Verfahren und System zur Konstruktion eines Bildes einer Probe

Details

Veröffentlichungs-Nr.

EP4787046

Anmeldetag

30. Januar 2025

Veröffentlichung

5. August 2026

Rechtsraum

EP

Abstract

The present inventive concept relates to a method (30) for training a machine learning model to construct an image of a sample. The method (30) comprises: acquiring (S300) a training set of images of a training sample (142) by: illuminating (S302) the training sample (142) with a plurality of illumination patterns formed by a first set of light-emitting elements (112) and a second set of light-emitting elements (114), wherein the first set of light-emitting elements (112) is configured to emit light having a first optical bandwidth and the second set of light-emitting elements (114) is configured to emit light having a second optical bandwidth narrower than the first optical bandwidth, and capturing (S304), for each illumination pattern of the plurality of illumination patterns, an image of the training sample (142); receiving (S310) a ground truth comprising one or more images of the training sample (142); and training (S320) the machine learning model to construct the image of a sample using the training set of images and the ground truth.

Anmelder

Firma

CellaVision AB

Land

🇸🇪 Schweden