Rahmenwerk mit Variablem Eingangssichttransformator zur Segmentierung Medizinischer 3D-Bilder

Details

Veröffentlichungs-Nr.

EP4797217

Anmeldetag

23. Februar 2026

Veröffentlichung

26. August 2026

Rechtsraum

EP

IPC

G06V10/44, G06T7/00

Abstract

Systems and methods for performing a medical imaging analysis task are provided. 1) One or more input medical images each in a domain and 2) a domain code for each of the one or more input medical images identifying its domain are received. Each particular input medical image of the one or more input medical images is downsampled using a dynamic convolutional layer based on the domain code for the particular input medical image. A first set of features is extracted from each particular downsampled input medical image of the one or more downsampled input medical images using the dynamic convolutional layer based on the domain code for the particular downsampled input medical image. The first set of features is encoded into a second set of features using a machine learning based encoder. A medical imaging analysis task is performed based on the second set of features using a machine learning based decoder. Results of the medical imaging analysis task are output.

Anmelder

Firmen

Siemens Healthineers AG
Friedrich-Alexander-Universität Erlangen-Nürnberg, in Vertretung des Freistaates Bayern

Land

🇩🇪 Deutschland

Fachgebiete