DEDector: Noninvasive Detection of Dry Eye Disease using Smartphones. (Preprint)

BACKGROUND Dry Eye Disease (DED) is a prevalent eye condition characterized by abnormalities in tear film stability. Despite its high prevalence, diagnosing DED remains challenging, primarily due to the invasive nature of most diagnostic tests, such as the Tear BreakUp Time (TBUT). OBJECTIVE The aim of this paper is to propose a novel smartphone-based methodology for non-invasive screening of Dry Eye Disease. METHODS In this work, we propose DEDector, a low-cost, smartphone-based automated Non-Invasive Break-Up Time (NIBUT) measurement methodology for DED diagnosis. Utilizing a 3D-printed placido ring attachment on a smartphone’s camera, DEDector projects concentric rings onto the cornea, capturing a video for subsequent analysis using our proposed video processing pipeline to identify tear film stability. RESULTS We conducted a real-world evaluation on 46 eyes from 23 patients, comparing the performance of DEDector against the traditional Fluorescein Break-Up Time (TBUT) method. Our results indicate a sensitivity of 77.78% and specificity of 82.14% for DEDector, outperforming the TBUT-based approach (Sensitivity = 72.22%, Specificity = 75%) CONCLUSIONS DEDector may be used for large-scale and point-of-care screening of Dry Eye Disease, due to its low cost, portability, and effectiveness in detecting Dry Eye Disease.