GEMINI aims to deliver validated multi-organ and multi-scale computational models for treatment decision support and improved fundamental understanding of acute strokes, both ischaemic and haemorrhagic. We will demonstrate the added benefit of these computational models in personalised disease management. Specifically, GEMINI will deliver validated, integrated multi-scale, multi-organ Digital Twin in Healthcare (DTH) models for cerebral blood and cerebrospinal fluid flow, brain perfusion and metabolism, and blood flow and thrombosis along the heart-brain axis by integrating available and newly developed dynamic, interoperable, and modular computational models. Building on these models, GEMINI will deliver validated population-based DTHs of ischemic and haemorrhagic stroke aetiology and onset, treatment, and disease progression. Utilising these population-based DTHs, GEMINI will validate five personalised subject-specific DTHs, (1) stroke treatment, and (2) disease progression DTHs for acute ischaemic stroke and (3) aneurysm treatment, (4) subarachnoid haemorrhage progression, and (5) unruptured intracranial aneurysm risk assessment DTHs for haemorrhagic stroke to guide patient care and long-term management. We will bring proof of value of digital twins by the evaluation of the ischaemic stroke treatment selection DTH in a multi-centre clinical trial, in which treatment and patient outcomes are compared in situations with and without the availability of a DTH. GEMINI will implement a project-wide structured approach for data harmonisation, curation, model validation, verification, and model certification of the DTHs. Several outcomes of GEMINI have a high value for clinical practice, medical device industry, and in enhancing research in the fields of (bio)medical and computer sciences, warranting an extensive valorisation strategy with adequate IP protection and versatile exploitation actions to enhance a wide adaptation of the results of GEMINI.

This proposal describes the creation in Kraków, Poland, of a centre for Computational Medicine. The Centre will be a major driver for European advancement in this rapidly growing sector, developing sophisticated engineering methods for the prevention, diagnosis and treatment of disease, and meeting the overarching worldwide need for radically streamlined healthcare systems. The Centre’s area of operation will be computational medicine: (1) Clinician-led activity can be radically reduced by migrating the majority of routine clinical assessments toward computationally-driven (2) Entirely new approaches to disease diagnosis and management follow the development of novel computed biomarkers of disease, personalised to the individual patient, and typically leading to dramatically reduced costs. The Centre will combine expertise in machine-learning based decision science with fundamental biomarker identification using advanced simulation and data analysis methods. The Centre will act as a core technology provider for industry, fostering modernisation, innovation and increasing productivity, and additionally it will act as a catalyst, attracting new investment and businesses. It will provide an essential link between computational, medical sciences and the industrial sector, thus fostering knowledge and technology transfer and accelerating innovation. The Centre is to be established in Kraków because of several local advantages with regard to the education and research infrastructure, the high concentration of research hospitals and a dynamic entrepreneurial community in life sciences, so being embedded in a highly dynamic and competitive ecosystem The Centre will provide a fully sustainable mix of three advanced technical services, loosely categorised as education, project participation and industrial product development, with these three strands ultimately contributing approximately equally to the Centre’s income stream.

The main goal is to design an Extreme near-data platform to enable consumption, mining and processing of dis- tributed and federated data without needing to master the logistics of data access across heterogeneous data locations and pools. We go beyond traditional passive or bulk data ingested from storage systems towards next generation near-data processing platforms both in the Cloud and in the Edge. In our platform, Extreme Data in- cludes both metadata and trustworthy data connectors enabling advanced data management operations like data discovery, mining, and filtering from heterogeneous data sources. The three core objectives are: O-1 Provide high-performance near-data processing for Extreme Data Types: The first objective is to create a novel intermediary data service (XtremeDataHub) providing serverless data connectors that optimize data management operations (partitioning, filtering, transformation, aggregation) and interactive queries (search, discovery, matching, multi-object queries) to efficiently present data to analytics platforms. Our data connectors facilitate a elas- tic data-driven process-then-compute paradigm which significantly reduces data communication on the data interconnect, ultimately resulting in higher overall data throughput. O-2 Support real-time video streams but also event streams that must be ingested and processed very fast to Object Storage: The second objective is to seamlessly combine streaming and batch data processing for analytics. To this end, we will develop stream data connectors deployed as stream operators offering very fast stateful computations over low-latency event and video streams. O-3 The third objective is to create a Data Broker service enabling trustworthy data sharing and confidential orchestration of data pipelines across the Compute Continuum. We will provide secure data orchestration, transfer, processing and access thanks to Trusted Execution Environments (TEEs) and federated learning architectures.

Thrombosis, an occlusive blood clot, underlies ischemic stroke, myocardial infarction, and venous thrombosis, causing 1 in 4 deaths globally. Its incidence is rising due to an increasingly aging population and increased cardiovascular diseases, along with more cardiovascular implants. Unfortunately, limited understanding of thrombus formation, growth, and rupture hampers patient-specific prognosis. Current treatment strategies include antithrombotic therapy and thrombectomy, but improvements are needed for patient-specific treatments due to common recurrence and unknown impacts on clot fragmentation and thromboembolism. To effectively understand disease mechanisms and accurately predict treatment outcomes, it is crucial to integrate knowledge across multiple scales, harnessing a diverse range of cutting-edge in silico, in vitro, and in vivo technologies. This powerful fusion of emerging technologies holds the potential to revolutionize targeted and personalized medicine. However, there remains a significant shortage of trained professionals capable of co-creating such comprehensive and holistic disease models. ThromboRisk will therefore train 18 exceptional doctoral candidates (DCs) in diverse scientific fields, including mechanobiology, biochemistry, pathophysiology, and modeling. These researchers will collaborate within an international and interdisciplinary consortium to develop a platform advancing our understanding of thrombosis across scales, hereby bridging the gap between micro-level thrombus processes and macro-level impacts on disease prognosis, enabling clinical application. Unlike traditional engineering curricula, which address well-defined problems, ThromboRisk tackles complex, multifactorial "wicked problems" in fields of thrombosis by implementing a Challenge-Based Learning (CBL) doctoral training program to foster creativity, innovation, and societal impact. CBL is growing in undergraduate education but remains underdeveloped in doctoral training, where it has the potential to encourage DCs to step outside their comfort zones, think creatively, manage risks, and use technology responsibly in problem-solving. The DCs will work on real-world problems with peers, supervisors, consortium members, and external stakeholders, who act as "real clients" and co-creators.

The overall aim of the In Silico World project is to accelerate the uptake of modelling and simulation technologies for the development and regulatory assessment of all kind of medical products. This will be achieved by supporting the trajectory of a number of In Silico Trials solutions through development, validation, regulatory approval, optimisation, and commercial exploitation. These solutions, already developed to different stages, target different medical specialities (endocrinology, orthopaedics, infectiology, neurology, oncology, cardiology), different diseases (osteoporosis, dynapenia-sarcopenia, tuberculosis, multiple sclerosis, mammary carcinoma, arterial stenosis, etc.), and different types of medical products (medicinal products, medical devices, and Advanced Therapeutic Medicinal Products). In parallel the consortium will work with a large multi-stakeholder advisory board to form a community of Practice around In Silico Trials, where academics, industry experts, regulators, clinicians, and patients can develop consensus around Good modelling Practices. As the solutions under development move toward their commercial exploitation, the ISW consortium will make available to the Community of Practice a number of resources (technologies, validation data, first in kind regulatory decisions, technical standardisation plans, good modelling practices, scalability and efficiency-improving solutions, exploitation business models, etc.) that will permanently lower barriers to adoption for any future development.