A residual dense network assisted sparse view reconstruction for breast computed tomography.
Ontology highlight
ABSTRACT:
SUBMITTER: Fu Z
PROVIDER: S-EPMC7713379 | biostudies-literature | 2020 Dec
REPOSITORIES: biostudies-literature
Publications
A residual dense network assisted sparse view reconstruction for breast computed tomography.
Fu Zhiyang Z Tseng Hsin Wu HW Vedantham Srinivasan S Karellas Andrew A Bilgin Ali A
Scientific reports 20201203 1
To develop and investigate a deep learning approach that uses sparse-view acquisition in dedicated breast computed tomography for radiation dose reduction, we propose a framework that combines 3D sparse-view cone-beam acquisition with a multi-slice residual dense network (MS-RDN) reconstruction. Projection datasets (300 views, full-scan) from 34 women were reconstructed using the FDK algorithm and served as reference. Sparse-view (100 views, full-scan) projection data were reconstructed using th ...[more]
Similar Datasets
Project description:BackgroundThe widespread application of X-ray computed tomography (CT) imaging in medical screening makes radiation safety a major concern for public health. Sparse-view CT is a promising solution to reduce the radiation dose. However, the reconstructed CT images obtained using sparse-view CT may suffer severe streaking artifacts and structural information loss.MethodsIn this study, a novel attention-based dual-branch network (ADB-Net) is proposed to solve the ill-posed problem of sparse-view CT image reconstruction. In this network, downsampled sinogram input is processed through 2 parallel branches (CT branch and signogram branch) of the ADB-Net to independently extract the distinct, high-level feature maps. These feature maps are fused in a specified attention module from 3 perspectives (channel, plane, and spatial) to allow complementary optimizations that can mitigate the streaking artifacts and the structure loss in sparse-view CT imaging.ResultsNumerical simulations, an anthropomorphic thorax phantom, and in vivo preclinical experiments were conducted to verify the sparse-view CT imaging performance of the ADB-Net. The proposed network achieved a root-mean-square error (RMSE) of 20.6160, a structural similarity (SSIM) of 0.9257, and a peak signal-to-noise ratio (PSNR) of 38.8246 on numerical data. The visualization results demonstrate that this newly developed network can consistently remove the streaking artifacts while maintaining the fine structures.ConclusionsThe proposed attention-based dual-branch deep network, ADB-Net, provides a promising alternative to reconstruct high-quality sparse-view CT images for low-dose CT imaging.
| S-EPMC10006128 | biostudies-literature
Project description:ObjectiveStimulated Raman projection tomography (SRPT), a recently developed label-free volumetric chemical imaging technology, has been reported to quantitatively reconstruct the distribution of chemicals in a three-dimensional (3D) complex system. The current image reconstruction scheme used in SRPT is based on a filtered back projection (FBP) algorithm that requires at least 180 angular-dependent projections to rebuild a reasonable SRPT image, resulting in a long total acquisition time. This is a big limitation for longitudinal studies on live systems.MethodsWe present a sparse-view data-based sparse reconstruction scheme, in which sparsely sampled projections at 180 degrees were used to reconstruct the volumetric information. In the scheme, the simultaneous algebra reconstruction technique (SART), combined with total variation regularization, was used for iterative reconstruction. To better describe the projection process, a pixel vertex driven model (PVDM) was developed to act as projectors, whose performance was compared with those of the distance driven model (DDM).ResultsWe evaluated our scheme with numerical simulations and validated it for SRPT by mapping lipid contents in adipose cells. Simulation results showed that the PVDM performed better than the DDM in the case of using sparse-view data. Our scheme could maintain the quality of the reconstructed images even when the projection number was reduced to 15. The cell-based experimental results demonstrated that the proposed scheme can improve the imaging speed of the current FBP-based SRPT scheme by a factor of 9-12 without sacrificing discernible imaging details.ConclusionOur proposed scheme significantly reduces the total acquisition time required for SRPT at a speed of one order of magnitude faster than the currently used scheme. This significant improvement in imaging speed would potentially promote the applicability of SRPT for imaging living organisms.
| S-EPMC7329365 | biostudies-literature
Project description:Summary Decreasing projection views to a lower X-ray radiation dose usually leads to severe streak artifacts. To improve image quality from sparse-view data, a multi-domain integrative Swin transformer network (MIST-net) was developed and is reported in this article. First, MIST-net incorporated lavish domain features from data, residual data, image, and residual image using flexible network architectures, where a residual data and residual image sub-network was considered as a data consistency module to eliminate interpolation and reconstruction errors. Second, a trainable edge enhancement filter was incorporated to detect and protect image edges. Third, a high-quality reconstruction Swin transformer (i.e., Recformer) was designed to capture image global features. The experimental results on numerical and real cardiac clinical datasets with 48 views demonstrated that our proposed MIST-net provided better image quality with more small features and sharp edges than other competitors. Highlights • A multi-domain integrative Swin transformer was proposed for spare-data imaging• A trainable image edge enhancement filter recovers sharp image edges and features• A modified vision transformer first demonstrates its great potential in tomography• Patient clinical cardiac and numerical data demonstrate our MIST power The bigger picture Decreasing projection views to a lower X-ray radiation dose usually leads to severe streak artifacts. To improve reconstructed image quality from sparse-view data, we develop a multi-domain integrative Swin transformer (MIST) network in this study. The proposed MIST-net incorporates lavish domain features from data, residual data, image, and residual image using flexible network architectures, which help deeply mine the data and image features. To detect image features and protect image edges, the trainable edge enhancement filter is further incorporated to the network for improving encode-decode ability. A high-quality reconstruction transformer was designed to improve the ability of global feature extraction. Our results from both simulation and real cardiac data demonstrated the great potential of MIST. A multi-domain integrative Swin transformer network (MIST-net) demonstrates feasibility, advantages, and great potential in sparse-view tomographic reconstruction. MIST-net mainly integrates data, image, residual data, and residual image domain information into a unified model with a trainable edge enhancement filter and advanced transformer AI technique to realize ultra-sparse-data (48 views) high-fidelity imaging.
| S-EPMC9214338 | biostudies-literature
Project description:Total generalized variation (TGV)-based computed tomography (CT) image reconstruction, which utilizes high-order image derivatives, is superior to total variation-based methods in terms of the preservation of edge information and the suppression of unfavorable staircase effects. However, conventional TGV regularization employs l1-based form, which is not the most direct method for maximizing sparsity prior. In this study, we propose a total generalized p-variation (TGpV) regularization model to improve the sparsity exploitation of TGV and offer efficient solutions to few-view CT image reconstruction problems. To solve the nonconvex optimization problem of the TGpV minimization model, we then present an efficient iterative algorithm based on the alternating minimization of augmented Lagrangian function. All of the resulting subproblems decoupled by variable splitting admit explicit solutions by applying alternating minimization method and generalized p-shrinkage mapping. In addition, approximate solutions that can be easily performed and quickly calculated through fast Fourier transform are derived using the proximal point method to reduce the cost of inner subproblems. The accuracy and efficiency of the simulated and real data are qualitatively and quantitatively evaluated to validate the efficiency and feasibility of the proposed method. Overall, the proposed method exhibits reasonable performance and outperforms the original TGV-based method when applied to few-view problems.
| S-EPMC4764011 | biostudies-literature
Project description:Tomographic imaging using penetrating waves generates cross-sectional views of the internal anatomy of a living subject. For artefact-free volumetric imaging, projection views from a large number of angular positions are required. Here we show that a deep-learning model trained to map projection radiographs of a patient to the corresponding 3D anatomy can subsequently generate volumetric tomographic X-ray images of the patient from a single projection view. We demonstrate the feasibility of the approach with upper-abdomen, lung, and head-and-neck computed tomography scans from three patients. Volumetric reconstruction via deep learning could be useful in image-guided interventional procedures such as radiation therapy and needle biopsy, and might help simplify the hardware of tomographic imaging systems.
| S-EPMC6858583 | biostudies-literature
Project description:BackgroundMulti-energy computed tomography (CT) provides multiple channel-wise reconstructed images, and they can be used for material identification and k-edge imaging. Nonetheless, the projection datasets are frequently corrupted by various noises (e.g., electronic, Poisson) in the acquisition process, resulting in lower signal-noise-ratio (SNR) measurements. Multi-energy CT images have local sparsity, nonlocal self-similarity in spatial dimension, and correlation in spectral dimension.MethodsIn this paper, we propose an image-spectral decomposition extended-learning assisted by sparsity (IDEAS) method to fully exploit these intrinsic priors for multi-energy CT image reconstruction. Particularly, a nonlocal low-rank Tucker decomposition (TD) is employed to utilize the correlation and nonlocal self-similarity priors. Moreover, considering the advantages of multi-task tensor dictionary learning (TDL) in sparse representation, an adaptive spatial dictionary and an adaptive spectral dictionary are trained during the iterative reconstruction process. Furthermore, a weighted total variation (TV) regularization term is employed to encourage local sparsity.ResultsNumerical simulation, physical phantom, and preclinical mouse experiments are performed to validate the proposed IDEAS algorithm. Specifically, in the simulation experiments, the proposed IDEAS reconstructed high-quality images that are very close to the references. For example, the root mean square error (RMSE) of IDEAS image in energy bin 1 is as low as 0.0672, while the RMSE of other methods are higher than 0.0843. Besides, the structural similarity (SSIM) of IDEAS reconstructed image in energy bin 1 is greater than 0.98. For material decomposition, the RMSE of IDEAS bone component is as low as 0.0152, and other methods are higher than 0.0199. In addition, the computational cost of IDEAS is as low as 98.8 s for one iteration, and the competing tensor decomposition method is higher than 327 s.ConclusionsTo further improve the quality of the reconstructed multi-energy CT images, multiple prior regularizations are introduced to the multi-energy CT reconstructed model, leading to an IDEAS method. Both qualitative and quantitative evaluation of our results confirm the outstanding performance of the proposed algorithm compared to the state-of-the-arts.
| S-EPMC9929415 | biostudies-literature
Project description:Although a substantial decrease in 2-[fluorine-18]fluoro-2-deoxy-d-glucose (FDG) uptake on positron emission tomography-computed tomography (PET-CT) indicates a promising metabolic response to treatment, predicting the pathologic status of lymph nodes (LN) remains challenging. We investigated the potential of a CT radiomics approach to predict the pathologic complete response of LNs showing residual uptake after neoadjuvant concurrent chemoradiotherapy (NeoCCRT) in patients with non-small cell lung cancer (NSCLC). Two hundred and thirty-seven patients who underwent NeoCCRT for stage IIIa NSCLC were included. Two hundred fifty-two CT radiomics features were extracted from LNs showing remaining positive FDG uptake upon restaging PET-CT. A multivariable logistic regression analysis of radiomics features and clinicopathologic characteristics was used to develop a prediction model. Of the 237 patients, 135 patients (185 nodes) met our inclusion criteria. Eighty-seven LNs were proven to be malignant (47.0%, 87/185). Upon multivariable analysis, metastatic LNs were significantly prevalent in females and patients with adenocarcinoma (odds ratio (OR) = 2.02, 95% confidence interval (CI) = 0.88-4.62 and OR = 0.39, 95% CI = 0.19-0.77 each). Metastatic LNs also had a larger maximal 3D diameter and higher cluster tendency (OR = 9.92, 95% CI = 3.15-31.17 and OR = 2.36, 95% CI = 1.22-4.55 each). The predictive model for metastasis showed a discrimination performance with an area under the receiver operating characteristic curve of 0.728 (95% CI = 0.654-0.801, p value < 0.001). The radiomics approach allows for the noninvasive detection of metastases in LNs with residual FDG uptake after the treatment of NSCLC patients.
| S-EPMC7761511 | biostudies-literature
Project description:Breast cancer detection largely relies on imaging characteristics and the ability of clinicians to easily and quickly identify potential lesions. Magnetic resonance imaging (MRI) of breast tumors has recently shown great promise for enabling the automatic identification of breast tumors. Nevertheless, state-of-the-art MRI-based algorithms utilizing deep learning techniques are still limited in their ability to accurately separate tumor and healthy tissue. Therefore, in the current work, we propose an automatic and accurate two-stage U-Net-based segmentation framework for breast tumor detection using dynamic contrast-enhanced MRI (DCE-MRI). This framework was evaluated using T2-weighted MRI data from 160 breast tumor cases, and its performance was compared with that of the standard U-Net model. In the first stage of the proposed framework, a refined U-Net model was utilized to automatically delineate a breast region of interest (ROI) from the surrounding healthy tissue. Importantly, this automatic segmentation step reduced the impact of the background chest tissue on breast tumors' identification. For the second stage, we employed an improved U-Net model that combined a dense residual module based on dilated convolution with a recurrent attention module. This model was used to accurately and automatically segment the tumor tissue from healthy tissue in the breast ROI derived in the previous step. Overall, compared to the U-Net model, the proposed technique exhibited increases in the Dice similarity coefficient, Jaccard similarity, positive predictive value, sensitivity, and Hausdorff distance of 3%, 3%, 3%, 2%, and 16.2, respectively. The proposed model may in the future aid in the clinical diagnosis of breast cancer lesions and help guide individualized patient treatment.
| S-EPMC9045980 | biostudies-literature
Project description:Sparse-view CT reconstruction is a fundamental task in computed tomography to overcome undesired artifacts and recover the details of textual structure in degraded CT images. Recently, many deep learning-based networks have achieved desirable performances compared to iterative reconstruction algorithms. However, the performance of these methods may severely deteriorate when the degradation strength of the test image is not consistent with that of the training dataset. In addition, these methods do not pay enough attention to the characteristics of different degradation levels, so solely extending the training dataset with multiple degraded images is also not effective. Although training plentiful models in terms of each degradation level can mitigate this problem, extensive parameter storage is involved. Accordingly, in this paper, we focused on sparse-view CT reconstruction for multiple degradation levels. We propose a single degradation-aware deep learning framework to predict clear CT images by understanding the disparity of degradation in both the frequency domain and image domain. The dual-domain procedure can perform particular operations at different degradation levels in frequency component recovery and spatial details reconstruction. The peak signal-to-noise ratio (PSNR), structural similarity (SSIM) and visual results demonstrate that our method outperformed the classical deep learning-based reconstruction methods in terms of effectiveness and scalability.
| S-EPMC8704775 | biostudies-literature
Project description:PURPOSE:Scatter is a major factor degrading the image quality of cone beam computed tomography (CBCT). Conventional scatter correction strategies require handcrafted analytical models with ad hoc assumptions, which often leads to less accurate scatter removal. This study aims to develop an effective scatter correction method using a residual convolutional neural network (CNN). METHODS:A U-net based 25-layer CNN was constructed for CBCT scatter correction. The establishment of the model consists of three steps: model training, validation, and testing. For model training, a total of 1800 pairs of x-ray projection and the corresponding scatter-only distribution in nonanthropomorphic phantoms taken in full-fan scan were generated using Monte Carlo simulation of a CBCT scanner installed with a proton therapy system. An end-to-end CNN training was implemented with two major loss functions for 100 epochs with a mini-batch size of 10. Image rotations and flips were randomly applied to augment the training datasets during training. For validation, 200 projections of a digital head phantom were collected. The proposed CNN-based method was compared to a conventional projection-domain scatter correction method named fast adaptive scatter kernel superposition (fASKS) method using 360 projections of an anthropomorphic head phantom. Two different loss functions were applied for the same CNN to evaluate the impact of loss functions on the final results. Furthermore, the CNN model trained with full-fan projections was fine-tuned for scatter correction in half-fan scan by using transfer learning with additional 360 half-fan projection pairs of nonanthropomorphic phantoms. The tuned-CNN model for half-fan scan was compared with the fASKS method as well as the CNN-based method without the fine-tuning using additional lung phantom projections. RESULTS:The CNN-based method provides projections with significantly reduced scatter and CBCT images with more accurate Hounsfield Units (HUs) than that of the fASKS-based method. Root mean squared error of the CNN-corrected projections was improved to 0.0862 compared to 0.278 for uncorrected projections or 0.117 for the fASKS-corrected projections. The CNN-corrected reconstruction provided better HU quantification, especially in regions near the air or bone interfaces. All four image quality measures, which include mean absolute error (MAE), mean squared error (MSE), peak signal-to-noise ratio (PSNR), and structural similarity (SSIM), indicated that the CNN-corrected images were significantly better than that of the fASKS-corrected images. Moreover, the proposed transfer learning technique made it possible for the CNN model trained with full-fan projections to be applicable to remove scatters in half-fan projections after fine-tuning with only a small number of additional half-fan training datasets. SSIM value of the tuned-CNN-corrected images was 0.9993 compared to 0.9984 for the non-tuned-CNN-corrected images or 0.9990 for the fASKS-corrected images. Finally, the CNN-based method is computationally efficient - the correction time for the 360 projections only took less than 5 s in the reported experiments on a PC (4.20 GHz Intel Core-i7 CPU) with a single NVIDIA GTX 1070 GPU. CONCLUSIONS:The proposed deep learning-based method provides an effective tool for CBCT scatter correction and holds significant value for quantitative imaging and image-guided radiation therapy.
| S-EPMC6684491 | biostudies-literature