Conditional Normalizing Flows for Low-Dose Computed Tomography Image Reconstruction

Denker, Alexander; Schmidt, Maximilian; Leuschner, Johannes; Maaß, Peter; Behrmann, Jens

doi:10.48550/arxiv.2006.06270

Cited by 4 publications

(4 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Data-driven regularized inversion methods for solving inverse problems in imaging have recently had great success in terms of reconstruction quality [6]. Three main classes of methods are: end-to-end learned methods [1,3,8,21,28,46], learned regularizers [34,37] and generative networks [2,7,13]. For the study described in this paper, we only focus on the end-to-end learned methods.…”

Section: Related Approaches and Motivationmentioning

confidence: 99%

Computed tomography reconstruction using deep image prior and learned reconstruction methods

2020

View full text Add to dashboard Cite

In this paper we describe an investigation into the application of deep learning methods for low-dose and sparse angle computed tomography using small training datasets. To motivate our work we review some of the existing approaches and obtain quantitative results after training them with different amounts of data. We find that the learned primal-dual method has an outstanding performance in terms of reconstruction quality and data efficiency. However, in general, end-to-end learned methods have two deficiencies: (a) a lack of classical guarantees in inverse problems and (b) the lack of generalization after training with insufficient data. To overcome these problems, we introduce the deep image prior approach in combination with classical regularization and an initial reconstruction. The proposed methods achieve the best results in the low-data regime in three challenging scenarios.

show abstract

Section: Related Approaches and Motivationmentioning

confidence: 99%

Computed tomography reconstruction using deep image prior and learned reconstruction methods

2020

View full text Add to dashboard Cite

show abstract

“…To date, Normalizing Flows have seen less adoption in medical and cancer imaging than GANs, but promising initial applications exist. For example, Normalizing Flows have been proposed for uncertainty estimation of lung lesion segmentation (Selvan et al, 2020), counterfactual inference on brain MRI (Pawlowski et al, 2020), and low-dose CT image reconstruction (Denker et al, 2020).…”

Section: Gan Alternatives and Complementary Methodsmentioning

confidence: 99%

Data synthesis and adversarial networks: A review and meta-analysis in cancer imaging

Osuala

Kushibar

Garrucho

et al. 2023

Medical Image Analysis

View full text Add to dashboard Cite

“…However, this process is challenging due to the limited and noisy information used to determine the original image, leading to structured uncertainty and correlations between nearby pixels in the reconstructed image [66]. To overcome this issue, current research in uncertainty quantification of inverse problems employs conditional deep generative models, such as cVAE, cGAN, and conditional normalizing flow models [2,29,31]. These methods utilize a lowdimensional latent space for image generation but may overlook unique data characteristics, such as structural constraints from domain physics in certain types of image data, such as remote sensing images, MRI images, or geological subsurface images [60,124,127].…”

Section: 21mentioning

confidence: 99%

A Survey on Uncertainty Quantification Methods for Deep Neural Networks: An Uncertainty Source Perspective

He¹,

Zhe²

2023

Preprint

View full text Add to dashboard Cite

Deep neural networks (DNNs) have achieved tremendous success in making accurate predictions for computer vision, natural language processing, as well as science and engineering domains. However, it is also well-recognized that DNNs sometimes make unexpected, incorrect, but overconfident predictions. This can cause serious consequences in high-stake applications, such as autonomous driving, medical diagnosis, and disaster response. Uncertainty quantification (UQ) aims to estimate the confidence of DNN predictions beyond prediction accuracy. In recent years, many UQ methods have been developed for DNNs. It is of great practical value to systematically categorize these UQ methods and compare their advantages and disadvantages. However, existing surveys mostly focus on categorizing UQ methodologies from a neural network architecture's perspective or a Bayesian perspective and ignore the source of uncertainty that each methodology can incorporate, making it difficult to select an appropriate UQ method in practice. To fill the gap, this paper presents a systematic taxonomy of UQ methods for DNNs based on the types of uncertainty sources (data uncertainty versus model uncertainty). We summarize the advantages and disadvantages of methods in each category. We show how our taxonomy of UQ methodologies can potentially help guide the choice of UQ method in different machine learning problems (e.g., active learning, robustness, and reinforcement learning). We also identify current research gaps and propose several future research directions.

show abstract

Conditional Normalizing Flows for Low-Dose Computed Tomography Image Reconstruction

Cited by 4 publications

References 20 publications

Computed tomography reconstruction using deep image prior and learned reconstruction methods

Computed tomography reconstruction using deep image prior and learned reconstruction methods

Data synthesis and adversarial networks: A review and meta-analysis in cancer imaging

A Survey on Uncertainty Quantification Methods for Deep Neural Networks: An Uncertainty Source Perspective

Contact Info

Product

Resources

About