Trustworthy Artificial Intelligence for Personalised Risk Assessment in Chronic Heart Failure

What is AI4HF

The “Trustworthy Artificial Intelligence for Personalised Risk Assessment in Chronic Heart Failure (AI4HF)” project is an innovative initiative that harnesses the power of Artificial Intelligence (AI) to provide personalized risk assessment and care plans for individuals living with Chronic Heart Failure. It utilizes advanced AI algorithms, global collaboration, and a patient-centered approach to improve healthcare outcomes. This four-year project is led by prof. dr. F.W. (Folkert) Asselbergs, coordinated by the Netherlands Heart Institute, and being conducted by a consortium of 16 international (associated) partners. AI4HF is funded by the European Commission.

Read more


The ambitious aim of the AI4HF project is, together with patients and health care professionals, to co-design, develop, evaluate and exploit an integrative and trustworthy AI-model for tailoring management of heart failure patients.

Read more


AI4HF empowers patients and caregivers by introducing an AI tool that predicts heart failure (HF) risk and offers personalized guidance on treatment. This tool enhances shared decision-making with healthcare providers, improves patient quality of life, and fosters digital health literacy and trust. In addition, AI4HF addresses the complexities of AI-mediated care, ensuring inclusivity, accountability, transparency, and fairness. The AI solutions benefit a diverse range of heart failure patients, delivering non-discriminatory AI solutions. Collaborative efforts with experts develop fairer approaches and methods. Moreover, AI4HF pioneers trustworthy AI in healthcare based on FUTUREAI guidelines. While other fields apply these guidelines, AI4HF demonstrates their framework in cardiology. This initiative sets community standards, promotes collaboration, and extends the work of global experts.

Read more

Research results

The AI4HF project is divided into subsections, called Work Packages (WPs). All project partners collaborate and work in a multidisciplinary matter on these different WPs. Every WP has to deliver their achieved results (also deliverables) to collectively drive the AI4HF project towards its ambitious goal of trustworthy AI in heart failure risk assessment. The obtained research results within the AI4HF project will be shared here.

Read more


The AI4HF project is being conducted by a consortium of 16 international (associated) partners from Spain, Greece, Turkey, Czechia, Tanzania, Peru, Portugal, Germany, France, Belgium, United Kingdom and the Netherlands. Our project brings together a multidisciplinary team of healthcare professionals, researchers, data scientists, and AI experts. We also collaborate with healthcare institutions and organizations globally to ensure a comprehensive and diverse dataset. AI4HF is an initiative of both clinical and technical partners as well as patient representatives.

Show all partners


Why is personalized risk assessment important in Chronic Heart Failure (CHF)?

CHF is a complex condition with varying degrees of severity and unique patient profiles. Personalized risk assessment enables healthcare providers to tailor interventions and treatment plans to each patient's specific needs, leading to more effective care and improved quality of life.

How does the AI algorithm work?

The AI algorithm will analyze a wide range of patient data, including medical history, diagnostic tests, lifestyle factors, and more. It uses this information to identify individual risk factors and predict the likelihood of CHF-related events. The algorithm continuously learns and improves from new data and research.

What does “trustworthy AI” mean?

Trustworthy AI refers to artificial intelligence systems that can be trusted to operate reliably, ethically, and securely, while also ensuring fairness and transparency in their decision-making processes.

Is patient data safe and secure in this project?

Absolutely. We prioritize the highest standards of data privacy and security. Patient data is anonymized and encrypted to protect confidentiality. We comply with all relevant data protection regulations and FUTURE guidelines.