy DIHM of Weakly-scattering Subjects

Published on: February 8, 2014

12.3K

Imaging inside highly scattering media using hybrid deep learning and analytical algorithm.

Ben Wiesel1, Shlomi Arnon1

  • 1Ben-Gurion University of the Negev, Department of Electrical and Computer Engineering, Beer-Sheva, Israel.

Journal of Biophotonics
|July 11, 2023
PubMed
Summary

We developed Hybrid-DOT, a novel imaging method combining analytical and deep learning techniques. This approach significantly improves image quality and resolution when imaging through scattering media, requiring less training data.

Keywords:
computational imagingdeep-learning-DOTdiffuse tomographytime-of-flight

More Related Videos

Simultaneous Evaluation of Cerebral Hemodynamics and Light Scattering Properties of the In Vivo Rat Brain Using Multispectral Diffuse Reflectance Imaging
07:06

Simultaneous Evaluation of Cerebral Hemodynamics and Light Scattering Properties of the In Vivo Rat Brain Using Multispectral Diffuse Reflectance Imaging

Published on: May 7, 2017

7.7K
Measuring Spatially- and Directionally-varying Light Scattering from Biological Material
11:57

Measuring Spatially- and Directionally-varying Light Scattering from Biological Material

Published on: May 20, 2013

13.6K

Related Experiment Videos

Last Updated: Jul 23, 2025

Digital Inline Holographic Microscopy DIHM of Weakly-scattering Subjects
10:16

Digital Inline Holographic Microscopy DIHM of Weakly-scattering Subjects

Published on: February 8, 2014

12.3K
Simultaneous Evaluation of Cerebral Hemodynamics and Light Scattering Properties of the In Vivo Rat Brain Using Multispectral Diffuse Reflectance Imaging
07:06

Simultaneous Evaluation of Cerebral Hemodynamics and Light Scattering Properties of the In Vivo Rat Brain Using Multispectral Diffuse Reflectance Imaging

Published on: May 7, 2017

7.7K
Measuring Spatially- and Directionally-varying Light Scattering from Biological Material
11:57

Measuring Spatially- and Directionally-varying Light Scattering from Biological Material

Published on: May 20, 2013

13.6K

Area of Science:

  • Optical imaging
  • Biomedical optics
  • Computational imaging

Background:

  • Imaging through highly scattering media is crucial for biomedical and remote-sensing applications.
  • Current methods using analytical or deep learning approaches face limitations like blurry images and extensive data requirements.

Purpose of the Study:

  • To introduce Hybrid-DOT, a hybrid imaging scheme combining analytical estimates and deep learning.
  • To overcome the limitations of existing methods for imaging in scattering environments.

Main Methods:

  • A hybrid scheme (Hybrid-DOT) integrating analytically derived image estimates with a deep learning network.
  • Comparative analysis against state-of-the-art Time-of-Flight Diffuse Optical Tomography (ToF-DOT) and standalone deep learning models.

Main Results:

  • Hybrid-DOT improved Peak Signal-to-Noise Ratio (PSNR) by 4.6 dB and reduced resolution by 2.5x compared to ToF-DOT.
  • Compared to deep learning models, Hybrid-DOT achieved a 0.8 dB PSNR increase, 1.5x better resolution, and reduced dataset needs by 1.6-3x.
  • The method maintained effectiveness at depths up to 160 mean-free paths.

Conclusions:

  • Hybrid-DOT offers a superior solution for imaging through scattering media.
  • The hybrid approach balances analytical rigor with deep learning efficiency, enhancing image quality and reducing data dependency.
  • This technique shows promise for advanced applications in deep tissue imaging and remote sensing.