Multi-dimensionAI: linking scales of information to improve care for patients with heart failure

Focus topic:	Artificial Intelligence
Type of funding:	Project funding programmes
Programme:	CZS Breakthroughs
Funded institution:	Universität Heidelberg

The project team is researching the treatment of heart failure using a combination of novel AI approaches and robotics. Comprehensive health data will be used to train an AI that will identify new correlations between the disease and the values and incorporate them into the therapy.

Goals

The project team at the Mannheim and Heidelberg sites focuses on patients with heart failure with preserved ejection fraction (HFpEF).

Health data, such as genomics, protein composition and metabolism, are collected over a longer period of time and used to train the AI. This is designed to identify correlations between the data and disease progression and use them to target interventions. For example, disease symptoms in patients typically lead to less physical activity, which further worsens the state of health. This is to be counteracted.

As part of a clinical study, the findings are being used to allow patients to exercise in an exosuit for a limited period of time. Exosuits usually consist of a lightweight frame that is attached to the user's body. Together with sensors and motors, these can control movement and provide support. This is intended to increase their mobility. Lifestyle improvements are repeatedly assessed to detect effects that range from the molecular to the macroscopic level. These results, in turn, will be used to improve the AI system.