Periodic Reporting for period 1 - ARISTOTELES (Applying ARtificial Intelligence to Define clinical trajectorieS for personalized predicTiOn and early deTEctiOn of comorbidiTy and muLtimorbidiTy pattErnS)
Período documentado: 2023-11-01 hasta 2025-04-30
ARISTOTELES aims to transform how we manage complex, chronic diseases evaluating the impact of information derived from a trustworthy artificial intelligence (AI) in the management of patients. As a proof of concept, the project focuses on atrial fibrillation (AF), a common heart rhythm disorder closely linked to ageing and the development of other health conditions (comorbidities). These comorbidities significantly increase the risk of poor health outcomes, including hospitalisations, disability, and mortality.
The project responds to a growing need for more personalized, preventive, and effective healthcare. Current approaches often treat conditions in isolation, missing opportunities to address the broader picture of a patient’s health. ARISTOTELES proposes a shift: using AI to provide a more holistic view that captures how different diseases interact over time. By doing so, it enables earlier and more targeted interventions that can slow disease progression and improve quality of life.
To achieve this, ARISTOTELES brings together diverse, high-quality health data from several European countries and harmonizes it into a common digital platform. This infrastructure will support the development of an AI tool capable of predicting a patient’s individual risk of developing or worsening comorbidities and adverse events — even in the presence of incomplete data — and will integrate clinical, imaging, lifestyle, and patient-reported information.
The project will test this tool both in simulated environments and in real-world clinical trials, comparing AI-informed care to standard medical approaches. Through this, ARISTOTELES aims to demonstrate measurable improvements in patient outcomes, adherence to care plans, and overall health system performance.
Social sciences play a key role in the project by ensuring that the tool is acceptable, ethical, and trusted by patients, healthcare professionals, and the public. By promoting responsible AI, the project also addresses regulatory, cultural, and practical barriers to adoption, laying the groundwork for wider implementation across European healthcare systems.
Ultimately, ARISTOTELES supports the vision of precision medicine — enabling citizens to live healthier, more independent lives through proactive, personalized, and data-driven healthcare.
The project responds to a growing need for more personalized, preventive, and effective healthcare. Current approaches often treat conditions in isolation, missing opportunities to address the broader picture of a patient’s health. ARISTOTELES proposes a shift: using AI to provide a more holistic view that captures how different diseases interact over time. By doing so, it enables earlier and more targeted interventions that can slow disease progression and improve quality of life.
To achieve this, ARISTOTELES brings together diverse, high-quality health data from several European countries and harmonizes it into a common digital platform. This infrastructure will support the development of an AI tool capable of predicting a patient’s individual risk of developing or worsening comorbidities and adverse events — even in the presence of incomplete data — and will integrate clinical, imaging, lifestyle, and patient-reported information.
The project will test this tool both in simulated environments and in real-world clinical trials, comparing AI-informed care to standard medical approaches. Through this, ARISTOTELES aims to demonstrate measurable improvements in patient outcomes, adherence to care plans, and overall health system performance.
Social sciences play a key role in the project by ensuring that the tool is acceptable, ethical, and trusted by patients, healthcare professionals, and the public. By promoting responsible AI, the project also addresses regulatory, cultural, and practical barriers to adoption, laying the groundwork for wider implementation across European healthcare systems.
Ultimately, ARISTOTELES supports the vision of precision medicine — enabling citizens to live healthier, more independent lives through proactive, personalized, and data-driven healthcare.
During the first reporting period, the project has achieved several important technical and scientific milestones. One of the main accomplishments was the development and deployment of the central data platform, which serves as the core infrastructure for data integration, harmonisation, and analysis. This platform enables the aggregation of diverse healthcare datasets into a unified structure, laying the groundwork for reliable and reproducible AI model development.
Parallel to this, the team initiated the development of the AI tool aimed at assessing the risk of developing or worsening chronic conditions in patients with complex health profiles. Early stages focused on feature engineering and testing of algorithms using publicly available clinical datasets. Multiple approaches to address missing or incomplete data were explored and implemented, including both traditional statistical techniques and more advanced generative models. While access to project-specific data is still being finalised, these preliminary efforts have provided a strong methodological basis for subsequent model training and validation.
In preparation for the real-world implementation of the AI solutions, interactive prototypes of the user interface were created. These were designed with a strong emphasis on usability and trustworthiness, providing healthcare professionals with predictive insights into disease progression and treatment options.
Efforts were also made to ensure the ethical and social robustness of the AI system. A large-scale survey has been launched across multiple European countries to gather perspectives from patients, citizens, and healthcare professionals regarding the use of AI in healthcare. The results will directly inform the design of the AI tool and its interface, ensuring alignment with user needs, values, and concerns.
Finally, work has started on the preparation of documentation and procedures necessary for launching the cluster-randomized controlled trial. This includes the definition of evaluation protocols and preliminary regulatory considerations to enable the testing of the AI tool in real-world clinical environments during the next phase of the project.
Overall, the project has advanced both the foundational infrastructure and the initial stages of AI model development, while embedding user-centric and trustworthy design principles from the outset.
Parallel to this, the team initiated the development of the AI tool aimed at assessing the risk of developing or worsening chronic conditions in patients with complex health profiles. Early stages focused on feature engineering and testing of algorithms using publicly available clinical datasets. Multiple approaches to address missing or incomplete data were explored and implemented, including both traditional statistical techniques and more advanced generative models. While access to project-specific data is still being finalised, these preliminary efforts have provided a strong methodological basis for subsequent model training and validation.
In preparation for the real-world implementation of the AI solutions, interactive prototypes of the user interface were created. These were designed with a strong emphasis on usability and trustworthiness, providing healthcare professionals with predictive insights into disease progression and treatment options.
Efforts were also made to ensure the ethical and social robustness of the AI system. A large-scale survey has been launched across multiple European countries to gather perspectives from patients, citizens, and healthcare professionals regarding the use of AI in healthcare. The results will directly inform the design of the AI tool and its interface, ensuring alignment with user needs, values, and concerns.
Finally, work has started on the preparation of documentation and procedures necessary for launching the cluster-randomized controlled trial. This includes the definition of evaluation protocols and preliminary regulatory considerations to enable the testing of the AI tool in real-world clinical environments during the next phase of the project.
Overall, the project has advanced both the foundational infrastructure and the initial stages of AI model development, while embedding user-centric and trustworthy design principles from the outset.
During this reporting period, the project has made significant progress in the foundational aspects critical to the success of the AI-based healthcare solution. The development and deployment of the data platform represent a major milestone, establishing the essential infrastructure for harmonized and secure management of diverse health data. Although this platform has not yet produced concrete predictive outcomes, it provides a robust and scalable foundation for subsequent AI model development and validation.
Parallel activities include the ongoing collection of stakeholder insights through surveys across several European countries, aimed at understanding patient and healthcare professional perspectives on AI acceptance and trustworthiness. These insights will be crucial to tailor the AI tools to user needs, thereby enhancing adoption and effectiveness.
While tangible results from AI model development and clinical validation are still forthcoming, the preparatory work conducted lays a strong groundwork for the upcoming phases. This includes preparations for a cluster randomized clinical trial, which will rigorously evaluate the clinical impact of the AI tools in real-world settings.
Key Needs for Further Uptake and Success
• Continued research and development to refine AI algorithms based on the integrated and harmonized data platform.
• Completion and analysis of stakeholder surveys to guide user-centric design and increase trustworthiness of AI tools.
• Advancement of clinical trial simulations and initiation of prospective clinical trials to demonstrate clinical efficacy and safety.
Overall, while final outcomes are pending, the project has established a solid technical and organizational base that supports future impactful developments in AI-driven personalized healthcare management.
Parallel activities include the ongoing collection of stakeholder insights through surveys across several European countries, aimed at understanding patient and healthcare professional perspectives on AI acceptance and trustworthiness. These insights will be crucial to tailor the AI tools to user needs, thereby enhancing adoption and effectiveness.
While tangible results from AI model development and clinical validation are still forthcoming, the preparatory work conducted lays a strong groundwork for the upcoming phases. This includes preparations for a cluster randomized clinical trial, which will rigorously evaluate the clinical impact of the AI tools in real-world settings.
Key Needs for Further Uptake and Success
• Continued research and development to refine AI algorithms based on the integrated and harmonized data platform.
• Completion and analysis of stakeholder surveys to guide user-centric design and increase trustworthiness of AI tools.
• Advancement of clinical trial simulations and initiation of prospective clinical trials to demonstrate clinical efficacy and safety.
Overall, while final outcomes are pending, the project has established a solid technical and organizational base that supports future impactful developments in AI-driven personalized healthcare management.