Brain MRI super resolution using 3D deep densely connected neural networks

Chen, Yuhua; Xie, Yibin; Zhou, Zhengwei; Shi, Feng; Christodoulou, Anthony G.; Li, Debiao

doi:10.1109/isbi.2018.8363679

Cited by 229 publications

(168 citation statements)

References 16 publications

Supporting

Mentioning

157

Contrasting

Order By: Relevance

“…22,30 The densely connected convolutional network uses a greatly reduced number of parameters compared to U-Net and provides a lightweight model with less overfitting due to the dense block. 28 This enables fast prediction at approximately 60 slices/s, which is an acceptable speed for clinical practice, and our model was able to process entire 512 9 512 CT images. Patched images are often used for training because of memory limitations of GPUs.…”

Section: Discussionmentioning

confidence: 96%

See 1 more Smart Citation

Deep learning‐based virtual noncontrast CT for volumetric modulated arc therapy planning: Comparison with a dual‐energy CT‐based approach

Koike¹,

Ohira²,

Akino

et al. 2019

Medical Physics

View full text Add to dashboard Cite

Purpose The aim of this study was to develop a deep learning (DL) method for generating virtual noncontrast (VNC) computed tomography (CT) images from contrast‐enhanced (CE) CT images (VNCDL) and to evaluate its performance in dose calculations for head and neck radiotherapy in comparison with VNC images derived from a dual‐energy CT (DECT) scanner (VNCDECT). Methods This retrospective study included data for 61 patients who underwent head and neck radiotherapy. All planning CT images were obtained with a single‐source DECT scanner (80 and 140 kVp) with rapid kVp switching. The DL‐based method used a pair of virtual monochromatic images (VMIs) at 70 keV with and without contrast materials. VMIs without contrast materials were used as reference true noncontrast (TNC) images. Deformable image registration was used between the TNC and CE images. We used the data of 45 patients, chosen randomly, for training (7922 paired images), and data from the other 16 patients as test data. We generated the VNCDL images with a densely connected convolutional network. As the VNCDECT images, we used VMIs with the iodine signal suppressed, reconstructed from the CE images of the 16 test patients. The CT numbers of the tumor, common carotid artery, internal jugular vein, muscle, fat, bone marrow, cortical bone, and mandible of each VNC image were compared with those of the TNC image. The dose of the reference TNC plan was recalculated using the CE, VNCDL, and VNCDECT images. Difference maps of the dose distributions and dose–volume histograms were evaluated. Results The mean prediction time for the VNCDL images was 3.4 s per patient, and the mean number of slices was 204. The absolute differences in CT numbers of the VNCDL images were significantly smaller than those of the VNCDECT images for the bone marrow (8.0 ± 6.5 vs 175.1 ± 40.9 HU; P < 0.001) and mandible (20.3 ± 19.3 vs 106.2 ± 80.5 HU; P = 0.002). The DL‐based model provided the dose distribution most similar to that of the TNC plan. With the VNCDECT plans, dose errors >1.0% were observed in bone regions. The dose–volume histogram analysis showed that the VNCDL plans yielded the smallest errors for the primary target, although dose differences were <1.0% for all the approaches. For the maximum dose to the mandible, the mean ± SD errors for the CE, VNCDL, and VNCDECT plans were –0.13% ± 0.23% (range: −0.46% to 0.31%; P = 0.037), –0.01% ± 0.22% (range: −0.40% to 0.36%; P = 1.0), and 0.53% ± 0.47% (range: −0.21% to 1.41%; P < 0.001), respectively. Conclusions In this study, we developed a method based on DL that can rapidly generate VNC images from CE images without a DECT scanner. Compared with the DECT approach, the DL‐based method improved the prediction accuracy of CT numbers in bone regions. Consequently, there was greater agreement between the VNCDL and TNC plan dose distributions than with the CE and VNCDECT plans, achieved by suppressing the contrast material signals while retaining the CT numbers of bone structures.

show abstract

Section: Discussionmentioning

confidence: 96%

“…1(b)]. 28 First, a convolutional layer with 48 filters was applied to the input images, followed by the densely connected block. This comprised four units of a batch normalization layer, exponential linear unit activation, and a 3 9 3 convolutional layer with 24 filters.…”

Section: B Deep Learning Methodsmentioning

confidence: 99%

Deep learning‐based virtual noncontrast CT for volumetric modulated arc therapy planning: Comparison with a dual‐energy CT‐based approach

Koike¹,

Ohira²,

Akino

et al. 2019

Medical Physics

View full text Add to dashboard Cite

show abstract

“…Although, currently, the usage of GAN is still limited to image processing and structure reconstruction [90,15], as a generative model, it has the potential for designing new molecules and drugs [132,122]. To achieve this goal, researchers need to make much more efforts.…”

Section: Biology Image Super-resolution Using Ganmentioning

confidence: 99%

“…We also require the super-resolution image to have more true details but less fake ones. This image super-resolution technique has achieved great success in fMRI image fast acquisition [15] and fluorescence microscopy super-resolution [90].…”

Section: Biology Image Super-resolution Using Ganmentioning

confidence: 99%

Deep learning in bioinformatics: Introduction, application, and perspective in the big data era

Liu

Huang

Ding

et al. 2019

Methods

286

107

View full text Add to dashboard Cite

Deep learning, which is especially formidable in handling big data, has achieved great success in various fields, including bioinformatics. With the advances of the big data era in biology, it is foreseeable that deep learning will become increasingly important in the field and will be incorporated in vast majorities of analysis pipelines. In this review, we provide both the exoteric introduction of deep learning, and concrete examples and implementations of its representative applications in bioinformatics. We start from the recent achievements of deep learning in the bioinformatics field, pointing out the problems which are suitable to use deep learning. After that, we introduce deep learning in an easy-to-understand fashion, from shallow neural networks to legendary convolutional neural networks, legendary recurrent neural networks, graph neural networks, generative adversarial networks, variational autoencoder, and the most recent state-of-the-art architectures. After that, we provide eight examples, covering five bioinformatics research directions and all the four kinds of data type, with the implementation written in Tensorflow and Keras. Finally, we discuss the common issues, such as overfitting and interpretability, that users will encounter when adopting deep learning methods and provide corresponding suggestions. The implementations are freely available at https://github.com/lykaust15/Deep_learning_examples.

show abstract

“…[7] and Chen et.al. [8], form 3D solutions. As analyzed in the authors' previous paper [9], the machine learning based techniques outperformed the traditional techniques in both the quality of the SR images and in the amount of time The SRGAN is composed of a generator and discriminator network trained in an adversarial fashion and processes 2D images but creates a 3D volume by stacking SR outputs required to acquire the result.…”

Section: Related Workmentioning

confidence: 99%

Anisotropic Super Resolution In Prostate Mri Using Super Resolution Generative Adversarial Networks

Sood

Rusu

2019

2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019)

View full text Add to dashboard Cite

Acquiring High Resolution (HR) Magnetic Resonance (MR) images requires the patient to remain still for long periods of time, which causes patient discomfort and increases the probability of motion induced image artifacts. A possible solution is to acquire low resolution (LR) images and to process them with the Super Resolution Generative Adversarial Network (SRGAN) to create a super-resolved version. This work applies SRGAN to MR images of the prostate and performs three experiments. The first experiment explores improving the in-plane MR image resolution by factors of 4 and 8, and shows that, while the PSNR and SSIM (Structural SIMilarity) metrics are lower than the isotropic bicubic interpolation baseline, the SRGAN is able to create images that have high edge fidelity. The second experiment explores anisotropic super-resolution via synthetic images, in that the input images to the network are anisotropically downsampled versions of HR images. This experiment demonstrates the ability of the modified SRGAN to perform anisotropic superresolution, with quantitative image metrics that are comparable to those of the anisotropic bicubic interpolation baseline. Finally, the third experiment applies a modified version of the SRGAN to super-resolve anisotropic images obtained from the through-plane slices of the volumetric MR data. The output super-resolved images contain a significant amount of high frequency information that make them visually close to their HR counterparts. Overall, the promising results from each experiment show that super-resolution for MR images is a successful technique and that producing isotropic MR image volumes from anisotropic slices is an achievable goal.

show abstract

Brain MRI super resolution using 3D deep densely connected neural networks

Cited by 229 publications

References 16 publications

Deep learning‐based virtual noncontrast CT for volumetric modulated arc therapy planning: Comparison with a dual‐energy CT‐based approach

Deep learning‐based virtual noncontrast CT for volumetric modulated arc therapy planning: Comparison with a dual‐energy CT‐based approach

Deep learning in bioinformatics: Introduction, application, and perspective in the big data era

Anisotropic Super Resolution In Prostate Mri Using Super Resolution Generative Adversarial Networks

Contact Info

Product

Resources

About