The DORIAN GRAY project: bridging cardiovascular disease and cognitive decline

Mild cognitive impairment (MCI) and cardiovascular disease (CVD) often co-occur in older adults, with cognitive decline affecting up to one-third of patients with CVD. According to the World Health Organization, over 55 million people worldwide were living with dementia in 2021, a number projected to rise to 78 million by 2030. In Europe, the median age-standardized prevalence of CVD is ∼7077 per 100 000 in men and 6026 per 100 000 in women.1 These striking figures underscore the growing public health challenge posed by the intersection of CVD and cognitive decline. The association between MCI and CVD goes beyond shared risk factors, as CVD itself can directly contribute to MCI. Mechanistic links include cerebral infarcts affecting cognitive-critical brain regions, arterial stiffness leading to end-organ damage, and chronic cerebral hypoperfusion from low cardiac output (e.g. heart failure or atrial fibrillation) causing neuroinflammation and white matter injury.2 Evidence from the European COGNITION.MATTERS HF study highlighted specific deficits in attention and verbal memory among heart failure patients, with structural correlates such as medial temporal lobe atrophy.3 Despite these findings, MCI remains underdiagnosed in CVD populations, with limited evidence-based strategies for its prevention and management.4 Addressing this gap requires integrating cardiovascular and cognitive data to develop personalized, proactive approaches for mitigating the dual burden of these conditions in ageing populations. In this context, the European Union, through the Horizon research and innovation action, has funded the ‘Devising a Personalized Risk Stratification and Holistic Management for Prevention of Cognitive Impairment in Patients with Different Cardiovascular Phenotypes’ (DORIAN GRAY) project (grant agreement no. 101156266). The consortium, led by the University of Brescia, brings together the diverse expertise of 24 leading institutions, including 12 academic partners, 5 hospitals, 5 companies, 1 scientific society, and 1 charity. The DORIAN GRAY project started in January 2025, and its programme of work is based on 17 work packages (WPs) to be delivered over a period of 5 years (Figure 1). Graphical abstract and work packages (WPs) of DORIAN GRAY project The DORIAN GRAY project aspires to uncover the mechanisms linking MCI with CVD and develop an integrated, personalized approach to mitigate dementia risk, promote resilience, and improve health in the ageing population. By focusing on data from CVD patients—where pathways to MCI are amplified—the project aims to identify mechanisms driving cognitive impairment and its progression in populations with cardiovascular risk. Timely and innovative, DORIAN GRAY leverages multi-modal AI technology to harmonize clinical and real-world data from diverse sources, including wearable devices and mobile apps (Figure 1). This approach enables precise risk stratification and tailored interventions across primary, secondary, and tertiary prevention strategies At the core of the project is the PORTRAIT platform, a cutting-edge framework that integrates genomics, biomarkers, imaging, and real-world data (RWD) to detect early signs of MCI. Complementing this is the MIRROR app, which delivers a personalized, adaptive intervention tailored to each patient. Through an avatar-based coaching system, the app dynamically adjusts cognitive and physical exercises based on user feedback and physiological data collected from smartwatches. This closed-loop approach not only enhances engagement and effectiveness but also enables real-time tracking of performance, generating valuable RWD that may serve as early markers for MCI diagnosis. In addition, the digital coaching combined with the avatar-based exergame enables the transfer of positive behaviours from the virtual world of the game to real life, avoiding the standardized, top-down format common in conventional health apps. The DORIAN GRAY project will harness data from multiple cohorts to inform and train advanced AI models, with a strong emphasis on harmonizing retrospective datasets to ensure interoperability. The project will leverage 10 observational datasets from concluded and ongoing studies, encompassing over 55 000 individuals from four European countries (UK, Netherlands, Germany, Czech Republic) and ∼235 000 individuals from USA. These datasets include nine population-based studies of middle-to-older age adults (40–90 years) and one clinical dataset representing a hospital population. To evaluate the feasibility of the DORIAN GRAY intervention, two clinical studies will be conducted, focusing on distinct cardiovascular phenotypes. The first is a pilot randomized controlled trial among heart failure patients with MCI, a population possibly more vulnerable to cognitive decline due to the interplay between advanced CVD and cognitive function. The second is an implementation study targeting individuals with MCI and cardiovascular risk factors, a population at lower cardiovascular risk compared to heart failure patients but still requiring early intervention to prevent progression towards dementia. Both studies will validate the adherence and effectiveness of the intervention delivered via the MIRROR app. The development of AI models in DORIAN GRAY will follow three phases. Phase 1 identifies key cardiovascular risk factors contributing to cognitive decline using interpretable techniques like Random Forest and causal analysis. Phase 2 applies advanced AI methods, including neural networks, to uncover complex patterns in unstructured data. Phase 3 integrates these insights into a refined system to enhance predictions and guide personalized interventions. By linking cardiovascular health and cognitive decline, this innovative approach aims to address unmet needs in personalized prevention and treatment strategies. To ensure effective communication and dissemination, DORIAN GRAY will engage in networking, mutual learning, and outreach activities. The project aims to deliver results to a wide range of stakeholders, including (i) clinicians, academics, and researchers; (ii) healthcare professionals (HCPs); (iii) patients, patient associations, and caregivers (formal and informal); (iv) policymakers and health authorities at regional and national levels, as well as regulatory and standardization bodies; (v) international societies involved in cardiovascular and cognitive health; and (vi) industry partners. The DORIAN GRAY project exemplifies collaboration between experts in health economics, market analysis, and AI-driven solutions, fostering a seamless transition from concept to implementation. By focusing on a market-oriented digital health solution for risk stratification and personalized multidomain prevention, the project addresses key healthcare system challenges. Through the optimization of existing resources and a focus on widespread adoption, the project aims to achieve long-term benefits and sustainable growth in healthcare innovation. A detailed socio-economic analysis will assess the cost-effectiveness of the intervention for healthcare systems, while addressing social, ethical, and legal aspects through a comprehensive health and technology assessment. The core concept of the DORIAN GRAY project is to leverage data from patients with CVD, such as heart failure—where mechanisms leading to MCI are amplified—to identify the drivers of cognitive decline in populations at cardiovascular risk. By addressing the continuum of prevention—primary (risk stratification for MCI), secondary (reducing progression towards dementia), and tertiary (mitigating the severity of MCI)—DORIAN GRAY aims to deliver meaningful health benefits to diverse subgroups of the European population, improving both cognitive and cardiovascular outcomes and promoting healthy ageing. Notably, while much of the focus has traditionally been on preventing cognitive decline, tools for cognitive enhancement are largely lacking. DORIAN GRAY seeks to bridge this gap, pioneering strategies that not only slow deterioration but also actively enhance cognitive function, ultimately fostering resilience and cognitive vitality in ageing populations. M.M. declares no conflicts of interest. R.P. declares no conflicts of interest. M.A. has received speaker fees from Abbott Vascular and Edwards Lifesciences. Funding source: European Union (European Health and Digital Executive Agency - HADEA, under the Horizon Europe programme). Grant agreement ID: 101156266. Type of support: research funding. Description: This work was funded by the European Union under the grant agreement n. 101156266. Views and opinions expressed are however those of the authors only and do not necessarily reflect those of the European Union or HADEA. Neither the European Union nor the granting authority HADEA can be held responsible for them.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.