Three-Dimensional Map of Descemet Membrane Endothelial Keratoplasty Detachment

Abstract

To develop and apply a neural network for quantification of endothelial corneal graft detachment using anterior segment (AS) OCT.Design: Training and validation of a neural network and application within a prospective cohort.Participants: Patients two weeks after Descemet membrane endothelial keratoplasty. Methods: Investigators manually labeled the posterior cornea and the graft in cross-sectional images of rotational AS OCT scans. Neural networks for image segmentation were trained to identify the area of graft detachment on cross-sectional images. The best-performing neural network with the lowest misclassification (Youden index) and highest spatial overlap with the ground truth (Dice coefficient) was selected and evaluated in a separate dataset. Three-dimensional maps of the area and volume of graft detachment were calculated. For application, the neural network's rating on the detachment was compared with slit-lampebased ratings of cornea specialists on the same day as the AS OCT imaging took place.Main Outcome Measures: Youden index and Dice coefficient.Results: Neural networks were trained on 27 AS OCT scans with 6912 labeled images. Among 48 combinations of probability thresholds and epoch states, the best-performing neural network showed a Youden index of 0.99 and a Dice coefficient of 0.77, indicating low misclassification and good spatial overlap on individual image segmentation. In the validation set unknown to the neural network with 20 scans (5120 images), the Youden index was 0.85 and the Dice coefficient was 0.73, and a high overall performance compared with the manually labeled ground truth (R 2 ¼ 0.90). In the application set with 107 eyes, the neural network estimated the mean percent detachment larger than the cornea specialist (mean difference, 8.2 percentage points; 95% confidence interval, 6.2e10.2). Masked review of 42 AS OCTs with more than AE10 percentage points difference in ratings showed that clinicians underestimated the true detachment in cases with significant detachment requiring intervention.Conclusions: Deep learning-based segmentation of AS OCT images quantified the percent and the volume of DMEK graft detachment with high precision. Fully automated 3-dimensional quantification of graft detachment is highly sensitive, particularly in corneas with a significant amount of detachment, and may support decision making.

Graft Detachment after Descemet Membrane Endothelial Keratoplasty with and without Cataract Surgery

“…Compared to clinical slit lamp-based experience, the median area of graft detachment at the first AS OCT (31%) may seem relatively large, which is fully expected as the neural network identifies graft detachments missed by corneal specialists on slit lamp exam. 13,16 Further, it quantifies depth (mainly flat) and location (mainly graft margins) of detachment. These measures contrast with the usual rules of thumb, such as one-third of area detached, which may not be nuanced enough as an indication for rebubbling.…”

“…2,15 To guide clinical decision making, we developed and validated a convolutional neural network to quantify the area and volume of graft detachment using a UNet++-based deep learning algorithm for image segmentation of AS OCT scans. 16 Quantifications were valid and able to identify detachments missed by clinicians on slit-lamp exam alone. 16…”

“…16 Quantifications were valid and able to identify detachments missed by clinicians on slit-lamp exam alone. 16…”

“…In brief, the validated neural network automatically segmented the posterior corneal surface and the graft in each image to quantify the area, the volume and height of graft detachment. 16 The area of detachment (range, 0-100%; continuous) was determined with respect to the total graft surface (range of trephine sizes, 7-8 mm).…”

Corneal Oedema: Aetiology, Diagnostic Testing, and Treatment

“…In patients with partial graft detachment, corneal oedema may not resolve. In these cases, AS-OCT has proven to be precise and sensitive in assessing graft adherence postoperatively [60].…”