Dataset Information

Deep Neural Network for Scleral Spur Detection in Anterior Segment OCT Images: The Chinese American Eye Study.

ABSTRACT: Purpose:To develop a deep neural network that detects the scleral spur in anterior segment optical coherence tomography (AS-OCT) images. Methods:Participants in the Chinese American Eye Study, a population-based study in Los Angeles, California, underwent complete ocular examinations, including AS-OCT imaging with the Tomey CASIA SS-1000. One human expert grader provided reference labels of scleral spur locations in all images. A convolutional neural network (CNN)-based on the ResNet-18 architecture was developed to detect the scleral spur in each image. Performance of the CNN model was assessed by calculating prediction errors, defined as the difference between the Cartesian coordinates of reference and CNN-predicted scleral spur locations. Prediction errors were compared with intragrader variability in detecting scleral spur locations by the reference grader. Results:The CNN was developed using a training dataset of 17,704 images and tested using an independent dataset of 921 images. The mean absolute prediction errors of the CNN model were 49.27 ± 42.07 µm for X-coordinates and 47.73 ± 39.70 µm for Y-coordinates. The mean absolute intragrader variability was 52.31 ± 47.75 µm for X-coordinates and 45.88 ± 45.06 µm for Y-coordinates. Distributions of prediction errors for the CNN and intragrader variability for the reference grader were similar for X-coordinates (P = 0.609) and Y-coordinates (P = 0.378). The mean absolute prediction error of the CNN was 73.08 ± 52.06 µm and the mean absolute intragrader variability was 73.92 ± 60.72 µm. Conclusions:A deep neural network can detect the scleral spur on AS-OCT images with performance similar to that of a human expert grader. Translational Relevance:Deep learning methods that automate scleral spur detection can facilitate qualitative and quantitative assessments of AS-OCT images.

SUBMITTER: Xu BY

PROVIDER: S-EPMC7395674 | biostudies-literature | 2020 Mar

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Deep Neural Network for Scleral Spur Detection in Anterior Segment OCT Images: The Chinese American Eye Study.

Xu Benjamin Y BY Chiang Michael M Pardeshi Anmol A AA Moghimi Sasan S Varma Rohit R

Translational vision science & technology 20200330 2

<h4>Purpose</h4>To develop a deep neural network that detects the scleral spur in anterior segment optical coherence tomography (AS-OCT) images.<h4>Methods</h4>Participants in the Chinese American Eye Study, a population-based study in Los Angeles, California, underwent complete ocular examinations, including AS-OCT imaging with the Tomey CASIA SS-1000. One human expert grader provided reference labels of scleral spur locations in all images. A convolutional neural network (CNN)-based on the Res ...[more]

PMID: 32818079

Similar Datasets

Project description:PURPOSE:To develop and test deep learning classifiers that detect gonioscopic angle closure and primary angle closure disease (PACD) based on fully automated analysis of anterior segment OCT (AS-OCT) images. METHODS:Subjects were recruited as part of the Chinese-American Eye Study (CHES), a population-based study of Chinese Americans in Los Angeles, California, USA. Each subject underwent a complete ocular examination including gonioscopy and AS-OCT imaging in each quadrant of the anterior chamber angle (ACA). Deep learning methods were used to develop 3 competing multi-class convolutional neural network (CNN) classifiers for modified Shaffer grades 0, 1, 2, 3, and 4. Binary probabilities for closed (grades 0 and 1) and open (grades 2, 3, and 4) angles were calculated by summing over the corresponding grades. Classifier performance was evaluated by 5-fold cross-validation and on an independent test dataset. Outcome measures included area under the receiver operating characteristic curve (AUC) for detecting gonioscopic angle closure and PACD, defined as either 2 or 3 quadrants of gonioscopic angle closure per eye. RESULTS:A total of 4036 AS-OCT images with corresponding gonioscopy grades (1943 open, 2093 closed) were obtained from 791 CHES subjects. Three competing CNN classifiers were developed with a cross-validation dataset of 3396 images (1632 open, 1764 closed) from 664 subjects. The remaining 640 images (311 open, 329 closed) from 127 subjects were segregated into a test dataset. The best-performing classifier was developed by applying transfer learning to the ResNet-18 architecture. For detecting gonioscopic angle closure, this classifier achieved an AUC of 0.933 (95% confidence interval, 0.925-0.941) on the cross-validation dataset and 0.928 on the test dataset. For detecting PACD based on 2- and 3-quadrant definitions, the ResNet-18 classifier achieved AUCs of 0.964 and 0.952, respectively, on the test dataset. CONCLUSION:Deep learning classifiers effectively detect gonioscopic angle closure and PACD based on automated analysis of AS-OCT images. These methods could be used to automate clinical evaluations of the ACA and improve access to eye care in high-risk populations.

Project description:PurposeTo develop a deep learning (DL) algorithm for predicting anterior chamber depth (ACD) from smartphone-acquired anterior segment photographs.MethodsFor algorithm development, we included 4,157 eyes from 2,084 Chinese primary school students (aged 11-15 years) from Mojiang Myopia Progression Study (MMPS). All participants had with ACD measurement measured with Lenstar (LS 900) and anterior segment photographs acquired from a smartphone (iPhone Xs), which was mounted on slit lamp and under diffuses lighting. The anterior segment photographs were randomly selected by person into training (80%, no. of eyes = 3,326) and testing (20%, no. of eyes = 831) dataset. We excluded participants with intraocular surgery history or pronounced corneal haze. A convolutional neural network was developed to predict ACD based on these anterior segment photographs. To determine the accuracy of our algorithm, we measured the mean absolute error (MAE) and coefficient of determination (R 2) were evaluated. Bland Altman plot was used to illustrate the agreement between DL-predicted and measured ACD values.ResultsIn the test set of 831 eyes, the mean measured ACD was 3.06 ± 0.25 mm, and the mean DL-predicted ACD was 3.10 ± 0.20 mm. The MAE was 0.16 ± 0.13 mm, and R 2 was 0.40 between the predicted and measured ACD. The overall mean difference was -0.04 ± 0.20 mm, with 95% limits of agreement ranging between -0.43 and 0.34 mm. The generated saliency maps showed that the algorithm mainly utilized central corneal region (i.e., the site where ACD is clinically measured typically) in making its prediction, providing further plausibility to the algorithm's prediction.ConclusionsWe developed a DL algorithm to estimate ACD based on smartphone-acquired anterior segment photographs. Upon further validation, our algorithm may be further refined for use as a ACD screening tool in rural localities where means of assessing ocular biometry is not readily available. This is particularly important in China where the risk of primary angle closure disease is high and often undetected.

Project description:PurposeTo assess the effects of age and refractive status on anterior segment anatomical structures, including the ciliary body, using a new swept-source anterior segment optical coherence tomography (AS-OCT) device.MethodsThis prospective observational study included 63 healthy volunteers (mean age: 44.2 years). Images of the anterior segment were obtained using a new swept-source AS-OCT (ANTERION, Heidelberg Engineering GmbH, Heidelberg, Germany) with tracking and image averaging from the right eye of all participants. Repeatability as well as inter- and intra-observer reliability of biometric measurements were evaluated. The impact of image tracking and averaging on ciliary muscle measurements was tested. Univariate and multivariable statistical models were developed to evaluate the relationship of age and refractive status on anterior segment biometric measurements.ResultsFor all test-retest repeatability and inter- and intra-observer reproducibility of swept-source AS-OCT measurements, high intraclass correlation (ICC) was noted (0.88-1.00). The nasal maximum ciliary muscle thickness (CMTMAX) and distance between scleral spur to the thickest point of the ciliary muscle (SSMAX) were larger than those on the temporal side (p<0.001 and p = 0.006, respectively). Nasal and temporal CMTMAX (p = 0.004 and p<0.001, respectively) and lens thickness (p<0.01) increased with age. Nasal and temporal SSMAX decreased with older age and increasing hyperopia (p = 0.01 and p<0.001, respectively). Image averaging resulted in improved ciliary muscle measurements (p = 0.008 to 0.02). Lens vault increased with older age and increased hyperopia (p<0.01). OCT measurements of the angle decreased with older age and increased hyperopia (p<0.001 to 0.03). Aqueous depth decreased with older age and increased hyperopia (p<0.01). Pupil diameter decreased with older age (p<0.01).ConclusionsRepeatability and reproducibility of biometric measurements using the ANTERION AS-OCT were excellent. Image averaging improved the accuracy of ciliary muscle measurements. The device produced measurements of biometric parameters that described superficial and deep structures including the ciliary body and full lens thickness from a single image.

Project description:This study reports the development of prototype swept-source optical coherence tomography (SS-OCT) technology for imaging the anterior eye. Advances in vertical-cavity surface-emitting laser (VCSEL) light sources, signal processing, optics and mechanical designs, enable a unique combination of high speed, long range, and deep penetration that addresses the challenges of anterior eye imaging. We demonstrate SS-OCT with a 325 kHz A-scan rate, 12.2 µm axial resolution (in air), and 15.5 mm depth range (in air) at 1310 nm wavelength. The ultrahigh 325 kHz A-scan rate not only facilitates biometry measurements by minimizing acquisition time and thus reducing motion, but also enables volumetric OCT for comprehensive structural analysis and OCT angiography (OCTA) for visualizing vasculature. The 15.5 mm (~ 11.6 mm in tissue) depth range spans all optical surfaces from the anterior cornea to the posterior lens capsule. The 1310 nm wavelength range enables structural OCT and OCTA deep in the sclera and through the iris. Achieving high speed and long range requires linearizing the VCSEL wavenumber sweep to efficiently utilize analog-to-digital conversion bandwidth. Dual channel recording of the OCT and calibration interferometer fringe signals, as well as sweep to sweep wavenumber compensation, is used to achieve invariant 12.2 µm (~ 9.1 µm in tissue) axial resolution and optimum point spread function throughout the depth range. Dynamic focusing using a tunable liquid lens extends the effective depth of field while preserving the lateral resolution. Improved optical and mechanical design, including parallax "split view" iris cameras and stable, ergonomic patient interface, facilitates accurate instrument positioning, reduces patient motion, and leads to improved imaging data yield and measurement accuracy. We present structural and angiographic OCT images of the anterior eye, demonstrating the unique imaging capabilities using representative scanning protocols which may be relevant to future research and clinical applications.

Dataset Information

Deep Neural Network for Scleral Spur Detection in Anterior Segment OCT Images: The Chinese American Eye Study.

Publications

Deep Neural Network for Scleral Spur Detection in Anterior Segment OCT Images: The Chinese American Eye Study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure