Paper
11 August 2023 Low-cost retinal imaging for disease progression monitoring at home
Ryo Kubota, Stefan Troller, Matthias Pfister
Author Affiliations +
Abstract
The unmet medical need of continuously monitoring retinal disease progression asks for an affordable and easy to use medical device that can be used multiple times per week at the patient’s home. A retinal monitor system has been designed for home use, has been built, and tested in a clinical study. The retinal monitor system consists of a handheld device operated by the patient, a tablet with an application, and cloud software. The retinal monitor system has been optimized for low production cost and ease of use for elderly patients while a minimum required performance has been assured to detect changes of retinal thickness and the presence of fluid. Component costs are optimized by relying on the wavelength-tuning capabilities of a VCSEL light source. The small form factor, the lightweight design and an auto-capture functionality enable patients to operate the retinal monitor on their own. The limited performance of low-cost components is compensated by sophisticated signal processing including a deep neural network. Clinical data has been acquired on 20 healthy and 50 diseased eyes to develop the algorithm and estimate the device performance. The clinical results indicate a 95% sensitivity and 95% specificity to detect a change of ±25 μm. Moreover, presence of fluid has been detected with a true positive rate of 97% and a true negative rate of 92%.
© (2023) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Ryo Kubota, Stefan Troller, and Matthias Pfister "Low-cost retinal imaging for disease progression monitoring at home", Proc. SPIE 12632, Optical Coherence Imaging Techniques and Imaging in Scattering Media V, 1263207 (11 August 2023); https://doi.org/10.1117/12.2670574
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Retinal diseases

Retinal scanning

Algorithm development

Design and modelling

Light sources

Medical devices

Neural networks

Back to Top