Low-dose CT with deep learning regularization via proximal forward–backward splitting

Ding, Qiaoqiao; Chen, Gaoyu; Zhang, Xiaoqun; Huang, Qiu; Gao, Hui

doi:10.1088/1361-6560/ab831a

Cited by 39 publications

(25 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By using GAN framework and incorporating information on target standard dose images from a data distribution, the proposed method has a potential to reduce noise in LDCT images for unpaired CT images or even for a complicated noise model. 30 Ding et al 31 proposed a method that unrolls the optimisation by PFBS (proximal forward-backward splitting) with datadriven image regularisation learned by DNN. To further improve image reconstruction quality, a preconditioned PFBS version was used with fused analytical and iterative reconstruction (AIR) and as a result, it was suggested that deep learning regularised methods PFBS-IR (proximal forward-backward splitting via iterative reconstruction) and PFBS-AIR (proximal forward-backward splitting via analytical and iterative reconstruction) provided better reconstruction quality compared to conventional wisdoms (AR or IR), and DL-based post processing method (FBPConvNet 31 ) (Table 2).…”

Section: Image Noise Reductionmentioning

confidence: 99%

“…It then learns the proper gradient of the low-dose CT in a pure unsupervised manner. 29 Another solution was developed by Ding et al 31 with a similar idea to the iterative reconstruction method, but changed to a deep learning based regularisation method for low-dose CT image reconstruction. It was shown to have a better reconstruction quality than analytical reconstruction or iterative reconstruction techniques.…”

Section: Capability Of Deep Learning Networkmentioning

confidence: 99%

“…The data contained normal-dose CT prostate image data from 100 anonymised scans to which Poisson noise was added to simulate lowdose CT scans. 31 Limitations AI technologies and solutions are advancing at unprecedented speed and a lot of new articles are published in different journals and conferences. Therefore, it is hard for this scoping review paper to cover all the reported works on AI technologies for dose optimisation in low-dose CT.…”

Section: Capability Of Deep Learning Networkmentioning

confidence: 99%

See 2 more Smart Citations

The use of deep learning towards dose optimization in low-dose computed tomography: A scoping review

et al. 2022

View full text Add to dashboard Cite

Introduction: Low-dose computed tomography tends to produce lower image quality than normal dose computed tomography (CT) although it can help to reduce radiation hazards of CT scanning. Research has shown that Artificial Intelligence (AI) technologies, especially deep learning can help enhance the image quality of low-dose CT by denoising images. This scoping review aims to create an overview on how AI technologies, especially deep learning, can be used in dose optimisation for low-dose CT. Methods: Literature searches of ProQuest, PubMed, Cinahl, ScienceDirect, EbscoHost Ebook Collection and Ovid were carried out to find research articles published between the years 2015 and 2020. In addition, manual search was conducted in SweMedþ, SwePub, NORA, Taylor & Francis Online and Medic. Results: Following a systematic search process, the review comprised of 16 articles. Articles were organised according to the effects of the deep learning networks, e.g. image noise reduction, image restoration. Deep learning can be used in multiple ways to facilitate dose optimisation in low-dose CT. Most articles discuss image noise reduction in low-dose CT. Conclusion: Deep learning can be used in the optimisation of patients' radiation dose. Nevertheless, the image quality is normally lower in low-dose CT (LDCT) than in regular-dose CT scans because of smaller radiation doses. With the help of deep learning, the image quality can be improved to equate the regulardose computed tomography image quality. Implications to practice: Lower dose may decrease patients' radiation risk but may affect the image quality of CT scans. Artificial intelligence technologies can be used to improve image quality in low-dose CT scans. Radiologists and radiographers should have proper education and knowledge about the techniques used.

show abstract

Section: Image Noise Reductionmentioning

confidence: 99%

Section: Capability Of Deep Learning Networkmentioning

confidence: 99%

Section: Capability Of Deep Learning Networkmentioning

confidence: 99%

See 1 more Smart Citation

The use of deep learning towards dose optimization in low-dose computed tomography: A scoping review

et al. 2022

View full text Add to dashboard Cite

show abstract

“…The fundamental reason is mainly through the calculation of any non-convex prior with a large amount of calculation, which leads to complex optimization algorithms. What's more, the parametric assumptions on blur kernels could greatly improve the robustness of blind image deblurring [19,20], a principled algorithm within the maximum a posterior framework to tackle image restoration with a partially known or inaccurate degradation model [21], and other deep learning and other methods that contain a large number of real image sets and training sets using self-supervised methods [22][23][24][25][26], the characteristics of training data sets are often It will greatly affect the performance of the entire model, and in many cases, the construction and processing of data sets are often very expensive and low feasibility.…”

Section: Image Deblurringmentioning

confidence: 99%

Fast Blind Deblurring of QR Code Images Based on Adaptive Scale Control

et al. 2021

View full text Add to dashboard Cite

With the development of 5G technology, the short delay requirements of commercialization and large amounts of data change our lifestyle day-to-day. In this background, this paper proposes a fast blind deblurring algorithm for QR code images, which mainly achieves the effect of adaptive scale control by introducing an evaluation mechanism. Its main purpose is to solve the out-of-focus caused by lens shake, inaccurate focus, and optical noise by speeding up the latent image estimation in the process of multi-scale division iterative deblurring. The algorithm optimizes productivity under the guidance of collaborative computing, based on the characteristics of the QR codes, such as the features of gradient and strength. In the evaluation step, the Tenengrad method is used to evaluate the image quality, and the evaluation value is compared with the empirical value obtained from the experimental data. Combining with the error correction capability, the recognizable QR codes will be output. In addition, we introduced a scale control parameter to study the relationship between the recognition rate and restoration time. Theoretical analysis and experimental results show that the proposed algorithm has high recovery efficiency and well recovery effect, can be effectively applied in industrial applications.

show abstract

“…It has been shown that various types of deep neural networks are capable of suppressing the noise in low-dose computed tomography (CT) as well as positron emission tomography (PET) images leading to dependable estimation of the standard-dose images [9,[22][23][24][25][26][27][28][29]. Likewise, a number of studies have been conducted in the field of lowdose SPECT-MPI.…”

Section: Introductionmentioning

confidence: 99%

Deep learning–based denoising of low-dose SPECT myocardial perfusion images: quantitative assessment and clinical performance

Olia

Kamali-Asl

Tabrizi

et al. 2021

Eur J Nucl Med Mol Imaging

View full text Add to dashboard Cite

Purpose This work was set out to investigate the feasibility of dose reduction in SPECT myocardial perfusion imaging (MPI) without sacrificing diagnostic accuracy. A deep learning approach was proposed to synthesize full-dose images from the corresponding low-dose images at different dose reduction levels in the projection space. Methods Clinical SPECT-MPI images of 345 patients acquired on a dedicated cardiac SPECT camera in list-mode format were retrospectively employed to predict standard-dose from low-dose images at half-, quarter-, and one-eighth-dose levels. To simulate realistic low-dose projections, 50%, 25%, and 12.5% of the events were randomly selected from the list-mode data through applying binomial subsampling. A generative adversarial network was implemented to predict non-gated standard-dose SPECT images in the projection space at the different dose reduction levels. Well-established metrics, including peak signal-to-noise ratio (PSNR), root mean square error (RMSE), and structural similarity index metrics (SSIM) in addition to Pearson correlation coefficient analysis and clinical parameters derived from Cedars-Sinai software were used to quantitatively assess the predicted standard-dose images. For clinical evaluation, the quality of the predicted standard-dose images was evaluated by a nuclear medicine specialist using a seven-point (− 3 to + 3) grading scheme. Results The highest PSNR (42.49 ± 2.37) and SSIM (0.99 ± 0.01) and the lowest RMSE (1.99 ± 0.63) were achieved at a half-dose level. Pearson correlation coefficients were 0.997 ± 0.001, 0.994 ± 0.003, and 0.987 ± 0.004 for the predicted standard-dose images at half-, quarter-, and one-eighth-dose levels, respectively. Using the standard-dose images as reference, the Bland–Altman plots sketched for the Cedars-Sinai selected parameters exhibited remarkably less bias and variance in the predicted standard-dose images compared with the low-dose images at all reduced dose levels. Overall, considering the clinical assessment performed by a nuclear medicine specialist, 100%, 80%, and 11% of the predicted standard-dose images were clinically acceptable at half-, quarter-, and one-eighth-dose levels, respectively. Conclusion The noise was effectively suppressed by the proposed network, and the predicted standard-dose images were comparable to reference standard-dose images at half- and quarter-dose levels. However, recovery of the underlying signals/information in low-dose images beyond a quarter of the standard dose would not be feasible (due to very poor signal-to-noise ratio) which will adversely affect the clinical interpretation of the resulting images.

show abstract

Low-dose CT with deep learning regularization via proximal forward–backward splitting

Cited by 39 publications

References 39 publications

The use of deep learning towards dose optimization in low-dose computed tomography: A scoping review

The use of deep learning towards dose optimization in low-dose computed tomography: A scoping review

Fast Blind Deblurring of QR Code Images Based on Adaptive Scale Control

Deep learning–based denoising of low-dose SPECT myocardial perfusion images: quantitative assessment and clinical performance

Contact Info

Product

Resources

About