Corneal Optical Coherence Tomography Images Research Articles

ScLNet: A cornea with scleral lens OCT layers segmentation dataset and new multi-task model

ObjectiveTo develop deep learning methods with high accuracy for segmenting irregular corneas and detecting the tear fluid reservoir (TFR) boundary under the scleral lens. Additionally, this study aims to provide a publicly available cornea with scleral lens OCT dataset, including manually labeled layer masks for training and validation of segmentation algorithms. This study introduces ScLNet, a dataset comprising cornea with Scleral Lens (ScL) optical coherence tomography (OCT) images with layer annotations, and a multi-task network designed to achieve rapid, accurate, automated segmentation of scleral lens with regular and irregular corneas. MethodsWe created a dataset comprising 31,360 OCT images with scleral lens annotations. The network architecture includes an encoder with multi-scale input and a context coding layer, along with two decoders for specific tasks. The primary task focuses on predicting ScL, TFR, and cornea regions, while the auxiliary task, aimed at predicting the boundaries of ScL, TFR, and cornea, enhances feature extraction for the main task. Segmentation results were compared with state-of-the-art methods and evaluated using Dice similarity coefficient (DSC), intersection over union (IoU), Matthews correlation coefficient (MCC), Precision, and Hausdorff distance (HD). ResultsScLNet achieves 98.22 % DSC, 96.50 % IoU, 98.13 % MCC, 98.35 % Precision, and 3.6840 HD (in pixels) in segmenting ScL; 97.78 % DSC, 95.66 % IoU, 97.71 % MCC, 97.70 % Precision, and 3.7838 HD (in pixels) in segmenting TFR; and 99.22 % DSC, 98.45 % IoU, 99.15 % MCC, 99.14 % Precision, and 3.5355 HD (in pixels) in segmenting cornea. The layer interfaces recognized by ScLNet closely align with expert annotations, as evidenced by high IoU scores. Boundary metrics further confirm its effectiveness. ConclusionWe constructed a dataset of corneal OCT images with ScL wearing, which includes regular and irregular cornea patients. The proposed ScLNet achieves high accuracy in extracting ScL, TFR, and corneal layer masks and boundaries from OCT images of the dataset.

Effect of fixational eye movements in corneal topography measurements with optical coherence tomography.

There is an increasing interest in applying optical coherence tomography (OCT) to quantify the topography of ocular structures. However, in its most usual configuration, OCT data is acquired sequentially while a beam is scanned through the region of interest, and the presence of fixational eye movements can affect the accuracy of the technique. Several scan patterns and motion correction algorithms have been proposed to minimize this effect, but there is no consensus on the ideal parameters to obtain a correct topography. We have acquired corneal OCT images with raster and radial patterns, and modeled the data acquisition in the presence of eye movements. The simulations replicate the experimental variability in shape (radius of curvature and Zernike polynomials), corneal power, astigmatism, and calculated wavefront aberrations. The variability of the Zernike modes is highly dependent on the scan pattern, with higher variability in the direction of the slow scan axis. The model can be a useful tool to design motion correction algorithms and to determine the variability with different scan patterns.

Deep Learning Model for the Detection of Corneal Edema Before Descemet Membrane Endothelial Keratoplasty on Optical Coherence Tomography Images.

Descemet membrane endothelial keratoplasty (DMEK) is the preferred method for treating corneal endothelial dysfunction, such as Fuchs endothelial corneal dystrophy (FECD). The surgical indication is based on the patients' symptoms and the presence of corneal edema. We developed an automated tool based on deep learning to detect edema in corneal optical coherence tomography images. This study aimed to evaluate this approach in edema detection before Descemet membrane endothelial keratoplasty surgery, for patients with or without FECD. We used our previously described model allowing to classify each pixel in the corneal optical coherence tomography images as "normal" or "edema." We included 1992 images of normal and preoperative edematous corneas. We calculated the edema fraction (EF), defined as the ratio between the number of pixels labeled as "edema," and those representing the cornea for each patient. Differential central corneal thickness (DCCT), defined as the difference in central corneal thickness before and 6 months after surgery, was used to quantify preoperative edema. AUC of EF for the edema detection was calculated for Several DCCT thresholds and a value of 20 µm was selected to define significant edema as it provided the highest area under the curve value. The area under the curve of the receiver operating characteristic curve for EF for the detection of 20 µm of DCCT was 0.97 for all patients, 0.96 for Fuchs and normal only and 0.99 for non-FECD and normal patients. The optimal EF threshold was 0.143 for all patients and patients with FECD. Our model is capable of objectively detecting minimal corneal edema before Descemet membrane endothelial keratoplasty surgery. Deep learning can help to interpret optical coherence tomography scans and aid the surgeon in decision-making.

A new approach for quantifying epithelial and stromal thickness changes after orthokeratology contact lens wear.

The aim of the study was to develop an automatic segmentation approach to optical coherence tomography (OCT) images and to investigate the changes in epithelial and stromal thickness profile and radius of curvature after the use of orthokeratology (Ortho-K) contact lenses. A total of 45 right eyes from 52 participants were monitored before, and after one month of, uninterrupted overnight Ortho-K lens wear. The tomography of their right eyes was obtained using optical OCT and rotating Scheimpflug imaging (OCULUS Pentacam). A custom-built MATLAB code for automatic segmentation of corneal OCT images was created and used to assess changes in epithelial thickness, stromal thickness, corneal and stromal profiles and radii of curvature before, and after one month of, uninterrupted overnight wear of Ortho-K lenses. In the central area (0–2 mm diameter), the epithelium thinned by 12.8 ± 6.0 µm (23.8% on average, p < 0.01) after one month of Ortho-K lens wear. In the paracentral area (2–5 mm diameter), the epithelium thinned nasally and temporally (by 2.4 ± 5.9 µm, 4.5% on average, p = 0.031). The stroma thickness increased in the central area (by 4.8 ± 16.1 µm, p = 0.005). The radius of curvature of the central corneal anterior surface increased by 0.24 ± 0.26 mm (3.1%, p < 0.01) along the horizontal meridian and by 0.34 ± 0.18 mm (4.2%, p < 0.01) along the vertical meridian. There were no significant changes in the anterior and posterior stromal radius of curvature. This study introduced a new method to automatically detect the anterior corneal surface, the epithelial posterior surface and the posterior corneal surface in OCT scans. Overnight wear of Ortho-K lenses caused thinning of the central corneal epithelium. The anterior corneal surface became flattered while the anterior and posterior surfaces of the stroma did not undergo significant changes. The results are consistent with the changes reported in previous studies. The reduction in myopic refractive error caused by Ortho-K lens wear was mainly due to changes in corneal epithelium thickness profile.

Corneal Edema Visualization With Optical Coherence Tomography Using Deep Learning: Proof of Concept.

Optical coherence tomography (OCT) is essential for the diagnosis and follow-up of corneal edema, but assessment can be challenging in minimal or localized edema. The objective was to develop and validate a novel automated tool to detect and visualize corneal edema with OCT. We trained a convolutional neural network to classify each pixel in the corneal OCT images as "normal" or "edema" and to generate colored heat maps of the result. The development set included 199 OCT images of normal and edematous corneas. We validated the model's performance on 607 images of normal and edematous corneas of various conditions. The main outcome measure was the edema fraction (EF), defined as the ratio between the number of pixels labeled as edema and those representing the cornea for each scan. Overall accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve were determined to evaluate the model's performance. Mean EF was 0.0087 ± 0.01 in the normal scans and 0.805 ± 0.26 in the edema scans (P < 0.0001). Area under the receiver operating characteristic curve for EF in the diagnosis of corneal edema in individual scans was 0.994. The optimal threshold for distinguishing normal from edematous corneas was 6.8%, with an accuracy of 98.7%, sensitivity of 96.4%, and specificity of 100%. The model accurately detected corneal edema and distinguished between normal and edematous cornea OCT scans while providing colored heat maps of edema presence.

Predicting the likelihood of need for future keratoplasty intervention using artificial intelligence

ObjectiveTo apply artificial intelligence (AI) for automated identification of corneal condition and prediction of the likelihood of need for future keratoplasty intervention from optical coherence tomography (OCT)-based corneal parameters. DesignCohort study. ParticipantsWe collected 12,242 corneal OCT images from 3162 subjects using CASIA OCT Imaging Systems (Tomey, Japan). We included 3318 measurements collected at the baseline visit of each patient. A total of 333 eyes had post-operative penetrating keratoplasty (PKP), lamellar keratoplasty (LKP), deep anterior keratoplasty (DALK), descemet's stripping automated endothelial keratoplasty (DSAEK) or descemet's membrane endothelial keratoplasty (DMEK) intervention. MethodWe developed a pipeline including linear and nonlinear data transformations followed by unsupervised machine learning and applied on corneal parameters from the baseline visit of each patient. Five non-overlapping clusters of eyes were identified. Post hoc analyses revealed that clusters corresponded to different likelihoods of need for future keratoplasty. These clusters on a 2-dimensional map can be used by clinicians and surgeons to identify patients with higher risk of need for future keratoplasty intervention. Main outcome measuresThe likelihood of the need for future surgery. ResultsThe mean age of participants was 69.7 (standard deviation; SD = 16.1) and 59% were female. The normalized likelihood of need for future corneal keratoplasty intervention for eyes mapped onto clusters one to five were 2.2%, 1.0%, 33.1%, 32.7%, and 31.0%, respectively. ConclusionsThe AI system can assist the (cornea) surgeon in identifying those patients who may be at higher risk for future keratoplasty using comprehensive corneal shape, thickness, and elevation parameters. Future research utilizing independent datasets is necessary to validate the proposed system.

Automated Detection and Classification of Corneal Haze Using Optical Coherence Tomography in Patients With Keratoconus After Cross-Linking.

To evaluate a proposed technology for offering objective grading and mapping of corneal haze as detected by corneal spectral domain optical coherence tomography after corneal cross-linking. This was a retrospective study to evaluate corneal optical coherence tomography images performed on 44 eyes of 44 patients who underwent corneal cross-linking between January 2014 and May 2015, at the American University of Beirut Medical Center. Overall average brightness of the cornea was markedly increased from 43.4% (±6.0) at baseline to 50.2% (±4.4) at 1 month, 47.9% (±4.4) at 3 months, and 46.4% (±5.7) at 6 months with P <0.001, <0.001, and 0.005, respectively. In the anterior stroma, the average brightness significantly increased at 1, 3, and 6 months with values of 54.8% (±3.9), 52.5% (±5.2), and 49.7% (±6.9) with P <0.001, <0.001, and 0.003, respectively. In the mid stroma, the change was clinically significant at 1 and 3 months, whereas in the posterior stroma, it was only significant at 1 month compared with baseline (P = 0.003). Overall, haze was mostly present at 1 month after surgery in all regions, especially in the anterior (32.1%; ±19.2) and mid stromal regions (9.1%; ±18.8), P <0.001 and 0.001, respectively. In contrast, haze in the posterior stromal region peaks at 3 and 6 months after surgery. Anterior stromal haze was the greatest in intensity and area and it was present for a longer time span than mid and posterior stromal haze. At 12 months, the anterior stroma had still more haze intensity than preoperatively. This image-based software can provide objective and valuable quantitative measurements of corneal haze, which may impact clinical decision-making after different corneal surgeries.

A Conditional Random Field Weakly Supervised Segmentation Approach for Segmenting keratocytes Cells in Corneal Optical Coherence Tomography Images

&lt;p&gt;Keratocytes are vital for maintaining the overall health of human&lt;br /&gt;cornea as they preserve the corneal transparency and help in healing&lt;br /&gt;corneal injuries. Manual segmentation of keratocytes is challenging,&lt;br /&gt;time consuming and also needs an expert. Here, we propose&lt;br /&gt;a novel semi-automatic segmentation framework, called Conditional&lt;br /&gt;Random FieldWeakly Supervised Segmentation (CRF-WSS)&lt;br /&gt;to perform the keratocytes cell segmentation. The proposed framework&lt;br /&gt;exploits the concept of dictionary learning in a sparse model&lt;br /&gt;along with the Conditional Random Field (CRF) modeling to segment&lt;br /&gt;keratocytes cells in Ultra High Resolution Optical Coherence&lt;br /&gt;Tomography (UHR-OCT) images of human cornea. The results&lt;br /&gt;show higher accuracy for the proposed CRF-WSS framework compare&lt;br /&gt;to the other tested Supervised Segmentation (SS) andWeakly&lt;br /&gt;Supervised Segmentation (WSS) methods.&lt;/p&gt;