Abstract
We aimed to classify early normal-tension glaucoma (NTG) and glaucoma suspect (GS) using Bruch’s membrane opening-minimum rim width (BMO-MRW), peripapillary retinal nerve fiber layer (RNFL), and the color classification of RNFL based on a deep-learning model. Discriminating early-stage glaucoma and GS is challenging and a deep-learning model may be helpful to clinicians. NTG accounts for an average 77% of open-angle glaucoma in Asians. BMO-MRW is a new structural parameter that has advantages in assessing neuroretinal rim tissue more accurately than conventional parameters. A dataset consisted of 229 eyes out of 277 GS and 168 eyes of 285 patients with early NTG. A deep-learning algorithm was developed to discriminate between GS and early NTG using a training set, and its accuracy was validated in the testing dataset using the area under the curve (AUC) of the receiver operating characteristic curve (ROC). The deep neural network model (DNN) achieved highest diagnostic performance, with an AUC of 0.966 (95%confidence interval 0.929–1.000) in classifying either GS or early NTG, while AUCs of 0.927–0.947 were obtained by other machine-learning models. The performance of the DNN model considering all three OCT-based parameters was the highest (AUC 0.966) compared to the combinations of just two parameters. As a single parameter, BMO-MRW (0.959) performed better than RNFL alone (0.914).
Highlights
The discrepancy between Bruch’s membrane opening-minimum rim width (BMO-MRW) and retinal nerve fiber layer (RNFL) color code classification[8]
We found that BMO-MRW may show a normal classification, whereas RNFL may show an abnormal classification in cases of large disc and myopia, which suggests the clinical usefulness of BMO-MRW in early glaucoma or glaucoma suspect when the RNFL color code classification may show false-positive findings[8]
Regarding global RNFL color code classifications, 94.3%, 4.4%, and 1.3% of the glaucoma suspect (GS) patients and 38.7%, 20.2%, and 41.4% of the early normal-tension glaucoma (NTG) patients showed within normal limits (WNL), borderline (BL), and outside normal limits (ONL), respectively
Summary
The discrepancy between BMO-MRW and RNFL color code classification[8]. We found that BMO-MRW may show a normal classification, whereas RNFL may show an abnormal classification in cases of large disc and myopia, which suggests the clinical usefulness of BMO-MRW in early glaucoma or glaucoma suspect when the RNFL color code classification may show false-positive findings[8]. Few previous studies using deep-learning investigated BMO-MRW for glaucoma diagnosis. One study by Park et al.[12] reported the diagnostic performance of combined BMO-MRW and RNFL using a neural network for glaucoma but early glaucoma was only partly included and they did not consider the RNFL color code classification. Previous deep-learning studies classifying glaucoma and normal subjects rarely included NTG and studies solely on NTG are difficult to find. In this retrospective cross-sectional study, we aimed to classify early NTG and GS using OCT imaging-based parameters including the new parameter, BMO-MRW, and peripapillary RNFL, along with the color classification of RNFL, based on a new model of deep-learning. We employed a deep-learning model to integrate all available data from the OCT images, which may be difficult for general physicians
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
