scholarly journals Deep Learning-Based Estimation of Axial Length and Subfoveal Choroidal Thickness From Color Fundus Photographs

2021 ◽  
Vol 9 ◽  
Author(s):  
Li Dong ◽  
Xin Yue Hu ◽  
Yan Ni Yan ◽  
Qi Zhang ◽  
Nan Zhou ◽  
...  
Keyword(s):  
Deep Learning ◽  
Axial Length ◽  
Choroidal Thickness ◽  
Absolute Error ◽  
Population Based ◽  
Fundus Images ◽  
Subfoveal Choroidal Thickness ◽  
Color Fundus Images ◽  
Fundus Photographs ◽  
Map Analysis

This study aimed to develop an automated computer-based algorithm to estimate axial length and subfoveal choroidal thickness (SFCT) based on color fundus photographs. In the population-based Beijing Eye Study 2011, we took fundus photographs and measured SFCT by optical coherence tomography (OCT) and axial length by optical low-coherence reflectometry. Using 6394 color fundus images taken from 3468 participants, we trained and evaluated a deep-learning-based algorithm for estimation of axial length and SFCT. The algorithm had a mean absolute error (MAE) for estimating axial length and SFCT of 0.56 mm [95% confidence interval (CI): 0.53,0.61] and 49.20 μm (95% CI: 45.83,52.54), respectively. Estimated values and measured data showed coefficients of determination of r2 = 0.59 (95% CI: 0.50,0.65) for axial length and r2 = 0.62 (95% CI: 0.57,0.67) for SFCT. Bland–Altman plots revealed a mean difference in axial length and SFCT of −0.16 mm (95% CI: −1.60,1.27 mm) and of −4.40 μm (95% CI, −131.8,122.9 μm), respectively. For the estimation of axial length, heat map analysis showed that signals predominantly from overall of the macular region, the foveal region, and the extrafoveal region were used in the eyes with an axial length of < 22 mm, 22–26 mm, and > 26 mm, respectively. For the estimation of SFCT, the convolutional neural network (CNN) used mostly the central part of the macular region, the fovea or perifovea, independently of the SFCT. Our study shows that deep-learning-based algorithms may be helpful in estimating axial length and SFCT based on conventional color fundus images. They may be a further step in the semiautomatic assessment of the eye.

scholarly journals Automatic Segmentation of Choroid Layer Using Deep Learning on Spectral Domain Optical Coherence Tomography

2021 ◽  
Vol 11 (12) ◽  
pp. 5488
Author(s):  
Wei Ping Hsia ◽  
Siu Lun Tse ◽  
Chia Jen Chang ◽  
Yu Len Huang
Keyword(s):  
Optical Coherence Tomography ◽  
Deep Learning ◽  
Choroidal Thickness ◽  
Automatic Segmentation ◽  
Average Error ◽  
Good Prediction ◽  
Optical Coherence ◽  
Learning Method ◽  
Subfoveal Choroidal Thickness ◽  
Fully Connected

The purpose of this article is to evaluate the accuracy of the optical coherence tomography (OCT) measurement of choroidal thickness in healthy eyes using a deep-learning method with the Mask R-CNN model. Thirty EDI-OCT of thirty patients were enrolled. A mask region-based convolutional neural network (Mask R-CNN) model composed of deep residual network (ResNet) and feature pyramid networks (FPNs) with standard convolution and fully connected heads for mask and box prediction, respectively, was used to automatically depict the choroid layer. The average choroidal thickness and subfoveal choroidal thickness were measured. The results of this study showed that ResNet 50 layers deep (R50) model and ResNet 101 layers deep (R101). R101 U R50 (OR model) demonstrated the best accuracy with an average error of 4.85 pixels and 4.86 pixels, respectively. The R101 ∩ R50 (AND model) took the least time with an average execution time of 4.6 s. Mask-RCNN models showed a good prediction rate of choroidal layer with accuracy rates of 90% and 89.9% for average choroidal thickness and average subfoveal choroidal thickness, respectively. In conclusion, the deep-learning method using the Mask-RCNN model provides a faster and accurate measurement of choroidal thickness. Comparing with manual delineation, it provides better effectiveness, which is feasible for clinical application and larger scale of research on choroid.

scholarly journals Characteristics of Fundal Changes in Fundus Tessellation in Young Adults

2021 ◽  
Vol 8 ◽  
Author(s):  
Hanyi Lyu ◽  
Qiuying Chen ◽  
Guangyi Hu ◽  
Ya Shi ◽  
Luyao Ye ◽  
...  
Keyword(s):  
Young Adults ◽  
Choroidal Thickness ◽  
Anterior Segment ◽  
Population Based ◽  
Cross Sectional ◽  
Swept Source ◽  
Lower Body Weight ◽  
Parapapillary Atrophy ◽  
Fundus Photographs ◽  
Body Weight Index

Purpose: To explore the characteristics and associated factors of fundus tessellation, especially the alternation of choroidal thickness among different degrees of tessellated fundus in young adults.Design: Cross-sectional, population-based study.Methods: A total of 796 students were included in the study and underwent comprehensive ophthalmic examinations, including anterior segment examinations and swept-source optical coherence tomography (OCT) measurements. The degree of tessellated fundus was assessed by fundus photographs applying an early treatment of diabetic retinopathy study grid to evaluate the location of fundus tessellation and then divided into five groups. The topographic variation and factors, tilted disc ratio, parapapillary atrophy (PPA), retinal thickness (ReT), choroidal thickness (ChT), and subfoveal scleral thickness (SST) related to tessellated fundus were analyzed.Results: Compared to normal fundus, tessellated fundus had a lower spherical equivalent (SE) (p < 0.0001), worse best-corrected visual acuity (BCVA)(p = 0.043), longer axial length (AL) (p < 0.0001), thinner retina (p < 0.0001), thinner (p < 0.0001) choroid, and thinner sclera in center fovea (p = 0.0035). Among all subfields of macular and peripapillary regions, center fovea and macula-papillary region showed the most significant decrease in choroidal thickness. The proportion of fundus tessellation significantly increased with lower body weight index (BMI) (p = 0.0067), longer AL (p < 0.0001), larger PPA(p = 0.0058), thinner choroid (p < 0.0001), and thinner sclera (p < 0.0001).Conclusions: Eyes showed more severe myopic morphological alternation with the increasement of proportion of fundus tessellation to the center fovea, including a significant decrease in both choroid and scleral thickness. Choroidal thinning may progress most rapidly in the macula-papillary region as fundus tessellation approaches to the center fovea.

Deep learning approach for classification of eye diseases based on color fundus images

2020 ◽  
pp. 25-57
Author(s):  
Bambang Krismono Triwijoyo ◽  
Boy Subirosa Sabarguna ◽  
Widodo Budiharto ◽  
Edi Abdurachman
Keyword(s):  
Deep Learning ◽  
Eye Diseases ◽  
Learning Approach ◽  
Fundus Images ◽  
Color Fundus Images

scholarly journals Changes in Subfoveal Choroidal Thickness and Axial Length in Children Wearing Hyperopic Glasses

2020 ◽  
Vol 49 (0) ◽  
pp. 127-135
Author(s):  
Chihiro Sato ◽  
Miwa Nitta ◽  
Ayaka Kasai ◽  
Takafumi Mori ◽  
Teiko Hashimoto ◽  
...  
Keyword(s):  
Axial Length ◽  
Choroidal Thickness ◽  
Subfoveal Choroidal Thickness

scholarly journals Choroidal Thickness in Chinese Children Aged 8 to 11 Years with Mild and Moderate Myopia

2018 ◽  
Vol 2018 ◽  
pp. 1-7
Author(s):  
Ya Qi ◽  
Li Li ◽  
Fengju Zhang
Keyword(s):  
Refractive Error ◽  
Axial Length ◽  
Choroidal Thickness ◽  
Narrow Range ◽  
The Other ◽  
Chinese Children ◽  
Healthy Children ◽  
Spherical Equivalent ◽  
Subfoveal Choroidal Thickness

Purpose. To investigate macular choroidal thickness (CT), topographical variation, and associations between subfoveal choroidal thickness (SFCT) and age, gender, spherical equivalent (SE), and axial length (AL) in Chinese healthy mild and moderate myopia children aged 8 to 11 years. Methods. One hundred twenty eyes from 120 healthy children were studied. Children were divided into mild and moderate myopia groups. AL and CT were evaluated. CTs were measured at the fovea, and 1 mm, 2 mm, and 3 mm nasal, temporal, superior, and inferior to the fovea. Results. SFCT was 252.80 ± 46.95 µm in the whole population. AL was shorter in the mild myopia group (24.18 ± 0.69 mm) than in the moderate myopia group (24.97 ± 0.68 mm, P<0.001), and SFCT was thicker in the mild myopia group (262.00 ± 40.57 µm) than in the moderate myopia group (236.00 ± 55.08 µm, P=0.005). The topographical variation was similar in refraction groups. CTs nasal to the fovea thinned gradually and were all significantly thinner than SFCT. CTs in the other three directions gradually thickened and peaked at locations of 2 mm to the fovea. Then, CTs thinned at 3 mm to the fovea. The thickest choroid is located temporal to the fovea. There were significant negative correlations between AL and SFCT in the mild myopia group and the whole population. No other correlations were found. Conclusions. The topographical variations of choroidal thickness were similar in mild and moderate myopia groups with the thickest locations temporal to the fovea. SFCT was relatively stable in children in narrow range of age and refractive error.

Sign in / Sign up

Export Citation Format

Share Document