Physics-informed neural networks for diffraction tomography

Saba, Amirhossein; Gigli, Carlo; Ayoub, Ahmed B.

doi:10.1117/1.ap.4.6.066001

Cited by 22 publications

(8 citation statements)

References 36 publications

(68 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To solve these reconstruction artifacts, convolutional neural networks have been trained in supervised or unsupervised settings, either using a fully-learned framework or in combination with a physical model [12][13][14][15][16][17]. Different approaches have been considered such as direct inversion, plug-and-play (PnP) priors, or physics-informed neural network (PINN) schemes.…”

Section: Introductionmentioning

confidence: 99%

“…Saba et al used a PINN framework as forward model for tomographic reconstructions, showing feasibility of light scattering prediction from standard adherent cell cultures. Their work, however, has not yet been extended to more complex 3D objects like embryos or multicellular spheroids [13]. By coupling a laser scanning confocal microscope with a quantitative phase imaging module, Chen et al used supervised learning to achieve confocal sectioning on label-free images of neurons and liver-cancer spheroids.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A CNN-based approach for 3D artefact correction of intensity diffraction tomography images

PIERRÉ,

BRIARD,

DESISSAIRE

et al. 2024

Preprint

View full text Add to dashboard Cite

3D reconstructions after tomographic imaging often suffer from elongation artifacts due to the limited-angle acquisitions. The retrieval of the original 3D shape is not an easy task, mainly due to the intrinsic morphological changes that biological objects undergo during their development. Here we present a novel approach for the correction of 3D artifacts after 3D reconstructions of intensity-only tomographic acquisitions. The method relies on a network architecture that combines a volumetric and a 3D finite object approach. The framework was applied on time-lapse images of a mouse preimplantation embryo developing from fertilization to the blastocyst stage, proving the correction of the axial elongation, the recovery of the spherical objects and improved quantitative refractive index values. This work paves the way for novel directions on a generalized non-supervised pipeline suited for different biological samples and imaging conditions.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A CNN-based approach for 3D artefact correction of intensity diffraction tomography images

PIERRÉ,

BRIARD,

DESISSAIRE

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Recently, computational imaging has become a research hotspot in optical field, especially phase retrieval [1][2][3][4] . Coherent diffraction imaging (CDI) [5,6] is a kind of phase retrieval technique using various iterative algorithms.…”

Section: Introductionmentioning

confidence: 99%

Linear mathematical model for the unique solution of 3D ptychographic iterative engine

Wu,

Qi,

Chang

et al. 2024

中国光学快报

View full text Add to dashboard Cite

Diffraction intensities of the 3D ptychographic iterative engine (3PIE) were written as a set of linear equations of the selfcorrelations of Fourier components of all sample slices, and an effective computing method was developed to solve these linear equations for the transmission functions of all sample slices analytically. With both theoretical analysis and numerical simulations, this study revealed the underlying physics and mathematics of 3PIE and demonstrated for the first time, to our knowledge, that 3PIE can generate mathematically unique reconstruction even with noisy data.

show abstract

“…The network corrects the errors of PDEs by adding them to the loss function and using gradient descent optimization. Saba et al predicted the scattered field accurately with PINN [19]. It has also been applied to retrieve the effective permittivity parameters in nano-optics [20], reconstruct the tomography of biologic samples [19], quantify the microstructure of polycrystalline [21].…”

Section: Introductionmentioning

confidence: 99%

“…Saba et al predicted the scattered field accurately with PINN [19]. It has also been applied to retrieve the effective permittivity parameters in nano-optics [20], reconstruct the tomography of biologic samples [19], quantify the microstructure of polycrystalline [21]. Additionally, Lu et al developed a Python library called DeepXDE that employs PINN to solve problems in computational science and engineering [22].…”

Section: Introductionmentioning

confidence: 99%

Ultrasonic guided wave imaging of pipelines based on physics embedded inversion neural network

Lv,

Chen,

Tong

et al. 2023

Meas. Sci. Technol.

View full text Add to dashboard Cite

Pipeline corrosion quantification plays a vital role in guaranteeing the safety of critical industrial structures and thus significant work has been carried out to address such an issue. Although quantitative imaging is crucial for non-destructive testing, research in guided wave pipeline testing has primarily centered on qualitative approaches. Here, we propose a deep neural network built upon physical model to reconstruct pipe wall thickness from ultrasonic guided wave (UGW) signals. The workflow of reconstruction contains three layers, where each layer consists of a fixed forward network and a residual inversion network. The forward model is represented by an agent convolutional neural network which would be embedded into the entire inversion network. The residuals between data from the forward model and real signals are then mapped into velocity profile differences through sub-inversion network. Numerical experiments were conducted to verify the inversion performance of the deep neural network using thickness maps obtained from guided wave frequency domain information. Results show that inversion images are capable to reveal the positions, shapes, and depths of corrosion with high resolution and precision, yielding an average inversion of 87.37% in the test set. In addition, by utilizing the periodicity of the pipeline, the inversion accuracy of eight pairs of transducers were improved from 67.7% to 89.43% with high-order helical guided wave. Compared with traditional high-precision inversion methods such as full waveform inversion, the proposed method achieved approximately 300 times faster inversion speed at the cost of some accuracy. The research demonstrates that real-time quantitative imaging of defects on pipes can be achieved accurately by physics embedded network. Furthermore, an experimental verification of the method was carried out through UGW pipeline testing, demonstrating its feasibility. The mean squared error of wall thickness reconstruction was 0.0070, achieving a high level of precision.

show abstract

Physics-informed neural networks for diffraction tomography

Cited by 22 publications

References 36 publications

A CNN-based approach for 3D artefact correction of intensity diffraction tomography images

A CNN-based approach for 3D artefact correction of intensity diffraction tomography images

Linear mathematical model for the unique solution of 3D ptychographic iterative engine

Ultrasonic guided wave imaging of pipelines based on physics embedded inversion neural network

Contact Info

Product

Resources

About