To evaluate the performance of a federated learning framework for deep neural network-based retinal microvasculature segmentation and referable diabetic retinopathy (RDR) classification using OCT angiography (OCTA).Retrospective analysis clinical OCTA scans control participants patients with diabetes.The 153 en face images used were acquired from 4 instruments fields view ranging 2 × 2-mm to 6 6-mm. The 700 eyes RDR consisted structural projections commercial systems.OCT delineated manually...

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 fast axis is much higher than positions slow in wide-field OCTA scan, a non-orthogonal scheme introduced.Two en face angiograms vertical priority (2 y-fast) are divided into microsaccade-free parallel strips.A gross based on large vessels and fine...

Accurate identification and segmentation of choroidal neovascularization (CNV) is essential for the diagnosis management exudative age-related macular degeneration (AMD). Projection-resolved optical coherence tomographic angiography (PR-OCTA) enables both cross-sectional

Objective: Optical coherence tomography (OCT) and its angiography (OCTA) have several advantages for the early detection diagnosis of diabetic retinopathy (DR). However, automated, complete DR classification frameworks based on both OCT OCTA data not been proposed. In this study, a convolutional neural network (CNN) method is proposed to fulfill framework using en face OCTA. Methods: A densely continuously connected with adaptive rate dropout (DcardNet) designed classification. addition,...

To evaluate wide-field optical coherence tomography angiography (OCTA) for detection of clinically unsuspected neovascularization (NV) in diabetic retinopathy (DR).This prospective observational single-center study included adult patients with a clinical diagnosis nonproliferative DR. Participants underwent examination, standard 7-field color photography, and OCTA commercial prototype swept-source devices. The was achieved by montaging five 6 × 10-mm scans from device into 25 image three...

Optical coherence tomographic angiography (OCTA) can image the retinal blood flow but visualization of capillary caliber is limited by low lateral resolution. Adaptive optics (AO) be used to compensate ocular aberrations when using high numerical aperture (NA), and thus improve However, previously reported AO-OCTA instruments were large complex, have a small sub-millimeter field view (FOV) that hinders extraction biomarkers with clinical relevance. In this manuscript, we developed sensorless...

Reliable classification of referable and vision threatening diabetic retinopathy (DR) is essential for patients with diabetes to prevent blindness. Optical coherence tomography (OCT) its angiography (OCTA) have several advantages over fundus photographs. We evaluated a deep-learning-aided DR framework using volumetric OCT OCTA.Four hundred fifty-six OCTA volumes were scanned from eyes 50 healthy participants 305 diabetes. Retina specialists labeled the as non-referable (nrDR), (rDR), or...

Timely diagnosis of eye diseases is paramount to obtaining the best treatment outcomes. OCT and angiography (OCTA) have several advantages that lend themselves early detection ocular pathology; furthermore, techniques produce large, feature-rich data volumes. However, full clinical potential both OCTA stymied when complex acquired using must be manually processed. Here, we propose an automated diagnostic framework based on structural volumes could substantially support application these...

<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Objective:</i> Deep learning classifiers provide the most accurate means of automatically diagnosing diabetic retinopathy (DR) based on optical coherence tomography (OCT) and its angiography (OCTA). The power these models is attributable in part to inclusion hidden layers that complexity required achieve a desired task. However, also render algorithm outputs difficult interpret. Here we...

Quantitative analysis of the peripapillary retinal layers and capillary plexuses from optical coherence tomography (OCT) OCT angiography images depend on two segmentation tasks - delineating boundary optic disc boundaries between layers. Here, we present a method combining neural network graph search to perform these tasks. A comparison this novel method's showed good agreement with ground truth, achieving an overall Dice similarity coefficient 0.91 ± 0.04 in healthy glaucomatous eyes. The...

Abstract We developed a high‐speed, swept source OCT system for widefield angiography (OCTA) imaging. The has an extended axial imaging range of 6.6 mm. An electrical lens is used fast, automatic focusing. recently split‐spectrum amplitude and phase‐gradient allow high‐resolution OCTA with only two B‐scan repetitions. improved post‐processing algorithm effectively removed trigger jitter artifacts reduced noise in the flow signal. demonstrated high contrast 3 mm×3 mm image 400×400 pixels...

To improve optic disc boundary detection and peripapillary retinal layer segmentation, we propose an automated approach for structural angiographic optical coherence tomography.The algorithm was performed on radial cross-sectional B-scans.The detected by searching the position of Bruch's membrane opening, boundaries were using a dynamic programming-based graph search each B-scan without region.A comparison our method with that determined manual delineation showed good accuracy, average Dice...

We propose a three-dimensional (3-D) registration method to correct motion artifacts and construct the volume structure for angiographic structural optical coherence tomography (OCT). This algorithm is particularly suitable nonorthogonal wide-field OCT scan acquired by ultrahigh-speed swept-source system ( > 200 ?? kHz A-scan rate). First, transverse are corrected between-frame based on en face angiography (OCTA). After translation between B-frames, axial motions rebuilt boundary of inner...

Purpose Accurate segmentation of pelvic organs in CT images is great importance external beam radiotherapy for prostate cancer. The aim this studying to develop a novel method automatic, multiorgan the male pelvis. Methods authors’ consists several stages. First, pretreatment includes parameterization, principal component analysis (PCA), and an established process region‐specific hierarchical appearance cluster (RSHAC) model which was executed on training dataset. After preprocessing, online...

PurposeRetinal ischemia is a major feature of diabetic retinopathy (DR). Traditional nonperfused areas measured by OCT angiography (OCTA) measure blood supply but not ischemia. We propose novel 3-dimensional (3D) quantitative method to derive measurements from OCTA data.DesignCross-sectional study.ParticipantsWe acquired 223 macular volumes 33 healthy eyes, eyes without retinopathy, 7 with nonreferable DR, 17 referable nonvision-threatening and 133 vision-threatening DR.MethodsEach eye was...

Deep learning classifiers provide the most accurate means of automatically diagnosing diabetic retinopathy (DR) based on optical coherence tomography (OCT) and its angiography (OCTA). The power these models is attributable in part to inclusion hidden layers that complexity required achieve a desired task. However, also render algorithm outputs difficult interpret. Here we introduce novel biomarker activation map (BAM) framework generative adversarial allows clinicians verify understand...

Objective: Optical coherence tomography (OCT) and its angiography (OCTA) have several advantages for the early detection diagnosis of diabetic retinopathy (DR). However, automated, complete DR classification frameworks based on both OCT OCTA data not been proposed. In this study, a convolutional neural network (CNN) method is proposed to fulfill framework using en face OCTA. Methods: A densely continuously connected with adaptive rate dropout (DcardNet) designed classification. addition,...

Abstract This paper proposes an automated registration method for multi‐modality retinal fundus photographs based on the directional vessel skeleton. The main purpose is to register two with different modalities of same scanning region, which can provide information clinicians diagnose diseases or make a treatment decision. skeleton each image first detected by bias field correction and Gabor filter. final registered are then obtained iterative affine between skeletons photographs. In this...