Competitive performance of a modularized deep neural network compared to commercial algorithms for low-dose CT image reconstruction

Shan, Hongming; Padole, Atul; Homayounieh, Fatemeh; Krüger, Uwe; Khera, Ruhani Doda; Nitiwarangkul, Chayanin; Kalra, Mannudeep K.; Wang, Ge

doi:10.1038/s42256-019-0057-9

Cited by 330 publications

(242 citation statements)

References 29 publications

Supporting

Mentioning

226

Contrasting

Unclassified

Order By: Relevance

“…Deep learning is powerful to enhance image quality by reducing image noise and artifacts, and neural network modeling has been extremely successful in image classification, identification, super-resolution imaging, and denoising [14][15][16][17]. It may find a regular pattern through learning and inference from a large amount of dataset to reflect real features faithfully and perform various types of intelligence-demanding tasks reliably against uncertainties in system and imaging models [18,19].…”

Section: *Wangg6@rpiedumentioning

confidence: 99%

Deep-learning-based breast CT for radiation dose reduction

Cong

Shan

Zhang

et al. 2019

Developments in X-Ray Tomography XII

Self Cite

View full text Add to dashboard Cite

Cone-beam breast computed tomography (CT) provides true 3D breast images with isotropic resolution and highcontrast information, detecting calcifications as small as a few hundred microns and revealing subtle tissue differences. However, breast is highly sensitive to x-ray radiation. It is critically important for healthcare to reduce radiation dose. Few-view cone-beam CT only uses a fraction of x-ray projection data acquired by standard cone-beam breast CT, enabling significant reduction of the radiation dose. However, insufficient sampling data would cause severe streak artifacts in CT images reconstructed using conventional methods. In this study, we propose a deep-learning-based method to establish a residual neural network model for the image reconstruction, which is applied for few-view breast CT to produce high quality breast CT images. We respectively evaluate the deep-learning-based image reconstruction using one third and one quarter of x-ray projection views of the standard cone-beam breast CT. Based on clinical breast imaging dataset, we perform a supervised learning to train the neural network from few-view CT images to corresponding full-view CT images. Experimental results show that the deep learning-based image reconstruction method allows few-view breast CT to achieve a radiation dose <6 mGy per cone-beam CT scan, which is a threshold set by FDA for mammographic screening.In this study, we developed a ResNet network for the image reconstruction of the few-view breast CT using Python with the Tensorflow library. The ResNet network is trained, validated, and tested based on real clinical breast CT dataset from Koning Inc. Our experimental results show that the optimization model of the neural network is highly cost-effective, and has an excellent convergent behavior in the learning process. The ResNet produces an output of high-quality breast CT images, effectively removing noise and artifacts and preserving structure details of breast images. For the image reconstructions with one third and one fourth of radiation dose delivered by a standard breast CT, we quantitatively evaluate breast images of output from network by the peak-to-noise ratio (PSNR) and structural similarity (SSIM). The proposed network model produces promising results and has an important application value in clinical breast imaging.

show abstract

Section: *Wangg6@rpiedumentioning

confidence: 99%

Deep-learning-based breast CT for radiation dose reduction

Cong

Shan

Zhang

et al. 2019

Developments in X-Ray Tomography XII

Self Cite

View full text Add to dashboard Cite

show abstract

“…These learning‐based methods have also been combined with reconstruction‐based methods to further improve SR performance (Yu et al, ). The sparse coding example–based SR technique, while inflexible, has been shown to outperform bicubic interpolation and has since been generalized (Dong et al, ) into the structurally analogous but highly flexible Super Resolution Convolutional Neural Network (SRCNN) for superresolving individual photographic images (Dong et al, ; Kim et al, ; Ledig et al, ; Lim et al, ; Wang et al, ; Yu et al, ), medical images (Umehara et al, ; You et al, ; Shan et al, ), and digital rock images (Wang et al, , ). Results from SRCNN have shown to consistently outperform previously utilized learning‐based and reconstruction‐based methods.…”

Section: Introductionmentioning

confidence: 99%

Boosting Resolution and Recovering Texture of 2D and 3D Micro‐CT Images with Deep Learning

Wang

Armstrong

Mostaghimi

2020

Water Resources Research

View full text Add to dashboard Cite

Simulation of flow directly at the pore scale depends on high-quality digital rock images but is constrained by detector hardware. A trade-off between the image field of view (FOV) and image resolution is made. This can be compensated for with superresolution (SR) techniques that take a wide FOV, low-resolution (LR) image, and superresolve a high resolution (HR). The Enhanced Deep Super Resolution Generative Adversarial Network (EDSRGAN) is trained on the DeepRock-SR, a diverse compilation of raw and processed micro-computed tomography (μCT) images in 2D and 3D. The 2D and 3D trained networks show comparable performance of 50% to 70% reduction in relative error over bicubic interpolation with minimal computational cost during usage. Texture regeneration with EDSRGAN shows superior visual similarity versus Super Resolution Convolutional Neural Network (SRCNN) and other methods. Difference maps show SRCNN recovers large-scale edge features while EDSRGAN regenerates perceptually indistinguishable high-frequency texture. Physical accuracy is measured by permeability and phase topology on consistently segmented images, showing EDSRGAN results achieving the closest match. Performance is generalized with augmentation, showing high adaptability to noise and blur. HR images are fed into the network, generating HR-SR images to extrapolate network performance to subresolution features present in the HR images themselves. Underresolution features are regenerated despite operating outside of trained specifications. Comparison with scanning electron microscopy (SEM) images shows details are consistent with the underlying geometry. Images that are normally constrained by the mineralogy of the rock, by fast transient imaging, or by the energy of the source can be superresolved accurately for further analysis. Plain Language SummaryWhen capturing an X-ray image of the insides of a rock sample (or any opaque object), hardware limitations on the image quality and size exist. These limitations can be overcome with the use of machine learning algorithms that "superresolve" a lower resolution image. Once trained, the machine algorithm can sharpen otherwise blurry features and regenerate the underlying texture of the imaged object. We train such an algorithm on a large and wide array of digital rock images and test its flexibility on some images that it had never seen before, as well as on some very high quality images that it was not trained to superresolve. The results of training and testing the algorithm shows a promising degree of accuracy and flexibility in handling a wide array of images of different quality and allows for higher quality images to be generated for use in other image-based analysis techniques.

show abstract

“…3 CNNbased noise reduction is very computationally efficient and there is evidence that it compares favorably to commercial iterative reconstruction methods. 4 Furthermore, a CNN trained to remove noise in low-dose CT images can then be applied to higher dose images to further reduce noise while maintaining image quality, suggesting that the rules learned by training a CNN on LDCT data generalize well to across dose levels. 5 However, maintaining spatial resolution and fine anatomic details may still be problematic with this approach.…”

Section: Introductionmentioning

confidence: 99%

Synthesizing images from multiple kernels using a deep convolutional neural network

Missert

Leng

et al. 2019

Medical Physics

View full text Add to dashboard Cite

Purpose Filtering measured projections with a particular convolutional kernel is an essential step in analytic reconstruction of computed tomography (CT) images. A tradeoff between noise and spatial resolution exists for different choices of reconstruction kernel. In a clinical setting, this often requires producing multiple images reconstructed with different kernels for a single CT exam, which increases the burden of computation, networking, archival, and reading. We address this problem by training a deep convolutional neural network (CNN) to synthesize multiple input images into a single output image which exhibits low noise while also preserving features in images reconstructed with the sharpest kernels. Methods A CNN architecture consisting of repeated blocks of residual units containing a total of 20 convolutional layers was used to combine features. The CNN inputs consisted of two images produced with different reconstruction kernels, one smooth and one sharp, which were stacked in the channel dimension. The network was trained using supervised learning with both full‐dose and simulated quarter‐dose abdominal CT images. After training, the performance was evaluated using a reserved set of full‐dose scans that were not used for network optimization. Noise reduction performance was measured by comparing root mean square (RMS) measurements in uniform regions. Spatial resolution was compared using line profiles of anatomic features. Results For the regions tested, the synthetic images feature noise levels slightly below those of the smooth input images, while maintaining the resolution of anatomic details found in the sharp input images. Conclusions A deep CNN can be used combine features from CT images reconstructed with different kernels to produce a single synthesized image series that exhibits both low noise and high spatial resolution. This approach has implications for improving image quality, reducing radiation dose, and simplifying the clinical workflow for CT imaging.

show abstract

Competitive performance of a modularized deep neural network compared to commercial algorithms for low-dose CT image reconstruction

Cited by 330 publications

References 29 publications

Deep-learning-based breast CT for radiation dose reduction

Deep-learning-based breast CT for radiation dose reduction

Boosting Resolution and Recovering Texture of 2D and 3D Micro‐CT Images with Deep Learning

Synthesizing images from multiple kernels using a deep convolutional neural network

Contact Info

Product

Resources

About