In this prospective, comparative, clinical study the authors’ aim was to assess the reliability and accuracy of smartphone ophthalmoscopy as compared with standard slit-lamp biomicroscopy in the grading of diabetic retinopathy. One hundred and twenty consecutive, new, diabetic patients underwent dilated fundoscopy first with the smartphone ophthalmoscope by one investigator and subsequently with slit-lamp biomicroscopy by another investigator, who was masked to the findings of the smartphone imaging. The D-eye device was used, which was attached to an iPhone 5, and captures images with the camera’s sensor. A single fundus image covers 20 degrees of field at a distance of 1cm from the patient’s eye. Colour images were obtained that covered the macula, optic disc and peripheral retina. Each investigator completed a template, reporting on the state of the pupil, media clarity and various clinical signs of diabetic retinopathy, which was then graded according to the International Clinical Diabetic Retinopathy Disease Severity Scale (five grades).
In 85% (CI 0.71-0.84) of cases there was total agreement between the two examination techniques and an agreement within one grade was achieved in 96.7%. Similarly, there was significant agreement regarding the presence or absence of diabetic macular oedema. The smartphone ophthalmoscope was, however, more sensitive to media opacities and pupil diameter. The authors felt that the smartphone device offers practical advantages in that it is easier to use than a direct ophthalmoscope, it is highly portable and with the wireless connectivity offered by smartphones, would lend itself to telemedicine particular in non-hospital settings in the community.
