Classifying Papilledema Using Machine Learning – Modernizing Frisén Grading (S12.001)

Abstract

<h3>Objective:</h3> We hypothesize that machine learning (ML) can quantify papilledema and detect a treatment effect on papilledema due to idiopathic intracranial hypertension (IIH). <h3>Background:</h3> ML can differentiate papilledema from normal optic discs. Currently, papilledema severity is assessed using the descriptive, ordinal Frisén scale. <h3>Design/Methods:</h3> We trained a convolutional neural network (CNN) to autonomously assign a Frisén grade using 2608 fundus photos from fellow eyes of 158 participants in the IIH Treatment Trial (IIHTT) and both eyes from eight clinic patients with grade 4 or 5 papilledema. Experts in classifying papilledema previously graded the photos. Our validation set consisted of 2969 photos of the IIHTT study eyes (for each participant, the eye with worse vision). To investigate the change over time, we divided study eyes into treatment groups, acetazolamide + diet (ACZ) vs. placebo + diet. <h3>Results:</h3> ML produced continuous values from 0–5, which identified photos that contained features from more than one Frisén grade. Activation maps showed the model focused on the optic disc. The average predicted Frisén grade correlated strongly with ground truth (r = 0.80, p &lt; 0.001; mean absolute error = 0.49). At presentation, treatment groups had similar ML Frisén grades. The average ML Frisén grade for the ACZ treatment group (1.6, 95% CI 1.5–1.8) was significantly lower (p &lt; 0.01) than for the placebo group (2.2, 95% CI 1.9–2.5) at the six month trial outcome. This difference was noted as early as one month (p = 0.03). <h3>Conclusions:</h3> Supervised ML of fundus photos successfully grades the degree of papilledema and tracks the changes, reflecting the effects of ACZ therapy. Given the increasing availability of fundus photography, neurologists will be able to utilize ML to quantify papilledema on a continuous scale that incorporates the descriptive features of the Frisén grade to monitor treatment of papilledema. <b>Disclosure:</b> Mr. Branco has nothing to disclose. Jui-Wang Wang has nothing to disclose. Dr. Elze has nothing to disclose. Mona Garvin has nothing to disclose. Mr. Szanto has nothing to disclose. Dr. Kardon has received personal compensation in the range of $50,000-$99,999 for serving as an Expert Witness for 19 different law firms. The institution of Dr. Kardon has received research support from NIH. The institution of Dr. Kardon has received research support from Veterans Administration RR&amp;D Division. Dr. Kardon has received intellectual property interests from a discovery or technology relating to health care. Dr. Pasquale has received personal compensation for serving as an employee of Eyenovia. Dr. Pasquale has received personal compensation for serving as an employee of Twenty Twenty. Dr. Pasquale has received personal compensation for serving as an employee of Skye Biosciences. Dr. Pasquale has received personal compensation for serving as an employee of Character Biosciences. Dr. Pasquale has received personal compensation in the range of $0-$499 for serving on a Scientific Advisory or Data Safety Monitoring board for The Glaucoma Foundation. The institution of Dr. Pasquale has received personal compensation in the range of $50,000-$99,999 for serving on a Scientific Advisory or Data Safety Monitoring board for The Glaucoma Foundation. The institution of Dr. Pasquale has received research support from NEI. Dr. Kupersmith has nothing to disclose.