Weakly Supervised Deep Learning-Based Optical Coherence Tomography Angiography

Sub-Second Optical Coherence Tomography Angiography Protocol for Intraoral Imaging Using an Efficient Super-Resolution Network

This study introduces a 200 kHz swept-source optical coherence tomography system-based fast optical coherence tomography angiography (OCTA) protocol for intraoral imaging by integrating an efficient Intraoral Micro-Angiography Super-Resolution Transformer (IMAST) model. This protocol reduces acquisition time to ~0.3 s by reducing the spatial sampling resolution, thereby minimizing motion artifacts while maintaining a field of view and image quality. The IMAST model utilizes a transformer-based architecture combined with convolutional operations to reconstruct high-resolution intraoral OCTA images from reduced-resolution scans. Experimental results from various intraoral sites and conditions show the model's robustness and high performance in enhancing image quality compared to existing deep-learning methods. Besides, IMAST shows advantages in model complexity, inference time, and computational cost, underscoring its suitability for clinical environments. These findings support the potential of our approach for noninvasive oral disease diagnosis, reducing patient discomfort and facilitating early detection of malignancies, thus serving as a valuable tool for oral assessment.

LS-Net: lightweight segmentation network for dermatological epidermal segmentation in optical coherence tomography imaging

Optical coherence tomography (OCT) can be an important tool for non-invasive dermatological evaluation, providing useful data on epidermal integrity for diagnosing skin diseases. Despite its benefits, OCT's utility is limited by the challenges of accurate, fast epidermal segmentation due to the skin morphological diversity. To address this, we introduce a lightweight segmentation network (LS-Net), a novel deep learning model that combines the robust local feature extraction abilities of Convolution Neural Network and the long-term information processing capabilities of Vision Transformer. LS-Net has a depth-wise convolutional transformer for enhanced spatial contextualization and a squeeze-and-excitation block for feature recalibration, ensuring precise segmentation while maintaining computational efficiency. Our network outperforms existing methods, demonstrating high segmentation accuracy (mean Dice: 0.9624 and mean IoU: 0.9468) with significantly reduced computational demands (floating point operations: 1.131 G). We further validate LS-Net on our acquired dataset, showing its effectiveness in various skin sites (e.g., face, palm) under realistic clinical conditions. This model promises to enhance the diagnostic capabilities of OCT, making it a valuable tool for dermatological practice.

Choroidal Optical Coherence Tomography Angiography: Noninvasive Choroidal Vessel Analysis via Deep Learning

Background: The choroid is the most vascularized structure in the human eye, associated with numerous retinal and choroidal diseases. However, the vessel distribution of choroidal sublayers has yet to be effectively explored due to the lack of suitable tools for visualization and analysis. Methods: In this paper, we present a novel choroidal angiography strategy to more effectively evaluate vessels within choroidal sublayers in the clinic. Our approach utilizes a segmentation model to extract choroidal vessels from OCT B-scans layer by layer. Furthermore, we ensure that the model, trained on B-scans with high choroidal quality, can proficiently handle the low-quality B-scans commonly collected in clinical practice for reconstruction vessel distributions. By treating this process as a cross-domain segmentation task, we propose an ensemble discriminative mean teacher structure to address the specificities inherent in this cross-domain segmentation process. The proposed structure can select representative samples with minimal label noise for self-training and enhance the adaptation strength of adversarial training. Results: Experiments demonstrate the effectiveness of the proposed structure, achieving a dice score of 77.28 for choroidal vessel segmentation. This validates our strategy to provide satisfactory choroidal angiography noninvasively, supportting the analysis of choroidal vessel distribution for paitients with choroidal diseases. We observed that patients with central serous chorioretinopathy have evidently (P < 0.05) lower vascular indexes at all choroidal sublayers than healthy individuals, especially in the region beyond central fovea of macula (larger than 6 mm). Conclusions: We release the code and training set of the proposed method as the first noninvasive mechnism to assist clinical application for the analysis of choroidal vessels.

Noise-imitation learning: unpaired speckle noise reduction for optical coherence tomography

Objective.Optical coherence tomography (OCT) is widely used in clinical practice for its non-invasive, high-resolution imaging capabilities. However, speckle noise inherent to its low coherence principle can degrade image quality and compromise diagnostic accuracy. While deep learning methods have shown promise in reducing speckle noise, obtaining well-registered image pairs remains challenging, leading to the development of unpaired methods. Despite their potential, existing unpaired methods suffer from redundancy in network structures or interaction mechanisms. Therefore, a more streamlined method for unpaired OCT denoising is essential.Approach.In this work, we propose a novel unpaired method for OCT image denoising, referred to as noise-imitation learning (NIL). NIL comprises three primary modules: the noise extraction module, which extracts noise features by denoising noisy images; the noise imitation module, which synthesizes noisy images and generates fake clean images; and the adversarial learning module, which differentiates between real and fake clean images through adversarial training. The complexity of NIL is significantly lower than that of previous unpaired methods, utilizing only one generator and one discriminator for training.Main results.By efficiently fusing unpaired images and employing adversarial training, NIL can extract more speckle noise information to enhance denoising performance. Building on NIL, we propose an OCT image denoising pipeline, NIL-NAFNet. This pipeline achieved PSNR, SSIM, and RMSE values of 31.27 dB, 0.865, and 7.00, respectively, on the PKU37 dataset. Extensive experiments suggest that our method outperforms state-of-the-art unpaired methods both qualitatively and quantitatively.Significance.These findings indicate that the proposed NIL is a simple yet effective method for unpaired OCT speckle noise reduction. The OCT denoising pipeline based on NIL demonstrates exceptional performance and efficiency. By addressing speckle noise without requiring well-registered image pairs, this method can enhance image quality and diagnostic accuracy in clinical practice.

Improved microvascular imaging with optical coherence tomography using 3D neural networks and a channel attention mechanism

Skin microvasculature is vital for human cardiovascular health and thermoregulation, but its imaging and analysis presents significant challenges. Statistical methods such as speckle decorrelation in optical coherence tomography angiography (OCTA) often require multiple co-located B-scans, leading to lengthy acquisitions prone to motion artefacts. Deep learning has shown promise in enhancing accuracy and reducing measurement time by leveraging local information. However, both statistical and deep learning methods typically focus solely on processing individual 2D B-scans, neglecting contextual information from neighbouring B-scans. This limitation compromises spatial context and disregards the 3D features within tissue, potentially affecting OCTA image accuracy. In this study, we propose a novel approach utilising 3D convolutional neural networks (CNNs) to address this limitation. By considering the 3D spatial context, these 3D CNNs mitigate information loss, preserving fine details and boundaries in OCTA images. Our method reduces the required number of B-scans while enhancing accuracy, thereby increasing clinical applicability. This advancement holds promise for improving clinical practices and understanding skin microvascular dynamics crucial for cardiovascular health and thermoregulation.

PSCAT: a lightweight transformer for simultaneous denoising and super-resolution of OCT images

Optical coherence tomography (OCT), owing to its non-invasive nature, has demonstrated tremendous potential in clinical practice and has become a prevalent diagnostic method. Nevertheless, the inherent speckle noise and low sampling rate in OCT imaging often limit the quality of OCT images. In this paper, we propose a lightweight Transformer to efficiently reconstruct high-quality images from noisy and low-resolution OCT images acquired by short scans. Our method, PSCAT, parallelly employs spatial window self-attention and channel attention in the Transformer block to aggregate features from both spatial and channel dimensions. It explores the potential of the Transformer in denoising and super-resolution for OCT, reducing computational costs and enhancing the speed of image processing. To effectively assist in restoring high-frequency details, we introduce a hybrid loss function in both spatial and frequency domains. Extensive experiments demonstrate that our PSCAT has fewer network parameters and lower computational costs compared to state-of-the-art methods while delivering a competitive performance both qualitatively and quantitatively.

VET: Vasculature Extraction Transformer for Single-Scan Optical Coherence Tomography Angiography

Optical coherence tomography angiography (OCTA) is a non-invasive imaging modality for analyzing skin microvasculature, enabling non-invasive diagnosis and treatment monitoring. Traditional OCTA algorithms necessitate at least two-repeated scans to generate microvasculature images, while image quality is highly dependent on the repetitions of scans (e.g., 4-8). Nevertheless, a higher repetition count increases data acquisition time, causing patient discomfort and more unpredictable motion artifacts, which can result in potential misdiagnosis. To address these limitations, we proposed a vasculature extraction pipeline based on the novelty vasculature extraction transformer (VET) to generate OCTA images by using a single OCT scan. Distinct from the vision Transformer, VET utilizes convolutional projection to better learn the spatial relationships between image patches. This study recruited 15 healthy participants. The OCT scans were performed in five various skin sites, i.e., palm, arm, face, neck, and lip. Our results show that in comparison to OCTA images obtained by the speckle variance OCTA (peak-signal-to-noise ratio (PSNR): 16.13) and eigen-decomposition OCTA (PSNR: 17.08) using four repeated OCT scans, OCTA images extracted by the proposed pipeline exhibit a better PSNR (18.03) performance while reducing the data acquisition time by 75%. Visual comparisons show that the proposed pipeline outperformed traditional OCTA algorithms, particularly in the imaging of lip and face areas, where artifacts are commonly encountered. This study is the first to demonstrate that the VET can efficiently extract high-quality vasculature images from a single, rapid OCT scan. This capability significantly enhances diagnostic accuracy for patients and streamlines the imaging process.

Insights into artificial intelligence in myopia management: from a data perspective

Given the high incidence and prevalence of myopia, the current healthcare system is struggling to handle the task of myopia management, which is worsened by home quarantine during the ongoing COVID-19 pandemic. The utilization of artificial intelligence (AI) in ophthalmology is thriving, yet not enough in myopia. AI can serve as a solution for the myopia pandemic, with application potential in early identification, risk stratification, progression prediction, and timely intervention. The datasets used for developing AI models are the foundation and determine the upper limit of performance. Data generated from clinical practice in managing myopia can be categorized into clinical data and imaging data, and different AI methods can be used for analysis. In this review, we comprehensively review the current application status of AI in myopia with an emphasis on data modalities used for developing AI models. We propose that establishing large public datasets with high quality, enhancing the model's capability of handling multimodal input, and exploring novel data modalities could be of great significance for the further application of AI for myopia.

U-shaped fusion convolutional transformer based workflow for fast optical coherence tomography angiography generation in lips

Oral disorders, including oral cancer, pose substantial diagnostic challenges due to late-stage diagnosis, invasive biopsy procedures, and the limitations of existing non-invasive imaging techniques. Optical coherence tomography angiography (OCTA) shows potential in delivering non-invasive, real-time, high-resolution vasculature images. However, the quality of OCTA images are often compromised due to motion artifacts and noise, necessitating more robust and reliable image reconstruction approaches. To address these issues, we propose a novel model, a U-shaped fusion convolutional transformer (UFCT), for the reconstruction of high-quality, low-noise OCTA images from two-repeated OCT scans. UFCT integrates the strengths of convolutional neural networks (CNNs) and transformers, proficiently capturing both local and global image features. According to the qualitative and quantitative analysis in normal and pathological conditions, the performance of the proposed pipeline outperforms that of the traditional OCTA generation methods when only two repeated B-scans are performed. We further provide a comparative study with various CNN and transformer models and conduct ablation studies to validate the effectiveness of our proposed strategies. Based on the results, the UFCT model holds the potential to significantly enhance clinical workflow in oral medicine by facilitating early detection, reducing the need for invasive procedures, and improving overall patient outcomes.

A hand-held optical coherence tomography angiography scanner based on angiography reconstruction transformer networks

Optical coherence tomography angiography (OCTA) has successfully demonstrated its viability for clinical applications in dermatology. Due to the high optical scattering property of skin, extracting high-quality OCTA images from skin tissues requires at least six-repeated scans. While the motion artifacts from the patient and the free hand-held probe can lead to a low-quality OCTA image. Our deep-learning-based scan pipeline enables fast and high-quality OCTA imaging with 0.3-s data acquisition. We utilize a fast scanning protocol with a 60 μm/pixel spatial interval rate and introduce angiography-reconstruction-transformer (ART) for 4× super-resolution of low transverse resolution OCTA images. The ART outperforms state-of-the-art networks in OCTA image super-resolution and provides a lighter network size. ART can restore microvessels while reducing the processing time by 85%, and maintaining improvements in structural similarity and peak-signal-to-noise ratio. This study represents that ART can achieve fast and flexible skin OCTA imaging while maintaining image quality.

Optical Coherence Tomography Angiography in Retinal Vascular Disorders

Traditionally, abnormalities of the retinal vasculature and perfusion in retinal vascular disorders, such as diabetic retinopathy and retinal vascular occlusions, have been visualized with dye-based fluorescein angiography (FA). Optical coherence tomography angiography (OCTA) is a newer, alternative modality for imaging the retinal vasculature, which has some advantages over FA, such as its dye-free, non-invasive nature, and depth resolution. The depth resolution of OCTA allows for characterization of the retinal microvasculature in distinct anatomic layers, and commercial OCTA platforms also provide automated quantitative vascular and perfusion metrics. Quantitative and qualitative OCTA analysis in various retinal vascular disorders has facilitated the detection of pre-clinical vascular changes, greater understanding of known clinical signs, and the development of imaging biomarkers to prognosticate and guide treatment. With further technological improvements, such as a greater field of view and better image quality processing algorithms, it is likely that OCTA will play an integral role in the study and management of retinal vascular disorders. Artificial intelligence methods-in particular, deep learning-show promise in refining the insights to be gained from the use of OCTA in retinal vascular disorders. This review aims to summarize the current literature on this imaging modality in relation to common retinal vascular disorders.

Deformation-Compensated Learning for Image Reconstruction Without Ground Truth

Deep neural networks for medical image reconstruction are traditionally trained using high-quality ground-truth images as training targets. Recent work on Noise2Noise (N2N) has shown the potential of using multiple noisy measurements of the same object as an alternative to having a ground-truth. However, existing N2N-based methods are not suitable for learning from the measurements of an object undergoing nonrigid deformation. This paper addresses this issue by proposing the deformation-compensated learning (DeCoLearn) method for training deep reconstruction networks by compensating for object deformations. A key component of DeCoLearn is a deep registration module, which is jointly trained with the deep reconstruction network without any ground-truth supervision. We validate DeCoLearn on both simulated and experimentally collected magnetic resonance imaging (MRI) data and show that it significantly improves imaging quality.

Advances in OCT Imaging in Myopia and Pathologic Myopia

Advances in imaging with optical coherence tomography (OCT) and optical coherence tomography angiography (OCTA) technology, including the development of swept source OCT/OCTA, widefield or ultra-widefield systems, have greatly improved the understanding, diagnosis, and treatment of myopia and myopia-related complications. Anterior segment OCT is useful for imaging the anterior segment of myopes, providing the basis for implantable collamer lens optimization, or detecting intraocular lens decentration in high myopic patients. OCT has enhanced imaging of vitreous properties, and measurement of choroidal thickness in myopic eyes. Widefield OCT systems have greatly improved the visualization of peripheral retinal lesions and have enabled the evaluation of wide staphyloma and ocular curvature. Based on OCT imaging, a new classification system and guidelines for the management of myopic traction maculopathy have been proposed; different dome-shaped macula morphologies have been described; and myopia-related abnormalities in the optic nerve and peripapillary region have been demonstrated. OCTA can quantitatively evaluate the retinal microvasculature and choriocapillaris, which is useful for the early detection of myopic choroidal neovascularization and the evaluation of anti-vascular endothelial growth factor therapy in these patients. In addition, the application of artificial intelligence in OCT/OCTA imaging in myopia has achieved promising results.

Low-dose CT reconstruction with Noise2Noise network and testing-time fine-tuning

PURPOSE

Deep learning-based image denoising and reconstruction methods demonstrated promising performance on low-dose CT imaging in recent years. However, most existing deep learning-based low-dose CT reconstruction methods require normal-dose images for training. Sometimes such clean images do not exist such as for dynamic CT imaging or very large patients. The purpose of this work is to develop a low-dose CT image reconstruction algorithm based on deep learning which does not need clean images for training.

METHODS

In this paper, we proposed a novel reconstruction algorithm where the image prior was expressed via the Noise2Noise network, whose weights were fine-tuned along with the image during the iterative reconstruction. The Noise2Noise network built a self-consistent loss by projection data splitting and mapping the corresponding filtered backprojection (FBP) results to each other with a deep neural network. Besides, the network weights are optimized along with the image to be reconstructed under an alternating optimization scheme. In the proposed method, no clean image is needed for network training and the testing-time fine-tuning leads to optimization for each reconstruction.

RESULTS

We used the 2016 Low-dose CT Challenge dataset to validate the feasibility of the proposed method. We compared its performance to several existing iterative reconstruction algorithms that do not need clean training data, including total variation, non-local mean, convolutional sparse coding, and Noise2Noise denoising. It was demonstrated that the proposed Noise2Noise reconstruction achieved better RMSE, SSIM and texture preservation compared to the other methods. The performance is also robust against the different noise levels, hyperparameters, and network structures used in the reconstruction. Furthermore, we also demonstrated that the proposed methods achieved competitive results without any pre-training of the network at all, that is, using randomly initialized network weights during testing. The proposed iterative reconstruction algorithm also has empirical convergence with and without network pre-training.

CONCLUSIONS

The proposed Noise2Noise reconstruction method can achieve promising image quality in low-dose CT image reconstruction. The method works both with and without pre-training, and only noisy data are required for pre-training.