Utility and Feasibility of Teleophthalmology Using a Smartphone-Based Ophthalmic Camera in Screening Camps in Nepal

Abstract

Purpose: To determine whether use of a mobile device-based ophthalmic camera by ophthalmic technicians (OTs) in village screening camps in Nepal followed by remote image interpretation by an ophthalmologist can improve detection of ocular pathology and medical decision-making.

Design: Evaluation of mobile device-based ophthalmic camera through study of before and after clinical decision-making.

Methods: One hundred forty patients over 18 years of age presenting to remote screening camps with best-corrected visual acuity ≤20/60 in one or both eyes were enrolled. Participants were examined by an OT with direct ophthalmoscopy. The technician recorded a diagnosis for each eye and a disposition for each patient. Patients then had anterior segment and fundus photos and/or videos taken using a smartphone-based ophthalmic camera system. Photos and videos were uploaded to a secure, HIPAA-compliant, cloud-based server, and interpreted by masked ophthalmologists from XXX, who independently recorded diagnoses and a disposition for each patient.

Results: The diagnoses given by OTs and ophthalmologists differed in 42.4% of eyes. Diagnosis agreement was highest for cataract [k = 0.732, 95% confidence interval (CI) 0.65-0.81], but much lower for posterior segment (retina/optic nerve) pathology (k = 0.057, 95% CI -0.03-0.14). Ophthalmologists and OTs suggested different dispositions for 68.6% of patients. Agreement was highest for cataract extraction (k = 0.623, 95% CI 0.49-0.75), whereas agreement for referral to XXX was lower (k = 0.12, 95% CI 0.00-0.24).

Conclusions: Remote ophthalmologist consultation utilizing a mobile device ophthalmic camera system is logistically feasible, easily scalable, and capable of capturing high-quality images in the setting of rural eye screening camps. Although OTs are well equipped to identify and triage anterior segment pathology, this technology may be helpful in the detection of and referral for posterior segment pathology.

FIGURE 1

(A) Paxos Scope ophthalmic camera system with removable light cover mounted to an iPhone, being used by author SC to take a fundus photo. (B) Example of normal fundus photo taken with Paxos scope, (C) Example of abnormal fundus photo taken with Paxos scope (toxoplasmosis scar).

FIGURE 2

Comparison of diagnoses between ophthalmologist and technician. Note that “Total Agreements” excludes cases where ophthalmologist and technician agreed on the absence of a condition. NAD indicates no abnormality detected.

FIGURE 3

Comparison of disposition between ophthalmologist and technician. “Total Agreements” excludes cases where ophthalmologist and technician agreed on the absence of a condition; CEC indicates community eye centers; TIO, Tilganga Institute of Ophthalmology.