Applying a Pix2Pix Generative Adversarial Network to a Fourier-Domain Optical Coherence Tomography System for Artifact Elimination

Huang, Chunming; Wijanto, Eddy; Cheng, Hsu-Chih

doi:10.1109/access.2021.3098865

Cited by 10 publications

(3 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To date, no previous study has reported the use of pix2pix GAN in this type of image translation task. The basic pix2pix framework has been used for medical image denoising, reconstruction, segmentation, and amplification of the original dataset [ 12 , 15 , 25 , 26 ]. In our study, we developed and evaluated the use of this network to obtain synthetic, clinically useful OCT and FA images that when used in concert, can assist the retinal specialist in decision making while treating patients with DME in case of machine failure/unavailability (either OCT or FA) or situations such as the presence of limited pupillary dilatation or significant media opacity that prevent performance or compromise the quality of FA, while OCT can still be performed with satisfactory image quality.…”

Section: Discussionmentioning

confidence: 99%

Bridging the resources gap: deep learning for fluorescein angiography and optical coherence tomography macular thickness map image translation

Abdelmotaal

Saad²,

Soliman³

et al. 2022

BMC Ophthalmol

View full text Add to dashboard Cite

Background To assess the ability of the pix2pix generative adversarial network (pix2pix GAN) to synthesize clinically useful optical coherence tomography (OCT) color-coded macular thickness maps based on a modest-sized original fluorescein angiography (FA) dataset and the reverse, to be used as a plausible alternative to either imaging technique in patients with diabetic macular edema (DME). Methods Original images of 1,195 eyes of 708 nonconsecutive diabetic patients with or without DME were retrospectively analyzed. OCT macular thickness maps and corresponding FA images were preprocessed for use in training and testing the proposed pix2pix GAN. The best quality synthesized images using the test set were selected based on the Fréchet inception distance score, and their quality was studied subjectively by image readers and objectively by calculating the peak signal-to-noise ratio, structural similarity index, and Hamming distance. We also used original and synthesized images in a trained deep convolutional neural network (DCNN) to plot the difference between synthesized images and their ground-truth analogues and calculate the learned perceptual image patch similarity metric. Results The pix2pix GAN-synthesized images showed plausible subjectively and objectively assessed quality, which can provide a clinically useful alternative to either image modality. Conclusion Using the pix2pix GAN to synthesize mutually dependent OCT color-coded macular thickness maps or FA images can overcome issues related to machine unavailability or clinical situations that preclude the performance of either imaging technique. Trial registration ClinicalTrials.gov Identifier: NCT05105620, November 2021. “Retrospectively registered”.

show abstract

Section: Discussionmentioning

confidence: 99%

Bridging the resources gap: deep learning for fluorescein angiography and optical coherence tomography macular thickness map image translation

Abdelmotaal

Saad²,

Soliman³

et al. 2022

BMC Ophthalmol

View full text Add to dashboard Cite

show abstract

“…Reconstruction of the under-sampled chest CT data was achieved via TL using a pre-trained Pix2Pix-GAN model ( 34 ). Repurposing an already-trained model for another task is known as TL.…”

Section: A Novel Covid Prediction Schemementioning

confidence: 99%

Prediction of the COVID disease using lung CT images by Deep Learning algorithm: DETS-optimized Resnet 101 classifier

Haennah,

Christopher,

King

2023

Front. Med.

View full text Add to dashboard Cite

As a result of the COVID-19 (coronavirus) disease due to SARS-CoV2 becoming a pandemic, it has spread over the globe. It takes time to evaluate the results of the laboratory tests because of the rising number of cases each day. Therefore, there are restrictions in terms of both therapy and findings. A clinical decision-making system with predictive algorithms is needed to alleviate the pressure on healthcare systems via Deep Learning (DL) algorithms. With the use of DL and chest scans, this research intends to determine COVID-19 patients by utilizing the Transfer Learning (TL)-based Generative Adversarial Network (Pix 2 Pix-GAN). Moreover, the COVID-19 images are then classified as either positive or negative using a Duffing Equation Tuna Swarm (DETS)-optimized Resnet 101 classifier trained on synthetic and real images from the Kaggle lung CT Covid dataset. Implementation of the proposed technique is done using MATLAB simulations. Besides, is evaluated via accuracy, precision, F1-score, recall, and AUC. Experimental findings show that the proposed prediction model identifies COVID-19 patients with 97.2% accuracy, a recall of 95.9%, and a specificity of 95.5%, which suggests the proposed predictive model can be utilized to forecast COVID-19 infection by medical specialists for clinical prediction research and can be beneficial to them.

show abstract

“…In most state-of-the-art GAN models whose objective is to synthesize images [6], [7], [8], [9], the D and the G are convolutional neural networks based on the DCGAN [10] and use the formula (1) as their loss function.…”

Section: Log(1 − D(g(z)))]mentioning

confidence: 99%

Stabilizing and Improving Training of Generative Adversarial Networks Through Identity Blocks and Modified Loss Function

2023

View full text Add to dashboard Cite

Generative adversarial networks (GANs) are a powerful tool for synthesizing realistic images, but they can be difficult to train and are prone to instability and mode collapse. This paper proposes a new model called Identity Generative Adversarial Network (IGAN) that addresses these issues. This model is based on three modifications to the baseline deep convolutional generative adversarial network (DCGAN). The first change is to add a non-linear identity block to the architecture. This will make it easier for the model to fit complex data types and cut down on the time it takes to train. The second change is to smooth out the standard GAN loss function by using a modified loss function and label smoothing. The third and final change is to use minibatch training to let the model use other examples from the same minibatch as side information to improve the quality and variety of generated images. These changes help to stabilize the training process and improve the model's performance. The performance of the GAN models is compared using the inception score (IS) and the Fréchet inception distance (FID), which are widely used metrics for evaluating the quality and diversity of generated images. The effectiveness of our approach was tested by comparing an IGAN model with other GAN models on the CelebA and stacked MNIST datasets. Results show that IGAN outperforms all the other models, achieving an IS of 13.95 and an FID of 43.71 after traning for 200 epochs. In addition to demonstrating the improvement in the performance of the IGAN, the instabilities, diversity, and fidelity of the models were investigated. The results showed that the IGAN was able to converge to a distribution of the real data more quickly. Furthermore, the experiments revealed that IGAN is capable of producing more stable and high-quality images. This suggests that IGAN is a promising approach for improving the training and performance of GANs and may have a range of applications in image synthesis and other areas.INDEX TERMS Generative adversarial network, deep learning, mode collapse, label smoothing, identity block.

show abstract

Applying a Pix2Pix Generative Adversarial Network to a Fourier-Domain Optical Coherence Tomography System for Artifact Elimination

Cited by 10 publications

References 26 publications

Bridging the resources gap: deep learning for fluorescein angiography and optical coherence tomography macular thickness map image translation

Bridging the resources gap: deep learning for fluorescein angiography and optical coherence tomography macular thickness map image translation

Prediction of the COVID disease using lung CT images by Deep Learning algorithm: DETS-optimized Resnet 101 classifier

Stabilizing and Improving Training of Generative Adversarial Networks Through Identity Blocks and Modified Loss Function

Contact Info

Product

Resources

About