Hybrid-Collaborative Noise2Noise Denoiser for Low-Dose CT Images

Hasan, Ahmed M.; Mohebbian, Mohammad Reza; Wahid, Khan A.; Babyn, Paul

doi:10.1109/trpms.2020.3002178

Cited by 34 publications

(14 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…31 A vast field of MLDs models and architectures have been established in the last decade including convolutional neural net (CNN) denoisers, [32][33][34][35] discrete CCN denoiser, 23,36,37 adaptive image denoiser, [38][39][40] least-squares denoiser, 41 multi-level wavelet convolution, 42 and tree-adapted wavelet shrinkage. 43 Specific approaches for medical CT imaging 44 and photoplethysmography (PPG) imaging 45 are complemented by denoising hyperspectral images for earth observation and target detection. 46 Fast blind image denoiser, 47,48 plug-and-play 49 and non-blind restoration 50 are further alternatives to the aforementioned techniques.…”

Section: Related Workmentioning

confidence: 99%

Adversarial attacks for machine learning denoisers and how to resisit them

Jain

Zongru²,

Veettil³

et al. 2022

Emerging Topics in Artificial Intelligence (ETAI) 2022

View full text Add to dashboard Cite

Adversarial attacks rely on the instability phenomenon appearing in general for all inverse problems, e.g., image classification and reconstruction, independently of the computational scheme or method used to solve the problem. We mathematically prove and empirically show that machine learning denoisers (MLD) are not excluded. That is to prove the existence of adversarial attacks given by noise patterns making the MLD run into instability, i.e., the MLD increases the noise instead of decreasing it. We further demonstrate that adversarial retraining or classic filtering do not provide an exit strategy for this dilemma. Instead, we show that adversarial attacks can be inferred by polynomial regression. Removing the underlying inferred polynomial distribution from the total noise distribution delivers an efficient technique yielding robust MLDs that make consistent computer vision tasks such as image segmentation or classification more reliable.

show abstract

Section: Related Workmentioning

confidence: 99%

Adversarial attacks for machine learning denoisers and how to resisit them

Jain

Zongru²,

Veettil³

et al. 2022

Emerging Topics in Artificial Intelligence (ETAI) 2022

View full text Add to dashboard Cite

show abstract

“…The structurally sensitive loss function is a hybrid that integrate the best characteristics of mean and feature-based methods, thus, allowing noise and artifacts suppression while preserving structure and texture [ 120 ]. Furthermore, noise2noise (N2N) methods have also been propose to denoise low-dose CT images; N2N methods reduce noise using pairs of noisy images [ 121 ]. Hasan et al [ 121 ] explored the potential collaboration between N2N generators for denoising of CT images by processing images from a phantom, and showed that these collaborative generators outperformed the common N2N method.…”

Section: Overview Of Deep Learning Applications In Medical Imagingmentioning

confidence: 99%

“…Furthermore, noise2noise (N2N) methods have also been propose to denoise low-dose CT images; N2N methods reduce noise using pairs of noisy images [ 121 ]. Hasan et al [ 121 ] explored the potential collaboration between N2N generators for denoising of CT images by processing images from a phantom, and showed that these collaborative generators outperformed the common N2N method. For denoising of single-channel CT images, an unsupervised learning approach known as REDAEP was introduced by Zhang et al in which variable-augmented denoising AEs were used to train higher-dimensional prior for the iterative reconstruction—this technique was tested using simulated and real data [ 122 ].…”

Section: Overview Of Deep Learning Applications In Medical Imagingmentioning

confidence: 99%

Application and Construction of Deep Learning Networks in Medical Imaging

Torres-Velázquez

Chen

et al. 2021

IEEE Trans. Radiat. Plasma Med. Sci.

View full text Add to dashboard Cite

Deep learning (DL) approaches are part of the machine learning (ML) subfield concerned with the development of computational models to train artificial intelligence systems. DL models are characterized by automatically extracting high-level features from the input data to learn the relationship between matching datasets. Thus, its implementation offers an advantage over common ML methods that often require the practitioner to have some domain knowledge of the input data to select the best latent representation. As a result of this advantage, DL has been successfully applied within the medical imaging field to address problems such as disease classification and tumor segmentation for which it is difficult or impossible to determine which image features are relevant. Therefore, taking into consideration the positive impact of DL on the medical imaging field, this paper reviews the key concepts associated with its evolution and implementation. The sections of this review summarize the milestones related to the development of the DL field, followed by a description of the elements of deep neural network and an overview of its application within the medical imaging field. Subsequently, the key steps necessary to implement a supervised DL application are defined, and associated limitations are discussed.

show abstract

“…Nevertheless, Hasan et al [8] made early attempts with Noise2Noise [13] for CT denoising. Here, several generators networks called hybrid-collaborative generators were proposed to predict NDCT from many LDCT pairs.…”

Section: Related Workmentioning

confidence: 99%

Self-Supervised Learning based CT Denoising using Pseudo-CT Image Pairs

Won,

Jung,

et al. 2021

Preprint

View full text Add to dashboard Cite

Recently, Self-supervised learning methods able to perform image denoising without ground truth labels have been proposed. These methods create low-quality images by adding random or Gaussian noise to images and then train a model for denoising. Ideally, it would be beneficial if one can generate high-quality CT images with only a few training samples via self-supervision. However, the performance of CT denoising is generally limited due to the complexity of CT noise. To address this problem, we propose a novel self-supervised learning-based CT denoising method. In particular, we train pre-train CT denoising and noise models that can predict CT noise from Low-dose CT (LDCT) using available LDCT and Normal-dose CT (NDCT) pairs. For a given test LDCT, we generate Pseudo-LDCT and NDCT pairs using the pre-trained denoising and noise models and then update the parameters of the denoising model using these pairs to remove noise in the test LDCT. To make realistic Pseudo LDCT, we train multiple noise models from individual images and generate the noise using the ensemble of noise models. We evaluate our method on the 2016 AAPM Low-Dose CT Grand Challenge dataset. The proposed ensemble noise model can generate realistic CT noise, and thus our method significantly improves the denoising performance existing denoising models trained by supervised-and self-supervised learning.

show abstract

Hybrid-Collaborative Noise2Noise Denoiser for Low-Dose CT Images

Cited by 34 publications

References 36 publications

Adversarial attacks for machine learning denoisers and how to resisit them

Adversarial attacks for machine learning denoisers and how to resisit them

Application and Construction of Deep Learning Networks in Medical Imaging

Self-Supervised Learning based CT Denoising using Pseudo-CT Image Pairs

Contact Info

Product

Resources

About