In a proof-of-concept study, researchers have determined that a smartphone tonometer that uses a machine-learning algorithm can accurately calculate IOP. After comparing its measurements with those recorded by Goldmann applanation tonometry (GAT) and other commonly used tonometers, the researchers found them grossly equivalent to the reference standard.
The study assessed 92 eyes of 81 participants. The smartphone tonometer uses fixed-force applanation in conjunction with an algorithm that detects the tonometer circle and the mire from each frame in the smartphone measurement, thereby calculating the IOP. The researchers also measured all subjects using GAT, while a subset were measured using the iCare tonometer, pneumotonometry and the Tono-Pen (Reichert).
The mean difference for IOP measurements of the smartphone tonometer vs. the other devices was +0.24mm Hg for GAT, −1.39mm Hg for the iCare, −3.71mm Hg for pneumotonometry and −1.30mm Hg for Tono-Pen. Overall, the researchers found that the smartphone tonometer results correlated best with GAT. Of the 92 measurements, 90 were within ±5mm Hg of GAT and 58 were within ±2mm Hg of GAT.
The researchers are certain that their prototype smartphone tonometer directly visualizes and measures the applanation surface. They believe that it could complement other existing smartphone hardware attachments and apps to facilitate a portable ophthalmological examination, which may improve access to eye care in resource-poor regions.
|Wu Y, Luttrell I, Feng S, et al. Development and validation of a machine learning, smartphone-based tonometer. Br J Ophthalmol. December 23, 2019. [Epub ahead of print].|