ATN-Res2Unet: an advanced deep learning network for the elimination of saturation artifacts in endoscopy optical coherence tomography.

Endoscopic optical coherence tomography (OCT) possesses the capability to non-invasively image internal lumens; however, it is susceptible to saturation artifacts arising from robust reflective structures. In this study, we introduce an innovative deep learning network, ATN-Res2Unet, designed to mitigate saturation artifacts in endoscopic OCT images. This is achieved through the integration of multi-scale perception, multi-attention mechanisms, and frequency domain filters. To address the challenge of obtaining ground truth in endoscopic OCT, we propose a method for constructing training data pairs. Experimental in vivo data substantiates the effectiveness of ATN-Res2Unet in reducing diverse artifacts while preserving structural information. Comparative analysis with prior studies reveals a notable enhancement, with average quantitative indicators increasing by 45.4-83.8%. Significantly, this study marks the inaugural exploration of leveraging deep learning to eradicate artifacts from endoscopic OCT images, presenting considerable potential for clinical applications.

Improved FAST algorithm for non-uniform rotational distortion correction in OCT endoscopic imaging.

Optical Coherence Tomography (OCT) is widely used for endoscopic imaging in endoluminal organs because of its high imaging accuracy and resolution. However, OCT endoscopic imaging suffers from Non-Uniform Rotational Distortion (NURD), which can be caused by many factors, such as irregular motor rotation and changes in friction between the probe and the sheath. Correcting this distortion is essential to obtaining high-quality Optical Coherence Tomography Angiography (OCTA) images. There are two main approaches for correcting NURD: hardware-based methods and algorithm-based methods. Hardware-based methods can be costly, challenging to implement, and may not eliminate NURD. Algorithm-based methods, such as image registration, can be effective for correcting NURD but can also be prone to the problem of NURD propagation. To address this issue, we process frames by coarse and fine registration, respectively. The new reference frame is generated by filtering out the A-scan that may have the NURD problem by coarse registration. And the fine registration uses this frame to achieve the final NURD correction. In addition, we have improved the Features from Accelerated Segment Test (FAST) algorithm and put it into coarse and fine registration process. Four evaluation functions were used for the experimental results, including signal-to-noise ratio (SNR), peak signal-to-noise ratio (PSNR), mean squared error (MSE), and structural similarity index measure (SSIM). By comparing with Scale-invariant feature transform (SIFT), Speeded up robust features (SURF), Oriented FAST and Rotated BRIEF (ORB), intensity-based (Cross-correlation), and Optical Flow algorithms, our algorithm has a higher similarity between the corrected frames. Moreover, the noise in the OCTA data is better suppressed, and the vascular information is well preserved. Our image registration-based algorithm reduces the problem of NURD propagation between B-scan frames and improves the imaging quality of OCT endoscopic images.

Development of a multi-scene universal multiple wavelet-FFT algorithm (MW-FFTA) for denoising motion artifacts in OCT-angiography in vivo imaging.

Optical coherence tomography angiography (OCTA) images suffer from inevitable micromotion (breathing, heartbeat, and blinking) noise. These image artifacts can severely disturb the visibility of results and reduce accuracy of vessel morphological and functional metrics quantization. Herein, we propose a multiple wavelet-FFT algorithm (MW-FFTA) comprising multiple integrated processes combined with wavelet-FFT and minimum reconstruction that can be used to effectively attenuate motion artifacts and significantly improve the precision of quantitative information. We verified the fidelity of image information and reliability of MW-FFTA by the image quality evaluation. The efficiency and robustness of MW-FFTA was validated by the vessel parameters on multi-scene in vivo OCTA imaging. Compared with previous algorithms, our method provides better visual and quantitative results. Therefore, the MW-FFTA possesses the potential capacity to improve the diagnosis of clinical diseases with OCTA.

Evaluation of Retinal Pigment Epithelium and Choroidal Neovascularization in Rats Using Laser-Scanning Optical-Resolution Photoacoustic Microscopy.

PURPOSE: To demonstrate the value of the laser-scanning optical-resolution (LSOR)-photoacoustic (PA) microscopy (PAM) system and the conventional multimodal imaging techniques in the evaluation of laser-induced retinal injury and choroidal neovascularization (CNV) in rats. METHODS: Different degrees of retinal injury were induced using laser photocoagulation. We compared the LSOR-PAM system with conventional imaging techniques in evaluating retinal injury with or without CNV. Six additional rats, treated with an anti-VEGF antibody or immunoglobulin G immediately after photocoagulation, were imaged 7 and 14 days after injection, and CNV lesion areas were compared. RESULTS: In the retinal injury model, fundus autofluorescence showed well-defined hyperreflection, while the lesion displayed abundant PA signals demonstrating nonuniform melanin distribution in retinal pigment epithelium (RPE). RPE was detected with higher contrast in the PAM B-scan image than optical coherence tomography (OCT). Additionally, the CNV lesion was present with multiple PA signal intensities which distinctly characterized the location and area of CNV as found in fundus fluorescein angiography. Furthermore, the decreased PA signals extending from the CNV lesion were similar to those of the vascular bud in ex vivo imaging, which was invisible in other in vivo images. When treated with anti-VEGF agents, statistically significant differences can be demonstrated by PAM similar to other modalities. CONCLUSIONS: LSOR-PAM can detect the melanin distribution of RPE in laser-induced retinal injury and CNV in rats. PAM imaging provides a potential new tool to evaluate the vitality and functionality of RPE in vivo as well as to monitor the development and treatment of CNV.

Acoustic Radiation Force Optical Coherence Elastography of Corneal Tissue.

We report on a real-time acoustic radiation force optical coherence elastography (ARF-OCE) system to map the relative elasticity of corneal tissue. A modulated ARF is used as excitation to vibrate the cornea while OCE serves as detection of tissue response. To show feasibility of detecting mechanical contrast using this method, we performed tissue-equivalent agarose phantom studies with inclusions of a different stiffness. We obtained 3-D elastograms of a healthy cornea and a highly cross-linked cornea. Finally we induced a stiffness change on a small portion of a cornea and observed the differences in displacement.

Automatic airway wall segmentation and thickness measurement for long-range optical coherence tomography images.

We present an automatic segmentation method for the delineation and quantitative thickness measurement of multiple layers in endoscopic airway optical coherence tomography (OCT) images. The boundaries of the mucosa and the sub-mucosa layers are accurately extracted using a graph-theory-based dynamic programming algorithm. The algorithm was tested with sheep airway OCT images. Quantitative thicknesses of the mucosal layers are obtained automatically for smoke inhalation injury experiments.

Fast subcellular optical coherence photoacoustic microscopy for pigment cell imaging.

We developed a fast ultrahigh resolution optical coherence photoacoustic microscopy (FU-OCPAM) system by combining two complementary imaging modes of optical coherence microscopy (OCM) and photoacoustic microscopy (PAM) for cellular/subcellular imaging. The system used optical scanning to realize fast imaging speed and provided ultrahigh resolution of 1.24 and 0.59 µm for OCM and PAM, respectively. We imaged the retinal pigment epithelium (RPE) to demonstrate the subcellular imaging capability of the FU-OCPAM system. The OCM and PAM images clearly showed the RPE cell morphology and reflected the complementary optical properties of scattering and absorption. A quantitative analysis of the RPE cells was made based on photoacoustic (PA) signals. The cell area mainly ranged from 80 to 300 µm2, and had a linear relationship with the sum intensity of PA signals which mainly reflected the melanin content of the cells. The morphology and the PA signal could be used to identify qualitatively and quantitatively the aging and healthy states of the RPE cells. The results show the potential applications in studying the real-time cellular response to external stimulations and the progress of aging and diseases at the cellular level with FU-OCPAM.

Dual-channel spectral-domain optical-coherence tomography system based on 3 × 3 fiber coupler for extended imaging range.

We have demonstrated a dual-channel multiplexing spectral-domain optical-coherence tomography (SD-OCT) system based on a 3×3 fiber coupler for extended imaging range of whole human eye depth, with a single light source and spectrometer. OCT images of anterior segments of a human eye were sequentially performed and constructed to demonstrate an extended depth range as large as 15 mm in air. A good quality OCT image of the whole anterior segment of an eye was present. Furthermore, whole eye segmental imaging was performed and ocular distances were calculated to show the validation of the system for whole eye morphological measurement.

Multi-label classification of retinal diseases based on fundus images using Resnet and Transformer.

Retinal disorders are a major cause of irreversible vision loss, which can be mitigated through accurate and early diagnosis. Conventionally, fundus images are used as the gold diagnosis standard in detecting retinal diseases. In recent years, more and more researchers have employed deep learning methods for diagnosing ophthalmic diseases using fundus photography datasets. Among the studies, most of them focus on diagnosing a single disease in fundus images, making it still challenging for the diagnosis of multiple diseases. In this paper, we propose a framework that combines ResNet and Transformer for multi-label classification of retinal disease. This model employs ResNet to extract image features, utilizes Transformer to capture global information, and enhances the relationships between categories through learnable label embedding. On the publicly available Ocular Disease Intelligent Recognition (ODIR-5 k) dataset, the proposed method achieves a mean average precision of 92.86%, an area under the curve (AUC) of 97.27%, and a recall of 90.62%, which outperforms other state-of-the-art approaches for the multi-label classification. The proposed method represents a significant advancement in the field of retinal disease diagnosis, offering a more accurate, efficient, and comprehensive model for the detection of multiple retinal conditions.

Image segmentation of mouse eye in vivo with optical coherence tomography based on Bayesian classification.

OBJECTIVES: Optical coherence tomography (OCT) is a new imaging technology that uses an optical analog of ultrasound imaging for biological tissues. Image segmentation plays an important role in dealing with quantitative analysis of medical images. METHODS: We have proposed a novel framework to deal with the low intensity problem, based on the labeled patches and Bayesian classification (LPBC) model. The proposed method includes training and testing phases. During the training phase, firstly, we manually select the sub-images of background and Region of Interest (ROI) from the training image, and then extract features by patches. Finally, we train the Bayesian model with the features. The segmentation threshold of each patch is computed by the learned Bayesian model. RESULTS: In addition, we have collected a new dataset of mouse eyes in vivo with OCT, named MEVOCT, which can be found at URL https://17861318579.github.io/LPBC. MEVOCT consists of 20 high-resolution images. The resolution of every image is 2048 × 2048 pixels. CONCLUSIONS: The experimental results demonstrate the effectiveness of the LPBC method on the new MEVOCT dataset. The ROI segmentation is of great importance for the distortion correction.

Diagnosis of dental caries based on attenuation coefficients analysis of optical coherence tomography images.

Quantitative analysis of optical attenuation based on optical coherence tomography images will offer an effective method to enhance diagnostic capabilities. In this paper, the optical attenuation in demineralized caries specimens was calculated to distinguish between normal teeth and carious teeth and further to differentiate the severity of caries, and thus come to the half-automated diagnosis of dental caries. Results show that the attenuation coefficient in carious regions is approximately 4.97 mm - 1 ± 0.206 , while that of normal teeth is about 3.69 mm - 1 ± 0.231 . Attenuation coefficient of carious regions is 35% higher than that of normal teeth. Moreover, five classes of caries were qualified and classified based on the optical attenuation coefficient. Compared with the healthy teeth, there is a noticeable disparity in the attenuation coefficients of carious teeth, both on the surface and at the dentinoenamel junction. This study provides a method for accurate caries diagnosis, particularly in detection of early lesions and subtle structural changes.

CSANet: a lightweight channel and spatial attention neural network for grading diabetic retinopathy with optical coherence tomography angiography.

Background: Diabetic retinopathy (DR) is one of the most common eye diseases. Convolutional neural networks (CNNs) have proven to be a powerful tool for learning DR features; however, accurate DR grading remains challenging due to the small lesions in optical coherence tomography angiography (OCTA) images and the small number of samples. Methods: In this article, we developed a novel deep-learning framework to achieve the fine-grained classification of DR; that is, the lightweight channel and spatial attention network (CSANet). Our CSANet comprises two modules: the baseline model, and the hybrid attention module (HAM) based on spatial attention and channel attention. The spatial attention module is used to mine small lesions and obtain a set of spatial position weights to address the problem of small lesions being ignored during the convolution process. The channel attention module uses a set of channel weights to focus on useful features and suppress irrelevant features. Results: The extensive experimental results for the OCTA-DR and diabetic retinopathy analysis challenge (DRAC) 2022 data sets showed that the CSANet achieved state-of-the-art DR grading results, showing the effectiveness of the proposed model. The CSANet had an accuracy rate of 97.41% for the OCTA-DR data set and 85.71% for the DRAC 2022 data set. Conclusions: Extensive experiments using the OCTA-DR and DRAC 2022 data sets showed that the proposed model effectively mitigated the problems of mutual confusion between DRs of different severity and small lesions being neglected in the convolution process, and thus improved the accuracy of DR classification.

Self-attention CNN for retinal layer segmentation in OCT.

The structure of the retinal layers provides valuable diagnostic information for many ophthalmic diseases. Optical coherence tomography (OCT) obtains cross-sectional images of the retina, which reveals information about the retinal layers. The U-net based approaches are prominent in retinal layering methods, which are usually beneficial to local characteristics but not good at obtaining long-distance dependence for contextual information. Furthermore, the morphology of retinal layers with the disease is more complex, which brings more significant challenges to the task of retinal layer segmentation. We propose a U-shaped network combining an encoder-decoder architecture and self-attention mechanisms. In response to the characteristics of retinal OCT cross-sectional images, a self-attentive module in the vertical direction is added to the bottom of the U-shaped network, and an attention mechanism is also added in skip connection and up-sampling to enhance essential features. In this method, the transformer's self-attentive mechanism obtains the global field of perception, thus providing the missing context information for convolutions, and the convolutional neural network also efficiently extracts local features, compensating the local details the transformer ignores. The experiment results showed that our method is accurate and better than other methods for segmentation of the retinal layers, with the average Dice scores of 0.871 and 0.820, respectively, on two public retinal OCT image datasets. To perform the layer segmentation of retinal OCT image better, the proposed method incorporates the transformer's self-attention mechanism in a U-shaped network, which is helpful for ophthalmic disease diagnosis.

High-precision retinal blood vessel segmentation based on a multi-stage and dual-channel deep learning network.

Objective.The high-precision segmentation of retinal vessels in fundus images is important for the early diagnosis of ophthalmic diseases. However, the extraction for microvessels is challenging due to their characteristics of low contrast and high structural complexity. Although some works have been developed to improve the segmentation ability in thin vessels, they have only been successful in recognizing small vessels with relatively high contrast.Approach.Therefore, we develop a deep learning (DL) framework with a multi-stage and dual-channel network model (MSDC_NET) to further improve the thin-vessel segmentation with low contrast. Specifically, an adaptive image enhancement strategy combining multiple preprocessing and the DL method is firstly proposed to elevate the contrast of thin vessels; then, a two-channel model with multi-scale perception is developed to implement whole- and thin-vessel segmentation; and finally, a series of post-processing operations are designed to extract more small vessels in the predicted maps from thin-vessel channels.Main results.Experiments on DRIVE, STARE and CHASE_DB1 demonstrate the superiorities of the proposed MSDC_NET in extracting more thin vessels in fundus images, and quantitative evaluations on several parameters based on the advanced ground truth further verify the advantages of our proposed DL model. Compared with the previous multi-branch method, the specificity and F1score are improved by about 2.18%, 0.68%, 1.73% and 2.91%, 0.24%, 8.38% on the three datasets, respectively.Significance.This work may provide richer information to ophthalmologists for the diagnosis and treatment of vascular-related ophthalmic diseases.

Deep-learning segmentation method for optical coherence tomography angiography in ophthalmology.

PURPOSE: The optic disc and the macular are two major anatomical structures in the human eye. Optic discs are associated with the optic nerve. Macular mainly involves degeneration and impaired function of the macular region. Reliable optic disc and macular segmentation are necessary for the automated screening of retinal diseases. METHODS: A swept-source OCTA system was designed to capture OCTA images of human eyes. To address these segmentation tasks, first, we constructed a new Optic Disc and Macula in fundus Image with optical coherence tomography angiography (OCTA) dataset (ODMI). Second, we proposed a Coarse and Fine Attention-Based Network (CFANet). RESULTS: The five metrics of our methods on ODMI are 98.91 % , 98.47 % , 89.77 % , 98.49 % , and 89.77 % , respectively. CONCLUSIONS: Experimental results show that our CFANet has achieved good performance on segmentation for the optic disc and macula in OCTA.

Unsupervised adversarial neural network for enhancing vasculature in photoacoustic tomography images using optical coherence tomography angiography.

Photoacoustic tomography (PAT) is a powerful imaging modality for visualizing tissue physiology and exogenous contrast agents. However, PAT faces challenges in visualizing deep-seated vascular structures due to light scattering, absorption, and reduced signal intensity with depth. Optical coherence tomography angiography (OCTA) offers high-contrast visualization of vasculature networks, yet its imaging depth is limited to a millimeter scale. Herein, we propose OCPA-Net, a novel unsupervised deep learning method that utilizes the rich vascular feature of OCTA to enhance PAT images. Trained on unpaired OCTA and PAT images, OCPA-Net incorporates a vessel-aware attention module to enhance deep-seated vessel details captured from OCTA. It leverages a domain-adversarial loss function to enforce structural consistency and a novel identity invariant loss to mitigate excessive image content generation. We validate the structural fidelity of OCPA-Net on simulation experiments, and then demonstrate its vascular enhancement performance on in vivo imaging experiments of tumor-bearing mice and contrast-enhanced pregnant mice. The results show the promise of our method for comprehensive vessel-related image analysis in preclinical research applications.

A new retinal OCT-angiography diabetic retinopathy dataset for segmentation and DR grading.

PURPOSE: Diabetic retinopathy (DR) is one of the most common diseases caused by diabetes and can lead to vision loss or even blindness. The wide-field optical coherence tomography (OCT) angiography is non-invasive imaging technology and convenient to diagnose DR. METHODS: A newly constructed Retinal OCT-Angiography Diabetic retinopathy (ROAD) dataset is utilized for segmentation and grading tasks. It contains 1200 normal images, 1440 DR images, and 1440 ground truths for DR image segmentation. To handle the problem of grading DR, we propose a novel and effective framework, named projective map attention-based convolutional neural network (PACNet). RESULTS: The experimental results demonstrate the effectiveness of our PACNet. The accuracy of the proposed framework for grading DR is 87.5% on the ROAD dataset. CONCLUSIONS: The information on ROAD can be viewed at URL https://mip2019.github.io/ROAD. The ROAD dataset will be helpful for the development of the early detection of DR field and future research. TRANSLATIONAL RELEVANCE: The novel framework for grading DR is a valuable research and clinical diagnosis method.

Time-Serial Evaluation of the Development and Treatment of Myopia in Mice Eyes Using OCT and ZEMAX.

Myopia is a significant cause of visual impairment which may lead to many complications. However, the understanding of the mechanisms of myopia is still limited. In this paper, in order to investigate the development and the treatment of myopia, we analyzed the biological structure parameters of mice eyes, obtained from optical coherence tomography (OCT), and the optical performance of mice eyes calculated using ZEMAX software (ZEMAX Development Corporation, Kirkland, WA, USA) in which the optical model was built on the segment-by-segment optically corrected OCT 3D-images. Time-serial evaluation of three groups of mice eyes (form-deprivation myopia mice eyes, normal mice eyes, and atropine-treated myopia mice eyes) was performed. In addition to the biological structure parameters, imaging performance with the development of root-mean-square wavefront aberration at six filed angles was compared and analyzed. Results show that the biological structure parameters of the eye are closely related to the development of myopia. The peripheral defocus of the retina has a significant impact on inducing myopia, which verifies the new theory of myopia development. The delaying effect of atropine solution on myopia development is shown to verify the therapeutic effect of the medicine. This study provides technical support for the investigation of the myopia mechanism.

Endoscopic optical coherence tomography angiography using an externally driving catheter.

Endoscopic optical coherence tomography (OCT) is an imaging modality that enables cross-sectional subsurface imaging of tubular organs and cavities. Recently, endoscopic OCT angiography (OCTA) was successfully achieved in distal scanning systems using an internal-motor-driving catheter. In conventional OCT systems using externally driving catheters, the mechanical instability in the proximal actuation causes difficulties for differentiating capillaries in tissues. In this study, OCTA in an endoscopic OCT system using an external-motor-driving catheter was proposed. Blood vessels were visualized by using a high-stability inter-A-scan scheme and the spatiotemporal singular value decomposition algorithm. It is not limited by nonuniform rotation distortion caused by the catheter and physiological motion artifacts. Results show that microvasculature in a custom-made microfluidic phantom and the submucosal capillaries in the mouse rectum are successfully visualized. Furthermore, OCTA using a catheter with a small size (outer diameter less than 1 mm) makes it possible for early diagnosis of narrow lumens, such as pancreatic and bile duct cancers.

Optical coherence tomography for whole eye segment imaging.

We proposed a dual focus dual channel spectral domain optical coherence tomography (SD-OCT) for simultaneous imaging of the whole eye segments from cornea to the retina. By using dual channels the system solved the problem of limited imaging depth of SD-OCT. By using dual focus the system solved the problem of simultaneous light focusing on the anterior segment of the eye and the retina. Dual focusing was achieved by adjusting the collimating lenses so the divergence of the two probing beams was tuned to make them focused at different depth in the eye. We further achieved full range complex (FRC) SD-OCT in one channel to increase the depth range for anterior segment imaging. The system was successfully tested by imaging a human eye in vivo.