Skip to main content
Aller à la page d’accueil de la Commission européenne (s’ouvre dans une nouvelle fenêtre)
français fr
CORDIS - Résultats de la recherche de l’UE
CORDIS

Context-Aware Generative AI for Accessible Medical Imaging Across Low-Resource Settings

Objectif

Conversational assistants powered by generative AI have great potential to improve healthcare in low-Conversational assistants powered by generative AI have great potential to improve healthcare in low-resource settings, where shortages of specialists limit access to quality imaging diagnostics. By combining image analysis with natural-language interaction, these systems can transform complex AI outputs into intuitive, user-friendly guidance, enabling minimally trained healthcare workers, such as nurses and technicians, to acquire images, interpret results, and generate structured reports within routine clinical workflows.

However, current generative AI technologies have been mostly developed in high-income countries, and are largely designed for well-resourced clinical centres. Their deployment in low-resource settings is constrained by high computational demands, reliance on cloud connectivity, limited support for local languages, and poor adaptation to local clinical workflows.

GenAIMIX will address these challenges by introducing a new class of context-aware generative AI assistants tailored to low-resource environments. The project will develop edge-deployable vision–language models capable of running fully offline on low-power devices, delivering real-time guidance during imaging acquisition by non-expert users. The system will integrate clinical reasoning aligned with local protocols and support multilingual interaction, including English and Swahili, ensuring seamless integration into frontline care.

The proposed solution will be deployed and evaluated in three rural maternal health clinics in Kenya, assessing its impact on obstetric ultrasound acquisition quality, clinical interpretation, and usability for minimally trained healthcare workers, such as midwives and nurses. In parallel, GenAIMIX will define exploitation strategies to enable future deployment across Africa and beyond, paving the way for more equitable access to diagnostic imaging worldwide.

Champ scientifique (EuroSciVoc)

Ce projet n'a pas encore été classé par EuroSciVoc.
Proposez les domaines scientifiques qui vous semblent les plus pertinents et aidez-nous à améliorer notre service de classification.

Programme(s)

Thème(s)

Régime de financement

HORIZON-ERC-POC - HORIZON ERC Proof of Concept Grants

Voir tous les projets financés dans le cadre de ce programme de financement

Appel à propositions

(s’ouvre dans une nouvelle fenêtre) ERC-2026-POC

Voir tous les projets financés au titre de cet appel

Institution d’accueil

UNIVERSITAT DE BARCELONA
Contribution nette de l'UE
€ 150 000,00
Adresse
GRAN VIA DE LES CORTS CATALANES 585
08007 Barcelona
Espagne
Région
Este Cataluña Barcelona
Type d’activité
Higher or Secondary Education Establishments
Liens
Coût total
Aucune donnée

Bénéficiaires (1)