In today’s data-rich healthcare landscape, the challenge is no longer access to information but making the right decisions at the right time. Decision Support Systems are stepping into this space, turning complex clinical, imaging and patient data into actionable insights for clinicians and tangible benefits for patients.
Across our Health Innovation Ecosystem, we are actively shaping this transformation by supporting a portfolio of projects that bring data-driven decision-making closer to clinical reality. By engaging across multiple initiatives and fostering exchange between them, our efforts extend beyond individual projects, contributing to a more connected ecosystem where knowledge, approaches, and lessons learned can be shared.
This collaborative perspective is essential to accelerate progress and ensure that Decision Support Systems deliver on their promise: better decisions, improved patient outcomes and a more resilient healthcare system.
From Data to Decisions
Projects such as REALM-AI and REDDIE focus on extracting meaningful insights from large-scale health data, enabling more informed and personalised decision-making. In areas like diabetes care, this means moving beyond standardised approaches towards tailored interventions based on real-world evidence.
At the same time, initiatives like IMMEDIATE, METASTRA and SafePolyMedpush decision support deeper into clinical workflows. Their focus lies on clinical decision support, where timing, accuracy and usability are critical.
SafePolyMed, for example, tackles one of healthcare’s most pressing challenges: polymedication. By modelling risks linked to multiple medications, the system helps clinicians better understand interactions and make safer prescribing decisions, directly benefiting both patients and care providers.
AI in the Clinical Loop
One of the most tangible shifts comes from projects like COMFORT, which places AI directly alongside clinicians. Currently advancing towards clinical validation, COMFORT is testing its tools using retrospective imaging data to assess how AI-supported decisions compare to those made by physicians.
The goal is not to replace clinical judgement, but to augment it. By running in parallel within clinical environments, such systems can highlight alternative interpretations, reduce uncertainty and ultimately support more confident decision-making. A first validation study marks an important step towards real-world integration.
Towards Trustworthy and Usable Systems
What connects all these efforts is a shared ambition: making Decision Support Systems not only powerful, but trustworthy, usable and clinically relevant.
This requires more than algorithms. It involves rigorous validation, integration into clinical workflows and a deep understanding of user needs. Whether supporting treatment choices, analysing imaging data or modelling medication risks, these systems must fit seamlessly into everyday healthcare practice.