Machine learning in optical coherence tomography angiography

Optical coherence tomography angiography (OCTA) offers a noninvasive label-free solution for imaging retinal vasculatures at the capillary level resolution. In principle, improved resolution implies a better chance to reveal subtle microvascular distortions associated with eye diseases that are asymptomatic in early stages. However, massive screening requires experienced clinicians to manually examine retinal images, which may result in human error and hinder objective screening. Recently, quantitative OCTA features have been developed to standardize and document retinal vascular changes. The feasibility of using quantitative OCTA features for machine learning classification of different retinopathies has been demonstrated. Deep learning-based applications have also been explored for automatic OCTA image analysis and disease classification. In this article, we summarize recent developments of quantitative OCTA features, machine learning image analysis, and classification.     

Improving cerebral microvascular image quality of optical coherence tomography angiography with deep learning-based segmentation

Optical coherence tomography angiography (OCTA) can map the microvascular networks of the cerebral cortices with micrometer resolution and millimeter penetration. However, the high scattering of the skull and the strong noise in the deep imaging region will distort the vasculature projections and decrease the OCTA image quality. Here, we proposed a deep learning-based segmentation method based on a U-Net convolutional neural network to extract the cortical region from the OCT image. The vascular networks were then visualized by three OCTA algorithms. The image quality of the vasculature projections was assessed by two metrics, including the peak signal-to-noise ratio (PSNR) and the contrast-to-noise ratio (CNR). The results show the accuracy of the cortical segmentation was 96.07%. The PSNR and CNR values increased significantly in the projections of the selected cortical regions. The OCTA incorporating the deep learning-based cortical segmentation can efficiently improve the image quality and enhance the vasculature clarity.     

Imaging sebaceous gland using optical coherence tomography with deep learning assisted automatic identification

Imaging sebaceous glands and evaluating morphometric parameters are important for diagnosis and treatment of serum problems. In this article, we investigate the feasibility of high-resolution optical coherence tomography (OCT) in combination with deep learning assisted automatic identification for these purposes. Specifically, with a spatial resolution of 2.3 μm × 6.2 μm (axial × lateral, in air), OCT is capable of clearly differentiating sebaceous gland from other skin structures and resolving the sebocyte layer. In order to achieve efficient and timely imaging analysis, a deep learning approach built upon ResNet18 is developed to automatically classify OCT images (with/without sebaceous gland), with a classification accuracy of 97.9%. Based on the result of automatic identification, we further demonstrate the possibility to measure gland size, sebocyte layer thickness and gland density.     

Quantitative evaluation of optical coherence tomography angiography images of diabetic retinopathy eyes before and after removal of projection artifacts

Projection artifacts (PAs) affect the quantification of vascular parameters in the deep layer optical coherence tomography (OCT) angiography image. This study eliminated PA and quantified its effect on imaging. 53 eyes (30 subjects) of normal Indian subjects and 113 eyes (92 patients) of type 2 diabetes mellitus with retinopathy (DR) underwent imaging with a scan area of 3 mm × 3 mm. In this study, a normalized cross‐correlation between superficial and deep layer was used to remove PA in deep layer. Local fractal analysis was done to compute vascular parameters such as foveal avascular zone area (mm²), vessel density (%), spacing between large vessels (%) and spacing between small vessels (%). Before PA removal, vessel density for mild nonproliferative (NPDR), moderate NPDR, severe NPDR and proliferative DR were 42.56 ±1.69%, 40.69 ±0.72%, 37.34 ±0.85% and 35.61 ±1.26%, respectively. After artifact removal, vessel density was 28.9 ±1.22%, 29.9 ±0.56%, 26.19 ±0.59% and 24.02 ±0.94%, respectively. All the vascular parameters were statistically significant (P <.001) between normal and disease eyes, irrespective of superficial and deep retinal layers. Parafoveal sectoral analyses showed that temporal zone had the lowest vessel density and may undergo DR‐related changes first. The current approach enabled rapid and accurate quantitative interpretation of DR eyes, without PA.      

N2NSR-OCT: Simultaneous denoising and super-resolution in optical coherence tomography images using semisupervised deep learning

Optical coherence tomography (OCT) imaging shows a significant potential in clinical routines due to its noninvasive property. However, the quality of OCT images is generally limited by inherent speckle noise of OCT imaging and low sampling rate. To obtain high signal-to-noise ratio (SNR) and high-resolution (HR) OCT images within a short scanning time, we presented a learning-based method to recover high-quality OCT images from noisy and low-resolution OCT images. We proposed a semisupervised learning approach named N2NSR-OCT, to generate denoised and super-resolved OCT images simultaneously using up- and down-sampling networks (U-Net (Semi) and DBPN (Semi)). Additionally, two different super-resolution and denoising models with different upscale factors (2× and 4× ) were trained to recover the high-quality OCT image of the corresponding down-sampling rates. The new semisupervised learning approach is able to achieve results comparable with those of supervised learning using up- and down-sampling networks, and can produce better performance than other related state-of-the-art methods in the aspects of maintaining subtle fine retinal structures.     

Detecting mouse squamous cell carcinoma from submicron full-field optical coherence tomography images by deep learning

The standard medical practice for cancer diagnosis requires histopathology, which is an invasive and time-consuming procedure. Optical coherence tomography (OCT) is an alternative that is relatively fast, noninvasive, and able to capture three-dimensional structures of epithelial tissue. Unlike most previous OCT systems, which cannot capture crucial cellular-level information for squamous cell carcinoma (SCC) diagnosis, the full-field OCT (FF-OCT) technology used in this paper is able to produce images at sub-micron resolution and thereby facilitates the development of a deep learning algorithm for SCC detection. Experimental results show that the SCC detection algorithm can achieve a classification accuracy of 80% for mouse skin. Using the sub-micron FF-OCT imaging system, the proposed SCC detection algorithm has the potential for in-vivo applications.     

Automated quantification of superficial retinal capillaries and large vessels for diabetic retinopathy on optical coherence tomographic angiography

Optical coherence tomography angiography (OCTA) is a relatively new technique with capillary‐level resolution, which has shown great potential for the diagnosis of diabetic retinopathy (DR). A fully automatic algorithm for the quantitative measurement of microcirculatory changes in sight‐threatening DR is presented. The foveal avascular zone (FAZ) segmentation was improved with a graph‐theoretic method and the large vessels and capillaries were separately identified and analyzed. The method was evaluated in healthy and diabetic eyes with various stages of retinopathy. Results showed that, compared with the healthy group, the diabetic group showed a significantly larger large vessel density, but a significantly smaller capillary density (P < .001). Circularity of FAZ was significantly smaller while nonperfusion area was significantly larger in the diabetic group. The combined variable of all image metrics reached an area under the ROC of 0.853 (95% CI, 0.784‐0.923) for mild to moderate nonproliferative DR and 0.950 (95% CI, 0.922‐0.979) for proliferative DR. Microvascular and FAZ changes with various DR stages can be accurately delineated using the developed automatic program. Quantitative metrics on OCTA serve as potential biomarkers for the staging of DR.     

Quantitative assessment of optical coherence tomography angiography algorithms for neuroimaging

Optical coherence tomography (OCT) angiography can noninvasively map microvascular networks and quantify blood flow in a cerebral cortex with a resolution of 1 to 10 μm and a penetration depth of 2 to 3 mm incorporating OCT signals and angiography algorithms. Different angiography algorithms have been developed in recent years; however, the performance of the algorithms has not been assessed quantitatively for neuroimaging applications. In this paper, we developed four metrics including vascular connectivity, contrast‐to‐noise ratio, signal‐to‐noise ratio and processing time to quantitatively assess the performance of OCT angiography algorithms in image quality and computation speed. After the imaging of a rat cortex using an OCT system, the cerebral microvascular networks were visualized by seven algorithms, and the performance of the algorithms was quantified and compared. Quantitative performance assessment of the algorithms can provide suggestions for the selection of appropriate OCT angiography algorithms in neuroimaging.     

Deep-learning-based motion correction in optical coherence tomography angiography

Optical coherence tomography angiography (OCTA) is a widely applied tool to image microvascular networks with high spatial resolution and sensitivity. Due to limited imaging speed, the artifacts caused by tissue motion can severely compromise visualization of the microvascular networks and quantification of OCTA images. In this article, we propose a deep-learning-based framework to effectively correct motion artifacts and retrieve microvascular architectures. This method comprised two deep neural networks in which the first subnet was applied to distinguish motion corrupted B-scan images from a volumetric dataset. Based on the classification results, the artifacts could be removed from the en face maximum-intensity-projection (MIP) OCTA image. To restore the disturbed vasculature induced by artifact removal, the second subnet, an inpainting neural network, was utilized to reconnect the broken vascular networks. We applied the method to postprocess OCTA images of the microvascular networks in mouse cortex in vivo. Both image comparison and quantitative analysis show that the proposed method can significantly improve OCTA image by efficiently recovering microvasculature from the overwhelming motion artifacts.     

A novel automatic 3D stitching algorithm for optical coherence tomography angiography and its application in dermatology

Optical coherence tomography (OCT) and OCT angiography (OCTA) techniques offer numerous advantages in clinical skin applications but the field of view (FOV) of current commercial systems are relatively limited to cover the entire skin lesion. The typical method to expand the FOV is to apply wide field objective lens. However, lateral resolution is often sacrificed when scanning with these lenses. To overcome this drawback, we developed an automated 3D stitching method for creating high-resolution skin structure and vascular volumes with large field of view, which was realized by montaging multiple adjacent OCT and OCTA volumes. The proposed stitching method is demonstrated by montaging 3 × 3 OCT and OCTA volumes (nine OCT/OCTA volumes as one data set with each volume covers 2.5 cm × 2.5 cm area) of healthy thin and thick skin from six volunteers. The proposed stitching protocol achieves high flexibility and repeatable for all the participants. Moreover, according to evaluation of structural similarity index and feature similarity index, our proposed stitched result has a superior similarity to single scanning protocol in large-scaled. We had also verified its improved performance through assessing metrics of vessel contrast-noise-ratio (CNR) from 2.07 ± 0.44 (single large-scaled scanning protocol) to 3.05 ± 0.51 (proposed 3 × 3 sub-volume stitching method).     

Cross-domain retinopathy classification with optical coherence tomography images via a novel deep domain adaptation method

Deep learning based retinopathy classification with optical coherence tomography (OCT) images has recently attracted great attention. However, existing deep learning methods fail to work well when training and testing datasets are different due to the general issue of domain shift between datasets caused by different collection devices, subjects, imaging parameters, etc. To address this practical and challenging issue, we propose a novel deep domain adaptation (DDA) method to train a model on a labeled dataset and adapt it to an unlabelled dataset (collected under different conditions). It consists of two modules for domain alignment, that is, adversarial learning and entropy minimization. We conduct extensive experiments on three public datasets to evaluate the performance of the proposed method. The results indicate that there are large domain shifts between datasets, resulting a poor performance for conventional deep learning methods. The proposed DDA method can significantly outperform existing methods for retinopathy classification with OCT images. It achieves retinopathy classification accuracies of 0.915, 0.959 and 0.990 under three cross-domain (cross-dataset) scenarios. Moreover, it obtains a comparable performance with human experts on a dataset where no labeled data in this dataset have been used to train the proposed DDA method. We have also visualized the learnt features by using the t-distributed stochastic neighbor embedding (t-SNE) technique. The results demonstrate that the proposed method can learn discriminative features for retinopathy classification.     

Cross‐scanning optical coherence tomography angiography for eye motion correction

We propose a cross‐scanning optical coherence tomography (CS‐OCT) system to correct eye motion artifacts in OCT angiography images. This system employs a dual‐illumination configuration with two orthogonally polarized beams, each of which simultaneously perform raster scanning in perpendicular direction with each other over the same area. In the reference arm, a polarization delay unit is used to acquire the two orthogonally polarized interferograms with a single photo detector by introducing different optical delay lines. The two cross‐scanned volume data are affected by the same eye motion but in two orthogonal directions. We developed a motion correction algorithm, which removes artifacts in the slow axis of each angiogram using the other and merges them through a nonrigid registration algorithm. In this manner, we obtained a motion‐corrected angiogram within a single volume scanning time without additional eye‐tracking devices.     

Endoscopic optical coherence tomography angiography using a forward imaging piezo scanner probe

A forward imaging endoscope for optical coherence tomography angiography (OCTA) featuring a piezoelectric fiber scanner is presented. Imaging is performed with an optical coherence tomography (OCT) system incorporating an akinetic light source with a center wavelength of 1300 nm, bandwidth of 90 nm and A‐line rate of 173 kHz. The endoscope operates in contact mode to avoid motion artifacts, in particular, beneficial for OCTA measurements, and achieves a transversal resolution of 12 μm in air at a rigid probe size of 4 mm in diameter and 11.3 mm in length. A spiral scan pattern is generated at a scanning frequency of 360 Hz to sample a maximum field of view of 1.3 mm. OCT images of a human finger as well as visualization of microvasculature of the human palm are presented both in two and three dimensions. The combination of morphological tissue contrast with qualitative dynamic blood flow information within this endoscopic imaging approach potentially enables improved early diagnostic capabilities of internal organs for diseases such as bladder cancer.      

Comparative study of optical coherence tomography angiography algorithms for rodent retinal imaging

Optical coherence tomography angiography (OCTA) is a functional extension of optical coherence tomography for non-invasive in vivo three-dimensional imaging of the microvasculature of biological tissues. Several algorithms have been developed to construct OCTA images from the measured optical coherence tomography signals. In this study, we compared the performance of three OCTA algorithms that are based on the variance of phase, amplitude, and the complex representations of the optical coherence tomography signals for rodent retinal imaging, namely the phase variance, improved speckle contrast, and optical microangiography. The performance of the different algorithms was evaluated by comparing the quality of the OCTA images regarding how well the vasculature network can be resolved. Quantities that are widely used in ophthalmic studies including blood vessel density, vessel diameter index, vessel perimeter index, vessel complexity index were also compared. Results showed that both the improved speckle contrast and optical microangiography algorithms are more robust than phase variance, and they can reveal similar vasculature features while there are statistical differences in the calculated quantities.     

Interplane bulk motion analysis and removal based on normalized cross‐correlation in optical coherence tomography angiography

Bulk motion seriously degrades the image quality of optical coherence tomography angiography (OCTA). Conventional correction methods focus on in‐plane displacement, while the bulk motion component perpendicular to B‐scans also introduces noise. This work first presents an evaluation of this component using a specific scan protocol and an approximate expression derived from peak‐normalized cross‐correlation values, and then quantitatively assesses how interplane bulk motion noise reduce the sensitivity of cross‐sectional angiograms. Finally, we developed a repetitive bulk motion correction method based on the estimated displacements and redundant volume scans. The correction does not require registration and angiogram reconstruction of low flow sensitivity frames, and the results of in vivo mice skin OCTA imaging experiments show that the proposed method can effectively reduce bulk motion noise caused by cardiac and respiratory motion and occasional shaking, and improve OCTA image quality, which has practical significance for clinical OCTA diagnosis and analysis.     

Optical coherence tomography angiography for mapping cerebral microvasculature based on normalized differentiation analysis

Optical coherence tomography angiography (OCTA) is a label‐free, noninvasive biomedical imaging modality for mapping microvascular networks and quantifying blood flow velocities in vivo. Simple computation and fast processing are critical for the OCTA in some applications. Herein, we report on a normalized differentiation method for mapping cerebral microvasculature with the advantages of simple analysis and high image quality, benefitting from computation of differentiation and characteristics of normalization. Normalized differentiation values are validated to have a nearly linear relationship with flow velocities in a range using a flow phantom. The measurements in a rat cerebral cortex show that the OCTA based on the normalized differentiation analysis can generate microvascular images with high quality and monitor spatiotemporal dynamics of blood flow with simple computation and fast processing before and after localized ischemia induced by arterial occlusion.     

Three-dimensional spatial reconstruction of coronary arteries based on fusion of intravascular optical coherence tomography and coronary angiography

We present a three-dimensional (3D) spatial reconstruction of coronary arteries based on fusion of intravascular optical coherence tomography (IVOCT) and digital subtraction angiography (DSA). Centerline of vessel in DSA images is exacted by multi-scale filtering, adaptive segmentation, morphology thinning and Dijkstra's shortest path algorithm. We apply the cross-correction between lumen shapes of IVOCT and DSA images and match their stenosis positions to realize co-registration. By matching the location and tangent direction of the vessel centerline of DSA images and segmented lumen coordinates of IVOCT along pullback path, 3D spatial models of vessel lumen are reconstructed. Using 1121 distinct positions selected from eight vessels, the correlation coefficient between 3D IVOCT model and DSA image in measuring lumen radius is 0.94% and 97.7% of the positions fall within the limit of agreement by Bland–Altman analysis, which means that the 3D spatial reconstruction IVOCT models and DSA images have high matching level.     

A new deep learning method for image deblurring in optical microscopic systems

Deconvolution is the most commonly used image processing method in optical imaging systems to remove the blur caused by the point‐spread function (PSF). While this method has been successful in deblurring, it suffers from several disadvantages, such as slow processing time due to multiple iterations required to deblur and suboptimal in cases where the experimental operator chosen to represent PSF is not optimal. In this paper, we present a deep‐learning‐based deblurring method that is fast and applicable to optical microscopic imaging systems. We tested the robustness of proposed deblurring method on the publicly available data, simulated data and experimental data (including 2D optical microscopic data and 3D photoacoustic microscopic data), which all showed much improved deblurred results compared to deconvolution. We compared our results against several existing deconvolution methods. Our results are better than conventional techniques and do not require multiple iterations or pre‐determined experimental operator. Our method has several advantages including simple operation, short time to compute, good deblur results and wide application in all types of optical microscopic imaging systems. The deep learning approach opens up a new path for deblurring and can be applied in various biomedical imaging fields.     

Simultaneous visible light optical coherence tomography and near infrared OCT angiography in retinal pathologies: A case study

A dual-channel optical coherence tomography system with wavelengths in the visible and near-infrared light ranges can provide both structural and functional information for retinal microvasculature simultaneously. We applied this integrated system in an ongoing clinical study of patients with various retinal pathologies. Here, we present case study results of patients with diabetic retinopathy, central retinal vein occlusion, and sickle cell retinopathy compared to a healthy subject. For the first time, this comparison validates the system’s ability to detect structural anomalies in both en face and B-scan images with simultaneous retinal optical coherence tomography angiography and measurement of sO₂ in parafoveal vessels that are around 20–30 µm in diameter. This integrated system represents a powerful instrument with potentially far-reaching clinical implications for the early detection and diagnosis of retinal vascular diseases.     

Jones matrix‐based speckle‐decorrelation angiography using polarization‐sensitive optical coherence tomography

We show that polarization‐sensitive optical coherence tomography angiography (PS‐OCTA) based on full Jones matrix assessment of speckle decorrelation offers improved contrast and depth of vessel imaging over conventional OCTA. We determine how best to combine the individual Jones matrix elements and compare the resulting image quality to that of a conventional OCT scanner by co‐locating and imaging the same skin locations with closely matched scanning setups. Vessel projection images from finger and forearm skin demonstrate the benefits of Jones matrix‐based PS‐OCTA. Our study provides a promising starting point and a useful reference for future pre‐clinical and clinical applications of Jones matrix‐based PS‐OCTA.