French AI driven biotech company Medinsights, in collaboration with Karolinska University Hospital in Sweden, has secured over €200,000 in Eureka Eurostars funding to develop CardioTwin, an AI-powered digital twin platform designed to transform the management of cardiometabolic diseases (CMD) and related multi-morbidities.
CMDs like heart disease and type 2 diabetes often coexist with other chronic conditions, creating complex cases that, currently, population-based clinical tools struggle to manage effectively. This project aims to build informed and reliable molecular digital twins for CMDs with multimorbidity profiles. This initiative includes a pilot study with the Karolinska University Hospital in the Swedish healthcare system, which will integrate diverse data from patient health records, laboratory studies, lifestyle and metabolite profiling across generations, and social demographics into a unified, interoperable system. A scalable pipeline using xAI algorithms will ensure high-quality data processing, regulatory compliance (GDPR), and adherence to FAIR principles, with a focus on data quality, validation and traceability.
Karolinska University Hospital brings clinical expertise and access to their curated and robust combined dataset of 2 million indexed patient records from Lp(A) Stockholm study and STRIREG datasets to validate the platform in real-world settings. The CardioTwin focuses on leveraging our NEXUS-Rx framework to identify diagnostic and therapeutic targets for CMDs. The project’s primary aim is to deploy MedInsights’ advanced artificial intelligence driven network pharmacological approaches, Gen AI assisted machine learning tools, and computer vision technologies to identify optimised targets and pathways along with compounds capable of targeting those specific biological pathways implicated in these stratified conditions.
The project also focuses on end-user engagement, which is crucial for ensuring that the models are clinically relevant, trustworthy, and sustainable. This collaborative model involves co-design with healthcare providers and patients from the early stages, ensuring the final product meets their needs. The integration of Human-in-the-loop (HITL) mechanisms allows for continuous feedback, where providers can adjust and fine-tune AI recommendations. By collaborating with both R&D experts and end-users, the project ensures the virtual twins are not only scientifically sound but also practical and trustworthy for real-world application. Once deployed, CardioTwin has the potential to support physicians with more accurate risk assessments, help pharmaceutical companies optimize drug development through virtual trials, and empower patients through more transparent and personalized treatment plans. Over time, the platform could reduce healthcare costs by enabling earlier interventions and improving long-term outcomes for those living with chronic diseases.
Working with Evolution Europe has been a revelation. Their support, from project evaluation to drafting the grant and Collaboration Agreement, was outstanding. It eased pressure on our team and facilitated smooth collaboration across Sweden and France.