Dataset Information

Enhanced Visualization of Retinal Microvasculature in Optical Coherence Tomography Angiography Imaging via Deep Learning.

ABSTRACT: BACKGROUND:To investigate the effects of deep learning denoising on quantitative vascular measurements and the quality of optical coherence tomography angiography (OCTA) images. METHODS:U-Net-based deep learning denoising with an averaged OCTA data set as teacher data was used in this study. One hundred and thirteen patients with various retinal diseases were examined. An OCT HS-100 (Canon inc., Tokyo, Japan) performed a 3 × 3 mm2 superficial capillary plexus layer slab scan centered on the fovea 10 times. A single-shot image was defined as the original image and the 10-frame averaged image and denoised image generated from the original image using deep learning denoising for the analyses were obtained. The main parameters measured were the OCTA image acquisition time, contrast-to-noise ratio (CNR), peak signal-to-noise ratio (PSNR), vessel density (VD), vessel length density (VLD), vessel diameter index (VDI), and fractal dimension (FD) of the original, averaged, and denoised images. RESULTS:One hundred and twelve eyes of 108 patients were studied. Deep learning denoising removed the background noise and smoothed the rough vessel surface. The image acquisition times for the original, averaged, and denoised images were 16.6 ± 2.4, 285 ± 38, and 22.1 ± 2.4 s, respectively (P < 0.0001). The CNR and PSNR of the denoised image were significantly higher than those of the original image (P < 0.0001). There were significant differences in the VLD, VDI, and FD (P < 0.0001) after deep learning denoising. CONCLUSIONS:The deep learning denoising method achieved high speed and high quality OCTA imaging. This method may be a viable alternative to the multiple image averaging technique.

SUBMITTER: Kadomoto S

PROVIDER: S-EPMC7290309 | biostudies-literature | 2020 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Enhanced Visualization of Retinal Microvasculature in Optical Coherence Tomography Angiography Imaging via Deep Learning.

Kadomoto Shin S Uji Akihito A Muraoka Yuki Y Akagi Tadamichi T Tsujikawa Akitaka A

Journal of clinical medicine 20200502 5

<h4>Background</h4>To investigate the effects of deep learning denoising on quantitative vascular measurements and the quality of optical coherence tomography angiography (OCTA) images.<h4>Methods</h4>U-Net-based deep learning denoising with an averaged OCTA data set as teacher data was used in this study. One hundred and thirteen patients with various retinal diseases were examined. An OCT HS-100 (Canon inc., Tokyo, Japan) performed a 3 × 3 mm<sup>2</sup> superficial capillary plexus layer slab ...[more]

PMID: 32370282

Similar Datasets

Project description:PurposeTo investigate factors associated with dropout of the parapapillary deep retinal layer microvasculature assessed by optical coherence tomography angiography (OCTA) in glaucomatous eyes.DesignCross-sectional study.ParticipantsSeventy-one eyes from 71 primary open-angle glaucoma (POAG) patients with β-zone parapapillary atrophy (βPPA) enrolled in the Diagnostic Innovations in Glaucoma Study.MethodsParapapillary deep-layer microvasculature dropout was defined as a complete loss of the microvasculature located within the deep retinal layer of the βPPA from OCTA-derived optic nerve head vessel density maps by standardized qualitative assessment. Circumpapillary vessel density (cpVD) within the retinal nerve fiber layer (RNFL) also was calculated using OCTA. Choroidal thickness and presence of focal lamina cribrosa (LC) defects were determined using swept-source optical coherence tomography.Main outcome measuresPresence of parapapillary deep-layer microvasculature dropout. Parameters including age, systolic and diastolic blood pressure, axial length, intraocular pressure, disc hemorrhage, cpVD, visual field (VF) mean deviation (MD), focal LC defects βPPA area, and choroidal thickness were analyzed.ResultsParapapillary deep-layer microvasculature dropout was detected in 37 POAG eyes (52.1%). Eyes with microvasculature dropout had a higher prevalence of LC defects (70.3% vs. 32.4%), lower cpVD (52.7% vs. 58.8%), worse VF MD (-9.06 dB vs. -3.83 dB), thinner total choroidal thickness (126.5 μm vs. 169.1 μm), longer axial length (24.7 mm vs. 24.0 mm), larger βPPA (1.2 mm2 vs. 0.76 mm2), and lower diastolic blood pressure (74.7 mmHg vs. 81.7 mmHg) than those without dropout (P < 0.05, respectively). In the multivariate logistic regression analysis, higher prevalence of focal LC defects (odds ratio [OR], 6.27; P = 0.012), reduced cpVD (OR, 1.27; P = 0.002), worse VF MD (OR, 1.27; P = 0.001), thinner choroidal thickness (OR, 1.02; P = 0.014), and lower diastolic blood pressure (OR, 1.16; P = 0.003) were associated significantly with the dropout.ConclusionsSystemic and ocular factors including focal LC defects more advanced glaucoma, reduced RNFL vessel density, thinner choroidal thickness, and lower diastolic blood pressure were factors associated with the parapapillary deep-layer microvasculature dropout in glaucomatous eyes. Longitudinal studies are required to elucidate the temporal relationship between parapapillary deep-layer microvasculature dropout and systemic and ocular factors.

Project description:PurposeTo examine whether deep-learning denoised optical coherence tomography angiography (OCTA) images could enhance automated macular ischemia quantification in branch retinal vein occlusion (BRVO).MethodsThis retrospective, single-center, cross-sectional study enrolled 74 patients with BRVO and 46 age-matched healthy subjects. The severity of macular ischemia was graded as mild, moderate, or severe. Denoised OCTA images were produced using a neural network model. Quantitative parameters derived from denoised images, including vessel density and nonperfusion area, were compared with those derived from the OCTA machine. The main outcome measures were correlations between quantitative parameters, and areas under receiver operating characteristic curves (AUCs) in classifying the severity of the macular ischemia.ResultsThe vessel density and nonperfusion area from denoised images were correlated strongly with the corresponding parameters from machine-derived images in control eyes and BRVO eyes with mild or moderate macular ischemia (all P < 0.001). However, no such correlation was found in eyes with severe macular ischemia. The vessel density and nonperfusion area from denoised images had significantly larger area under receiver operating characteristic curve than those derived from the original images in classifying moderate versus severe macular ischemia (0.927 vs 0.802 [P = 0.042] and 0.946 vs 0.797, [P = 0.022], respectively). There were no significant differences in the areas under receiver operating characteristic curve between the denoised images and the machine-derived parameters in classifying control versus BRVO, and mild versus moderate macular ischemia.ConclusionsA neural network model is useful for removing speckle noise on OCTA images and facilitating the automated grading of macular ischemia in eyes with BRVO.Translational relevanceDeep-learning denoised optical coherence tomography angiography images could enhance automated macular ischemia quantification.

Project description:Purpose: To evaluate the conjunctival and fundus retinal vessel density in patients with systemic lupus erythematosus (SLE) with optical coherence tomography angiography (OCTA), and to investigate the relationship between vessel density and clinical indicators. Methods: Twelve patients with SLE (24 eyes) and 12 healthy controls (24 eyes) were recruited. OCTA was used to examine the superficial retina layer (SRL) and deep retina layer (DRL) in the macular retina and conjunctival capillary plexus of each eye. We calculated the density of the temporal conjunctival vessels, fundus microvascular (MIR), macrovascular (MAR) and total MIR(TMI) and compared the results in both groups. We used annular partitioning (C1-C6), hemispheric quadrants, and Early Treatment Diabetic Retinopathy Study partitioning (ETDRS) to analyze changes in the retinal vascular density. Correlation analysis was used to investigate the association between blood capillary density and clinical indicators. Results: OCTA results showed significant differences in the conjunctival microvascular density (p < 0.001). There was no significant difference in MIR, TMI, and MAR in the superficial layers between the SLE and healthy group (p > 0.05). The DRL and DTMI (Deeper TMI) densities were decreased in the macular regions of SLE patients (p < 0.05). In the hemispheric segmentation analysis, the superficial MIR was significantly decreased in the IL (inferior left) region of the SLE patients (p < 0.05), and the deep MIR in the IR (inferior right) region was significantly reduced (p < 0.05). In the ETDRS partitioning analysis, the superficial MIR in the inferior, right, and left subdivisions was significantly decreased in the SLE patients (p < 0.05). In the circular segmentation analysis, the deep MIR in the C1 and C3 regions was significantly reduced in SLE patients (p < 0.05), while the superficial MIR density was decreased only in the C3 region (p < 0.05). The conjunctival vascular density was negatively correlated with the STMI (Superficial TMI) (r = -0.5107; p = 0.0108) and DTMI (r = -0.9418, p < 0.0001). There was no significant correlation between vascular density and SLEDAI-2k (Systemic Lupus Erythematosus Disease Activity Index-2000) (P > 0.05). Conclusion: Clinically, patients with SLE and patients suspected of SLE should receive OCTA examination in a comprehensive eye examination to detect changes in ocular microcirculation at an early stage.

Dataset Information

Enhanced Visualization of Retinal Microvasculature in Optical Coherence Tomography Angiography Imaging via Deep Learning.

Publications

Enhanced Visualization of Retinal Microvasculature in Optical Coherence Tomography Angiography Imaging via Deep Learning.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets