Retinal Image Processing Method Using Multispectral Imaging for ML Training

ChangeBridge: Patent Apps - Diagnosis & Surgery (A61B)

Published April 2nd, 2026

Detected April 2nd, 2026

Summary

USPTO published patent application US20260090719A1 for a computer-implemented method of processing retinal images using multispectral imaging and machine learning. The invention assesses image quality by determining artifact extent across different wavelengths and computing a quality index to determine suitability for training machine learning algorithms. Inventors include Claudia Chevrefils, Gabriel Dignard, Wei You, and Jean-Philippe Sylestre.

View original document View source feed page

What changed

This patent application describes a computer-implemented method and system for processing retinal images. The method involves accessing multispectral images of a retina captured at distinct wavelengths, determining the extent of artifacts in images at each wavelength, computing a quality index based on artifact measurements, and determining whether the multispectral image is suitable for training a machine-learning algorithm.

Patent applications do not create compliance obligations. No deadlines, penalties, or required actions apply to this publication. Entities in the medical imaging or AI healthcare sectors may review the application for competitive intelligence or to assess potential licensing opportunities.

Source document (simplified)

← USPTO Patent Applications

METHOD AND SYSTEM FOR PROCESSING RETINAL IMAGES

Application US20260090719A1 Kind: A1 Apr 02, 2026

Inventors

Claudia CHEVREFILS, Gabriel DIGNARD, Wei YOU, Jean-Philippe SYLVESTRE

Abstract

A computer-implemented method of processing retinal images comprises accessing a multispectral image of a retina, the multispectral image comprising a first image of the retina associated with a first wavelength and a second image of the retina associated with a second wavelength, the second wavelength being distinct from the first wavelength, determining a first extent of a first artefact associated with the first image, determining a second extent of the first artefact or of a second artefact associated with the second image, computing a quality index based on the first extent of the first artefact and the second extent of the first or second artefact; and determining, based on the quality index, whether the multispectral image is suitable for training a machine-learning algorithm (MLA).