Siemens Healthineers AG Patent for AI Pre-training Medical Images

ChangeBridge: EPO Bulletin - AI & Computing (G06N)

Published March 18th, 2026

Detected March 24th, 2026

Summary

The European Patent Office has published a new patent application (EP4712046A1) by Siemens Healthineers AG for a technique related to pre-training AI models for medical image analysis. The patent describes methods for enhancing the encoder and concept head components of AI systems used in processing medical images.

View original document View source feed page

What changed

The European Patent Office (EPO) has published patent application EP4712046A1, filed by Siemens Healthineers AG, detailing a novel technique for pre-training AI models used in medical image analysis. The invention focuses on improving the principal encoder and concept head components of AI systems, enabling more effective processing of medical images by generating latent representations and reconstructing the original image. This patent application is a new addition to the intellectual property landscape concerning AI in healthcare.

While this is a patent publication and not a regulatory rule, companies involved in developing or utilizing AI for medical imaging should be aware of this intellectual property. It signifies potential advancements in AI capabilities for diagnostic and analytical purposes within the healthcare sector. Compliance officers should monitor patent filings in AI and healthcare to understand evolving technological standards and potential licensing requirements.

Source document (simplified)

← EPO Patent Bulletin

TECHNIQUE FOR CONCEPT AND STYLE PRE-TRAINING FOR A PERCEPTION TASK

Publication EP4712046A1 Kind: A1 Mar 18, 2026

Applicants

Siemens Healthineers AG

Inventors

Serban, Alexandru Constantin, Passerini, Tiziano, Ciusdel, Costin Florian

Abstract

The invention relates to a technique for pre-training a principal encoder and a concept head. A method comprises receiving, at a principal encoder (302), a medical image (502) and processing it for obtaining a principal latent representation (504), which is provided to a concept head (306) and to a style head (310) to obtain a first vector of discretized anatomical concepts and an associated further first vector of continuous styles per discretized anatomical concept in the medical image (502), respectively. An auxiliary feature decoder determines, based on the obtained first vector of discretized anatomical concepts, an auxiliary latent representation (10), based on which an auxiliary image decoder (316) performs a reconstruction (512) of the medical image (502). The principal encoder (302) and concept head (306) are pre-trained based a reconstruction loss between the received medical image (502) and the first reconstruction (512).

IPC Classifications

G06V 10/82 20220101AFI20251006BHEP G06N 3/0455 20230101ALI20251006BHEP

Designated States

AL, AT, BE, BG, CH, CY, CZ, DE, DK, EE, ES, FI, FR, GB, GR, HR, HU, IE, IS, IT, LI, LT, LU, LV, MC, ME, MK, MT, NL, NO, PL, PT, RO, RS, SE, SI, SK, SM, TR

View original document →