scholarly journals Realistic High-resolution Body Computed Tomography Image Synthesis Using Progressive Growing Generative Adversarial Network: A Visual Turing Test (Preprint)

10.2196/23328 ◽  
2020 ◽  
Author(s):  
Ho Young Park ◽  
Hyun-Jin Bae ◽  
Gil-Sun Hong ◽  
Minjee Kim ◽  
JiHye Yun ◽  
...  
Keyword(s):  
Computed Tomography ◽  
High Resolution ◽  
Image Synthesis ◽  
Turing Test ◽  
Generative Adversarial Network ◽  
Adversarial Network ◽  
Tomography Image ◽  
Computed Tomography Image

scholarly journals Realistic high-resolution body CT image synthesis using progressive growing generative adversarial network: A Visual Turing Test (Preprint)

2020 ◽  
Author(s):  
Ho Young Park ◽  
Hyun-Jun Bae ◽  
Gil-Sun Hong ◽  
JiHye Yun ◽  
Sung Won Park ◽  
...  
Keyword(s):  
High Resolution ◽  
Image Synthesis ◽  
Turing Test ◽  
Ct Images ◽  
Body Images ◽  
Generative Adversarial Network ◽  
Adversarial Network ◽  
Image Set ◽  
Validation Set ◽  
Body Ct

BACKGROUND Generative Adversarial Network (GAN)-based synthetic images can be viable solutions to current supervised deep learning challenges. However, generating highly realistic images is a prerequisite for these approaches. OBJECTIVE We investigated and validated the unsupervised synthesis of highly realistic body CT images using a progressive growing GAN (PGGAN) trained to learn the probability distribution of normal data. METHODS We trained the PGGAN using 11 755 body CT scans. Ten radiologists then evaluated the results in a binary approach using an independent validation set of 300 images (150 real, 150 synthetic) to judge the authenticity of each image. RESULTS Mean accuracy for the entire image set was low (59.4%), and accuracy among three reader groups with different experience levels was not significantly different (58.0% - 60.5%, P = 0.36). Inter-reader agreements were poor (κ = 0.11) for the entire image set. In subgroup analysis, the discrepancies between real and synthetic CT images occurred mainly in the thoracoabdominal junction and in anatomical details. CONCLUSIONS The GAN can synthesize highly realistic high-resolution body CT images, which are indistinguishable from real images; however, it has limitations in generating body images in the thoracoabdominal junction and lacks accuracy in anatomical details.

scholarly journals Text to Realistic Image Generation with Attentional Concatenation Generative Adversarial Networks

2020 ◽  
Vol 2020 ◽  
pp. 1-10
Author(s):  
Linyan Li ◽  
Yu Sun ◽  
Fuyuan Hu ◽  
Tao Zhou ◽  
Xuefeng Xi ◽  
...  
Keyword(s):  
High Resolution ◽  
Image Synthesis ◽  
Semantic Space ◽  
Generative Adversarial Networks ◽  
Generative Adversarial Network ◽  
Fine Grained ◽  
Adversarial Network ◽  
Adversarial Networks ◽  
Cascade Structure ◽  
High Resolution Images

In this paper, we propose an Attentional Concatenation Generative Adversarial Network (ACGAN) aiming at generating 1024 × 1024 high-resolution images. First, we propose a multilevel cascade structure, for text-to-image synthesis. During training progress, we gradually add new layers and, at the same time, use the results and word vectors from the previous layer as inputs to the next layer to generate high-resolution images with photo-realistic details. Second, the deep attentional multimodal similarity model is introduced into the network, and we match word vectors with images in a common semantic space to compute a fine-grained matching loss for training the generator. In this way, we can pay attention to the fine-grained information of the word level in the semantics. Finally, the measure of diversity is added to the discriminator, which enables the generator to obtain more diverse gradient directions and improve the diversity of generated samples. The experimental results show that the inception scores of the proposed model on the CUB and Oxford-102 datasets have reached 4.48 and 4.16, improved by 2.75% and 6.42% compared to Attentional Generative Adversarial Networks (AttenGAN). The ACGAN model has a better effect on text-generated images, and the resulting image is closer to the real image.

scholarly journals Colonoscopic image synthesis with generative adversarial network for enhanced detection of sessile serrated lesions using convolutional neural network

2022 ◽  
Vol 12 (1) ◽  
Author(s):  
Dan Yoon ◽  
Hyoun-Joong Kong ◽  
Byeong Soo Kim ◽  
Woo Sang Cho ◽  
Jung Chan Lee ◽  
...  
Keyword(s):  
High Resolution ◽  
Detection System ◽  
Morphological Characteristics ◽  
Image Synthesis ◽  
Polyp Detection ◽  
Generative Adversarial Network ◽  
Adversarial Network ◽  
Serrated Lesions ◽  
Low Performance ◽  
Cade System

AbstractComputer-aided detection (CADe) systems have been actively researched for polyp detection in colonoscopy. To be an effective system, it is important to detect additional polyps that may be easily missed by endoscopists. Sessile serrated lesions (SSLs) are a precursor to colorectal cancer with a relatively higher miss rate, owing to their flat and subtle morphology. Colonoscopy CADe systems could help endoscopists; however, the current systems exhibit a very low performance for detecting SSLs. We propose a polyp detection system that reflects the morphological characteristics of SSLs to detect unrecognized or easily missed polyps. To develop a well-trained system with imbalanced polyp data, a generative adversarial network (GAN) was used to synthesize high-resolution whole endoscopic images, including SSL. Quantitative and qualitative evaluations on GAN-synthesized images ensure that synthetic images are realistic and include SSL endoscopic features. Moreover, traditional augmentation methods were used to compare the efficacy of the GAN augmentation method. The CADe system augmented with GAN synthesized images showed a 17.5% improvement in sensitivity on SSLs. Consequently, we verified the potential of the GAN to synthesize high-resolution images with endoscopic features and the proposed system was found to be effective in detecting easily missed polyps during a colonoscopy.

Digital rock physics: Segmentation of sub-resolution features

2020 ◽  
Author(s):  
Martin Balcewicz ◽  
Erik H. Saenger
Keyword(s):  
Computed Tomography ◽  
High Resolution ◽  
Rock Physics ◽  
Rock Properties ◽  
X Ray ◽  
Digital Rock Physics ◽  
Tomography Image ◽  
X Ray Computed ◽  
Computed Tomography Image ◽  
Elastic Rock

<p>Digital rock physics (DRP) became a complementary part in reservoir characterization during the last two decades. Deriving transport, thermal, or effective elastic rock properties from a digital twin requires a three-step workflow: (1) Preparation of a high-resolution X-ray computed tomography image, (2) segmentation of pore and grain phases, respectively, and (3) solving equations due to the demanded properties. Despite the high resolution µ-CT images, the numerical predictions of rock properties have their specific uncertainties compared to laboratory measurements. Missing unresolved features in the µ-CT image might be the key issue. These findings indicate the importance of a full understanding of the rocks microfabrics. Most digital models used in DRP treat the rock as a heterogeneous, isotropic, intact medium which neglect unresolved features. However, we expect features within the microfabrics like micro-cracks, small-scale fluid inclusions, or stressed grains which may influence the elastic rock properties but have not been taken into account in DRP, yet. Former studies have shown resolution-issues in grain-to-grain contacts within sandstones or inaccuracies due to micro-porosity in carbonates, this means the micritic phase. Within the scope of this abstract, we image two different sandstone samples, Bentheim and Ruhrsandstone, as well as one carbonate sample. Here, we compare the mentioned difficulties of X-ray visualization with further analytical methods, this means thin section and focused ion beam measurements. This results into a better understanding of the rocks microstructures and allows us to segment unresolved features in the X-ray computed tomography image. Those features can be described with effective properties at the µ-scale in the DRP workflow to reduce the uncertainty of the predicted rock properties at the meso- and fieldscale.</p>

scholarly journals Visual and Quantitative Evaluation of Amyloid Brain PET Image Synthesis with Generative Adversarial Network

2020 ◽  
Vol 10 (7) ◽  
pp. 2628 ◽  
Author(s):  
Hyeon Kang ◽  
Jang-Sik Park ◽  
Kook Cho ◽  
Do-Young Kang
Keyword(s):  
Quantitative Evaluation ◽  
Data Augmentation ◽  
Image Synthesis ◽  
Turing Test ◽  
Training Data ◽  
Generative Adversarial Network ◽  
Visual Evaluation ◽  
Image Distribution ◽  
Adversarial Network ◽  
Quantitative Measurements

Conventional data augmentation (DA) techniques, which have been used to improve the performance of predictive models with a lack of balanced training data sets, entail an effort to define the proper repeating operation (e.g., rotation and mirroring) according to the target class distribution. Although DA using generative adversarial network (GAN) has the potential to overcome the disadvantages of conventional DA, there are not enough cases where this technique has been applied to medical images, and in particular, not enough cases where quantitative evaluation was used to determine whether the generated images had enough realism and diversity to be used for DA. In this study, we synthesized 18F-Florbetaben (FBB) images using CGAN. The generated images were evaluated using various measures, and we presented the state of the images and the similarity value of quantitative measurement that can be expected to successfully augment data from generated images for DA. The method includes (1) conditional WGAN-GP to learn the axial image distribution extracted from pre-processed 3D FBB images, (2) pre-trained DenseNet121 and model-agnostic metrics for visual and quantitative measurements of generated image distribution, and (3) a machine learning model for observing improvement in generalization performance by generated dataset. The Visual Turing test showed similarity in the descriptions of typical patterns of amyloid deposition for each of the generated images. However, differences in similarity and classification performance per axial level were observed, which did not agree with the visual evaluation. Experimental results demonstrated that quantitative measurements were able to detect the similarity between two distributions and observe mode collapse better than the Visual Turing test and t-SNE.

A novel computed tomography image synthesis method for correcting the spectrum dependence of CT numbers

2020 ◽  
Vol 65 (2) ◽  
pp. 025013
Author(s):  
Yi-Wen Chen ◽  
Hsin-Yuan Fang ◽  
Yi-Chun Wang ◽  
Shin-Lei Peng ◽  
Cheng-Ting Shih
Keyword(s):  
Computed Tomography ◽  
Synthesis Method ◽  
Image Synthesis ◽  
Tomography Image ◽  
Computed Tomography Image
Sign in / Sign up

Export Citation Format

Share Document