An Educated Warm Start For Deep Image Prior-Based Micro CT Reconstruction

Barbano, Riccardo; Leuschner, Johannes; Schmidt, Maximilian; Denker, Alexander; Hauptmann, Andreas; Maaß, Peter; Jin, Bangti

doi:10.48550/arxiv.2111.11926

Cited by 3 publications

(7 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The main drawback of the DIP is that it must be optimised separately for each new measurement y δ . This can be mitigated by pretraining, as demonstrated by Barbano et al (2021b) and Knopp & Grosser (2021).…”

Section: The Deep Image Priormentioning

confidence: 99%

“…Since its first proposal (Ulyanov et al, 2018;, DIP has been widely applied and improved in several aspects, e.g. with early stopping (Liu et al, 2019;Baguer et al, 2020) or via pretraining (Barbano et al, 2021b;Knopp & Grosser, 2021). The present work builds on (Liu et al, 2019;Baguer et al, 2020) and provides a scalable probabilistic version of the DIP + TV approach which delivers uncertainty together with the reconstruction.…”

Section: Related Workmentioning

confidence: 99%

“…This allows us to leverage advances in DIP optimisation, e.g. TV regulariser (Baguer et al, 2020) and educated DIP (Barbano et al, 2021b). We use Linearised Laplace to approximate the finite width DIP with a GP prior over images, for which inference is tractable.…”

Section: Related Workmentioning

confidence: 99%

“…Figure 6. A schematic illustration of the U-Net architecture in the numerical experiments, taken from (Barbano et al, 2021b). Each light-blue box corresponds to a multi-channel feature map.…”

Section: A3 Setup For X-ray Walnut Data Experimentsmentioning

confidence: 99%

“…6. Following the approach in (Barbano et al, 2021b), we pretrain the network to perform post-processing of filtered back-projection (FBP) reconstructions on synthetic data. The dataset consists of pairs of images containing random ellipses, and corresponding FBPs from observations simulated according to (1) with 5% noise.…”

Section: A3 Setup For X-ray Walnut Data Experimentsmentioning

confidence: 99%

See 4 more Smart Citations

A Probabilistic Deep Image Prior for Computational Tomography

Antorán¹,

Barbano²,

Leuschner³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Existing deep-learning based tomographic image reconstruction methods do not provide accurate estimates of reconstruction uncertainty, hindering their real-world deployment. To address this limitation, we construct a Bayesian prior for tomographic reconstruction, which combines the classical total variation (TV) regulariser with the modern deep image prior (DIP). Specifically, we use a change of variables to connect our prior beliefs on the image TV semi-norm with the hyperparameters of the DIP network. For the inference, we develop an approach based on the linearised Laplace method, which is scalable to highdimensional settings. The resulting framework provides pixel-wise uncertainty estimates and a marginal likelihood objective for hyperparameter optimisation. We demonstrate the method on synthetic and real-measured high-resolution µCT data, and show that it provides superior calibration of uncertainty estimates relative to previous probabilistic formulations of the DIP.

show abstract

Section: The Deep Image Priormentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: A3 Setup For X-ray Walnut Data Experimentsmentioning

confidence: 99%

Section: A3 Setup For X-ray Walnut Data Experimentsmentioning

confidence: 99%

See 3 more Smart Citations

A Probabilistic Deep Image Prior for Computational Tomography

Antorán¹,

Barbano²,

Leuschner³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

PatchNR: learning from very few images by patch normalizing flow regularization

et al. 2023

View full text Add to dashboard Cite

Learning neural networks using only few available information is an important ongoing research topic with tremendous potential for applications. In this paper, we introduce a powerful regularizer for the variational modeling of inverse problems in imaging. Our regularizer, called patch normalizing flow regularizer (patchNR), involves a normalizing flow learned on small patches of very few images. In particular, the training is independent of the considered inverse problem such that the same regularizer can be applied for different forward operators acting on the same class of images. By investigating the distribution of patches versus those of the whole image class, we prove that our model is indeed a MAP approach. Numerical examples for low-dose and limited-angle computed tomography (CT) as well as superresolution of material images demonstrate that our method provides very high quality results. The training set consists of just six images for CT and one image for superresolution. Finally, we combine our patchNR with ideas from internal learning for performing superresolution of natural images directly from the low-resolution observation without knowledge of any high-resolution image.

show abstract

A Self-Supervised Deep Learning Reconstruction for Shortening the Breathhold and Acquisition Window in Cardiac Magnetic Resonance Fingerprinting

Hamilton

2022

Front. Cardiovasc. Med.

View full text Add to dashboard Cite

The aim of this study is to shorten the breathhold and diastolic acquisition window in cardiac magnetic resonance fingerprinting (MRF) for simultaneous T1, T2, and proton spin density (M0) mapping to improve scan efficiency and reduce motion artifacts. To this end, a novel reconstruction was developed that combines low-rank subspace modeling with a deep image prior, termed DIP-MRF. A system of neural networks is used to generate spatial basis images and quantitative tissue property maps, with training performed using only the undersampled k-space measurements from the current scan. This approach avoids difficulties with obtaining in vivo MRF training data, as training is performed de novo for each acquisition. Calculation of the forward model during training is accelerated by using GRAPPA operator gridding to shift spiral k-space data to Cartesian grid points, and by using a neural network to rapidly generate fingerprints in place of a Bloch equation simulation. DIP-MRF was evaluated in simulations and at 1.5 T in a standardized phantom, 18 healthy subjects, and 10 patients with suspected cardiomyopathy. In addition to conventional mapping, two cardiac MRF sequences were acquired, one with a 15-heartbeat(HB) breathhold and 254 ms acquisition window, and one with a 5HB breathhold and 150 ms acquisition window. In simulations, DIP-MRF yielded decreased nRMSE compared to dictionary matching and a sparse and locally low rank (SLLR-MRF) reconstruction. Strong correlation (R2 > 0.999) with T1 and T2 reference values was observed in the phantom using the 5HB/150 ms scan with DIP-MRF. DIP-MRF provided better suppression of noise and aliasing artifacts in vivo, especially for the 5HB/150 ms scan, and lower intersubject and intrasubject variability compared to dictionary matching and SLLR-MRF. Furthermore, it yielded a better agreement between myocardial T1 and T2 from 15HB/254 ms and 5HB/150 ms MRF scans, with a bias of −9 ms for T1 and 2 ms for T2. In summary, this study introduces an extension of the deep image prior framework for cardiac MRF tissue property mapping, which does not require pre-training with in vivo scans, and has the potential to reduce motion artifacts by enabling a shortened breathhold and acquisition window.

show abstract

An Educated Warm Start For Deep Image Prior-Based Micro CT Reconstruction

Cited by 3 publications

References 37 publications

A Probabilistic Deep Image Prior for Computational Tomography

A Probabilistic Deep Image Prior for Computational Tomography

PatchNR: learning from very few images by patch normalizing flow regularization

A Self-Supervised Deep Learning Reconstruction for Shortening the Breathhold and Acquisition Window in Cardiac Magnetic Resonance Fingerprinting

Contact Info

Product

Resources

About