☆ 3.8 Editorial Material

Addressing inter-device variations in optical coherence tomography angiography: will image-to-image translation systems help?

INTERNATIONAL JOURNAL OF RETINA AND VITREOUS (2023)

Related references

Note: Only part of the references are listed.

Review Medicine, General & Internal

Deep Learning in Optical Coherence Tomography Angiography: Current Progress, Challenges, and Future Directions

Dawei Yang et al.

Summary: Optical coherence tomography angiography (OCT-A) provides non-invasive visualization of retinal microvasculature, and deep learning (DL) has been applied in OCT-A image analysis to enhance its clinical values. However, the deployment of this combination in real-world clinics is still in the proof-of-concept stage due to limitations in training sample size, data preprocessing, external dataset testing, and results interpretation standardization. In this review, we introduce the applications of DL in OCT-A, summarize the challenges of clinical deployment, and discuss future research directions.

DIAGNOSTICS (2023)

Add to Collection

Review Engineering, Biomedical

Deep learning based unpaired image-to-image translation applications for medical physics: a systematic review

Junhua Chen et al.

Summary: There is a lack of systematic review on the application of unpaired image-to-image translation in medical imaging for medical physicists. This article provides a comprehensive review of the challenges and opportunities for applying this technique in practice. The results show that unpaired image-to-image translation has various applications in medical imaging, such as segmentation, domain adaptation, and denoising. However, the scarcity of external validation studies and publicly available pre-trained models limits the immediate applicability of these methods.

PHYSICS IN MEDICINE AND BIOLOGY (2023)

Add to Collection

Article Computer Science, Artificial Intelligence

Diffusion models in medical imaging: A comprehensive survey

Amirhossein Kazerouni et al.

Summary: Denoising diffusion models have attracted great interest in deep learning problems. These models gradually perturb input data with Gaussian noise and learn to reverse the process to retrieve noise-free data. The field of medical imaging has also shown growing interest in diffusion models. This survey provides a comprehensive overview of diffusion models in medical imaging, covering theoretical foundations, different modeling frameworks, application taxonomy, practical use cases, limitations, and future directions. The overviewed studies and open-source implementations are also available on GitHub.

MEDICAL IMAGE ANALYSIS (2023)

Add to Collection

Article Multidisciplinary Sciences

A multimodal comparison of latent denoising diffusion probabilistic models and generative adversarial networks for medical image synthesis

Gustav Mueller-Franzes et al.

Summary: This study introduces Medfusion, a conditional latent DDPM for medical image generation, which outperforms GANs in terms of diversity and fidelity. By comparing with StyleGAN-3, ProGAN, and cGAN on three datasets, Medfusion achieves superior scores in diversity and fidelity, demonstrating its potential as a promising alternative for generating high-quality medical images.

SCIENTIFIC REPORTS (2023)

Add to Collection

Review Ophthalmology

Application of generative adversarial networks (GAN) for ophthalmology image domains: a survey

Aram You et al.

Summary: GAN has a wide range of applications in ophthalmology image domains, including segmentation, data augmentation, denoising, domain transfer, super-resolution, post-intervention prediction, and feature extraction. However, GAN also has limitations such as mode collapse, spatial deformities, unintended changes, and the generation of high-frequency noises and artifacts of checkerboard patterns.

EYE AND VISION (2022)

Add to Collection

Article Engineering, Biomedical

Domain Adaptation for Medical Image Analysis: A Survey

Hao Guan et al.

Summary: This paper surveys the recent advances of domain adaptation methods in medical image analysis, presents the motivation of introducing domain adaptation techniques to tackle domain heterogeneity issues in medical image analysis, and reviews the recent domain adaptation models in various medical image analysis tasks.

IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING (2022)

Add to Collection

Proceedings Paper Computer Science, Interdisciplinary Applications

Unsupervised Denoising of Retinal OCT with Diffusion Probabilistic Model

Dewei Hu et al.

Summary: In this study, an unsupervised diffusion probabilistic model is proposed to learn from noise for OCT image restoration, which significantly improves the image quality.

MEDICAL IMAGING 2022: IMAGE PROCESSING (2022)

Add to Collection

Article Ophthalmology

Microaneurysms visualisation using five different optical coherence tomography angiography devices compared to fluorescein angiography

Salvatore Parrulli et al.

Summary: This study compared the performance of FA and five different OCTA devices in evaluating MAs secondary to DR, and found that FA remains the best imaging modality while Spectralis OCTA was able to detect more MAs compared to other devices. However, the high variability between OCTA devices should be taken into account in future clinical trials and practice.

BRITISH JOURNAL OF OPHTHALMOLOGY (2021)

Add to Collection

Article Computer Science, Interdisciplinary Applications

Feasibility study to improve deep learning in OCT diagnosis of rare retinal diseases with few-shot classification

Tae Keun Yoo et al.

Summary: The study demonstrated the improvement of applying deep learning (DL) in diagnosing rare ophthalmic diseases through few-shot learning (FSL) using generative adversarial network (GAN). Synthetic images helped extract rare disease features and the DL model showed significant enhancement in OCT diagnosis, outperforming traditional models and reducing diagnosis delay.

MEDICAL & BIOLOGICAL ENGINEERING & COMPUTING (2021)

Add to Collection

Article Computer Science, Interdisciplinary Applications

Unsupervised CT Metal Artifact Learning Using Attention-Guided β-CycleGAN

Junghyun Lee et al.

Summary: Metal artifact reduction is a significant research topic in CT, with deep learning methods being widely used. Although supervised learning methods are popular, it is challenging to obtain matched image pairs in real CT acquisition. A recent unsupervised learning approach was proposed, but the complex network architecture struggled with large clinical images. This study introduces a simple and effective unsupervised learning method based on a novel beta-cycleGAN architecture, which, combined with CBAM layers in the generator, achieves improved metal artifact reduction.

IEEE TRANSACTIONS ON MEDICAL IMAGING (2021)

Add to Collection

Article Computer Science, Interdisciplinary Applications

Leveraging Regular Fundus Images for Training UWF Fundus Diagnosis Models via Adversarial Learning and Pseudo-Labeling

Lie Ju et al.

Summary: This study proposes the use of an improved CycleGAN model to enhance limited UWF fundus data and annotations using regular fundus images for more efficient training. Our method does not require paired images or the same semantic labels, and it is robust to noise and errors introduced by the generated unlabeled data with the pseudo-labeling technique.

IEEE TRANSACTIONS ON MEDICAL IMAGING (2021)

Add to Collection

Article Computer Science, Interdisciplinary Applications

Adopting low-shot deep learning for the detection of conjunctival melanoma using ocular surface images

Tae Keun Yoo et al.

Summary: The study developed a low-shot deep learning model using GAN-based augmentation for conjunctival melanoma detection, achieving significant improvement in classification accuracy. Among the models tested, MobileNetV2 showed the highest accuracy in four-class classification and 3D melanoma phantom image detection, highlighting the potential of deep learning in ocular surface image analysis.

COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE (2021)

Add to Collection

Article Oncology

CT-Based Pelvic T1-Weighted MR Image Synthesis Using UNet, UNet plus plus and Cycle-Consistent Generative Adversarial Network (Cycle-GAN)

Reza Kalantar et al.

Summary: This study utilized deep learning techniques to synthesize pelvic T-1 weighted MRI images, with results showing Cycle-GAN excelling in generating high-quality images compared to UNet and UNet++. Radiologists' qualitative assessments revealed variability in test scores, but they were able to correctly identify real images from synthetic ones. Additionally, successful unsupervised segmentation of pelvic bones on T1-weighted images was achieved in several test cases.

FRONTIERS IN ONCOLOGY (2021)

Add to Collection

Article Radiology, Nuclear Medicine & Medical Imaging

Investigation of Low-Dose CT Image Denoising Using Unpaired Deep Learning Methods

Zeheng Li et al.

Summary: The study proposes an image-domain denoising method based on CycleGAN that can effectively learn image translation from low-dose domain to full-dose domain without aligning FDCT and LDCT images. Experimental results show that this method achieves good denoising performance in LDCT image processing, comparable to or better than other state-of-the-art denoising methods.

IEEE TRANSACTIONS ON RADIATION AND PLASMA MEDICAL SCIENCES (2021)

Add to Collection

Article Ophthalmology

COMPARISON BETWEEN SEVERAL OPTICAL COHERENCE TOMOGRAPHY ANGIOGRAPHY DEVICES AND INDOCYANINE GREEN ANGIOGRAPHY OF CHOROIDAL NEOVASCULARIZATION

Federico Corvi et al.

RETINA-THE JOURNAL OF RETINAL AND VITREOUS DISEASES (2020)

Add to Collection

Article Biochemical Research Methods

Reducing image variability across OCT devices with unsupervised unpaired learning for improved segmentation of retina

David Romo-Bucheli et al.

BIOMEDICAL OPTICS EXPRESS (2020)

Add to Collection

Article Ophthalmology

CycleGAN-based deep learning technique for artifact reduction in fundus photography

Tae Keun Yoo et al.

GRAEFES ARCHIVE FOR CLINICAL AND EXPERIMENTAL OPHTHALMOLOGY (2020)

Add to Collection

Article Engineering, Electrical & Electronic

Unsupervised Super-Resolution of OCT Images Using Generative Adversarial Network for Improved Age-Related Macular Degeneration Diagnosis

Vineeta Das et al.

IEEE SENSORS JOURNAL (2020)

Add to Collection

Article Multidisciplinary Sciences

Discordant vascular parameter measurements in diabetic and non-diabetic eyes detected by different optical coherence tomography angiography devices

Yi Chen et al.

PLOS ONE (2020)

Add to Collection

Article Computer Science, Interdisciplinary Applications

Deep learning can generate traditional retinal fundus photographs using ultra-widefield images via generative adversarial networks

Tae Keun Yoo et al.

COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE (2020)

Add to Collection

Article Computer Science, Information Systems

Dehaze of Cataractous Retinal Images Using an Unpaired Generative Adversarial Network

Yuhao Luo et al.

IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS (2020)

Add to Collection

Article Ophthalmology

Comparison of two different optical coherence tomography angiography devices in detecting healthy versus glaucomatous eyes - an observational cross-sectional study

A. R. Kee et al.

BMC OPHTHALMOLOGY (2020)

Add to Collection

Article Computer Science, Artificial Intelligence

MCMT-GAN: Multi-Task Coherent Modality Transferable GAN for 3D Brain Image Synthesis

Yawen Huang et al.

IEEE TRANSACTIONS ON IMAGE PROCESSING (2020)

Add to Collection

Article Ophthalmology

Reproducibility of Vessel Density, Fractal Dimension, and Foveal Avascular Zone Using 7 Different Optical Coherence Tomography Angiography Devices

Federico Corvi et al.

AMERICAN JOURNAL OF OPHTHALMOLOGY (2018)

Add to Collection

Review Ophthalmology

Optical coherence tomography angiography

Richard F. Spaide et al.

PROGRESS IN RETINAL AND EYE RESEARCH (2018)

Add to Collection

Article Multidisciplinary Sciences

The Impact of Utilizing Different Optical Coherence Tomography Devices for Clinical Purposes and in Multiple Sclerosis Trials

Christina V. Warner et al.

PLOS ONE (2011)

Add to Collection

© Peeref 2019-2024. All rights reserved.