Updated: May 11, 2026

Clinical Imaging of Microwave Mammography
Published on: November 14, 2025
Maya Aviv1, Eran Gur, Zeev Zalevsky
1Bar-Ilan University, Faculty of Engineering, Ramat-Gan 52900, Israel. maya.shalev@gmail.com
Researchers tested a new image processing method to see through biological tissues like skin. By capturing multiple images at different depths, they successfully reconstructed clear pictures of objects hidden behind scattering layers.
You might also read
Articles linked to this work by shared authors, journal, and citation graph.
Area of Science:
Background:
Biological tissues often degrade optical clarity by scattering light, which obscures underlying structures. No prior work had fully resolved how to maintain high resolution when imaging through static random perturbation media. That uncertainty drove interest in computational techniques for recovering lost visual information. It was already known that traditional photography fails to penetrate these complex, non-homogeneous layers effectively. Prior research has shown that phase retrieval methods offer potential for correcting such distortions. This gap motivated the development of specialized algorithms to improve visibility in medical diagnostics. Scientists frequently struggle to distinguish deep-seated features from the surrounding noise of soft tissue. That challenge persists because light paths become randomized before reaching the detector.
Purpose Of The Study:
The aim of this study is to present experimental results of a revised deblurring approach for imaging objects behind static random perturbation media. Researchers sought to address the significant reduction in image quality caused by scattering biological tissues. They specifically targeted the problem of visualizing structures like bones located behind skin or flesh. The motivation for this work stems from the need to improve diagnostic clarity in non-transparent environments. By adapting iterative computation, the team intended to recover high spatial frequencies lost during light propagation. This project explores whether phase retrieval techniques can mitigate the distortions inherent in soft tissue imaging. The authors focused on developing a practical method that utilizes multiple focal planes to reconstruct hidden scenes. They aimed to provide a robust computational solution for overcoming the challenges posed by complex, non-homogeneous scattering layers.
The researchers propose that the technique functions by capturing multiple images at varying focal planes. This allows the algorithm to extract high spatial frequencies and reconstruct the complex field, which is then numerically propagated to the target's estimated location to reveal the hidden object.
The team utilizes a modified version of the Gerchberg-Saxton algorithm. This iterative computational tool is specifically adapted to retrieve phase distributions from the captured focal planes, enabling the correction of distortions caused by the random perturbation of the medium.
The researchers state that focusing the camera onto three or more distinct planes between the detector and the scattering layer is necessary. This multi-plane acquisition provides the required data depth to accurately retrieve the phase distribution for subsequent reconstruction.
Main Methods:
Review Approach involved testing a newly developed deblurring technique on objects hidden behind scattering barriers. The investigators utilized an iterative computational framework inspired by established phase retrieval procedures. They captured intensity data by focusing the camera on at least three distinct planes. These planes were positioned strategically between the imaging sensor and the random perturbation layer. The team then processed the recorded intensity values to extract the phase distribution of each plane. Numerical free-space propagation was applied to the recovered complex field to estimate the object's true intensity. This mathematical transformation allowed the researchers to reconstruct the target's appearance at its specific depth. The entire procedure was validated through experimental trials to confirm the efficacy of the modified iterative approach.
Main Results:
Key Findings From the Literature show that the proposed iterative method successfully retrieves high spatial frequencies from scenes obscured by static scattering media. The researchers achieved improved image reconstruction by focusing on three or more planes located between the sensor and the perturbation. Their results indicate that the complex field can be accurately extracted using this multi-plane acquisition strategy. By numerically propagating this field, they successfully visualized objects positioned behind the scattering layer. The data confirms that the iterative computation effectively mimics phase retrieval principles to overcome tissue-induced blurring. This approach provides a clear improvement over traditional imaging methods that fail to account for random light scattering. The experimental evidence supports the feasibility of using this computational model to see through soft tissue. These findings highlight the potential of phase-based reconstruction for enhancing visibility in complex biological environments.
Conclusions:
The authors demonstrate that their modified iterative technique successfully recovers high spatial frequency data from obscured scenes. Synthesis and Implications suggest this computational framework effectively mitigates the blurring effects caused by static scattering layers. The researchers propose that capturing multiple focal planes provides sufficient information to reconstruct the hidden object's intensity. Their findings indicate that numerical propagation of the complex field allows for accurate localization of the target. This approach offers a viable path for enhancing imaging quality in scenarios involving biological barriers. The study confirms that phase retrieval principles can be adapted for complex, non-transparent environments. Future applications might benefit from the ability to see through skin or flesh without invasive procedures. The team concludes that their method provides a robust solution for retrieving visual data from degraded optical signals.
The camera serves as the primary data acquisition device, recording the intensity patterns at specified depths. These intensity measurements act as the input for the iterative phase retrieval process, which ultimately allows for the recovery of the hidden object's spatial information.
The study measures the intensity of the object after numerical free-space propagation. This phenomenon relies on the extracted complex field, which is mathematically shifted to the object's estimated position to produce a clearer, deblurred representation of the target.
The authors propose that this approach significantly improves image quality when viewing objects behind static random perturbation media. They suggest that this method could be applied to various biological applications where soft tissue obscures underlying structures like bones.