No-Reference Denoising Of Low-Dose Ct Projections

Elvira, Zainulina,; Chernyavskiy, Alexey; Dylov, Dmitry V.

doi:10.1109/isbi48211.2021.9433825

Cited by 8 publications

(6 citation statements)

References 12 publications

(11 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We observe that, for the proposed method, IGC is able to enhance the sharpness of the overblurred regions and suppress the streak artifacts near high contrast edges. For TD method, 12 IGC helps suppress more noise. We attribute this to the role of IGC for maintaining gradient between sequential projections, which suppresses noise and outliers and preserves sharpness.…”

Section: Resultsmentioning

confidence: 99%

“…These methods use the same network architecture shown in Figure 2 and adopt the MAE loss for model training. The time‐distributed model 12 in projection domain (TD‐PD) with IGC or without IGC is also investigated.…”

Section: Methodsmentioning

confidence: 99%

“…Noise2Context 11 pairs adjacent LD image slices for training, but the noise in reconstructed images is non‐independent, which does not meet the assumption of independent observations in Noise2Noise training. In a recent study, 12 an unsupervised learning model considering correlations between adjacent LD projections has been developed, which is similar to the idea of our study. However, the model excludes features of the central projection and sums up that of the rest ones in the second last layer.…”

Section: Introductionmentioning

confidence: 91%

See 2 more Smart Citations

Learning CT projection denoising from adjacent views

Hong

2022

Medical Physics

View full text Add to dashboard Cite

Background: Many learning-based low-dose (LD) computed tomography (CT) imaging methods require large paired full-and low-dose datasets for training, which are usually unavailable in clinic. Whereas models trained on simulated data often face the generalization problem on real clinical data. Purpose: To develop an unsupervised learning technique to acquire clean CT projection from its adjacent LD projections. Methods: Given a sequential LD projection set, the method extracts out the middle projection as the target and treats the rest ones as the input. The model is trained with the mean absolute error with proposed inter-view gradient constraint term, which helps to suppress outliers and preserve edges in the denoised projection. The simulated low-dose CT grand challenge dataset and a real physical torso phantom dataset were employed for experiment. The peak signal to noise ratio (PSNR) and structural similarity index measure (SSIM) were calculated for quantitative evaluation. Results:In experiments with both the simulated and real datasets, visual comparisons reveal that the proposed method obtained images superior to unsupervised and supervised methods working in both image and projection domain. For numerical comparison, our method obtains larger SSIMs than other unsupervised methods at quarter and eighth dose levels. As for PSNR, our method obtains larger value at eighth dose whereas smaller value at quarter dose. The supervised models obtain better numerical results than all unsupervised models on simulated datasets. Conclusion:The proposed method can reduce the noise in CT projections effectively, making it an attractive tool for practical LDCT pre-processing.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 91%

See 1 more Smart Citation

Learning CT projection denoising from adjacent views

Hong

2022

Medical Physics

View full text Add to dashboard Cite

show abstract

“…Inspired by previous work [16,22,21] and our understanding of CT imaging. We propose to employ a self-supervised learning model,Noise2Projection, to do low-dose CT image reconstruction using raw projection data.…”

Section: Introductionmentioning

confidence: 98%

“…As a step further, selfsupervised learning, which requires neither clean targets nor noisy image pairs in denoising tasks, has been proposed [2,12,13,16,22], but they add complexity to the model, and training process making it less efficient. In recent work [21], a self-supervised learning models have been developed to do projection data noise reduction with a promising result. However, the method has only been evaluated on simulated data and has yet to be validated on clinical data.…”

Section: Introductionmentioning

confidence: 99%

Low-dose CT reconstruction by self-supervised learning in the projection domain

Zhang¹,

Wang²,

Hu³

et al. 2022

Preprint

View full text Add to dashboard Cite

In the intention of minimizing excessive X-ray radiation administration to patients, low-dose computed tomography (LDCT) has become a distinct trend in radiology. However, while lowering the radiation dose reduces the risk to the patient, it also increases noise and artifacts, compromising image quality and clinical diagnosis. In most supervised learning methods, paired CT images are required, but such images are unlikely to be available in the clinic. We present a self-supervised learning model (Noise2Projection) that fully exploits the raw projection images to reduce noise and improve the quality of reconstructed LDCT images. Unlike existing self-supervised algorithms, the proposed method only requires noisy CT projection images and reduces noise by exploiting the correlation between nearby projection images. We trained and tested the model using clinical data and the quantitative and qualitative results suggest that our model can effectively reduce LDCT image noise while also drastically removing artifacts in LDCT images.

show abstract

Benchmarking deep learning‐based low‐dose CT image denoising algorithms

Eulig,

Ommer,

Kachelrieß

2024

Medical Physics

View full text Add to dashboard Cite

BackgroundLong‐lasting efforts have been made to reduce radiation dose and thus the potential radiation risk to the patient for computed tomography (CT) acquisitions without severe deterioration of image quality. To this end, various techniques have been employed over the years including iterative reconstruction methods and noise reduction algorithms.PurposeRecently, deep learning‐based methods for noise reduction became increasingly popular and a multitude of papers claim ever improving performance both quantitatively and qualitatively. However, the lack of a standardized benchmark setup and inconsistencies in experimental design across studies hinder the verifiability and reproducibility of reported results.MethodsIn this study, we propose a benchmark setup to overcome those flaws and improve reproducibility and verifiability of experimental results in the field. We perform a comprehensive and fair evaluation of several state‐of‐the‐art methods using this standardized setup.ResultsOur evaluation reveals that most deep learning‐based methods show statistically similar performance, and improvements over the past years have been marginal at best.ConclusionsThis study highlights the need for a more rigorous and fair evaluation of novel deep learning‐based methods for low‐dose CT image denoising. Our benchmark setup is a first and important step towards this direction and can be used by future researchers to evaluate their algorithms.

show abstract

No-Reference Denoising Of Low-Dose Ct Projections

Cited by 8 publications

References 12 publications

Learning CT projection denoising from adjacent views

Learning CT projection denoising from adjacent views

Low-dose CT reconstruction by self-supervised learning in the projection domain

Benchmarking deep learning‐based low‐dose CT image denoising algorithms

Contact Info

Product

Resources

About