Search

Purpose: To report an image analysis pipeline, DDLSNet, consisting of a rim segmentation (RimNet) branch and a disc size classification (DiscNet) branch to automate estimation of the disc damage likelihood scale (DDLS). Design: Retrospective observational. Participants: RimNet and DiscNet were developed with 1208 and 11 536 optic disc photographs (ODPs), respectively. DDLSNet performance was evaluated on 120 ODPs from the RimNet test set, for which the DDLS scores were graded by clinicians. Reproducibility was evaluated on a group of 781 eyes, each with 2 ODPs taken within 4 years apart. Methods: Disc damage likelihood scale calculation requires estimation of optic disc size, provided by DiscNet (VGG19 network), and the minimum rim-to-disc ratio (mRDR) or absent rim width (ARW), provided by RimNet (InceptionV3/LinkNet segmentation model). To build RimNet’s dataset, glaucoma specialists marked optic disc rim and cup boundaries on ODPs. The ''ground truth'' mRDR or ARW was calculated. For DiscNet’s dataset, corresponding OCT images provided ''ground truth'' disc size. Optic disc photographs were split into 80/10/10 for training, validation, and testing, respectively, for RimNet and DiscNet. DDLSNet estimation was tested against manual grading of DDLS by clinicians with the average score used as ''ground truth.'' Reproducibility of DDLSNet grading was evaluated by repeating DDLS estimation on a dataset of nonprogressing paired ODPs taken at separate times. Main Outcome Measures: The main outcome measure was a weighted kappa score between clinicians and the DDLSNet pipeline with agreement defined as ± 1 DDLS score difference. Results: RimNet achieved an mRDR mean absolute error (MAE) of 0.04 (± 0.03) and an ARW MAE of 48.9 (± 35.9) degrees when compared to clinician segmentations. DiscNet achieved 73% (95% confidence interval [CI]: 70%, 75%) classification accuracy. DDLSNet achieved an average weighted kappa agreement of 0.54 (95% CI: 0.40, 0.68) compared to clinicians. Average interclinician agreement was 0.52 (95% CI: 0.49, 0.56). Reproducibility testing demonstrated that 96% of ODP pairs had a difference of ≤ 1 DDLS score. Conclusions: DDLSNet achieved moderate agreement with clinicians for DDLS grading. This novel approach illustrates the feasibility of automated ODP grading for assessing glaucoma severity. Further improvements may be achieved by increasing the number of incomplete rims sample size, expanding the hyperparameter search, and increasing the agreement of clinicians grading ODPs. Financial Disclosure(s): Proprietary or commercial disclosure may be found after the references.

Purpose: Accurate neural rim measurement based on optic disc imaging is important to glaucoma severity grading and often performed by trained glaucoma specialists. We aim to improve upon existing automated tools by building a fully automated system (RimNet) for direct rim identification in glaucomatous eyes and measurement of the minimum rim-to-disc ratio (mRDR) in intact rims, the angle of absent rim width (ARW) in incomplete rims, and the rim-to-disc-area ratio (RDAR) with the goal of optic disc damage grading. Design: Retrospective cross sectional study. Participants: One thousand and twenty-eight optic disc photographs with evidence of glaucomatous optic nerve damage from 1021 eyes of 903 patients with any form of primary glaucoma were included. The mean age was 63.7 (± 14.9) yrs. The average mean deviation of visual fields was −8.03 (± 8.59). Methods: The images were required to be of adequate quality, have signs of glaucomatous damage, and be free of significant concurrent pathology as independently determined by glaucoma specialists. Rim and optic cup masks for each image were manually delineated by glaucoma specialists. The database was randomly split into 80/10/10 for training, validation, and testing, respectively. RimNet consists of a deep learning rim and cup segmentation model, a computer vision mRDR measurement tool for intact rims, and an ARW measurement tool for incomplete rims. The mRDR is calculated at the thinnest rim section while ARW is calculated in regions of total rim loss. The RDAR was also calculated. Evaluation on the Drishti-GS dataset provided external validation (Sivaswamy 2015). Main Outcome Measures: Median Absolute Error (MAE) between glaucoma specialists and RimNet for mRDR and ARW. Results: On the test set, RimNet achieved a mRDR MAE of 0.03 (0.05), ARW MAE of 31 (89)°, and an RDAR MAE of 0.09 (0.10). On the Drishti-GS dataset, an mRDR MAE of 0.03 (0.04) and an mRDAR MAE of 0.09 (0.10) was observed. Conclusions: RimNet demonstrated acceptably accurate rim segmentation and mRDR and ARW measurements. The fully automated algorithm presented here would be a valuable component in an automated mRDR-based glaucoma grading system. Further improvements could be made by improving identification and segmentation performance on incomplete rims and expanding the number and variety of glaucomatous training images.

Background: Clinically, there are many systems of visual field staging parallel exist. The agreement between glaucoma function and appearance is still uncertain. This study was to compare visual field staging classifications (as function staging) with disc damage likelihood scale (DDLS as appearance staging) in primary open angle glaucoma (POAG). Methods: Cross-section study on 55 patients (90 eyes) examined and treated in Vietnam National Institute of Ophthalmology from 9/2016 – 9/2017. The MD (mean deviation), PSD (pattern standard deviation), VFI (visual field index) indexes were recorded and were used to classify glaucoma staging according to categories of eGSS (enhance glaucoma severity staging), mGSS (modified glaucoma severity staging), HPA (Hoddaps – Palmela – Adersons), AGIS (advance glaucoma intervention study) and DDLS. Using SPSS 16.0 software, statistically significance was tested with proper tests as Chi – square test, Fisher exact test, Kappa (to assess the agreement - disagreement), Spearman Rank (to assess the level of correlation). Results: Mean visual field indexes were -11.49 dB (MD), 5.85 dB (PSD), 74.2% (VFI). Compared to the DDLS, all systems had slight agreement (K < 0.2). The disagreement percentage was highest in eGSS (70%) then followed by mGSS (50%) and HPA (48.9%). In terms of definitive diagnosis, HPA and mGSS showed a substantial agreement (K > 0.6) with AGIS which is higher than that between eGSS and AGIS (retrospectively 0.773 and 0.75 vs 0.399) with p < 0.001. In terms of staging detection, the agreement between HPA &mGSS and AGIS was substantial (K > 0.6) while that between eGSS and AGIS was fair (K < 0.4). Conclusion: mGSS and HPA tend to show the stronger agreement with standard classifications than eGSS. mGSS should be used in clinical practice and research.