Learning-Based Artifact Removal via Image Decomposition for Low-Dose CT Image Processing

Cui, Xueying; Gui, Zhiguo; Zhang, Quan; Shangguan, Hong; Wang, Anhong

doi:10.1109/tns.2016.2565604

Cited by 26 publications

(7 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Postprocessing categories, in contrast, do not rely on projection data directly; instead, they operate directly on the images reconstructed from the raw data. These postprocessing methods include dictionary learning methods, NLM filtering methods, block‐matching three‐dimensional (3D) filtering methods, and deep neural network‐based methods . However, the denoizing ability of these traditional postprocessing algorithms is not suitable for reducing both noise and artifacts.…”

Section: Introductionmentioning

confidence: 99%

Artifact correction in low‐dose dental CT imaging using Wasserstein generative adversarial networks

Jiang

Sun

et al. 2019

Medical Physics

View full text Add to dashboard Cite

Purpose: In recent years, health risks concerning high-dose x-ray radiation have become a major concern in dental computed tomography (CT) examinations. Therefore, adopting low-dose computed tomography (LDCT) technology has become a major focus in the CT imaging field. One of these LDCT technologies is downsampling data acquisition during low-dose x-ray imaging processes. However, reducing the radiation dose can adversely affect CT image quality by introducing noise and artifacts in the resultant image that can compromise diagnostic information. In this paper, we propose an artifact correction method for downsampling CT reconstruction based on deep learning. Method: We used clinical dental CT data with low-dose artifacts reconstructed by conventional filtered back projection (FBP) as inputs to a deep neural network and corresponding high-quality labeled normal-dose CT data during training. We trained a generative adversarial network (GAN) with Wasserstein distance (WGAN) and mean squared error (MSE) loss, called m-WGAN, to remove artifacts and obtain high-quality CT dental images in a clinical dental CT examination environment. Results: The experimental results confirmed that the proposed algorithm effectively removes lowdose artifacts from dental CT scans. In addition, we showed that the proposed method is efficient for removing noise from low-dose CT scan images compared to existing approaches. We compared the performances of the general GAN, convolutional neural networks, and m-WGAN. Through quantitative and qualitative analysis of the results, we concluded that the proposed m-WGAN method resulted in better artifact correction performance preserving the texture in dental CT scanning. Conclusions: The image quality evaluation metrics indicated that the proposed method effectively improves image quality when used as a postprocessing technique for dental CT images. To the best of our knowledge, this work is the first deep learning architecture used with a commercial cone-beam dental CT scanner. The artifact correction performance was rigorously evaluated and demonstrated to be effective. Therefore, we believe that the proposed algorithm represents a new direction in the research area of low-dose dental CT artifact correction.

show abstract

Section: Introductionmentioning

confidence: 99%

Artifact correction in low‐dose dental CT imaging using Wasserstein generative adversarial networks

Jiang

Sun

et al. 2019

Medical Physics

View full text Add to dashboard Cite

show abstract

“…In a specific part of the continuous scan of the human body, the specific situation of the patient's high-resolution scan results is used. Transmit the scan results to a computer and use the computer for editing to provide clear CT images [11][12]. CT imaging technology is a medical imaging technology based on four-dimensional angiography, and the preprocessing of scan results can make the patient's image clearer [13].…”

Section: Ct Image Processing Technologymentioning

confidence: 99%

Construction of Early Risk Factors and Intervention Programs of Lower Limb Lymph-edema in Patients with Gynecological Malignant Tumors Combined with CT Images

2021

IJPHPM

View full text Add to dashboard Cite

Lymph-edema of lower limbs is one of the common complications of gynecologic cancer patients. Once lymph-edema of lower extremities appears after treatment, it is difficult to cure, which seriously affects the quality of treatment of patients. Lymph-edema early detection and timely intervention is important. In order to prevent and treat the lymph-edema of lower limbs in patients with gynecological malignancies in the early stage, this paper uses the computer-aided diagnosis method to scan the lower limbs of patients with gynecological malignant tumors by using CT imaging technology, and identify the location of lesions in patients with gynecological malignant tumors, and carry out image enhancement and denoising on the scanned CT images to mark the lesion sites. The incidence rate of incidence of lymph-edema in lower extremities was analyzed according to the patients' basic characteristics. The influence of different treatment methods on the incidence rate of lymph-edema and the incidence rate of lymph-edema in different stages of gynecologic malignancies were also analyzed. Finally, a drop of congestive therapy based on the research literature, three drugs therapy and surgical therapy for lower extremity lymph-edema intervention programs, through the patient's body recovery, lower limb swelling analysis of these three methods to effect the lower extremity lymph-edema, for patients to find a suitable intervention programs. The final results showed that the number of patients between the ages of 40-49 and 60-69 were the largest, reaching 427 and 417 respectively, and the heavier the patient, the higher the incidence. Radiotherapy, chemotherapy and pelvic lymph node dissection resulted in 89.73%, 89.76% and 86.96% of patients with lower limb lymph-edema, which is one of the risk factors for lower limb lymph-edema. Among the three intervention programs for lymph-edema of the lower limbs, decongestant therapy has a slower treatment speed but low cost and no side effects. Drug treatment can only be applied to some patients, and surgical treatment has good results but high cost.

show abstract

“…Thus, these postprocessing methods could be conveniently integrated into any CT system. Traditional postprocessing methods can be divided into dictionary learning methods [22]- [25], NLM filtering methods [26], [27], and block-matching three-dimensional filtering (BM3D) methods [28], [29]. More recently, deep learning (DL) methods [30]- [51] have gradually become popular methods for LDCT.…”

Section: Introductionmentioning

confidence: 99%

Deep Cascade Residual Networks (DCRNs): Optimizing an Encoder–Decoder Convolutional Neural Network for Low-Dose CT Imaging

Huang

Chen

Quan

et al. 2022

IEEE Trans. Radiat. Plasma Med. Sci.

View full text Add to dashboard Cite

To suppress noise and artifacts caused by the reduced radiation exposure in low-dose computed tomography, several deep learning (DL)-based image restoration methods have been proposed over the past few years. Many of these popular DL-based methods adopt an encoder-decoder framework, for instance, the residual encoder-decoder convolutional neural network. However, this popular framework may suffer from information loss for continual downsampling operations. In this article, deep cascaded residual networks (DCRNs) are proposed to optimize the popular encoder-decoder network. First, cross up-and downsampling operations as well as attention extraction are substitutes for the strict "downsampling and then upping" principle. What is more, four hybrid loss functions, namely, mean absolute error, edge loss, perceptual loss and adversarial loss, are engaged to achieve better visual effects and suppress noise. The experiments are conducted on three individual clinical CT datasets: dental CT data collected with a scanner manufactured by Zhongke Tianyue Company (ZTC), data from the American Association of Physicists in Medicine (AAPM) Challenge, and data collected with a commercial CT scanner from United Imaging Healthcare (UIH). The experimental results indicate the effective noise reduction and detail preservation capabilities of the proposed methods under different radiation dose-reduction strategies.

show abstract

Learning-Based Artifact Removal via Image Decomposition for Low-Dose CT Image Processing

Cited by 26 publications

References 36 publications

Artifact correction in low‐dose dental CT imaging using Wasserstein generative adversarial networks

Artifact correction in low‐dose dental CT imaging using Wasserstein generative adversarial networks

Construction of Early Risk Factors and Intervention Programs of Lower Limb Lymph-edema in Patients with Gynecological Malignant Tumors Combined with CT Images

Deep Cascade Residual Networks (DCRNs): Optimizing an Encoder–Decoder Convolutional Neural Network for Low-Dose CT Imaging

Contact Info

Product

Resources

About