PurposeThe purpose of this study was to evaluate an automated algorithm for detecting avascular area (AA) in optical coherence tomography angiograms (OCTAs) separated into three individual plexuses using a projection-resolved technique.MethodsA 3 × 3 mm macular OCTA was obtained in 13 healthy and 13 mild nonproliferative diabetic retinopathy (NPDR) participants. A projection-resolved algorithm segmented OCTA into three vascular plexuses: superficial, intermediate, and deep. An automated algorithm detected AA in each of the three plexuses that were segmented and in the combined inner-retinal angiograms. We assessed the diagnostic accuracy of extrafoveal and total AA using segmented and combined angiograms, the agreement between automated and manual detection of AA, and the within-visit repeatability.ResultsThe sum of extrafoveal AA from the segmented angiograms was larger in the NPDR group by 0.17 mm2 (P < 0.001) and detected NPDR with 94.6% sensitivity (area under the receiver operating characteristic curve [AROC] = 0.99). In the combined inner-retinal angiograms, the extrafoveal AA was larger in the NPDR group by 0.01 mm2 (P = 0.168) and detected NPDR with 26.9% sensitivity (AROC = 0.62). The total AA, inclusive of the foveal avascular zone, in the segmented and combined angiograms, detected NPDR with 23.1% and 7.7% sensitivity, respectively. The agreement between the manual and automated detection of AA had a Jaccard index of >0.8. The pooled SDs of AA were small compared with the difference in mean for control and NPDR groups.ConclusionsAn algorithm to detect AA in OCTA separated into three individual plexuses using a projection-resolved algorithm accurately distinguishes mild NPDR from control eyes. Automatically detected AA agrees with manual delineation and is highly repeatable.

Abstract:We propose an innovative registration method to correct motion artifacts for widefield optical coherence tomography angiography (OCTA) acquired by ultrahigh-speed sweptsource OCT (>200 kHz A-scan rate). Considering that the number of A-scans along the fast axis is much higher than the number of positions along slow axis in the wide-field OCTA scan, a non-orthogonal scheme is introduced. Two en face angiograms in the vertical priority (2 y-fast) are divided into microsaccade-free parallel strips. A gross registration based on large vessels and a fine registration based on small vessels are sequentially applied to register parallel strips into a composite image. This technique is extended to automatically montage individual registered, motion-free angiograms into an ultrawide-field view. Werner, "En face projection imaging of the human choroidal layers with tracking SLO and swept source OCT angiography methods," Proc. SPIE 9541, 954112 (2015).

IntroductionOptical coherence tomography angiography (OCTA) is a novel technique that provides depth-resolved images of circulation in the retina and choroid (1-4). Using the motion of blood cells as intrinsic imaging contrast, it produces highresolution angiography without the need for dye injection (5,6). Recent advances in computational efficiency (7), removal of projections between different vascular plexuses (8), precise segmentation of retinal layers (9) and quantification of ocular pathologies (5,10,11) have strengthened the potential of OCTA for clinical evaluation. All OCTA algorithms, however, are susceptible to motion artifacts (12).Although small axial motion can usually be tolerated, transverse eye movements such as microsaccades introduce uncorrectable motion artifacts that manifest as white stripes on en face OCTA. These artifacts can be removed by registering and merging OCT volumes collected at perpendicular scanning directions. Background: Motion artifacts degrade the quality of optical coherence tomography angiography (OCTA).

Coronavirus disease 2019 (COVID-19) has spread rapidly worldwide. The rapid and accurate automatic segmentation of COVID-19 infected areas using chest computed tomography (CT) scans is critical for assessing disease progression. However, infected areas have irregular sizes and shapes. Furthermore, there are large differences between image features. We propose a convolutional neural network, named 3D CU-Net, to automatically identify COVID-19 infected areas from 3D chest CT images by extracting rich features and fusing multiscale global information. 3D CU-Net is based on the architecture of 3D U-Net. We propose an attention mechanism for 3D CU-Net to achieve local cross-channel information interaction in an encoder to enhance different levels of the feature representation. At the end of the encoder, we design a pyramid fusion module with expanded convolutions to fuse multiscale context information from high-level features. The Tversky loss is used to resolve the problems of the irregular size and uneven distribution of lesions. Experimental results show that 3D CU-Net achieves excellent segmentation performance, with Dice similarity coefficients of 96.3% and 77.8% in the lung and COVID-19 infected areas, respectively. 3D CU-Net has high potential to be used for diagnosing COVID-19.

Automated detection and grading of angiographic high-risk features in diabetic retinopathy can potentially enhance screening and clinical care. We have previously identified capillary dilation in angiograms of the deep plexus in optical coherence tomography angiography as a feature associated with severe diabetic retinopathy. In this study, we present an automated algorithm that uses hybrid contrast to distinguish angiograms with dilated capillaries from healthy controls and then applies saliency measurement to map the extent of the dilated capillary networks. The proposed algorithm agreed well with human grading.

No abstract

No abstract

Active contour model (ACM) which has been extensively studied recently is one of the most successful methods in image segmentation. The present paper advances an improved hybrid model based on Region-Scalable Fitting Model by combining global convex segmentation method with edge detector operator. The proposed model not only inherits the ability of RSF model to deal with the images with intensity inhomogeneity, but also overcomes such a drawback: existence of local minima because of non-convexity that makes the segmentation result highly dependent of the initial position of the contour. In addition, the paper exploits two fast numerical implementation schemes to overcome a huge amount of level set methods. The duality projection method is implemented by introducing dual variables which lead to semi-implicit iterative scheme of dual variables as well as exact formulation of primal variables. The Split-Bregman method is implemented by introducing auxiliary variables which transform the relaxed convex model into solving simple poisson equations and exact soft thresholding formulation. Experimental results for synthetic and real medical images prove that the proposed model is featured by greater numerical accuracy and faster division speed.