EyeWorld India December 2020 Issue

NEWS & OPINION 56 EWAP DECEMBER 2020 network(BNN)” for the two key lesion components in PCV. The workgroup recommended the term “polypoidal lesion” as opposed to “polyps” or “aneurysmal lesions” as both terms had histological connotations that were not completely representative of the nature of these lesions. The term “branching neovascular network” was preferred over “branching vascular network” to better reflect the neovascular nature of the vascular network within a PCV complex. These recommendations in nomenclature aimed to reflect the clinical and histological characteristics of PCV as we currently understand it. A new set of recommended diagnostic criteria which does not require the use of ICGA to differentiate the subtype was devised and validated. All 15 members of the workgroup independently graded the presence or absence of nine shortlisted signs in a test set comprising CFP, enhanced depth imaging OCT, and enface OCT of the eye with and without PCV. These nine signs included six signs on OCT (sharp-peaked pigment epithelial detachment (PED), sub-retinal pigment epithelial (RPE) ring-like lesion, multilobular PED, double layer sign, thick choroid with dilated Haller’s layer vessels, fluid compartment), one sign on enface OCT (complex RPE elevation)—and two signs on CFP (extensive subretinal hemorrhage and orange nodules). The signs were divided into major and minor criteria based on an area under the curve (AUC) of >0.75 or <0.75, respectively. Accuracy was further enhanced by combining major and minor criteria. The combination of three OCT- based “major criteria”—sub-RPE ring-like lesion, enface OCT complex RPE elevation, and sharp-peaked PED emerged as the most accurate in diagnosing PCV without ICGA with an AUC of 0.90. A further validation exercise based on the above combination was performed. This exercise comprised six readers from Singapore and Milan (three specialists and three trainees) who applied the recommendation to an independent test set of 80 eyes (50 from Singapore and 30 from Italy) and achieved an accuracy of 75–82%. These results will be very useful for both retina specialist and non-retinal specialist alike and represent a set of practical, clinically applicable OCT based diagnostic criteria for PCV with high accuracy. This is especially helpful in settings where ICGA is not routinely performed. With this OCT-based criteria, we hope to empower clinicians with a simple tool to differentiate PCV from typical nAMD so alternative treatment modalities can be considered especially in eyes that respond poorly to VEGF inhibitor monotherapy. EWAP Editors’ note: The authors acknowledge that this research is supported by the National Medical Research Council Singapore Open Fund Large Collaborative Grant: NMRC/ LCG/004/2018. In addition, Dr. Cheung reports non-financial support from Bayer during the conduct of the study, as well as, outside the submitted work, grants, personal fees, and non-financial support from Bayer; grants, personal fees, and non-financial support from Novartis; grants from Roche; grants from GlaxoSmith Kline; non-financial support form Allergan; non-financial support from Topcon. Dr. Teo reports consultancy fees, honorarium, travel support, and speaker fees from Bayer and Novartis outside the submitted work. nature of an AI system is that machine learning makes it difficult to predict at what point the algorithm should be reviewed,” he said. There is also a risk for false negatives. Impact on clinical trials “I think we’re going through a learning phase about how AI can be used in clinical trials,” Dr. Stoller said. He thinks AI could help identify patients most likely to develop disease or progress and could impact dosing regimens used in clinical trials. “I think there’s a great deal of promise,” he said. “It’s not something that’s been widely adopted yet in the ophthalmic community, but I think we’re in the early phases of seeing that paradigm shift and embracing the technology.” AI could pinpoint groups at higher risk of progression, identifying subgroups that would be best suited for preventive therapies, Dr. Lim said. In addition, AI could analyze the images for treatment response instead of relying on human graders. Lastly, AI may find patterns of disease that respond better to a treatment, she added. Dr. Joseph also thinks that AI could have an impact in clinical trials. “I think clinical trials are more rigorous in their controls and designs than regular clinical practice,” he said. “I think it’s a good opportunity for gathering and curating data and clinical images that would be useful in developing machine learning tools and accurate algorithms.” Dr. Joseph added that he thinks researchers will realize when designing clinical trials that they may want to capture data in a fashion that will lend itself well to developing AI tools, in addition to testing whatever medications or interventions that they’re trying to study. EWAP Editors’ note: Dr. Ho is Director, Retina Research, Wills Eye Hospital, Philadelphia, Pennsylvania. Dr. Joseph practices at Ophthalmic Consultants of Boston, Boston, Massachusetts. Dr. Lim is Director, Retina Service, University of Illinois at Chicago, Chicago, Illinois. Dr. Stoller practices at Ophthalmic Consultants of Long Island, Rockville Centre, New York. None of the doctors declared conflicting interests. Opportunities for artificial intelligence - from page 54