In ophthalmology, traditional color-based cameras are commonly used for eye fundus examination but lack spectral resolution, which may cause certain alterations to go unnoticed. To address this issue, multispectral imaging systems have been developed. These systems capture images across multiple spectral bands, from the visible (VIS) to the near-infrared (NIR), providing more detailed information about fundus structures. This study presents a prototype towards a portable multispectral fundus camera that captures images in the 470–850 nm spectral range, offering both high spatial and spectral resolution. The camera consists of a modified PanOpticTM ophthalmoscope with fiber-coupled LEDs and a monochrome CMOS camera, all controlled with a software developed in Matlab. Reflectance values are extracted from fundus structures, and a contrast parameter is used for their discrimination. Results show that the prototype is able to obtain spectral images in the VIS and NIR, and quantitative spectral data can be extracted from the fundus images.

This publication is part of the project PID2023-147541OB-I00, funded by MCIU /AEI /10.13039/501100011033 and by FEDER, UE. The first author gratefully acknowledges the Universitat Politècnica de Catalunya and Banco Santander for the financial support of her predoctoral grant FPIUPC.

Peer Reviewed