Pilot Clinical Validation of a Machine Learning Platform for Noninvasive Smartphone-Based Assessment of Corneal Epithelial Integrity

Abstract

Purpose: Fluorescein staining (FS) is a standard method of assessing corneal epithelium (CE) integrity. However, the equipment and personnel required for FS may be unavailable in low-resource environments. We developed and validated a low-cost, noninvasive, and quantitative CE evaluation pipeline using a custom smartphone attachment and convolutional neural networks (CNNs). Methods: A 3D-printed smartphone attachment and placido disk illumination module was attached to a OnePlus 7 Pro smartphone. 26 smartphone-acquired images were obtained from 15 subjects, comprising a dataset including healthy eyes and corneal epitheliopathies of Oxford grade I-V. A classifier CNN was trained on 8 subjects (23,173 image patches) to identify areas of suspected epithelial disruption, and validated on 7 subjects (10,883 image patches). The fraction of disrupted corneal surface area (FDSA) was computed for each subject from the model output. Results were compared with FS slit lamp photos which were independently graded by two clinicians using the Oxford scheme. Results: FDSA showed promise as a non-invasive marker of CE integrity, with mean FDSA in the Oxford >II cohort being higher than the Oxford ≤II cohort (p = 0.04 and p = 0.09 using Oxford scores from each clinician, respectively). Additionally, areas of CE disruption identified by our smartphone-based technique showed qualitative concordance with those revealed by FS. Conclusions: Our technique for smartphone-based CE imaging and automated analysis is a promising low-cost, noninvasive method to quantitatively evaluate the CE. Translational Relevance: This tool can be used to evaluate ocular surface disease in low-resource regions.

Integration of smartphone technology and artificial intelligence for advanced ophthalmic care: A systematic review

“…to validate a corneal epithelium (CE) evaluation pipeline using a custom smartphone attachment and CNNs. 33 The results showed that the smartphone-based CE evaluation tool in calculating areas of CE disruption had qualitative concordance with those revealed by fluorescein staining slit-lamp photos graded by two clinicians. Apart from AI models, it should be noted that specialized equipment like slit-lamp is still necessary for smartphone-based anterior segment imaging and the smartphone camera is in need of a short focus distance to scan the ocular surface and high resolutions to capture the images.…”